My thanks, though I doubt that much further study is necessary as you've about exhausted all there is to know about what amounted to a juggling game of confidence selling. Its like video game consoles for the fellas and gals who can't afford full-blown PC, gaming comuters, they all sort of do the same thing, but who's to care as long as they can bring it to market and sell more than anyone else at least for a while.
I was waiting for you to drag the files into VLC after you were able to read the disc and somewhat disappointed that you didn't. Thanks for mentioning that detail in the description!
Great Video and detail!!!... Techmoan you put the BBC to shame with your content!!!!! Would love to hear if those involved with the development of this format came forward with further details of the format and firmware!!!!!! Come on HD-VMD developers i know you have watched this video!!!!!!! Come forward with some details!!!!!!
I happened to be watching an older video of yours right before this, and I have to say, while the audio in that one was plenty listenable, this is even better, I appreciate the work you’ve put in to improving that over the years!
It bothers me when TH-camrs have big huge microphones in front of their faces and the audio is still terrible and you can also hear them swallow their saliva and gross stuff like that, when they just need a quality lavalier mic to clip on their shirt and it can sound great.
@@JaredConnell Or videos that have a loudness variation between -60 dB and 0 dB. Aka music at 0 and their speech at -60 dB... This is why I trust Techmoan videos and turn up the volume (It's a bit low) because I know I won't be ear-raped by unnecessary loud music.
I definitely don't get it right all the time - but I'll keep trying. I've just bought another new mic that I'm trying out on the next video after this.
Considering the HD-VMD player using a DVDRW drive it stands to reason that the format is based on multi-layer DVDs, so chances are the layers aren't anywhere near 5GB, but closer to 4.7GB (or 4.35GiB), meaning that the total capacity would max out at 14.1GB / 13.05GiB. And that's not even taking into account a dual layer is 9GB/8.5GiB, so chances are the size per layer are 4.5GB/4.25GiB, making the total disc size 13.5GB/12.75GiB. So I doubt it was less not being able to fit 15GB on a disc due to limitations, but rather the manufacturer not being entirely honest about the actual capacity of the discs.
Exactly and a 50GB Blu-ray isn't usable 50GB its around 46GB, the concept of using better codecs for DVD's wasn't new but the encoder tech was too immature, MPEG-2 was so common in early BD years because movie studios just stuck with what they knew, today the x264 encoder can push out crisp HD 1080p on DVD no problem at all (if you know what your doing). HD-VMD had a decent idea but not the expertise to pull it off.
You exactly right and over this you need to consider that the provisioning of space is secondary to the provisioning of bandwidth limitation. See the comment I did up the thread.
@@voidtoast4880 Well, on the pirate scene, more or less DVD-quality movies in CD-sized files, and HD movies in DVD-sized files were the norm from the start. And they usually looked somewhat decent. Although of course they had no special features and just a single audio track, often just in stereo rather than 5.1. But for an actual legitimate commercial format to be so focused on cost... It doesn't make much sense. Most of the legitimate release cost is licensing, not manufacturing. So the difference in raw media cost between a DVD and a Blu-ray is probably rather insignificant relative to the entire release cost.
@@jfolz I am one of the people that do care. I've turned things off that I was streaming because of the poor picture quality. Also I have a surround sound system, and very little streaming includes 5.1 sound. I either watch things on Freeview HD or I buy it on Blu Ray, pretty much everything else annoys me with poor picture and mediocre sound due to bit rate limitations. Unfortunately I'm not a large enough market, and an increasing amount of stuff never makes it off streaming onto Blu Ray.
18:26 I professionaly mastered DL DVDs back in the 2000. There's a reason why the file don't use the maximum capacity. This not a device developpememt issue. The mastering of DL layer imply to choose where the layer switching point on the movie will be so that it don't reveal itself in a fast paced part. Thus layers utilisation varies. More relevent here is the fact that the max bandwidth of a DVD, (which it is after all) is 9MB but less in reality depending on drives. I use to master VBR min 4 to 8 at 6 average. SO, it's not how big can you make your movie file but how much space do you have, balanced the layer spaces, the length of the movie and the bandwidth limit per sec for your video, audio tracks and subtitles. A final ISO mastered using all the capacity can make you go overboard if the movie is to short. You're provisioning for bandwidth not for space.
As an early adopter of the DVD format in the UK I can attest to the layer switching aspect. At some point mid film you would frequently get a brief pause. I put up with it because it was preferable to having to turn the disc over.
@@vespasian606 Some films hid it better. The Matrix on the original UK DVD release put the layer change in a fade to black in the film, the only way you could tell is that the audio suddenly dropped to total silence rather than a little bit of hiss.
With the DVD burner in the player, I would assume that bandwidth would not be as much of a problem as burners are usually able to go much faster than normal speed.
@@ArchieHalliwell That's a fair good point. 😊I would still wonder about the specs of the video decoder chip on this contraption of player.🤔 This not a PC, even if it has a repurposed burner with an hacked firmware to read a mirrored third layer pressed like a double sided DVD. This player is engineered on the weekend on a dollar store budget🤣
That's glue residue you're feeling on the edges of the disc. Discs are bonded together with glue and the glue can slightly leak out the sides. It was a fairly common issue with Laserdiscs due to poor quality control in the plants, and while rarer with other multimedia formats, it does still happen occasionally. Given the cost cutting practices of VMD, I'm sure they had their discs pressed by the lowest bidder they could find resulting in these sorts of happenings.
Yes, I was thinking layer pressing tailings. It's interesting to think that in multilayer discs, that it's actually separate parts pressed together with glue. Up until recently I didn't think that multi-layer or even single layer blank RW DVD's or BD's have the spiral tracks already pressed into them. I thought the laser tracked it's own way as it burned the disc, but it's in fact following a blank track. Like a record LP with a groove that has no sound, just a spiral. Laserdiscs are amazing in how large they are, and they too are glued together. Like you said, poor conditions for pressings, resulted in delamination of the glue layer most often, and I believe oxidation began to set it. A similar issue arised with the HD-DVD's Warner Brother pressings. Every demo I've seen on TH-cam usually shows the disc failing to playback after the layer change, or choppily if that.
How safe is it to cut or grind away the residue? Would it be okay to span the disk in a grinder (or dremel or other rotating device, after all, the discs are made to rotate and have a mounting hole) and carefully sand off the glue?
Amazing video, Mat! Kudos for the extra effort in reading the disk: connecting the internal drive of the HD-VMD disk to your computer was a genius move (and quite lucky one too! I got impressed it worked, didn't it even ask for drivers?? plug-n-play like that??! wow!) Thanks for giving us this, otherwise lost, piece of history!
Yes fortunately the drive just popped up with no problem. It no doubt helped that they used an off the shelf Asus Lightscribe burner. This also confirms that they were using custom firmware for the drive.
@@DripDripDrip69 Interesting... and turn other similar drives into HD-VMD drives?! wow! I don't know if extracting the firmware is something easily done... do you know how to do it??
@@GeekDetour : You dump the roms (usually eproms, eeproms, or similar), which is either trivial for those with the knowledge & equipment (external rom chips), semi-trivial (read-unprotected internal rom chips), or very difficult (read-protected internal rom chips). Having said that, it's fully possible to have more than one rom chip (in which case each one can have a different dumping difficulty from the others), and it may well be that LightScribe partially customized the drives, in which case there might be some custom hardware that actually does something important. The only way to know is for someone to get one or more of the drives (preferably starting with a broken one), and start reverse engineering it to document the thing.
Just as a tease for anyone considering this, if someone manages to dump the roms and figure out what the processor is, then it'll be possible (though unlikely) for people to write VMD firmware for _other_ drives as well, maybe even including BluRay drives.
"You haven't really seen Lucky Number Sleven until you've watched it in it's open map format and of course with it's original klingon soundtrack" I can imagine the faces of some movie 'enthusiasts' who now think they're missing out and their life is incomplete 😂 Great video again 👍
"... and I don't have the budget to fake this in CGI." I cracked up at this! I *really* love your sense of humour; it is the icing on the cake of your videos about technology! BTW, I also really love my "Taping the Hiss" t-shirt, which has drawn a few comments. People here in Oz (incl. me!) don't get it, until they say it out aloud ...
Interesting that one layer seemed to be mirrored - I wonder if this was to improve the discrimination between layers - are there any patents on this format? I'd also be interested to see inside the drive to see of there are any obvious mods - if they were tight on money it could be that early drives were modified rather than fully production engineered.
I think a firmware update was all they needed. Modern DVD burners are capable of focussing the lense pretty well, and burners have stronger lasers of course, which also helps penetrating the other layers. But it seems to me like penetrating two layers was about the maximum they could achieve (so only 3 layers), I would think that the HD-VMD 20 never came to light. Also: Nice to see you here!
If I remember the articles from back in the day properly, there is a "M" shaped geometry that was used to give the laser a good view. Imagine the outer legs of the M at steeper angles. Now imagine encoding from the angle of each of the faces. Rather than stop, traverse the head back to the beginning, they just reversed the motor and kept reading the next face backwards.
On a normal dual layer DVD the lower layer is read from the outside to the inside. The top layer is then read from the inside to the outside. This was done to keep the switch between layers as short as possible.
could it be that the layer 2(3.) being reversed comes down to the manufacturing? I mean like glueing a DL and a SL together? that might also result in some rough glued edges. but it probably would require the disk to spin in reverse some time so IDK 🤷
Opposite track path (OTP) is standard for DVDs and Blu-rays as well. From Wikipedia: "A dual-layer DVD or Blu-ray with Opposite Track Path means that layer 0 is written in a spiral track starting at the inside of the disc reading outwards, whereas the layer 1 spiral starts at the outside of the disc reading inwards. Both tracks are written so that the DVD player will rotate the disc in the same direction."
Somewhere around here I have a couple of Blu-Rays and HD-DVDs that turned out to be upscaled (barely) dvd transfers, and I remember *so many* DVDs that looked like they were copied from clapped out VHS tapes. Even a lot the widescreen ones looked terrible for the first few years. And there was the era before anamorphic DVDs where widescreen was just 4:3 with black bars on the top and bottom, again like old VHS transfers. Anyway, my point is, I'm not surprised if the transfer is not up to modern standards, leaving aside the whole bit rate issue. And a part of me dies every time I remember that 2006 isn't modern anymore, I was a 30-year-old stay at home mom by then.
There is so much video upscaled and smoothed with a filter out there. It does get rid of artifacts, but also washes out details, so skin looks like plastic. And the windowboxing (letterboxing + pillowboxing) on hard coded video with embedded bars is another sin. Even the good old PAL format had a piece in the signal to call if the video was widescreen or not so that your player and TV could dynamically adjust and send the right aspect ratio.
This is the most detailed look into a dead, unknown format I've ever seen. Props to hooking up the internal drive to your Mac to get the file size off. The real win here is reminding me that I only have a DVD of this film, so time to get a Blu-ray.
