Q500, The Weirdest Optical Mouse
ฝัง
- เผยแพร่เมื่อ 29 ส.ค. 2024
- I stumbled on this mouse a while back and have been trying to figure out exactly how it works for a while. I think I worked it out, so here are my guesses, plus some elaboration about how conventional mice work
Support me on Patreon: / cathoderaydude
Tip me: ko-fi.com/cath...
When you opened that mouse up and I saw those light pipes, I thought “that damn thing is a ball mouse emulator”
I immediately read “that damn thing is a ball mouse emulator” in Hank Hill's voice
yeah the parts add up, it all makes sense, I feel like I should have picked up on it a lot faster than I did hahaha
@@Fuzy2K “No, you are not tripping. That is an emu.” - Hank Hill
I thought exactly the same thing!
how didnt i noticed it!. Its essentially the same but cheaper! reading which axis and which direction it moves in to make movement.
A friend of mine once stole all the trackballs from our jr. high's computer lab. Later that day we chucked them all at kids on the playground. People were still finding them years later in random places. Good times.
And that's why some mice had them under a screw-locked holder plate! ;-)
Kenny Lauderdale, your story sounds like it came from a sukeban anime!
Hah, when I did tech support in high school for a couple years we were instructed to use nail varnish to glue the access hatch closed forever so that balls wouldn't get stolen. Unfortunately then all mice weren't serviceable and were soon total garbage, all gunked up. Then again we had a LOT of surplus mice sitting around, so the gross half-working ones would just be sent to e-waste Or maybe back then literally into the trash, but don't tell anyone... A couple computer labs got new leased computers with optical mice which was very convenient. All in all, you could also rejuvenate a clunky ball mouse, a little, by whacking it on your hand (or the desk, shhhhhh) nice and... firmly.
@@alecjahn the IT people at my old school did something similar except they used hot glue rather than nail varnish!
@@alecjahn How wasteful! Glad we're past that. Our school found a way to buy just the track balls in bulk.
I'm reminded of how trackballs are *basically* ball mice put upside down, and the modern optical trackballs conceptually just treat the ball as a globular mousepad that you move around instead of the mouse itself.
They’ve gone optical -
And you still have to bloody clean them
@@AllonKirtchik Which reminds me I haven't cleaned mine this week yet
I love my trackballs.
That’s the one thing about the Orbit that gets annoying, cleaning the damn thing out fully.
They should have eyelids that blink 👁️
I bought five of these NOS off eBay as a teenager in the late 90's. Mice would outlast the pads so I used Paint Shop Pro to make wider pads since I was using Q500's well into the widescreen era. Great vid as usual!
thank you so much for watching, and it's cool to hear from someone who actually used one of these - I can't believe I didn't think of extending the pad, I even tested scanning and printing a replacement, but it never occurred to me to change the geometry.
@@CathodeRayDude Live near an art institute so had access to wide format printers and laminators. Student I was dating liked these stupid mice, gave me the idea, and I ran with it.
See, I love this kind of thing. Modern solutions to old school problems :)
How well did that work? That’s such a great idea!
@@siliconinsect perfect for doom
The "zinc trophy' line killed me 😆
six pounds of ZAMAK! solders OK if you break off one of the handles! will let you talk shop with the hot wheels boys!
We ain't wasting no money on "gold" trophies here!
Me too, I'm just waiting for an opportunity to throw it at one of our penny pinchers at work :o)
I remember a friend of mine back in college calling those early optical mice as "female mice". I have never heard anybody else use the term and I wonder if somebody was joking with her and she didn't realize it. Like somebody told her they were female mice because they have no balls and she just assume they were being serious.
and this one in particular would be a trans female mouse because it uses a system built for balls but has none
Or perhaps she had a sense of humor.
Keeping that mousepad out of sight up to the right moment was a stroke of genius.
thank you, yeah, it was critical to drop it at the exact moment it needed to be there
hello hi, I would just like to note that after reviewing the text of the patent again, I believe I may not have correctly described how the mouse systems design works. You can see the patent number in the video if you'd like to read it yourself, but I believe it is stating that the cells are read at three levels, not simple binary, and that custom motion prediction algorithms are used instead of the simple Grey code approach that I was picturing. there may be other misinterpretations. I do think they're semantic however, overall my description of the process is pretty close to accurate I believe.
Also! I mixed up my dates, the intellimouse of 96 was a ball mouse; intellieye came out in 99, so this thing was way ahead of it's time!
