ATI capture card comparison: how they handle a VHS tape with flagging

I played my VHS tape with the most flagging on a regular Sony VHS player to see how three ATI capture devices handle it. It was a kind of stress test. These are the capture devices I tested:
- ATI USB 2.0
- ATI TV Wonder 600
- ATI All-In-Wonder 9600XT
The VHS tape contains a cable television recording from 1986, recorded in EP (SLP) mode on a Panasonic Omnivision VHS VCR model PV-1265. It was recorded onto a VHS tape over a previous recording. As a result, there is flagging and tearing, making it great example of a terrible recording.
In addition to the three ATI capture devices, I also tested a MiniDV camcorder. Most/all experts at DigitalFAQ don't consider this a good method because of its compressed color space. But others on TH-cam, such as @TechTVusa claim that the ATI cards are crap compared to the DV method.
The ATI samples were captured as AVI files (HuffYUV) and deinterlaced TFF with QTGMC in Hybrid, and resized to 1080p and exported to ProRes 422 MOV. The DV file was deinterlaced BFF with QTGMC in Hybrid and resized to 1080p and exported to ProRes 422 MOV. No other filters were used in Hybrid.
00:00 Introduction
00:25 - Test 1: ATI USB 2.0N
00:57 - Test 2: ATI AIW 9600XT
01:16 - Test 3: ATI TV Wonder 600
01:41 - Test 4: Sony TRV-17 MiniDV camcorder
01:57 - Test 5: Sony CRT television KV-1363R
02:25 - Summary of results
02:50 - Sample video comparing ATI 600 and Sony MiniDV
03:05 - MiniDV workflow with good VHS player
03:56 - Further testing required
04:10 - Conclusion
#ati600 #atiusb2.0 #ati9600XT #VHS #VHScapture

[@TechTVusa]

Thanks for the shout out. It used to be so simple to use Premiere Pro and a Firwewire DV converter to capture and edit video but in the year 2024 Premiere Pro can no longer captures video. On the Apple side iMovie and FCPX still work with Firewire DV converters.

[@electronicwaves]

Keep those videos going. One year earlier that I was on the search of a device with TBC (DVD recorder with pass through capabilities, miniDV or digital 8 camcorders with svideo/RCA input and firewire output and line TBC) I couldn't find any comparison like that on youtube. This is extremely helpful. Btw I don't remember reading anywhere on digitalfaq/videohelp that those ATI cards have some sort of TBC. They were regarded as overall excellent in capturing analog VHS signal and keeping audio and video in sync. None recommended to use them without TBC if I remember correct.

[@videocaptureguide]

you are correct that these capture devices require a TBC. I was surprised, however, that there is such a difference between the ability to capture very bad VHS tapes. I had assumed that the recommended ATI capture devices were equivalent in all these areas. But in this one test on this one tape, it appears that the ATI 600 is best.

[@electronicwaves]

@@videocaptureguide it would be very interesting to know what version of ATI (rage) theater IC each one uses. BTW the fact that you own 3 different ATI capture cards and know how to do useful comparisons like that is very rare. Could you please upload more videos testing those 3 cards with known good VHS tapes and then also comparing to DV method? You could try good VHS tapes and good VHS VCR + ATI cards without line TBC and also good VHS tapes and good VHS VCR + ATI cards With line TBC in the middle. I am really jealous that you have those ATI cards. I have an AIW 9800 SE that started outputting artifacts due to GPU issue or RAM issue, I don't know and for that reason I cannot use it to capture. I might try to fix it soon because your videos show how good those cards are in VHS capturing. Thanks again!

[@orihalcon8693]