@Techmoan, I'd guess that the noted difference in running time in both discs when compared with the ImDB page is the result of the 25fps conversion. Europe uses 25fps for TV and film, and while modern electronics can display pretty much any framerate, it was very common to convert movies from 24fps to 25fps when releasing them for European home video. It is done by simplying displaying the 24fps movie 4.16% faster, and applying pitch adjustment to the audio. That seems to be the case here: 25/24 = 0.96, and multiplying the 110min duration of Lucky Number Slevin by 0.96 you get approximately 105 minutes.
You should probably work with someone to extract the custom firmware from that VMD/DVD SATA drive in the interest of data preservation. It would definitely be useful for people who would wanna play around with VMD files in a computer and don't wanna go through the rigamarole of having to search for a VMD player. One could just, in theory, buy the same DVD drive model and burn that firmware in. This could also give an opportunity for someone to reverse engineer the .vmd media container and make it playable in a computer.
@@Hans-gb4mv The documentation Mat has shown mentions DRM protection in form of AACS, which I know has been broken for bluray, but I'm not sure if it's applicable to HD-VMDs too
Ok, question is, how can I extract firmware from that drive? Do I need some special tools and proper experience, or is it possible just with single mouse click? And how helpful would be sharing that firmware, if there are basically no HD VMD discs to find? Also there are no VMD-R/RW discs (and probably there were no plans to make burnable VMDs, anyway) as is no chance to make 3L VMD from DVD-R DL. Also, you probably won't need a VMD drive to view/run .vmd files downloaded from some torrent/warez, if available (which I highly doubt, because collectors know how HD VMD discs are rare/expensive and they would be dumb, if they share even ISO from the worst movie ever produced on HD VMD disc). I own HD VMD player, but I have no HD VMD discs. Collectors, who own VMDs, won't even sell just one disc, if they have duplicates. Or they won't reply at all. Or they would make fun of me. "Oh, you bought HD VMD player for 17€ and you have no discs? Nobody told you that player is worthless without discs? Boo-hoo." This is how it works between hardcore collectors, who have all VMDs produced in multiple copies. The only gamechanger would be if someone will share VMD authoring software, to make your own dual-layer VMDs.
You can only use the firmware of the DVD drive to rip the VMD disks. The drive doesn't decode the VMD files, it's only responsible to read the physical data on the disk. Given that the disk shows up on a regular computer means that it uses a standard file system (probably UDF or less likely ISO 9660). In the player there is another circuit that reads the file system and reads the contents of individual files, and another circuit is responsible to decode the video data stream. These can be integrated into a single chip or can be implemented on several chips.
@@baksatibi Yes, that's the one and only thing you can do: open/grab/extract content from HD VMD disc, unless you'll be extremely lucky to get one. But I have a bad news for you, at least 18.000 VMD discs were discarded/destroyed in my country after a company, which distributed VMD players and discs here, filled for bankcrupcy/insolvency in May 2010.
@@tarstarkusz If you can't see a significant improvement in picture quality between VHS and DVD then you either need a new TV, new glasses, or stop watching from the other end of a football pitch. Either that or you don't care at all about picture quality and would be just as satisfied with the old 405 line TV or 8mm cine.
@@tarstarkusz I agree that vhs was good enough. I never enjoyed a movie less because it was a lesser resolution. And quality epuipment helps. High end vhs player and a nice Sony crt still look great to me. Assuming the vhs tape isn't junk.
@@tarstarkusz I'm talking about PAL VHS and DVD, I assumed you were since this is Techmoan. I have always had a multi standard TV back to CRT days, and PAL DVDs have noticeably better picture quality than NTSC DVDs. VHS on the other hand looks about the same on both standards. So your being underwhelmed with DVD is likely because they're NTSC and hence lower resolution.
@@tarstarkusz NTSC video from 2” quad tape looked amazing. 1” tape wasn’t so great. Early DVD transfers weren’t that good, but later on they got very good, especially from 2” master tapes.
Since the first video about HD-VMD i asked myself: If there is a DVD-Burner in the HD-VMD-Player, could you hook it up to a computer and use it as an external drive? Thank you for answering that question with this video!
I was thinking the same thing, maybe he can hook it up directly to his laptop instead. Sure enough. I wish he had shown 'Get Info', it should say total capacity and used capacity I would think. I don't have a Mac, don't use them much, but I think that's basically the same as Properties in windows.
@@marcusdamberger I'm pretty sure that once burnt/pressed, on most systems optical media will show up as 100% full, the capacity being the amount of data on it rather than the theoretical/practical maximum.
Thanks tor the follow-up video showing the real HD-VMD disk functioning. Considering that they were using DVD readers to play HD movies, I now consider this format quite an achievement
What i love about this channel is that you take the time to get fully into things! You even bought a blu-ray version from about the same time to check if that too was soft looking. Like you can be sure that if there is any information on the subject, we are getting it! This is some quality content!
It's interesting that the disc has visible layer identifiers - since they don't mean much to a user of the final physical disc, they must have been there to be read by people working in the manufacturing process, perhaps to ensure that the right set of 3 layer masters were installed in the stamping machines - or the right set of layers glued together after stamping. I haven't looked closely at any non-CD optical discs, so I don't know if this is normal on multilayer discs, or a sign of the small scale of their prototype (and only) process.
@SteelRodent Indeed, almost all of the Blu-ray and DVD discs that I have contain visible marked rings for the different layers. You can clearly tell 3-layer UHD Blu-rays apart from the normal 1- or 2-layer Blu-rays, for example.
You make a good point about trying to compare the HD-VMD to BD. So many factors to choose from: video compression codec, target bitrate of the final feature, and of course our friend digital noise reduction which while used unfortunately all too often these days to reduce the presence of film grain (and even sensor noise on digitally shot movies), was in much stronger use then than say now. As you mention this was shot in Super35 format which is full aperture 35mm negative film with spherical lenses and then cropped to the desired AR which was 2.35/2.40:1 and by 2007 it's almost certainly finished to a 2K digital intermediate (DI). This means both discs utilized that DI as that is the finished movie, so the film to video transfer was the same as that would be done during the DI process when the original negatives were scanned which definitely helps your comparison here. The company responsible for the HD-VMD transfer may have altered the color timing slightly in addition to the wrong aspect ratio, and added digital noise reduction or other filtering to soften the image further in addition to softening as a result of low bitrate. One thing you can do to see if more DNR was added and not just compression softening the image is to look for areas that have relatively solid backgrounds like a skyline or especially a white wall where film grain would be most naturally visible. If the HD-VMD has more "sluggish", "sticky", or otherwise unnaturally looking film grain, bam there's some more DNR added compared to say the Blu-ray version. Even a DVD, if not hit with more DNR will display film grain that is more naturally present and moving but due to the low resolution and bit rate, will appear clumpy and less defined. With a trained eye, you can spot this. I hope this helps explain the differences you may be seeing! As a whole though, the low bitrate and codec would be the main culprits here to explain the softness of the image compared to the BD.
I'm tinkering around with video game capturing for years now and noticed a similarity that confirms your theory about the softness. In short: yes, if you have a softer image, it reduces encoder stress and you can potentially avoid those nasty compression artifacts (image blocks). The softness is not produced by the encoder, you add it artificially, before sending the video to the encoder. When capturing games, I have set my encoder to VBR Quality. So basically I'm telling the encoder: I want THIS level of image quality. Take as much space as you like to achieve it. What really blew my mind was, the resulting video sizes are DRASTICALLY different depending on what game I capture, although I'm using excatly the same encoder settings. 1 Hour of Cyberpunk 2077 in 1440p60 barely scratches the 10GB mark, while 1 hour of Phantasy Star Online 2 New Genesis happily will eat up 50GB and more. Most funny thing: Cyberpunk even looks better. So what gives? The main difference is the image sharpness. Cyberpunk has got some really advanced post-processing, trying to soften unnatural sharp lines, some DLSS resolution trickery to achieve more FPS, and all that stuff which potentially makes the image softer. Phantasy Star has a more straightfoward Sega-Like basic "arcade" style engine with all kinds of hard edges, zero f***s given. This is giving the encoder a hard time to stay efficient.
All widespread video compression descends from JPEG image compression format. It was formulated for "photographic" images. This was specifically a design goal, in differentiation to compression of "synthetic" images, and it was actually an insane breakthrough, the sheer insight that went into it is difficult to grasp. Cyberpunk leans closer to the ideal "photographic" while PSO2 leans closer to ideal "synthetic". Unfortunately "synthetic" image compression is well behind, suitable approaches exist but they tend to be very heavy in terms of encoding and decoding system requirements. Like sinc wavelets could provide feature extraction and vector quantisation (VQ) should be good for actual compression to encode self-similarity within the image and to past images, VQ is just insanely slow to encode (multi dimensional hill climb) and sinc requires approximations for good performance that we're currently just slowly edging towards. I do think you can help the encoder. One thing i would do is always apply as much anti-aliasing as possible. I think maybe i should port FXAA to OBS so you can apply it before the encoding just to the video, while it attenuates some high frequency detail, it should actually help preserve detail at a given constrained encoding bitrate, and it doesn't eat nearly as much detail as a simple blur.
@@SianaGearz Wow that was thorough, thanks. I'm already using all AA the game is giving me, but i can still add more by forcing it in the nVidia driver. I will try to tinker with that a bit. I'm also using HEVC as it's currently the most efficient codec I can practically use (GPU Encoding). I am also not super dependant of the quality outcome, i just started to wonder how far I can push it. TH-cam is massacring it anyway, but I wanted the best possible quality I can provide from my end. That's also why I'm using 1440p as TH-cam grants way more Bitrate and the blocking is not as bad as with 1080p. I'm currently recording with Bandicam, but I also have OBS ready to go.
@@Raveheart Word of note, depending on capture mode (i.e. game capture), NVidia forced FXAA might be visible during gameplay, but might not reach the capture and the encoder. I don't remember the details, just something to look out for. I do think it gets captured in ShadowPlay though. Also why i want to make it an OBS filter is because while FXAA works invariably better especially on a livestream, people may not want to actually play with it on, so a filter would be ideal.
I tend to record with a ridiculously high fixed bitrate using GPU encoding (to have the initial recording done fast, sacrificing compression for quality) and after cutting use CBR x264 and after that end up with around 2-4 mbps for 1080p video depending on content. The issue with todays AA implementations is that they either look blurry or flicker or both, I wish there was some more advanced, adaptive tech that would work on all games and could simply be added on driver level to not run into issues with anticheats.
@@HappyBeezerStudios speaking for all the game engine developers out there, a drop-in antialiasing which doesn't introduce its own issues would be the holy grail. The driver angle has become increasingly difficult, because the driver doesn't have a lot of insight and reasoning into what the application is doing any longer. The more programmability became possible from the application side, the less leverage they got. So that's another level of complexity.