Interesting.. Do i also remember those mice would sometimes move backwards if you moved them too fast as the pulsing got "out of sync"?
Interesting indeed - I also was a bit surprised by your description, because for these 4 (or even 2) bit systems to describe a direction, usually the sequence is like 1100 (and then either 0110 if moved right or 1001 if moved left) - that's how encoders work in general (eg in ball mice) and in fact I think a variant of this is also used for optical mice (much simpler than analyzing a real image, may explain why it's more reliable than your instinct expected)?
EDIT: Oh, the 2-bit encoder-style thing is exactly what this mouse uses, apparently, nice!
They may use a few buffers to remember previous bits, idk just a guess.
@@RichardDzien - That's exactly why it happened. The effect is called "stroboscopic aliasing", or more colorfully, "The Wagon-Wheel Effect". If he'd been able to put an oscilloscope on one of the LEDs to measure how many times it flashes per second, we would know its Nyquist Frequency. Then we'd know the point at which moving the mouse faster made it think it was actually going slower - until it thought it was actually going backwards. And if you kept going faster, eventually the mouse would think you started going forward again.
The reason for the sensors in the Sun mouse being located far away is not because they could not install them upside down, it is to the have more angle of separation for the light that is hitting it. It should be far tinier to be upside down on top of the holes and i'm fairly certain that would require some optical stuff, like lenses.. and it would be nightmare to adjust its focal distance just right. Installing them further away makes it cheaper and easier to build.
The three holes could be for probing, Q&A testing. They seem to line up with three solder pads.
"Mom can we have optical mouse?"
"No, we have optical mouse at home."
Optical mouse at home:
LoL 😂
Those sensors are probaby phototransistors - labelled "PT" on the board, and cheaper as they need minimal electronics to read the output as the signal from a strong source is higher than photodiodes.
Mike was here: now I know for sure that this channel is very promising
Also, a key difference between a phototransistor and a photodiode is that the latter is digital*; above a certain amount of light, it passes current, below it it doesn't. Whereas a phototransistor is an amplifier; the more light it gets, the more current passes through. This means the IC in the middle _might_ be doing analog sampling of the light, rather than strictly lit/unlit. That means it _might_ be able to tell the difference between both fibers being illuminated and just one being illuminated, allowing for might finer detection.
I suspect this is how the LMOX2 in the more recent video works, too, and possibly why it takes a bit to recover from being on the wrong pad (its internal model of the world is deeply confused by the switch back).
* Technically it's more complicated. A photodiode _can_ act as an amp, it's just far less sensitive and has a much steeper response curve.
Ben from Applied Science said in one of his videos that he designed a mouse that was entierly based on fiberoptics so it could be used inside a MRI machine
why buy crap on Ebay, post it on TH-cam?????
Need a real job??????
@@lucasrem1870 what
This is by far way more entertaining and engaging than it has any right to be. Your clever, witty, sometimes sarcastic and clear/concise/informative presentation is the selling point of your videos. You're genuinely easily one of the best retro tech youtubers right now, mark my words, your channel will explode in subscriber count sooner than later, you deserve it by way of merit.
I too cannot believe I spent half an hour watching a video about a mouse.
@@emmeryncariglino4983 It was such an epic tale! I laughed, I cried.... I had to pause it to go pee. I smiled really hard at the face-reflected-in-mousepad scene. ◡̈
Fully agreed. There's only about half a dozen channels I stop whatever tf I'm doing to watch, and this is one of them.
This reminds me a lot of TechnologyConnections, I had the pleasure to see that channel grow to what it is today, and I see the same thing happening here. The production value is very high lately. I like how CRD used the reflective mousepad to keep his face in shot for instance, I can't believe that that was not coincidence ^^
Indeed. LGR has been blerbing his main channel too much. 8-bit-guy seems to be more interested in his side projects than his channel. Technology Connections is mostly a talking head. Tech Tengens is often 40 minutes of talking over a wobbly table. This is the best channel of retro tech right now.
The engineers behind the mouse are absolute madlads, huge props to them.
I’ve been watching this channel since “The NES is a radio” video. Watching it grow and evolve has been awesome.
I've been watching since around the same time! A little earlier, the one about PBX phones with no PBX
"They actually do not help the mouse figure out where it is... Or even where it isn't."
I see what you did there :)
Whichever is greater.
The missile knows where it is at all times, because it knows where it isn't.
Standard Si photodiodes have their peak responsivity around 900 nm so IR makes sense for cost cutting since you can use lower power LEDs.