Keep the comparisons coming!
On the DV camcorder - did you save that file as DV, or did you do something with passthrough? I believe the typical camcorder line TBC workflow uses something like a Hi8 camcorder for passthrough back to S-Video and then capturing S-Video as usual.
Camcorder seemed to add A LOT of horizontal chroma noise on random individual lines - like randomly having a mostly green or pink line noticeable seemingly every second at least. For that reason, I wouldn't use that specific camera - could be a capacitor issue maybe.
What I'd describe as "film grain"/sharpness/blockiness looks pretty unique to each, though I can't say which is "best" since that's somewhat subjective - but if you used a digital source like a DVD, you could compare the capture to the original to see what looks closest to the source.
Each chipset will have a different tolerance for line errors, so in that specific setup, the 600 may be adequate without other devices inline if you also aren't getting any dropped frames.
In future comparisons, a count of dropped frames over say a 15 minute period would be pretty cool to have for data as well on each card tested.
I plan to do my own capture testing and post results, but designing the test is probably the hardest part - ideally you'd want the test to be reproducible by others so that others could post results of the same test. I imagine it would involve using something like a DVD with test footage and various patterns such as "Video Essentials" which is quite cheap and readily available and I'm pretty sure that it does not have macrovision. You can't do any time base error testing with DVD (since there are no time base errors), but you can see how the captured video compares to the original digital source. The main variability in using a DVD like that would be that different players will output at a bit different levels - Like mine tends to have the 100% white at 110IRE (instead of 100IRE) just as an example, and you really can't tell that without a waveform monitor or a rather accurate/calibrated oscilloscope. S-Video waveform monitors are also rather uncommon, but I think you can run just the luma into a regular vectorscope (scope will see it as black and white composite I believe) to determine that as well.
Comparisons of S-Video to composite for each device might also be interesting. I've seen some capture cards seemingly do just as well with composite as S-Video and theoretically it's possible for composite to be better than S-Video if the comb filter within the VCR in use is poor.
AGC (Automatic gain control) issues get brought up a lot as well, but I haven't come up with how you'd reliably test that - open to ideas if anyone has any. That should be part of any comprehensive test as well I think.

[@videocaptureguide]

I was satisfied earlier to finally get this comparison video out the door. And now your excellent suggestions for comparison tests has me in a cold sweat. I wish my brain had a build-in line TBC to keep my vertical jitter in check. 🤣 I have added these ideas to my list of future videos. Many thanks.

[@TechTVusa]

@orihalcon8693
You wrote the following below.
"Camcorder seemed to add A LOT of horizontal chroma noise on random individual lines - like randomly having a mostly green or pink line noticeable seemingly every second at least. For that reason, I wouldn't use that specific camera - could be a capacitor issue maybe".
How do you know that what you saw was not identical to the original tape? Could capturing uncompressed 10 bit at 4:2:2 yield the same result? Do you kind of see my point?
AGC (automatic gain control)? All audio and video corrections should be done through software as opposed to using hardware. Software gets better and better but hardware becomes obsolete.

[@cactoosdh0]

hi, it might be silly question, but how did you connect your vhs to your sony camcorder ? i just checked trv-17 specs, and can't really see any inputs there...I am not sure if i understood this concept : vhs player -> sony trv17 to fix footage ( TBC ) -> captured on the PC ?

[@videocaptureguide]

On the TRV-17, you set the POWER switch to VCR (as opposed to CAMERA). Typically you would do this in order to play back one of your MiniDV tapes. You'd connect an S-Video cable from the TRV-17 to your television and watch it. So the flow is TRV-17---SVideo--->TV.
But the SVideo port seems to be an in and an out. I'm not sure how it switches, but it does.
You you can also connect any analog device to your TRV-17 and the TRV-17 will display it on its LCD screen. You can then press the record button and it will make a copy of the footage entering its SVideo port. But you can go one step further and connect the firewire cable and send the video out to your computer firewire card in real time.

[@cactoosdh0]

@@videocaptureguide THanks for taking time to answer!

[@harryshuman9637]

Where did you even find the drivers for the AGP card...

[@videocaptureguide]

The indispensable DigitalFAQ forum has copies of all the various drivers.

[@jarecki83]

Is there any advantage to using the built-in TBC in SVHS together with the TBC in the camcorder? Should I only use the built-in TBC in SVHS or only the one in the camcorder when capturing via MiniDV?

[@videocaptureguide]

That is an excellent question and I've added it to my ever-growing lists of comparison tests. Based on my experience/expert advice from DigitalFAQ with the similar situation of combining SVHS w/lineTBC + ES-15 w/lineTBC, I think you would be best to turn off the SVHS TBC. But on the JVC models, that also turns off the noise reduction, which many recommend leaving on. I have checked the menu and I think that there is way to turn off the Line-TBC in the Sony MiniDV camcorder. Ultimately, this would need to be tested to see if leaving them both on creates weird issues, like vertical jitter. If it doesn't, then I guess you can keep on the line TBC on the SVHS player.

[@neongreenwool]

Get rid of the AI voice. Other than that, it's a very informative video.

[@harryshuman9637]

Agreed, I don't get why people use AI voice. Albeit this is not the typical robot voice.