Another really interesting and well-researched video, Mat, cheers. I always look forward to sitting down to the latest Techmoan video with a coffee at the weekends. Out of interest, I just checked my HD-DVD release of Lucky Number Slevin (a 2006 disc, as best as I can tell), and it's AVC encoded, has special features (Making of, Deleted Scenes/Alternate Ending, two commentaries, trailer and "An Intimate Conversation") and English and French DD+ 5.1 soundtracks. It's also 2.35:1, and, interestingly, IS 1hr 50mins in length (the credits end 10 seconds shy of the 110 minute mark). This really did seem, for the many reasons you mentioned, a really half-baked format.
@Techmoan I have a recording from this film when it was broadcasted on the Swiss public television. The Video-Stream is 720p 50fps and probably sped up from 24 to 25 frames a second that was standard for DVDs and still is for TV. On BD they started to encode them in 24 fps. My recording without the end titles is 1:41:22 long. Probably this must be converted from 25 to 24 frames. That would lead to 1:45:35 seconds without the end titles. The movie is also in 2:35 and the car park scene matches your 2:35 version. My version is even softer with 720p and only 6.409 Mbit/s (including German Audio in 5.1 AC-3)
As a guy in his mid 30’s I randomly found this channel a few years back and have been absolutely enamored with the content you create. I’m a huge A/V nerd and am always tweaking, calibrating or and/or upgrading my own setup and find it absolutely fascinating to explore the history of all of these lost formats. Keep up the phenomenal work!
I'm surprised you haven't done a proper dive in to UMD's. Or if you have, I didn't find it. The PSP probably had some interesting compression or something noteworthy.
Yeah, I recently watched the classic "Goodfellas" on UMD. it's around 3 hours long and the picture quality is nearly as good as DVD. But due to the battery life not being good on my PSP and I could never watch it all the way through without charging.
Yeah. UMD's. That format never did make sense for me as a home video platform. But it would be fun to see what can be gained with oroper research into what the plans were as quick wikipedia skimming reveals plans were at one point there for multidisc standalone systems.
You can read the discs over USB with a hacked PSP. I remember checking one of my UMD movies, and it was H.264 encoded, I believe at normal SD resolution. Approximately DVD quality, maybe slightly worse (better efficiency compression, but less bytes to go around). The movies will obviously look sharper on that small (and low-resolution) screen of the PSP. You can connect it up to a TV with a component cable though. Or just rip it over USB with a hacked PSP (probably the sharpest image you're going to get out of it).
I never did understand movies on umd, it always felt like a bit of a hassle. I did put some anime shows on the psp back in the day but at least then I had a whole show.
Glad to get this coda. And I'm really glad to see you interfaced with the disc drive in the player to see what was on the disc, I would surely have suggested that :P
@@Techmoan i know you like to mix and match withint fan fair. To be honest i loved your ending in the past with more bits of video and a longer sample of the music from a couple of years ago so thus often when I watch your videos I will boot up the end of a random old one just to get the extra credits! I also ended up looking up the free source music and sound. Oddly even listening to the music by itself doesnt do it without you taking apart things in the background slowly or spinning things on a lazy Susan it just loses some of the magic. As for the sound I got to find out it was from the old 7 million dollar man show I never watched an episode but I had heard of it. As you might say overly long explanation but I felt compelled to share it. Thanks for the work.
Matt, the sheer technolgical knowledge and deep diving has brightened my slighty dark Saturdays, from my previous projectionist past, you have made me seriously chuckle. You are a gift to us Sir. And I thank you. (I particularly enjoyed this one.)
Me: Finally! THIS is the LAST of the physical media formats that I will ever have to compulsively seek out to collect. Techmoan 2022: I just got this new old stock XYZVD player from a seller in Japan, let's have a look.
You have a Super Video CD player? Dunno if those will also play CVD (China Video Disc, a variant of SVCD) or if a CVD requires its own player. A SVCD with some care taken in its encoding (2 or more pass) can look quite good at 480x480 with anamorphic squeeze. Same vertical resolution as NTSC DVD.
Good luck! It looks like movies are virtually impossible to find in this format. The movie Techmoan shows is a Scandinavian release - makes you wonder if they had a better rollout in Europe. Fellow collector, I am the proud owner of a MovieCD that plays in Windows 3.1 👍
I have fun imagining techmoan with a time machine like quantum leap but he would go back and rescue rare hi fi and video equipment and never miss a discontinued line sale .and rescue all the obscure formats from landfill fighting off sea gulls as he goes .But what to call this unlikely super hero .Techmoan man. Oh dear
Theorising that one could examine different media during his own lifetime, Doctor Matt Techmoan stepped into the Quantum Leap accelerator. He didn't vanish, but a bunch of different machines appeared, and driven by a force of TH-cam users to review these machines for the better. And so, Matt finds himself leaping from format to format, striving to clear up all the malarky surrounding each format's history, and hoping each time his next leap will be leap home...
From a pure storage perspective, using video with AVC/H.264 encoding at 2.6 mbps (which is fine in terms of quality for the codec) would allow around 3 hrs runtime with 2 audio languages in 6 channel 384 kbps AC3 audio onto a single layer DVD. For dual layer it goes up to around 5:36 hrs. So DVD is totally capable to deliver FullHD 1920x1080p video at acceptable quality purely by storage. And for 2.35:1 cinema format it could either be anamorphic or in 1920x832 As a bonus: youtube recommends 8 mbps for 1080p upload, at which you'll get around 1:14 runtime for a single layer or 2:15 for a dual layer, so even at that bitrate it would be fine to run put HD video on a dual layer DVD with enough storage for a movie. Squashing a bit more down to 6 mbps and a single audio set would land us at 1:40 and 3 hrs respectively, which should be enough for basically anything.
I wonder if someone has dumped the firmware of one of these to see how it works? Since it seems to be an off the shelf drive, just with custom firmware, so it should be possible to make other DVD drives support the HD-VMD, and maybe even modify a BD drive to support it as well? There are apparently ways of dumping the firmware of disc drives, though I've never done it myself. If nothing else, it would be good for there to be an archive of the firmware for posterity.
I bet it looks soft because they most likely used 1440x1080 instead of 1920x1080 in order to reduce bit rate, it was a technique used a lot in digital satellite broadcasts to reduce bit rate and making digital compression artifacts less noticeable.
It was also used in the early days of Freeview HD as I bought the first released to market DVB-T2 set-top-boxes and DVB-T2 dongle to work with Linux. Thankfully they've reverted back to 1920x1080.
Ciao Mr. TechMoan 🙋🏻♂️ You are THE MAN for tech and hi-tech matters. I can only say thank you for running this channel fulfilled with you strong passion and depp competence. Plus you explain things in a way that also a monkey can get them. Magnifico 🥇 Cheers from Monza (Italy)
This has been an interesting journey through a format I never knew about. That's a lot of work, and I'm very grateful that you put all this together. I think you've really created some durable canon on the topic, here. Well-done.
The people who worked on this format must be delighted you’ve spent all this time explaining how awful it is and how it’s nowhere near as good as what they had hoped it would be
No one can say Techmoan isn't willing to do whatever it takes to make a video; he had to travel back in time 14 years to 2007 just to find a specific bluray version of a mid-2000s movie. I'm just a little disappointed that he decided to that off-camera, however.
Another possibility for the low size of the file is that many companies use standard templates instead of adjusting the bitrate to the movie, so if with one template the movie is too big they'll use another one, and if this one don't use all the available space well, bad luck.
I really think they have a 11.5GB capacity disk and filled it to the brim. You can't do fixed bitrate anyway, it just looks invariably bad, and at 2 pass encoding which distributes the bitrate where it's needed, sniping for correct size isn't really an effort. Because have you noticed the capacity of DVD-9 is 10% smaller at 8.5GB than two DVD-5s at 4.7GB each? If i assume that marginal capacity introduced by each layer is 20% lower than the last, so the per-layer capacity regression continues, the 3-layer disc comes out to 11.5GB, which i think is a little much of a coincidence to just dismiss. 4 and 5 layers could house approximately 14 and 16GB respectively.
13:40 (writing before I watch any further) Regarding the softer image - applying a bit of blur before encoding will remove high frequency information, and allow to use a lower bit rate without introducing visible blocking artifacts, so I think it was done on purpose. Some JPEG encoders have this option built-in (GIMP). This however obviously diminishes the benefits of using a higher resolution, since the point of that to have sharp detail.
I’ve been encoding movies for the past 20 years or so, and what I’ve been doing is applying a noise filter on some titles where there was a lot of film grain or noise. This reduces the size of the file significantly. Granted it can make the movie a bit soft at times, but I do a few tests before a complete encode to balance size and quality. In the end, when you have collected a large library, this will save you quite a bit of space. It’s possible that they did this as well to get the movie to fit on the HD-VMD properly. Anyway, thanks for all the hard work you have done. Cheers from Canada Eh! :-)
@@AlexanderBellOpelika For sure, in some cases where you have a really old b&W movie, removing the grain or noise actually makes it worse. In those cases it's best to leave the processing off.
@@robertromero8692 It's not a terrible practice, You're not meant to see the grain that makes up the film picture in the same sense you're not suppose to see pixels in digital representations. Grain or rather noise is a pain to keep retained because it adds unnecessary pixel data making it exponentially bigger with no real practical use outside of archival needs. Were not even talking about the hell it can play on compression methods just the raw storage method alone, it costs nothing to store a frame of pure black but it costs everything to store noise with a 3 bit variation of black that you're never going to see. The more solid pixel color you can use to make up the image the more cheap it is to store.
@@robertromero8692 It's not a justification nor is it a data challenge that wheel has been invented many times for image storage. Grain is noise physically and even in a digital representation, This has the same energy of needing to see the shimmer of cells under a microscope in order to see that my blood is the color red. When you start to see grain it stops being the image and becomes an artifact of chemistry there is no more image the further "in" you go, the more in and higher res you digitally represent it the more it becomes microscope slide of crystals there's nothing left. What im trying to convey is that it's a massive inconvenience to store noise like grain nor is it necessary to expand its fidelity outside of archival needs. The sad part about this is that it's easier to randomize grain/noise in realtime than it is to retain it. I get that you want that "film" look but when you transfer mediums all it will ever be is just a representation of it and you have to draw the line somewhere and accomplish the goal the original medium set out to do.
@@robertromero8692 No I do not care to preserve a visual artifact at cost unless it's part of a goal nor am I going to pretend a Mona Lisa painted in acrylic is the real deal. Im well aware grain can and has been captured to a point but it's just fat that can be trimmed and adds no new visual information from my point of view.
Lower bitrates normally would result in a more blocky image, especially with VC-1 which is just a variation of MPEG-4 ASP (aka. DivX). I think they might have blurred the image to hide blocking artifacts before encoding, to make it easier for codec.
@@DJlegionuk Hmmm... Could be, but you often still see "staircase" effects around pixels with that. But it depends. Maybe they use a gaussian upscaler. They're very soft and blur out pixel artifacts.