It also means any ambient light will have less of an effect since they used filtered photodiodes (which adds no cost, just uses a different plastic).
ohhhh, of course! thank you!
@@CathodeRayDude As well historically IR and red LED's led the way in luminous efficiency, so you could get a much lower power consumption on the emitter for a specific received intensity, so you could make the drive current low enough that you can drive the LED's from the microcontroller output pins directly, instead of needing to have external NPN transistors to drive them. Saves having to find space to place and route 2 1c transistors, and also space to put 2 0.1c resistors on the board. Thus both a smaller cheaper board and a 2.2c saving on the costs.
Having the light output being a close match to silicon diode response also helps, you just have to have a packaging that is transparent to IR and the red light, so the receiver has less problems with strong ambient light sources.
I would guess if you replaced the microcontroller on the board with the identical version, that was introduced for PS/2 operation, and put the power supply directly into the microcontroller, instead of the resistor zener diode supply they used, it would probably work. The single transistor is there as an interface to the RS232 Txd line, converting the 5V output of the micro to a RS232 voltage level, using the phantom power stolen off the other control lines. I still have one ball mouse, this one is different, in being an IR remote type, with a small purple plastic puck you placed on the desk to detect the light, and the mouse runs off 2 AA cells, with the computer side being PS/2. Another is wired PS/2, and the last one is a serial mouse, which I used for many years on an old PC, deemed too slow to run Win98, but which was very snappy running Redhat.
Is the filtering that blue plastic, that some IR LEDs also have? Particularly sometimes in TV remote controls.
@@greenaum Yup, that plastic is typically transparent in the ~800-1000 nm wavelength range and does a reasonably good job blocking visible light.
A 27 minute video about an out of date computer mouse? Yes please. I was genuinely excited to see this pop up today 🙌
16:59 this is exactly how a rotary encoder for a mouse scroll works! the binary output is 00 01 11 10 00 for clockwise and 00 10 11 01 00 for anti-clockwise. i found this out after thinking a bit on how to use one of those rotary encoders for a steering wheel controller
Opening up that weird mouse was the second best reveal I've ever seen on youtube. It would be first except for the four SD card memory video camera shot you did.
The holes in the bottom could be used for some sort of factory test/programming if they line up with pads on the PCB.
I was just getting ready to write the same comment :-)
Same though as well.
I though the same. That means it was even cheaper .LOL.
No programming is needed here. And for testing they would need at least 4 contact points. Not to mention the holes do not line up with any test points on the PCB.
@@111chicane Also that PCB is so janky there's no way it's got test pads on the underside.
Funny story about rotary encoder's and a trackball I bought my mom in the late 90's. She kept complaining that the trackball would just stop and start working, but she lived two states over so I couldn't figure out what was going on...until I visited. Sure enough, I was on her computer and the trackball just quit working. I traced it down to her desk lamp. It was so bright that enough light leaked through the plastic to flood the optical sensor with light and it couldn't tell the shutter wheel was spinning.
This design is unique enough that it likely was patented somewhere. The Microsoft Mouse RS-232 protocol only ran at 1200 baud, so time sharing the sensor at 2x that rate for readings was well within the capabilities of the tech of the time.
Love the face reflection off the aluminium pad while you're explaining things. Subtle yet quality!
So, are you telling me I could have made my own optical mouse at home using my ball mouse?! The levels of frustration I went through until I could afford an optical mouse!
Hell, replace the two infrared leds with a red one and a blue one, then print a grid of red and blue lines as big as you want.
A neat idea now, but a blue LED in the mid 90s would have cost more than the whole mouse.
@@jimmy21584 So use a green LED, the only other color available in the 1970s. I believe I have used such a mouse a long time ago on someone else's computer.
I was thinking this. But it's so obvious that they must have considered it
My favorite jokes that you make are the ones you cut off mid word.. “I’m sorry, I promise, that’s the last su-“ 😂😂😂
Betting that the three holes are for test points to test the electronics once assembled
They line up with three solder points, so I guess the same.
my first thought after opening was, that they where to lazy to shorten the leads of the parts, so the made a cut out in the housing to avoid trimming leads.
Could also be a previous design iteration where it was easier to amend the mold with new cutouts than to fill in the now-unused features.
The squarishness of your circles might be due to the "Enhance pointer precision" option in Windows, which in spite of its name makes the pointer less precise by squaring off movements that are almost-but-not-quite vertical or horizontal. I hate it. It's especially annoying for CAD or graphics.
Strictly speaking, that _is_ precision. It’s just less _accurate_.
You can make a clock 100% precise by taking out the battery, at the obvious expense of its accuracy.
@@tomgidden I'm pretty sure that's not what anyone reading the description expects it to do.
@@djcsdy2 Sounds about right. Windows never does what I expect it to do anyway.
Well, apart from make me more at peace with Apple's pricing model.
@@tomgidden oh don't get me started on Apple's treatment of the mouse. Moving the cursor in macOS feels like wading through custard, and there's not even a misleadingly named option you can turn off to fix it.
mmmm custard
Damn. I remember those old Sun mice - my dad had one and although we’ve updated at the factory, we still have the mouse pads on our ion implanters!
oh that's adorable!
What do you implant
@@AgentOffice I work in a semiconductor fab; we implant ions into the wafers to make the transistors in chips!
@@Jessica.Amelia wow
Funny, I have a coaster on ion implanter.
you have no idea how much i needed a new CRD today (I saw the pre-screen version pop up in Discord at like 5 AM but I fell asleep and dreamed of cleaning out the blades of a lawn mower).
also 0:14 I'm glad I didn't get around to suggesting you put that obnoxious tiktok song about "the boy" and how he is coming and later is here, because omg.
i remember early SGI machines super computers and Sun machines using these kind of optical mice with the special pad late 80's early 90's
Yep, they were standard on Sun workstations at least as far back as 1984.
Boeing had a buttload of I think SGI (maybe Sun) machines in like 1991 or 1992 that used those stupid 3 button optical mice. You could have fun with someone by rotating the mouse pad 90 degrees, it would swap vertical and horizontal movements on the mouse. Everyone hated them, so they replaced them will I think IBM standard ball mice at the time. They had a surplus warehouse to sell, surplused stuff. We went alot and once we came upon a giant box, 6ft x 6ft x 6ft, full of those mice. And a metal skid (metal carrier box, 5ft or 6ft on a side - 2 1/2 foot tall with fork slots to pick it up, for heavy crap), full of the mouse pads. I think the mice were a buck each and the mouse pads were per pound (whatever AL was going for then, pretty cheap). Thousands and thousands of them basically thrown away. That was the Boeing way then.
Yeah, had these at college. You could use your leg if the mousepad was missing, if you were wearing slightly faded jeans.
The aluminum mouse pads were horrible, not only it was uncomfortable but there were often dents and scratches or the pad was slightly bent.
Oh yeah and I believe Sun had enough complaints at some point because I think they started to ship just "regular" mouses even for the computer models that originally came with these mouses. It also took a bit longer for them to join the more modern optical mouse gang.
Man I love the well researched old weird technology videos you make.
“…Let’s make a short story long”
That’s why I am here.
The mouse knows where it is at all times.
Before you took of the case I thought they had put fiber optics and connected it to a regular ball mouse PCB, but was still shocked when they actually used fiber optics! Crazy design!
I'm almost always impressed at the lengths we go through for cheapness.
Of course, it's not as useful as cheap lights and other cheap commodities, but still, damn.
Yea the cheaper something is the higher the chances are of someone buying it. Even more so in poorer countries.
“Even though there’s only 100 lines, each one produces 4 state transitions”; you know, weirdly, this helped me get a handle on the principle behind QAM and other similar modulations which cram multiple bits into one state transition… it’s kinda sorta this but in reverse.
And instead of double-sampling the lines with the sensors separated by a little space, it gets the extra “sampling” and “space” inside that complicated mathematical structure which always made me go “hmm ok it’s just magic” when looking into QAM before.
I mean I’m sure that’s wrong in subtle ways but just seeing it in space as a real artefact let me get my head around how they can be not-always 1:1.
Also even before you got here, in fact as soon as you said “it’s not really an optical mouse at all”, I screamed at my TV “it just hooks right up to a ball mouse’s photodiodes in lieu of the spotted wheels!? :D” and then felt vindicated when you popped the top off!
Wow, that double wheel mouse. Such value! It’s even got a real middle mouse button too! FIVE BUTTONS **and** TWO wheels!!
oh crap did I doge? I wasn’t trying to, honest
or, instead of 4 state transitions, you could just measure the phase of the sinewave intensity profile this would produce, and get "infinite" resolution. Or at least more than 4.
The linear optical sensors in the MSC mouse _might_ be proprietary, but that type of sensor is used a lot in automated production lines to sense conveyor speed, product position on a conveyor, etc.
"Joystick mode" reminds me of anchored auto-scrolling through documents.