If you look at the settings for Xvid (which it's been a while since I opened that up) it lets you choose between h.263 quantization for a softer image that works better with low bitrates without blocking and MPEG quantization which gives a crisper image. Using a crisper quantization matrix works better for high bitrates but results in the classic blocky video compression at low bitrates. Modern codecs include an in-loop deblocking filter that effectively makes adjustments on the fly based on how much bitrate it has and the complexity of the video.
Good investigation! Three layer DVD... makes sense that it could be read on a normal DVD with special firmware. the second layer on a dual layer DVD is smaller capacity than the one layer one. Probably the third layer on this is even smaller, to make the symbols easier to extract.
A 1-layer DVD is 4.7GB (in billions of bytes) and a dual-layer is 8.5GB. Therefore, the second layer is only about 3.85GB. Multiply that by three and you get 11.5GB, which is 10.75GiB as seen by Windows explorer. Therefore, I suspect the disk is three 3.85GB layers . I guess they just couldn't make it with a nice 4.7GB first layer, and it really should be called VMD-12.
funny thing is with today's h265 compression, if you have the cores and time you can make masterfully good quality 1080p, obviously non HDR backups of blurays for around 10gb. My father was a laserdisc collector, so have backed up every bluray i've owned out of pure paranoia of disc rot, and sure they are far more robust but it takes a tiny fraction in damage with blurays to render them unplayable next to a dvd, let along laserdisc. Such is time.
21:04 - That's not true! VC-1 with 13 Mbps in 1080p would not degrade image quality by totally blurring all frames! That's not how compression works. Lower bitrate could introduce compression artefacts like macroblocking in complex scenes. Im 100% sure that 13Mbps is more than enough to produce super crisp static scenes. The reason why image is blurry is due to bad transfer. Check for example blu-ray version of "Last Man Standing" with Bruce Willis. It is also very blurry and it looks like upscaled SD.
I studied compression in college and wether x bitrate results in a crisp image depends entirely on the codec's efficiency. 13Mbps at 1080p with H.264 will look absolutely fine, but VC-1 is less efficient, which means you'll need more than 13Mbps for a clear image. Compression artefacts are easy to see if you go really low on the bitrate, for example transcoding a video into 1080p H.264 with 1Mbps, you'll immediatly notice the squares in the video. But if the bitrate is just a bit on the low-side, the compression artefacts will be small enough that the video will appear to be slightly blurry and this is what Mat noticed. 1080p H.264 suffers from the same problem at 4-5Mbps. At 4-5Mbps, only the newer HEVC H.265 is going to provide a decently crisp image.
@@r34ztune11 you are forgetting other critical factors like video image complexity and the encoder implementation, and in this regard most VC1 implementations are really poor. Also the mastering and the tranfer are also very important, I have seen that in one movie that while the USA version had more bitrate than the european from the same year the image quality was much lower but in a posterior re-edition of the USA version it had the same quality. I recommend you to read the Netflix article about per-title encoding and shot-based encoding.
Either way VC-1 isn’t very good and was quickly dropped in favor of the AVC/MPEG4/H.264 compression. Very interested to see what it could look like in H.266/VVC codec. Could it essentially fit well on a 12gb disc?
IMHO I wasn’t to excited about the first video, but then with all additional videos and revelations it turned in to quitte an interesting format adventure! Very nice!
I'm 23 years old, but have been following the channel for the past 5 years or so, and I cant describe how exited I am to come home after my morning swim on a Saturday, pour myself a good cup of coffee and watch a video from Mat. It's just proper relaxing, and for me I'm also associating his channel with the feeling of safety and calmness. Well done and thanks!
Great video as always! I love learning about things like this, and you did a fantastic job of sleuthing out all the details. And your soothing voice always picks me up. Thank you!
There is no way a VMD 15 is going to be 15gb. they are rounding ~4.5gb up to 5gb per layer, but that adds up and the capacity shrinks per layer. So based on some loose maths, going from 4.7gb to 8.5gb is about 90% of double capacity. So if we just apply that again I get a rough estimate of about 12/13gb for a VMD 15. That seems logical.
Another fantastic video and really interesting deep dive into a format most of the world never knew existed. Watching a techmoan video has become a part of my Saturday ritual so thanks very much for keeping up this schedule! I know it can't always be done but I look forward to them every week and you always deliver! I really enjoy the minidisc episodes and was wondering what you planned to do with all those hundreds of players you recently got? Could be an auction (or do you put stuff on eBay ever?)
By getting that drive working on your PC, you might have done something very few people have done before, used a "15"gb DVD drive on one. That's quite cool.
17:53 in case anyone's wondering about the date (31 - 12 - 1969 @ 16:59) That's the date you get when you put 0 in unix time... minus 7 hours & 1 minute. (unix time starts from 01 - 01 - 1970 @ 00:00) UTC -07:00 is a time zone for somewhere in North America. NME was incorporated in Nevada... and the time zone that Nevada is in is in fact UTC -07:00 The more you know! :D
I remember when that movie came out on DVD at the shop it was stylized on the box as "Lucky # Slevin" Then when Twitter and their hashtag system arrived, I thought "Kids these days won't understand a movie title like "Lucky Hashtag Slevin"
I never called that symbol 'number' when referring to a button on a phone - everyone I knew called it the hash key. I used to program call centres and it was always - 'press the hash key to continue'.
Every Friday night after a long night I like to sit down and watch Techmoan while I finish my work, so thanks Mat for being punctual with your uploads, your videos always give me something to look forward to and enjoy. Then I usually watch it again the next day because I usually miss a few parts here and there because I'm doing something else. When you don't publish it on time I get worried lol.
I think the "REVERSE" written next to Layer 1 indicates that the player will read this layer in reverse compared to layer 0 and 2, that way the drive can switch layers quickly without moving the lens all the way back to the start of the disc again. That way the disc can be read FWD>REV>FWD. I'm sure if you had a 4 layer disc, the fourth layer would be REV too. (This is probably something that dual layer DVD's already do, too, though)
I noticed the watch....I wouldn't think it would be his style, I can't imagine matt spending many thousands of pounds on a watch, it doesn't fit into to his homely northen charm, and artex front room, hasn't even got a nixie tube in it!
@@sadnerd397 hahah, the same for me, i remember the old video of his with the Seiko Sea Urchin and i thought it was that one, until i noticed it's actually a Sub
@@solidamber A sub' fits everybody everywhere. Period 😎. And this is the most economical watch on the Planet. You buy one for the rest of your life and it doesn't need any batery. Consequently a sub' is good for Our Mother Gaia. Period bis.😄
We're you ever contacted by people actually working on those "forgotten formats" back then? Especially with this VMD it would be incredibly interesting to interview someone involved.
*There’s a bit more information about this video in the video description.* 🔝👍
My thanks, though I doubt that much further study is necessary as you've about exhausted all there is to know about what amounted to a juggling game of confidence selling. Its like video game consoles for the fellas and gals who can't afford full-blown PC, gaming comuters, they all sort of do the same thing, but who's to care as long as they can bring it to market and sell more than anyone else at least for a while.
Thanks a lot for the great video!
It is probably unintentional, Matt, but that is a really bad pun. The lecture itself was interesting.
I was waiting for you to drag the files into VLC after you were able to read the disc and somewhat disappointed that you didn't. Thanks for mentioning that detail in the description!
Great Video and detail!!!... Techmoan you put the BBC to shame with your content!!!!! Would love to hear if those involved with the development of this format came forward with further details of the format and firmware!!!!!! Come on HD-VMD developers i know you have watched this video!!!!!!! Come forward with some details!!!!!!
"Last video on HD-VMD not quite nerdy enough for you? Not to worry, here's one that definitly is!" Thanks Matt!
It's why I'm here.
I happened to be watching an older video of yours right before this, and I have to say, while the audio in that one was plenty listenable, this is even better, I appreciate the work you’ve put in to improving that over the years!
It bothers me when TH-camrs have big huge microphones in front of their faces and the audio is still terrible and you can also hear them swallow their saliva and gross stuff like that, when they just need a quality lavalier mic to clip on their shirt and it can sound great.
@@JaredConnell ?
@@JaredConnell Or videos that have a loudness variation between -60 dB and 0 dB. Aka music at 0 and their speech at -60 dB... This is why I trust Techmoan videos and turn up the volume (It's a bit low) because I know I won't be ear-raped by unnecessary loud music.
I definitely don't get it right all the time - but I'll keep trying. I've just bought another new mic that I'm trying out on the next video after this.
The world needs an AI-based anti-saliva audio filter. Listening to clear recordings of people with dentures is a nightmare when you've got misophonia.
Considering the HD-VMD player using a DVDRW drive it stands to reason that the format is based on multi-layer DVDs, so chances are the layers aren't anywhere near 5GB, but closer to 4.7GB (or 4.35GiB), meaning that the total capacity would max out at 14.1GB / 13.05GiB. And that's not even taking into account a dual layer is 9GB/8.5GiB, so chances are the size per layer are 4.5GB/4.25GiB, making the total disc size 13.5GB/12.75GiB.
So I doubt it was less not being able to fit 15GB on a disc due to limitations, but rather the manufacturer not being entirely honest about the actual capacity of the discs.
Exactly and a 50GB Blu-ray isn't usable 50GB its around 46GB, the concept of using better codecs for DVD's wasn't new but the encoder tech was too immature, MPEG-2 was so common in early BD years because movie studios just stuck with what they knew, today the x264 encoder can push out crisp HD 1080p on DVD no problem at all (if you know what your doing). HD-VMD had a decent idea but not the expertise to pull it off.
@@voidtoast4880 not to mention HEVC bitrates!
You exactly right and over this you need to consider that the provisioning of space is secondary to the provisioning of bandwidth limitation. See the comment I did up the thread.
@@voidtoast4880 Well, on the pirate scene, more or less DVD-quality movies in CD-sized files, and HD movies in DVD-sized files were the norm from the start. And they usually looked somewhat decent. Although of course they had no special features and just a single audio track, often just in stereo rather than 5.1.
But for an actual legitimate commercial format to be so focused on cost... It doesn't make much sense. Most of the legitimate release cost is licensing, not manufacturing. So the difference in raw media cost between a DVD and a Blu-ray is probably rather insignificant relative to the entire release cost.
@@jfolz I am one of the people that do care. I've turned things off that I was streaming because of the poor picture quality. Also I have a surround sound system, and very little streaming includes 5.1 sound. I either watch things on Freeview HD or I buy it on Blu Ray, pretty much everything else annoys me with poor picture and mediocre sound due to bit rate limitations. Unfortunately I'm not a large enough market, and an increasing amount of stuff never makes it off streaming onto Blu Ray.
18:26 I professionaly mastered DL DVDs back in the 2000. There's a reason why the file don't use the maximum capacity. This not a device developpememt issue.
The mastering of DL layer imply to choose where the layer switching point on the movie will be so that it don't reveal itself in a fast paced part. Thus layers utilisation varies.
More relevent here is the fact that the max bandwidth of a DVD, (which it is after all) is 9MB but less in reality depending on drives. I use to master VBR min 4 to 8 at 6 average.
SO, it's not how big can you make your movie file but how much space do you have, balanced the layer spaces, the length of the movie and the bandwidth limit per sec for your video, audio tracks and subtitles.
A final ISO mastered using all the capacity can make you go overboard if the movie is to short. You're provisioning for bandwidth not for space.
As an early adopter of the DVD format in the UK I can attest to the layer switching aspect. At some point mid film you would frequently get a brief pause. I put up with it because it was preferable to having to turn the disc over.
@@vespasian606 Some films hid it better. The Matrix on the original UK DVD release put the layer change in a fade to black in the film, the only way you could tell is that the audio suddenly dropped to total silence rather than a little bit of hiss.
With the DVD burner in the player, I would assume that bandwidth would not be as much of a problem as burners are usually able to go much faster than normal speed.
@@ArchieHalliwell That's a fair good point. 😊I would still wonder about the specs of the video decoder chip on this contraption of player.🤔
This not a PC, even if it has a repurposed burner with an hacked firmware to read a mirrored third layer pressed like a double sided DVD.
This player is engineered on the weekend on a dollar store budget🤣
@@ArchieHalliwell Not to mention burners carry a larger buffer most of the time.
That's glue residue you're feeling on the edges of the disc. Discs are bonded together with glue and the glue can slightly leak out the sides. It was a fairly common issue with Laserdiscs due to poor quality control in the plants, and while rarer with other multimedia formats, it does still happen occasionally. Given the cost cutting practices of VMD, I'm sure they had their discs pressed by the lowest bidder they could find resulting in these sorts of happenings.
Yes, I was thinking layer pressing tailings. It's interesting to think that in multilayer discs, that it's actually separate parts pressed together with glue. Up until recently I didn't think that multi-layer or even single layer blank RW DVD's or BD's have the spiral tracks already pressed into them. I thought the laser tracked it's own way as it burned the disc, but it's in fact following a blank track. Like a record LP with a groove that has no sound, just a spiral.
Laserdiscs are amazing in how large they are, and they too are glued together. Like you said, poor conditions for pressings, resulted in delamination of the glue layer most often, and I believe oxidation began to set it. A similar issue arised with the HD-DVD's Warner Brother pressings. Every demo I've seen on TH-cam usually shows the disc failing to playback after the layer change, or choppily if that.
It usually is only an issue on LDs that came from plants that had a high propensity for releasing future ROT victims.
How safe is it to cut or grind away the residue? Would it be okay to span the disk in a grinder (or dremel or other rotating device, after all, the discs are made to rotate and have a mounting hole) and carefully sand off the glue?
@@HappyBeezerStudios I think Mat could use his CD-shaver to clean up the edge.
Is that how disc rot happens?
This is a very enjoyable addition to the story, and you presented it very well. Thank you for your efforts!
Thanks old chap.
@@Techmoan 5 days ago?
@@riponrip4574 Patreon I would presume.
@@riponrip4574 people that support Matt on patreon can watch and comment on videos earlier before they go live on TH-cam.
@@riponrip4574 You really should read the description, he explains it there because somebody makes this kind of comment with every video...
Amazing video, Mat! Kudos for the extra effort in reading the disk: connecting the internal drive of the HD-VMD disk to your computer was a genius move (and quite lucky one too! I got impressed it worked, didn't it even ask for drivers?? plug-n-play like that??! wow!) Thanks for giving us this, otherwise lost, piece of history!
Yes fortunately the drive just popped up with no problem. It no doubt helped that they used an off the shelf Asus Lightscribe burner. This also confirms that they were using custom firmware for the drive.
@@Techmoan Is it possible for you to dump the drive's firmware?
@@DripDripDrip69 Interesting... and turn other similar drives into HD-VMD drives?! wow! I don't know if extracting the firmware is something easily done... do you know how to do it??
@@GeekDetour : You dump the roms (usually eproms, eeproms, or similar), which is either trivial for those with the knowledge & equipment (external rom chips), semi-trivial (read-unprotected internal rom chips), or very difficult (read-protected internal rom chips).
Having said that, it's fully possible to have more than one rom chip (in which case each one can have a different dumping difficulty from the others), and it may well be that LightScribe partially customized the drives, in which case there might be some custom hardware that actually does something important. The only way to know is for someone to get one or more of the drives (preferably starting with a broken one), and start reverse engineering it to document the thing.
Just as a tease for anyone considering this, if someone manages to dump the roms and figure out what the processor is, then it'll be possible (though unlikely) for people to write VMD firmware for _other_ drives as well, maybe even including BluRay drives.
"You haven't really seen Lucky Number Sleven until you've watched it in it's open map format and of course with it's original klingon soundtrack"
I can imagine the faces of some movie 'enthusiasts' who now think they're missing out and their life is incomplete 😂
Great video again 👍
I love that reference to Star Trek VI! "You've not experienced Shakespeare until you have read him in the original Klingon." Bravo!
taH pagh taHbe!
I must admit, the moment he said "You haven't experienced....." I was expecting "...in the original klingon"
I was not disappointed.
Oh I did not get the refecne, very funny :).
"...in the original Klingon." I'm assuming the theme music is sung by... (wait for it...) *_Slevin of Nine?_*
@@bobblum5973 **headdesk**
I’m impressed the disc stayed in shape after all this time. Especially witnessing HD-DVDs just rotting away.
"... and I don't have the budget to fake this in CGI." I cracked up at this! I *really* love your sense of humour; it is the icing on the cake of your videos about technology!
BTW, I also really love my "Taping the Hiss" t-shirt, which has drawn a few comments. People here in Oz (incl. me!) don't get it, until they say it out aloud ...
It's gone to a good home.
Still prefer the non-cgi intro better… the CG feels strangely unsatisfying considering the original was an optical illusion.
I mix and match them now. It makes life just that bit more exciting.
Interesting that one layer seemed to be mirrored - I wonder if this was to improve the discrimination between layers - are there any patents on this format?
I'd also be interested to see inside the drive to see of there are any obvious mods - if they were tight on money it could be that early drives were modified rather than fully production engineered.
I think a firmware update was all they needed. Modern DVD burners are capable of focussing the lense pretty well, and burners have stronger lasers of course, which also helps penetrating the other layers. But it seems to me like penetrating two layers was about the maximum they could achieve (so only 3 layers), I would think that the HD-VMD 20 never came to light.
Also: Nice to see you here!
If I remember the articles from back in the day properly, there is a "M" shaped geometry that was used to give the laser a good view. Imagine the outer legs of the M at steeper angles.
Now imagine encoding from the angle of each of the faces.
Rather than stop, traverse the head back to the beginning, they just reversed the motor and kept reading the next face backwards.
On a normal dual layer DVD the lower layer is read from the outside to the inside. The top layer is then read from the inside to the outside. This was done to keep the switch between layers as short as possible.
could it be that the layer 2(3.) being reversed comes down to the manufacturing?
I mean like glueing a DL and a SL together? that might also result in some rough glued edges.
but it probably would require the disk to spin in reverse some time so IDK 🤷
Opposite track path (OTP) is standard for DVDs and Blu-rays as well. From Wikipedia: "A dual-layer DVD or Blu-ray with Opposite Track Path means that layer 0 is written in a spiral track starting at the inside of the disc reading outwards, whereas the layer 1 spiral starts at the outside of the disc reading inwards. Both tracks are written so that the DVD player will rotate the disc in the same direction."
Somewhere around here I have a couple of Blu-Rays and HD-DVDs that turned out to be upscaled (barely) dvd transfers, and I remember *so many* DVDs that looked like they were copied from clapped out VHS tapes. Even a lot the widescreen ones looked terrible for the first few years. And there was the era before anamorphic DVDs where widescreen was just 4:3 with black bars on the top and bottom, again like old VHS transfers.
Anyway, my point is, I'm not surprised if the transfer is not up to modern standards, leaving aside the whole bit rate issue. And a part of me dies every time I remember that 2006 isn't modern anymore, I was a 30-year-old stay at home mom by then.
Horrible DVDs with the letterbox bars as part of the video so when played on a widescreen TV you got a postage stamp in the middle of a black field.
Cool that you can tell! Not many people can tell that level of quality even if it’s obvious!
@@greggv8 Ohhhh! THAT'S why some movies look like that. Great to know now! Thanks!
@@greggv8 non anamorphic. Aka THE ABSOLUTE WORST!!!
There is so much video upscaled and smoothed with a filter out there. It does get rid of artifacts, but also washes out details, so skin looks like plastic.
And the windowboxing (letterboxing + pillowboxing) on hard coded video with embedded bars is another sin. Even the good old PAL format had a piece in the signal to call if the video was widescreen or not so that your player and TV could dynamically adjust and send the right aspect ratio.
I’m always amazed at the lengths you go to to tell the story of these forgotten formats. My hat’s off to you sir!
This is the most detailed look into a dead, unknown format I've ever seen. Props to hooking up the internal drive to your Mac to get the file size off. The real win here is reminding me that I only have a DVD of this film, so time to get a Blu-ray.
@Techmoan, I'd guess that the noted difference in running time in both discs when compared with the ImDB page is the result of the 25fps conversion.
Europe uses 25fps for TV and film, and while modern electronics can display pretty much any framerate, it was very common to convert movies from 24fps to 25fps when releasing them for European home video. It is done by simplying displaying the 24fps movie 4.16% faster, and applying pitch adjustment to the audio. That seems to be the case here: 25/24 = 0.96, and multiplying the 110min duration of Lucky Number Slevin by 0.96 you get approximately 105 minutes.
23.976
I just wanted to say thank you for making this video and all the hard work you've put in Matt. I really appreciate it and I really enjoyed this video.
You should probably work with someone to extract the custom firmware from that VMD/DVD SATA drive in the interest of data preservation. It would definitely be useful for people who would wanna play around with VMD files in a computer and don't wanna go through the rigamarole of having to search for a VMD player. One could just, in theory, buy the same DVD drive model and burn that firmware in. This could also give an opportunity for someone to reverse engineer the .vmd media container and make it playable in a computer.
@@Hans-gb4mv The documentation Mat has shown mentions DRM protection in form of AACS, which I know has been broken for bluray, but I'm not sure if it's applicable to HD-VMDs too
@@Hans-gb4mv That's honestly possible, we'd need to see the files to know for sure though
Ok, question is, how can I extract firmware from that drive? Do I need some special tools and proper experience, or is it possible just with single mouse click? And how helpful would be sharing that firmware, if there are basically no HD VMD discs to find? Also there are no VMD-R/RW discs (and probably there were no plans to make burnable VMDs, anyway) as is no chance to make 3L VMD from DVD-R DL.
Also, you probably won't need a VMD drive to view/run .vmd files downloaded from some torrent/warez, if available (which I highly doubt, because collectors know how HD VMD discs are rare/expensive and they would be dumb, if they share even ISO from the worst movie ever produced on HD VMD disc).