The '90s were such a weird time when it came to input devices. There was a big street sale near downtown Dallas many years ago called the First Saturday Sale, and I remember buying all sorts of weird mice, trackpads, and "pen-mice" for pocket change. All the bizarre, unwanted technology you could ever want, laid out on card tables in a parking lot. That's where I bought my BeOS package (that would only freeze my Mac even though it was supposed to be compatible), lots of Amiga mice, Atari Portfolio accessories, laserdiscs, Bandai Pippin keyboards and modems, and so many trash Mac CD-ROM games made with Adobe Shockwave. Sigh. Those were the days.
It's important to remove the cat before testing a mouse.
I have been wanting to work that Metalocalypse NOTHING clip into a video for years now. You have bested me at my own game.
I basically became a youtuber so that I could start doing something with these things that live in my head for endless years.
There's a possibility (remote) that those 3 holes could be some sort of manufacturing artifact like a test point or jig connection location. Excellent analysis!
Pretty remote, normally you'd do that on the bare PCB before you put it in the case. For one thing it saves having to throw away a good case for a faulty PCB, or else pay someone to de-case them again.
maybe they put them there to make it look more like a transitional optical mouse so that people wouldn't think they were scammed?
I bought one of those about two years ago. I was trying to revive a “vintage” computer without a working ps/2 mouse port (for nostalgia’s sake) and these things are all I could find for “serial mouse” at the time. I’m probably that one guy who actually did want a new ball mouse, for authenticity or something, but I settled for this. I assumed at the time they were some sort of industrial design, meant as replacements for that ancient pc running a production line somewhere that nobody’s bothered to update in 30 years. The lack of moving parts might be convenient in a dirty shop. Didn’t know the design dated back to the 90s. Thanks for the video.
I remember the sun mice on grid pad. They were bad as well when you were tilted off x/y axis.
the reflection in the aluminium mousepad, the fancy title screens motif and your enthusiasm about this video. amazing!!
I was one of the rare people who preferred ball mice - I got very used to using the inertia of the ball to flick the cursor around (which combined with ‘sloppy focus’ on Linux), and this was also integral to my Quake strategy. I’m still not fond of optical mice, and generally use trackballs whenever I can.
You are so correct.
I bought a couple of these close to 10 years ago off the bay. Up until right now I thought this was a cutting-edge design, the first optical mouse design ever, but hey, I was a little kid when these came out.. I didn't know they made these more sophisticated in the 80's even. Great video. I really didn't expect it to be this in-depth. You really know your stuff!
I’ve never seen someone try to explain an optical mouse without saying “optical flow” and instead calling it a weird trick that shouldn’t work. Quite interesting
it's not really a weird trick at all. Just convolution.
Between the hypnotic Belly Shot™ cam, and the humor, and way you explain things(4 sensors in one trench coat lmao), I could not stop watching this. I NEED MOAR!!!!
Speaking of optical mice: I'd genuinely love to know why optical mice are coming back now after we finally started ditching them for laser mice back in the late 00's. Back in the late 00's and early 10's, all the high resolution gaming mice (like the legendary Logitech G5) were moving away from inaccurate, low resolution, finicky optical sensors and becoming high resolution, incredibly accurate laser sensors that would work on any surface (even glass). Now recently it's near impossible to find a laser mouse, and instead the market is once again flooded with outdated optical mice. Why did we move backwards to an inaccurate technology we'd left behind?
There is a lack of chips for new mice, same for graphics cards, cars, ps5's. The factories can't keep up, also one huge factory burned down, it takes months to create chips.
@@roberts3423 It's just sad that the chip shortage extends so far to even include parts for mice. Yea, if a factory was destroyed as well, then that would have a flow-on effect for at least half a year (if not longer)
People are used to mice being cheap.
I came to the same conclusion about the ball mouse parallel. I actually expected you to open it up and there be optical fibres connected to the interrupter, simulating the rotary encoder. It's pretty clever
Someone did his homework. Nice vid. Personally i love optic mice since the day they came, though older models seem have lower resolution, i collect them nowadays because of the shape - all modern mice too small for my hand (and don't get me started on gamer mice with too many buttons that accidentaly hit objects, like keyboards, while operatin. i love the old fashioned button-on-top-design), so i'm actually considering retrofitting an old MS model with modern opticts. That all said, in my box i have a '3D' mouse (as in - works hovering in the air, capturing motion). Has PS/2 compatabilty. It sort of worked, but in practice lacked precision. Not (only) because of the mouse tended to drift into one of the corners on the screen while being idle, but more because it missed the tactile feedback. Quite sure it predates nintendo controllers with 3D sensors. But this vid, with this hack, made me curious again. Gonna dig. Will ship it to you if you are willing to ship it back. Regards.