I own HD VMD player, but I have no HD VMD discs. Collectors, who own VMDs, won't even sell just one disc, if they have duplicates. Or they won't reply at all. Or they would make fun of me. "Oh, you bought HD VMD player for 17€ and you have no discs? Nobody told you that player is worthless without discs? Boo-hoo."
This is how it works between hardcore collectors, who have all VMDs produced in multiple copies.
The only gamechanger would be if someone will share VMD authoring software, to make your own dual-layer VMDs.
You can only use the firmware of the DVD drive to rip the VMD disks. The drive doesn't decode the VMD files, it's only responsible to read the physical data on the disk. Given that the disk shows up on a regular computer means that it uses a standard file system (probably UDF or less likely ISO 9660). In the player there is another circuit that reads the file system and reads the contents of individual files, and another circuit is responsible to decode the video data stream. These can be integrated into a single chip or can be implemented on several chips.
@@baksatibi Yes, that's the one and only thing you can do: open/grab/extract content from HD VMD disc, unless you'll be extremely lucky to get one. But I have a bad news for you, at least 18.000 VMD discs were discarded/destroyed in my country after a company, which distributed VMD players and discs here, filled for bankcrupcy/insolvency in May 2010.
Fascinating - and some very good technological archeology indeed!
When used to watching analogue broadcast using composite, even DVD or standard def digital looks like HD
@@tarstarkusz If you can't see a significant improvement in picture quality between VHS and DVD then you either need a new TV, new glasses, or stop watching from the other end of a football pitch. Either that or you don't care at all about picture quality and would be just as satisfied with the old 405 line TV or 8mm cine.
@@tarstarkusz I agree that vhs was good enough. I never enjoyed a movie less because it was a lesser resolution.
And quality epuipment helps. High end vhs player and a nice Sony crt still look great to me. Assuming the vhs tape isn't junk.
@@tarstarkusz I'm talking about PAL VHS and DVD, I assumed you were since this is Techmoan. I have always had a multi standard TV back to CRT days, and PAL DVDs have noticeably better picture quality than NTSC DVDs. VHS on the other hand looks about the same on both standards. So your being underwhelmed with DVD is likely because they're NTSC and hence lower resolution.
@@tarstarkusz NTSC video from 2” quad tape looked amazing. 1” tape wasn’t so great. Early DVD transfers weren’t that good, but later on they got very good, especially from 2” master tapes.
Since the first video about HD-VMD i asked myself:
If there is a DVD-Burner in the HD-VMD-Player, could you hook it up to a computer and use it as an external drive?
Thank you for answering that question with this video!
and the next question is, could those files be read with any existing software?
It was so satisfying to have that question answered!
@@martijn208 He tried in the previous video - no dice. Not to say it isn’t possible, but nobody seems to have cracked it yet.
I was thinking the same thing, maybe he can hook it up directly to his laptop instead. Sure enough. I wish he had shown 'Get Info', it should say total capacity and used capacity I would think. I don't have a Mac, don't use them much, but I think that's basically the same as Properties in windows.
@@marcusdamberger I'm pretty sure that once burnt/pressed, on most systems optical media will show up as 100% full, the capacity being the amount of data on it rather than the theoretical/practical maximum.
Thanks tor the follow-up video showing the real HD-VMD disk functioning. Considering that they were using DVD readers to play HD movies, I now consider this format quite an achievement
What i love about this channel is that you take the time to get fully into things! You even bought a blu-ray version from about the same time to check if that too was soft looking. Like you can be sure that if there is any information on the subject, we are getting it! This is some quality content!
I really really love how thorough you've been with this detective work, so much fun to peel away at this product.
It's interesting that the disc has visible layer identifiers - since they don't mean much to a user of the final physical disc, they must have been there to be read by people working in the manufacturing process, perhaps to ensure that the right set of 3 layer masters were installed in the stamping machines - or the right set of layers glued together after stamping. I haven't looked closely at any non-CD optical discs, so I don't know if this is normal on multilayer discs, or a sign of the small scale of their prototype (and only) process.
Yes I don’t normally look closely at discs either - I might start doing so from now on.
Yes, multilayer discs have Layer information printed on the center ring.
@SteelRodent Indeed, almost all of the Blu-ray and DVD discs that I have contain visible marked rings for the different layers. You can clearly tell 3-layer UHD Blu-rays apart from the normal 1- or 2-layer Blu-rays, for example.
You make a good point about trying to compare the HD-VMD to BD. So many factors to choose from: video compression codec, target bitrate of the final feature, and of course our friend digital noise reduction which while used unfortunately all too often these days to reduce the presence of film grain (and even sensor noise on digitally shot movies), was in much stronger use then than say now. As you mention this was shot in Super35 format which is full aperture 35mm negative film with spherical lenses and then cropped to the desired AR which was 2.35/2.40:1 and by 2007 it's almost certainly finished to a 2K digital intermediate (DI). This means both discs utilized that DI as that is the finished movie, so the film to video transfer was the same as that would be done during the DI process when the original negatives were scanned which definitely helps your comparison here.
The company responsible for the HD-VMD transfer may have altered the color timing slightly in addition to the wrong aspect ratio, and added digital noise reduction or other filtering to soften the image further in addition to softening as a result of low bitrate. One thing you can do to see if more DNR was added and not just compression softening the image is to look for areas that have relatively solid backgrounds like a skyline or especially a white wall where film grain would be most naturally visible. If the HD-VMD has more "sluggish", "sticky", or otherwise unnaturally looking film grain, bam there's some more DNR added compared to say the Blu-ray version. Even a DVD, if not hit with more DNR will display film grain that is more naturally present and moving but due to the low resolution and bit rate, will appear clumpy and less defined. With a trained eye, you can spot this. I hope this helps explain the differences you may be seeing!
As a whole though, the low bitrate and codec would be the main culprits here to explain the softness of the image compared to the BD.
I'm tinkering around with video game capturing for years now and noticed a similarity that confirms your theory about the softness. In short: yes, if you have a softer image, it reduces encoder stress and you can potentially avoid those nasty compression artifacts (image blocks). The softness is not produced by the encoder, you add it artificially, before sending the video to the encoder.
When capturing games, I have set my encoder to VBR Quality. So basically I'm telling the encoder: I want THIS level of image quality. Take as much space as you like to achieve it. What really blew my mind was, the resulting video sizes are DRASTICALLY different depending on what game I capture, although I'm using excatly the same encoder settings. 1 Hour of Cyberpunk 2077 in 1440p60 barely scratches the 10GB mark, while 1 hour of Phantasy Star Online 2 New Genesis happily will eat up 50GB and more. Most funny thing: Cyberpunk even looks better. So what gives? The main difference is the image sharpness. Cyberpunk has got some really advanced post-processing, trying to soften unnatural sharp lines, some DLSS resolution trickery to achieve more FPS, and all that stuff which potentially makes the image softer. Phantasy Star has a more straightfoward Sega-Like basic "arcade" style engine with all kinds of hard edges, zero f***s given. This is giving the encoder a hard time to stay efficient.
All widespread video compression descends from JPEG image compression format. It was formulated for "photographic" images. This was specifically a design goal, in differentiation to compression of "synthetic" images, and it was actually an insane breakthrough, the sheer insight that went into it is difficult to grasp. Cyberpunk leans closer to the ideal "photographic" while PSO2 leans closer to ideal "synthetic". Unfortunately "synthetic" image compression is well behind, suitable approaches exist but they tend to be very heavy in terms of encoding and decoding system requirements. Like sinc wavelets could provide feature extraction and vector quantisation (VQ) should be good for actual compression to encode self-similarity within the image and to past images, VQ is just insanely slow to encode (multi dimensional hill climb) and sinc requires approximations for good performance that we're currently just slowly edging towards.
I do think you can help the encoder. One thing i would do is always apply as much anti-aliasing as possible. I think maybe i should port FXAA to OBS so you can apply it before the encoding just to the video, while it attenuates some high frequency detail, it should actually help preserve detail at a given constrained encoding bitrate, and it doesn't eat nearly as much detail as a simple blur.
@@SianaGearz Wow that was thorough, thanks. I'm already using all AA the game is giving me, but i can still add more by forcing it in the nVidia driver. I will try to tinker with that a bit. I'm also using HEVC as it's currently the most efficient codec I can practically use (GPU Encoding). I am also not super dependant of the quality outcome, i just started to wonder how far I can push it. TH-cam is massacring it anyway, but I wanted the best possible quality I can provide from my end. That's also why I'm using 1440p as TH-cam grants way more Bitrate and the blocking is not as bad as with 1080p. I'm currently recording with Bandicam, but I also have OBS ready to go.
@@Raveheart Word of note, depending on capture mode (i.e. game capture), NVidia forced FXAA might be visible during gameplay, but might not reach the capture and the encoder. I don't remember the details, just something to look out for. I do think it gets captured in ShadowPlay though. Also why i want to make it an OBS filter is because while FXAA works invariably better especially on a livestream, people may not want to actually play with it on, so a filter would be ideal.
I tend to record with a ridiculously high fixed bitrate using GPU encoding (to have the initial recording done fast, sacrificing compression for quality) and after cutting use CBR x264 and after that end up with around 2-4 mbps for 1080p video depending on content.
The issue with todays AA implementations is that they either look blurry or flicker or both, I wish there was some more advanced, adaptive tech that would work on all games and could simply be added on driver level to not run into issues with anticheats.
@@HappyBeezerStudios speaking for all the game engine developers out there, a drop-in antialiasing which doesn't introduce its own issues would be the holy grail.
The driver angle has become increasingly difficult, because the driver doesn't have a lot of insight and reasoning into what the application is doing any longer. The more programmability became possible from the application side, the less leverage they got. So that's another level of complexity.
Another really interesting and well-researched video, Mat, cheers. I always look forward to sitting down to the latest Techmoan video with a coffee at the weekends. Out of interest, I just checked my HD-DVD release of Lucky Number Slevin (a 2006 disc, as best as I can tell), and it's AVC encoded, has special features (Making of, Deleted Scenes/Alternate Ending, two commentaries, trailer and "An Intimate Conversation") and English and French DD+ 5.1 soundtracks. It's also 2.35:1, and, interestingly, IS 1hr 50mins in length (the credits end 10 seconds shy of the 110 minute mark). This really did seem, for the many reasons you mentioned, a really half-baked format.
@Techmoan I have a recording from this film when it was broadcasted on the Swiss public television. The Video-Stream is 720p 50fps and probably sped up from 24 to 25 frames a second that was standard for DVDs and still is for TV. On BD they started to encode them in 24 fps. My recording without the end titles is 1:41:22 long. Probably this must be converted from 25 to 24 frames. That would lead to 1:45:35 seconds without the end titles. The movie is also in 2:35 and the car park scene matches your 2:35 version. My version is even softer with 720p and only 6.409 Mbit/s (including German Audio in 5.1 AC-3)
Your calm voice always manages to make my day even better! (:
As a guy in his mid 30’s I randomly found this channel a few years back and have been absolutely enamored with the content you create. I’m a huge A/V nerd and am always tweaking, calibrating or and/or upgrading my own setup and find it absolutely fascinating to explore the history of all of these lost formats. Keep up the phenomenal work!
Wasn't expecting a ST:6 reference, and yet here we are.
Great video as always Mat!
I'm surprised you haven't done a proper dive in to UMD's. Or if you have, I didn't find it. The PSP probably had some interesting compression or something noteworthy.
Yeah, I recently watched the classic "Goodfellas" on UMD. it's around 3 hours long and the picture quality is nearly as good as DVD. But due to the battery life not being good on my PSP and I could never watch it all the way through without charging.
Yeah. UMD's. That format never did make sense for me as a home video platform. But it would be fun to see what can be gained with oroper research into what the plans were as quick wikipedia skimming reveals plans were at one point there for multidisc standalone systems.
You can read the discs over USB with a hacked PSP. I remember checking one of my UMD movies, and it was H.264 encoded, I believe at normal SD resolution. Approximately DVD quality, maybe slightly worse (better efficiency compression, but less bytes to go around).
The movies will obviously look sharper on that small (and low-resolution) screen of the PSP. You can connect it up to a TV with a component cable though. Or just rip it over USB with a hacked PSP (probably the sharpest image you're going to get out of it).
I never did understand movies on umd, it always felt like a bit of a hassle. I did put some anime shows on the psp back in the day but at least then I had a whole show.
I feel like it's close to the minidisc format it's just how it was formatted
You make the most obscure, insignificant things interesting. Thanks for quality content.
I’m watching Lucky Number Slevin now because of your recommendation. For the record, it’s streaming at 2.35:1.
Edit: Corrected aspect ratio
Glad to get this coda. And I'm really glad to see you interfaced with the disc drive in the player to see what was on the disc, I would surely have suggested that :P
The intro somehow felt different, then I noticed it’s because it’s actually a completely new intro and the whole video is in 60 FPS, nice :)
I like to mix and match to keep things confusing.
@@Techmoan i know you like to mix and match withint fan fair. To be honest i loved your ending in the past with more bits of video and a longer sample of the music from a couple of years ago so thus often when I watch your videos I will boot up the end of a random old one just to get the extra credits!
I also ended up looking up the free source music and sound. Oddly even listening to the music by itself doesnt do it without you taking apart things in the background slowly or spinning things on a lazy Susan it just loses some of the magic.
As for the sound I got to find out it was from the old 7 million dollar man show I never watched an episode but I had heard of it.
As you might say overly long explanation but I felt compelled to share it. Thanks for the work.
i thought the same thing!! I thought it was me!!
Matt, the sheer technolgical knowledge and deep diving has brightened my slighty dark Saturdays, from my previous projectionist past, you have made me seriously chuckle. You are a gift to us Sir. And I thank you. (I particularly enjoyed this one.)
Me: Finally! THIS is the LAST of the physical media formats that I will ever have to compulsively seek out to collect.
Techmoan 2022: I just got this new old stock XYZVD player from a seller in Japan, let's have a look.
You have a Super Video CD player? Dunno if those will also play CVD (China Video Disc, a variant of SVCD) or if a CVD requires its own player. A SVCD with some care taken in its encoding (2 or more pass) can look quite good at 480x480 with anamorphic squeeze. Same vertical resolution as NTSC DVD.
Good luck! It looks like movies are virtually impossible to find in this format. The movie Techmoan shows is a Scandinavian release - makes you wonder if they had a better rollout in Europe.
Fellow collector, I am the proud owner of a MovieCD that plays in Windows 3.1 👍
@@johnfriction4919 Was it that It's a Wonderful Life CD-ROM Kinesoft made?
I have fun imagining techmoan with a time machine like quantum leap but he would go back and rescue rare hi fi and video equipment and never miss a discontinued line sale .and rescue all the obscure formats from landfill fighting off sea gulls as he goes .But what to call this unlikely super hero .Techmoan man. Oh dear
Theorising that one could examine different media during his own lifetime, Doctor Matt Techmoan stepped into the Quantum Leap accelerator. He didn't vanish, but a bunch of different machines appeared, and driven by a force of TH-cam users to review these machines for the better. And so, Matt finds himself leaping from format to format, striving to clear up all the malarky surrounding each format's history, and hoping each time his next leap will be leap home...
From a pure storage perspective, using video with AVC/H.264 encoding at 2.6 mbps (which is fine in terms of quality for the codec) would allow around 3 hrs runtime with 2 audio languages in 6 channel 384 kbps AC3 audio onto a single layer DVD. For dual layer it goes up to around 5:36 hrs.
So DVD is totally capable to deliver FullHD 1920x1080p video at acceptable quality purely by storage. And for 2.35:1 cinema format it could either be anamorphic or in 1920x832
As a bonus: youtube recommends 8 mbps for 1080p upload, at which you'll get around 1:14 runtime for a single layer or 2:15 for a dual layer, so even at that bitrate it would be fine to run put HD video on a dual layer DVD with enough storage for a movie. Squashing a bit more down to 6 mbps and a single audio set would land us at 1:40 and 3 hrs respectively, which should be enough for basically anything.
I wonder if someone has dumped the firmware of one of these to see how it works? Since it seems to be an off the shelf drive, just with custom firmware, so it should be possible to make other DVD drives support the HD-VMD, and maybe even modify a BD drive to support it as well?
There are apparently ways of dumping the firmware of disc drives, though I've never done it myself.
If nothing else, it would be good for there to be an archive of the firmware for posterity.
Great video as always, you did a superb job of explaining the technicalities.
I bet it looks soft because they most likely used 1440x1080 instead of 1920x1080 in order to reduce bit rate, it was a technique used a lot in digital satellite broadcasts to reduce bit rate and making digital compression artifacts less noticeable.
It was also used in the early days of Freeview HD as I bought the first released to market DVB-T2 set-top-boxes and DVB-T2 dongle to work with Linux. Thankfully they've reverted back to 1920x1080.
Anamorphic video was also used in the analog days and stayed into digital.
Ciao Mr. TechMoan 🙋🏻♂️
You are THE MAN for tech and hi-tech matters.
I can only say thank you for running this channel fulfilled with you strong passion and depp competence.
Plus you explain things in a way that also a monkey can get them.
Magnifico 🥇
Cheers from Monza (Italy)
A fascinating story. Thanks for your persistence!
This has been an interesting journey through a format I never knew about. That's a lot of work, and I'm very grateful that you put all this together. I think you've really created some durable canon on the topic, here. Well-done.
The people who worked on this format must be delighted you’ve spent all this time explaining how awful it is and how it’s nowhere near as good as what they had hoped it would be
The idea was good, but the implementation showed that it was unpractical.
What a fun adventure in failed video format. Always look forward to your videos. Thanks
No one can say Techmoan isn't willing to do whatever it takes to make a video; he had to travel back in time 14 years to 2007 just to find a specific bluray version of a mid-2000s movie. I'm just a little disappointed that he decided to that off-camera, however.
Sensational change of shirt at 2:23
"...you haven't really seen Lucky Number Sleven until you've watched it...with its original Klingon soundtrack."
Reference acknowledged.
its been years since i saw ya videos ,,, we are glad your channel grew over a million subs , im impressed bro
Another possibility for the low size of the file is that many companies use standard templates instead of adjusting the bitrate to the movie, so if with one template the movie is too big they'll use another one, and if this one don't use all the available space well, bad luck.
I really think they have a 11.5GB capacity disk and filled it to the brim. You can't do fixed bitrate anyway, it just looks invariably bad, and at 2 pass encoding which distributes the bitrate where it's needed, sniping for correct size isn't really an effort.
Because have you noticed the capacity of DVD-9 is 10% smaller at 8.5GB than two DVD-5s at 4.7GB each? If i assume that marginal capacity introduced by each layer is 20% lower than the last, so the per-layer capacity regression continues, the 3-layer disc comes out to 11.5GB, which i think is a little much of a coincidence to just dismiss. 4 and 5 layers could house approximately 14 and 16GB respectively.
13:40 (writing before I watch any further) Regarding the softer image - applying a bit of blur before encoding will remove high frequency information, and allow to use a lower bit rate without introducing visible blocking artifacts, so I think it was done on purpose. Some JPEG encoders have this option built-in (GIMP). This however obviously diminishes the benefits of using a higher resolution, since the point of that to have sharp detail.
I’ve been encoding movies for the past 20 years or so, and what I’ve been doing is applying a noise filter on some titles where there was a lot of film grain or noise. This reduces the size of the file significantly. Granted it can make the movie a bit soft at times, but I do a few tests before a complete encode to balance size and quality. In the end, when you have collected a large library, this will save you quite a bit of space. It’s possible that they did this as well to get the movie to fit on the HD-VMD properly. Anyway, thanks for all the hard work you have done. Cheers from Canada Eh! :-)
Noise reduction can also remove film grain, which may or may not be desirable
@@AlexanderBellOpelika For sure, in some cases where you have a really old b&W movie, removing the grain or noise actually makes it worse. In those cases it's best to leave the processing off.
@@robertromero8692 It's not a terrible practice, You're not meant to see the grain that makes up the film picture in the same sense you're not suppose to see pixels in digital representations. Grain or rather noise is a pain to keep retained because it adds unnecessary pixel data making it exponentially bigger with no real practical use outside of archival needs. Were not even talking about the hell it can play on compression methods just the raw storage method alone, it costs nothing to store a frame of pure black but it costs everything to store noise with a 3 bit variation of black that you're never going to see. The more solid pixel color you can use to make up the image the more cheap it is to store.
@@robertromero8692 It's not a justification nor is it a data challenge that wheel has been invented many times for image storage. Grain is noise physically and even in a digital representation, This has the same energy of needing to see the shimmer of cells under a microscope in order to see that my blood is the color red. When you start to see grain it stops being the image and becomes an artifact of chemistry there is no more image the further "in" you go, the more in and higher res you digitally represent it the more it becomes microscope slide of crystals there's nothing left. What im trying to convey is that it's a massive inconvenience to store noise like grain nor is it necessary to expand its fidelity outside of archival needs. The sad part about this is that it's easier to randomize grain/noise in realtime than it is to retain it. I get that you want that "film" look but when you transfer mediums all it will ever be is just a representation of it and you have to draw the line somewhere and accomplish the goal the original medium set out to do.
@@robertromero8692 No I do not care to preserve a visual artifact at cost unless it's part of a goal nor am I going to pretend a Mona Lisa painted in acrylic is the real deal. Im well aware grain can and has been captured to a point but it's just fat that can be trimmed and adds no new visual information from my point of view.
I loved the Undiscovered Country reference at the end. It made me immensely happy.