I totally had some weird optical mouse with the special pad in the mid 90s. I can't remember if I bought it at a hamfest or a DAK catalogue. It might have even came in a package with GEM Desktop and Ventura Publisher.
I wouldn’t call it a master piece of cheapness. I consider it the highest form of ingenuity and thinking outside the box literally. That only a true master of creative thinking could accomplish. And I think a lot of companies would benefit from this kind of engineering.
Damn I love this channel. Just candy for the eyes and ears for a computer engineer when you deep dive into designs like this. Thanks.
"Why does TH-cam keep recommending me this video about a weird mouse?"
"Oh hey someone I'm subscribed to uploaded a video. I should watch it."
"Oh..." lol
This was pretty interesting. Not exploring weird mice too much myself, I was using cheapy OEM ball mice until like 2010, a cheap wireless optical mouse for a bit, and then been on laser gaming mice ever since. While I like the idea of a sniper button, I constantly forget to make any use of it. I do manually turn DPI up and down sometimes midgame though, like cranking it up for tanks in Battlefield, or down for weird old games that freak out at high DPI.
Yeah I don’t like the press and hold sniper buttons, but I do have a shortcut to toggle between, IIRC, 400 and 1000dpi or something like that. I was tweaking the sliders a lot by feel for the settings so I don’t have exact figures but yeah
@@kaitlyn__L ooh, that's a thought, I could maybe set a macro key to toggling a sniper mode on/off 🤔
I have been using a "gaming mouse" for CAD work for years. I never thought of the gaming implications for it, but it has a button on the top for toggling between three different DPI modes. I almost always have it on medium, occasionally slow, and fast only if I'm doing data entry. Its handy.
I’ve been using Macs since the early 90s and there were some interesting after market mice and trackballs by Kensington for the Mac but we didn’t think about it too hard. When Windows 3.x came out is when the hardware manufacturers lost their shit. Our CompUSA had a whole aisle devoted to mice. Sounds crazy now, but most PC users didn’t have to deal with mice in DOS. When Windows dropped it all changed.
Ah yeah. I remember changing mice frequently in the late nineties/early 00s because they wore out quickly. Before Windows 95, the only use for mice was in drawing programs and some games.
Yeah! I thought it was using a stock standard ball mouse controller IC and components, too!! The way you explained how the four fibre optic cables were getting X and Y movement made me think about how the encoder wheels in a ball mouse works.
Very cool mouse, I'm so glad you documented it. Could have disappeared into history and never ever been seen again!
two of them
I hope your cat is behaving better than mine is today!
The smallest mouse pad I've ever seen was clearly a piece cut from a regular mouse pad. It was at a point-of-sale terminal, in the small area on top of the unit next to the display pedestal. It was, basically, the same size as the mouse. There was no room to push the mouse in any direction. I thought it was the silliest thing!
The ball mouse controller chip theory is interesting, I wonder if it would actually work if you temporarily wired the photocell pins of the chip to a regular ball mouse.
Or put that controller chip in a regular ball mouse
You make such amazing videos with such high production value. I cannot for the life of me understand why you have not blown up yet!! Its gotta be soon!!
HAHAHAHA "is there an adult here" my sides
I remember using on of these on an SGI machine as a kid. My Dad brought it home from work for a few weeks and whenever he let me I'd marvel at the Button Fly interface and playing with a virtual lathe. Compared the the family Amstrad CPC it was like magic.
Generic ODMs Zany Double Sensor Mouse!
Also would like to add that I wouldn't want to imagine a world without zinc!
COME BACK, ZINC
2:50 Fun fact, the Mars Helicopter doesn't have GPS to tell it where it is or how far and in what direction it is moving so it uses a more complex version of this method that takes altitude over the surface into account as part of it's navigation system along with some angle sensors and accelerometers.
the 3 holes look like they like up with the three metal pads, my bet is they are used for testing
Fantastic! This turned out way more entertaining than I thought it would be. Excellent job, my man.
The Boy
The Boy
The Boy
Uncanny!
I used a mouse just like that mouse systems one during my work experience, back in the 90s.
It was attached to a Silicon Graphics workstation, which was running a cad program called DOGS.
Same 3 button design, same mouse mat, and I remember thinking at the time, this has to be better than using a rubber ball with 3 rollers.