I actually have some DVDs with such few special features where they also list Scene Selection as well as ‘picture on the disc’.
Great follow up Matt keep up the great work you do. I can't ever even remember this format back when I was 26/27 I always had DVDs
Lower bitrates normally would result in a more blocky image, especially with VC-1 which is just a variation of MPEG-4 ASP (aka. DivX). I think they might have blurred the image to hide blocking artifacts before encoding, to make it easier for codec.
I wonder if it's a lower resolution and the player is upscaling ?
@@DJlegionuk seems more plausible to me than the first option. Would not be surprised if they encoded it at like 720p then upscaled
@@DJlegionuk Hmmm... Could be, but you often still see "staircase" effects around pixels with that. But it depends. Maybe they use a gaussian upscaler. They're very soft and blur out pixel artifacts.
If you look at the settings for Xvid (which it's been a while since I opened that up) it lets you choose between h.263 quantization for a softer image that works better with low bitrates without blocking and MPEG quantization which gives a crisper image. Using a crisper quantization matrix works better for high bitrates but results in the classic blocky video compression at low bitrates. Modern codecs include an in-loop deblocking filter that effectively makes adjustments on the fly based on how much bitrate it has and the complexity of the video.
@@ravewulf Yeah! I just don't know if VC-1 lets you do this. It's a proprietary implementation of MPEG4 ASP after all.
Man I love this stuff. Your videos are top notch brother
I was hoping that you’d connect the actual drive to your computer… then you did! 🎉
I like the way your shirt magically changes through this video.
Good investigation! Three layer DVD... makes sense that it could be read on a normal DVD with special firmware.
the second layer on a dual layer DVD is smaller capacity than the one layer one. Probably the third layer on this is even smaller, to make the symbols easier to extract.
Absolutely phenomenal. Love the level of detail you go to.
My man Matt rocking the new Sub 😏
Thank you for documenting all these formats - with style!
Great enjoyable and detailed conclusion. Thank you!
Ta.
The Sherlock Holmes of obscure audio and video formats.
Your a trip my friend, I always enjoy watching you!
PS, I miss the puppets.
A 1-layer DVD is 4.7GB (in billions of bytes) and a dual-layer is 8.5GB. Therefore, the second layer is only about 3.85GB. Multiply that by three and you get 11.5GB, which is 10.75GiB as seen by Windows explorer. Therefore, I suspect the disk is three 3.85GB layers . I guess they just couldn't make it with a nice 4.7GB first layer, and it really should be called VMD-12.
"Windows explorer" SORRY? WTF is that? Animal, vegetable, or mineral? 😉
@@a.peterrogers3488 it was a rhetorical question, hence the emoticon.
funny thing is with today's h265 compression, if you have the cores and time you can make masterfully good quality 1080p, obviously non HDR backups of blurays for around 10gb.
My father was a laserdisc collector, so have backed up every bluray i've owned out of pure paranoia of disc rot, and sure they are far more robust but it takes a tiny fraction in damage with blurays to render them unplayable next to a dvd, let along laserdisc.
Such is time.
@@rahb1 "emoticon" WTF is that? Animal, vegetable or mineral?
@@revivedfears "Animal, vegetable or mineral" WTF is that? Windows explorer, emoticon, or rhetorical question?
Just use VLC - drag the video file into it, then bring up the stats window. Very insightful!
*There’s a bit more information about this video in the video description.* 🔝👍 Includes updates, links etc...
Brilliant detective work. Hugely enjoyable!
Wonderful due diligence. Thanks for the video!
I just loved the ball bearing explosion at the start of swordfish. It was the first DVD I played on a 5.1 surrounded sound setup years and years ago.
My local theater had upgraded their sound system only the week it came out, and that scene was so much more impressive than it had any right to be
Now that's a rabbit hole I never intended to go down and learn about. Bravo.
21:04 - That's not true! VC-1 with 13 Mbps in 1080p would not degrade image quality by totally blurring all frames! That's not how compression works. Lower bitrate could introduce compression artefacts like macroblocking in complex scenes. Im 100% sure that 13Mbps is more than enough to produce super crisp static scenes.
The reason why image is blurry is due to bad transfer. Check for example blu-ray version of "Last Man Standing" with Bruce Willis. It is also very blurry and it looks like upscaled SD.
I studied compression in college and wether x bitrate results in a crisp image depends entirely on the codec's efficiency. 13Mbps at 1080p with H.264 will look absolutely fine, but VC-1 is less efficient, which means you'll need more than 13Mbps for a clear image. Compression artefacts are easy to see if you go really low on the bitrate, for example transcoding a video into 1080p H.264 with 1Mbps, you'll immediatly notice the squares in the video. But if the bitrate is just a bit on the low-side, the compression artefacts will be small enough that the video will appear to be slightly blurry and this is what Mat noticed. 1080p H.264 suffers from the same problem at 4-5Mbps. At 4-5Mbps, only the newer HEVC H.265 is going to provide a decently crisp image.
@@r34ztune11 you are forgetting other critical factors like video image complexity and the encoder implementation, and in this regard most VC1 implementations are really poor. Also the mastering and the tranfer are also very important, I have seen that in one movie that while the USA version had more bitrate than the european from the same year the image quality was much lower but in a posterior re-edition of the USA version it had the same quality.
I recommend you to read the Netflix article about per-title encoding and shot-based encoding.
Poor Bruce willis
@@r34ztune11 what was that class called and where at?
Either way VC-1 isn’t very good and was quickly dropped in favor of the AVC/MPEG4/H.264 compression. Very interested to see what it could look like in H.266/VVC codec. Could it essentially fit well on a 12gb disc?
Going deep into the forensics of things.... impressed! As always!
IMHO I wasn’t to excited about the first video, but then with all additional videos and revelations it turned in to quitte an interesting format adventure! Very nice!
@@CommentFrom yeah that’s nice too
Brilliant explanation of all this techy stuff. Really enjoyed it.
I'm 23 years old, but have been following the channel for the past 5 years or so, and I cant describe how exited I am to come home after my morning swim on a Saturday, pour myself a good cup of coffee and watch a video from Mat. It's just proper relaxing, and for me I'm also associating his channel with the feeling of safety and calmness. Well done and thanks!
Great video as always! I love learning about things like this, and you did a fantastic job of sleuthing out all the details. And your soothing voice always picks me up. Thank you!
I’d be curious to know what they did for DRM. Did they even get that far?
Is there a HDCP protocol?
Really great to see a wrapup on this, kudos for revisiting the subject!
Will you ever bring back your hilariously funny puppets again? I miss seeing them during or at the end of your videos!
I miss them too, but in case you don't know Mat removed them due to TH-cam categorizing his videos as "kid's content" hence no ad revenue
@@MrJef06 Thank you for telling me. I didn't know that. That's a stupid TH-cam rule IMO.
@@MrJef06 That's really stupid. Maybe if he made them say adult jokes that'd fix the problem
Love the detective work and mystery-solving. Really interesting stuff!
There is no way a VMD 15 is going to be 15gb. they are rounding ~4.5gb up to 5gb per layer, but that adds up and the capacity shrinks per layer. So based on some loose maths, going from 4.7gb to 8.5gb is about 90% of double capacity. So if we just apply that again I get a rough estimate of about 12/13gb for a VMD 15. That seems logical.
Another fantastic video and really interesting deep dive into a format most of the world never knew existed. Watching a techmoan video has become a part of my Saturday ritual so thanks very much for keeping up this schedule! I know it can't always be done but I look forward to them every week and you always deliver!
I really enjoy the minidisc episodes and was wondering what you planned to do with all those hundreds of players you recently got? Could be an auction (or do you put stuff on eBay ever?)
By getting that drive working on your PC, you might have done something very few people have done before, used a "15"gb DVD drive on one. That's quite cool.
17:53 in case anyone's wondering about the date (31 - 12 - 1969 @ 16:59)
That's the date you get when you put 0 in unix time... minus 7 hours & 1 minute. (unix time starts from 01 - 01 - 1970 @ 00:00)
UTC -07:00 is a time zone for somewhere in North America. NME was incorporated in Nevada... and the time zone that Nevada is in is in fact UTC -07:00
The more you know! :D
I remember when that movie came out on DVD at the shop it was stylized on the box as "Lucky # Slevin"
Then when Twitter and their hashtag system arrived, I thought "Kids these days won't understand a movie title like "Lucky Hashtag Slevin"
Whereas my agèd brain automatically tries to parse it "lucky pound slevin", which doesn't work either. :)
and I’m still on Lucky Octothorp Slevin
And C Sharp programmers were just as confused.
I never called that symbol 'number' when referring to a button on a phone - everyone I knew called it the hash key. I used to program call centres and it was always - 'press the hash key to continue'.
Surely the title is just "Lucky" and the rest of the line is commented out by hash.
Every Friday night after a long night I like to sit down and watch Techmoan while I finish my work, so thanks Mat for being punctual with your uploads, your videos always give me something to look forward to and enjoy. Then I usually watch it again the next day because I usually miss a few parts here and there because I'm doing something else. When you don't publish it on time I get worried lol.
I’m beginning to feel like Mat needs an honorary PhD in Media History.
He deserves it for sure. So much info in all those years.
Agreed!
Wholeheartedly agree
For sure.
I think the "REVERSE" written next to Layer 1 indicates that the player will read this layer in reverse compared to layer 0 and 2, that way the drive can switch layers quickly without moving the lens all the way back to the start of the disc again. That way the disc can be read FWD>REV>FWD. I'm sure if you had a 4 layer disc, the fourth layer would be REV too.
(This is probably something that dual layer DVD's already do, too, though)
And Techmoan finaly gets a proper watch! 😉 Congrats for your submariner, Tech
I was just checking the comments to see if anyone had spotted that too.
I noticed the watch....I wouldn't think it would be his style, I can't imagine matt spending many thousands of pounds on a watch, it doesn't fit into to his homely northen charm, and artex front room, hasn't even got a nixie tube in it!
@@sadnerd397 hahah, the same for me, i remember the old video of his with the Seiko Sea Urchin and i thought it was that one, until i noticed it's actually a Sub
@@solidamber A sub' fits everybody everywhere. Period 😎.
And this is the most economical watch on the Planet. You buy one for the rest of your life and it doesn't need any batery. Consequently a sub' is good for Our Mother Gaia. Period bis.😄
@@FLH3official Using the phone you already have is better for the planet, and a phone is a better watch too.
Tremendous detective work that clarifies the death of a format.
Brilliant Video and amazing detective work, but why isn't anyone talking about Matt's Rolex drip 🤩🥶🧊💧
I saw that too!
Great presentation of the technical aspect
a fresh new video about an incredibly obscure format?? yes please!
Thanks for another great video Mat. Fascinating stuff.
We're you ever contacted by people actually working on those "forgotten formats" back then? Especially with this VMD it would be incredibly interesting to interview someone involved.
There was one guy commenting on previous video and he said he was involved in authoring VMDs.