The design especially the 3 buttons which was super unusual at the time, reminded me a lot of the Archimedes mouse design which I was very familiar with as we had those at school.
Fascinating video! Thank you for making it.
The mouse knows where it is at all times. It knows this because it knows where it isn't. By subtracting where it is from where it isn't, or where it isn't from where it is (whichever is greater), it obtains a difference, or deviation. The guidance subsystem uses deviations to generate corrective commands to drive the mouse from a position where it is to a position where it isn't, and arriving at a position where it wasn't, it now is. Consequently, the position where it is, is now the position that it wasn't, and it follows that the position that it was, is now the position that it isn't.
I was looking for this as soon as he said it
I've seen these for sale on eBay for many years, and almost bought one at one point, but what turned me off was the design of the pad. I have been familiar with the Mouse Systems design for a long time, so I always figured this used similar technology, but I thought there was no way it would be any good with that pad design. The lines just looked so thick, so I didn't think it could be precise, and I thought that going too far up or down was likely going to be a problem. Awesome to finally see a review of one, and now I'm wondering if I should pick one up.
With this mouse, it seems as though you could control the input by changing the layout of the lines. You could print a linear "mousepad" with lines of calculated thicknesses where dragging the mouse across at a constant speed does a set of specific mouse movements.
hahaha, you're right. an optical cam track.
Hmm. I wonder if you could make a track that lets you draw perfect circles and arcs.
Thanks for the explanation of the reflective mouse mat. I too saw the Sun version of this mouse when I went into university as a kid (my dad worked there). I always wondered why there were two holes on the bottom of the mouse when only one was illuminated - of course, I couldn't see the infrared light!
You may not have intended this, but you've actually convinced me to consider this for my super retro PC build. Since i can't find the actual retro style optical mouse i want in stock.
Like, this thing is perfect for what I was building. It's old, (in design), it's silly, it's db9, and the mechanics are beautiful.
And i can just print out the "mouse pad".
I am hoping I am not the only one who, during this entire video, sat and moved their mouse pointer around and was amazed like a lummox.
It was when you said "no specialised components" that it truly dawned on me. Brb, gonna go get some fiber optic cable and retrofit all my old ball mice to optical.
The publications department in my UK company used desk top publishing with Xerox computers. These were pioneer users of mice - and they were optical. I don't know how they worked but the mouse mat carried a grid. Unlike the one in the video the grid was simple printed lines or dots on paper, so fine that you couldn't see it without a very close look - there was a file on the computer that allowed you to print a new mat whenever you liked.
I remember a mouse like that in the mid-90s. The local library had one of those attached to one of the Macs in the kids' section. No idea why it was there. I now assume the old mouse had the ball stolen, and this is what they replaced it with. Never saw another optical mouse for years, until high-school.
Came here to watch from Patreon, gotta support my favorite creators every way I can! Thank you for this. I look forward to the weekends and the bit higher possibility of CRD drop on the horizon.
You started talking about the lines and how they appeared and that brought me back to when I had a headlamp that could change between white, red, and blue light, and using that headlamp, I'd shine the different colors onto my comic books to obscure the different color inks on the page to see how it would change the art.
>Nobody uses an old ball mouse on purpose"
Me and the Compaq ball mouse next to me, simultaneously:
"I took that personally"
Speaking of the modern design, The 8-bit Guy did a video where he got an earlier iteration of it where the USB interface IC and the optical sensor IC were separate parts and hacked an interface direct to the sensor IC's output so you can see what it's seeing. Great stuff.
Loved the infographics on this one. I think part of why the dots work is, when moving in the off-axis, 90% of the time you're going to see either no illumination change or you'll see both sensors change illumination at the same time. The microcontroller could be programmed to ignore these.
yeah, I can't quite figure it out, because this would feel really jittery if so. I have this picture in my head of how they use the offset of the columns to make this not happen, but I can't really illustrate it, and even so I can't really figure out how that would fix all the problems either
@@CathodeRayDude I have a feeling that Clint's mouse may not have many more sensors than yours- I think those fibre optics are pooled internally- 4 rows of two- if you space the two rows a half grid pitch apart, then while one is over white, the other is over black.
If you consider the sum of the illumination from one pair one pair from each row it would be functionally blind to the change when movong in the wrong direction.
Given the cheapness, I'm almost sure that's what they've done, and I see no other reason to have two rows
That just leaves 4 columns instead of two...and if my previous point holds that's confusing because they're seemingly equally spaced as the rows, meaning the same thing applies to them, 1,3 and 2,4 should light up at the same time, so they're probably pooled too, why I don't know, perhaps to improve resilience to rotations somehow?
@@CathodeRayDude remember the grid of the other pad? White or bright with a dark or black grid. Effectively it is white dots or rather squares. Well your black dots are actually a black pad with white grid lines. In other words, the grid contrast is inverted. That also means it would be cheaper to produce, uses less ink. It works the exact same way, just inverted the contrast.
My father had a mouse like this, back in the 80's.
It only works when the metal plate is orientated one way, not the other.
His mouse version had three buttons and it was for his Atari 800 xl computer.
I never thought I'd see another mouse like this again.
You're correct on it being the same as rotary encoder, this would likely be called "a multiplexed dual linear quadrature encoder with external code stripes". The encoders in ball mice aren't exactly photo interrupter, its actually a pair of them in a single device. The offset between the two sensors creates a stepped rising and falling signal, depending on which leads or lags determines the direction.
I've heard quadrature encoding also called "gray code", but think that was more of an old generic term for many different forms of encoding which it also falls under.
Dedicated counting ICs for quadrature encoding used to be fairly common, but now its all done in software on microcontrollers or ASICs like here. Similar, these optical encoders for both linear and rotary used to be commonly available from suppliers along with code wheels/strips, now they're rather specialty devices and just getting the sensors themselves is kind of difficult. Has made some projects a bit challenging.
Ink jet printers used to have similar linear quadrature encoders, though with much high DPI, for tracking the position of the print head. Modern ones just seem to use cheaper tricks now.
Pretty new to your channel but as a big fan of stuff like LGR and Techmoan you have joined my retro tech lineup for those nights where I just want to vibe! Thanks so much for your vids!
I've come here after binge-watching ALL of your newer content, looks like I opened a wormhole and ended up at This Old Tony but for old tech! xD
The optical mouse design is so successful, i recently bought a trackball and when i took out the ball i saw an optical mouse sensor thats just pointed at the ball from below
I have only seen the design where a metal plate is used in the early-mid 1990s. I was in high school, and one of the teachers, or teacher's aide had one.
Ball mice are so high maintenance, so this was interesting. Considering even ball mice needed a mouse pad, whether it was metal or fabric, it didn't matter.
Something else unusual, were the ball mice cleaner products. A Velcro ball and mouse-pad. The ball was wrapped in the hooks, and the pad was made of the loops. A special cleaner was applied to the ball, and the mouse was moved around on the pad. The hooks would scrub the rollers.
An alternative was always a carefully applied pocket knife.
i feel like the reason few consumers had the three button sun mice is because they were probably marketed towards practical applications. i imagine you could probably go to a machine shop somewhere where there's still a cnc mill controlled by a computer cart that has one of the reflective mouse pads stuck down on it with double sided tape.
I really enjoyed the more humorous editing and presentation this time around. Great work 👍
Awesome. I just received the same mouse as I needed a serial device - so stoked that we have the cath deep dive on it! Wait, that sounds....
this was way more interesting than I expected. I'm glad I stuck around for the reveal! I wonder if you are wrong about how the decoding works, though. You suggest that it's basically a binary threshold and the hardware just counts black/white transitions. That would be very low resolution. Instead, if they measured the relative intensity of the light reflected into the two fiber optic strands you would get a sinewave signal out of a square wave source, and could actually pick up the phase of the sensor's position within a single pair of white and black lines, since the intensity of the light reflected back into the strand would depend on how close to the center of the line that strand was. The test is simple: does the mouse move in a jerky stairstep way when slowly pushed very slowly over the pad?
This was the only mouse/pad I was given to use when I was doing CAD work for a startup.
I was 'using up' a pad every three weeks or so!
I was wiping the pad down with acetone or ethanol...had to keep it clean!
The faint stripes are what the mouse 'sees'; I though they were decorative.
Grandfather had a similar mouse called a Mouse Systems Optical Mouse in 1989. Using the same internal LED and sensors found in double sided disk drives of that era, reflective light captured resulted in bit translations.
About the MouseSystems style optical mouse, the pads were available in a reflective plastic as well - I have one - but the aluminum pad was certainly nicer.
The odd trick is that you can actually just print a fine grid on a sheet of paper and tape it to your desk. This was a common hack when university surplus workstations filtered out to the community sometimes without their special and expensive pads.
WOW, I didn't know anything about how any type of mouse worked, but this was wonderfully explained and I'm now fascinated by this weird little mouse!