Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

Video: The Grey Scanner For Computer (Or: A Long And Incomplete History Of Scanning) (Patreon)

Published:
2024-06-25 15:10:24
Imported:
2026-06
Tags:

Content

I absolutely did not intend for this to be an hour and forty minutes long but once I got to a certain point it just didn't make sense for it to be any shorter. It's basically two videos, but the thing doesn't make sense without the history and the history isn't meaningful without the thing, and then in order to explain the history I had to explain how three-tube video cameras work, so. Here you go. It took way, way, way too long to create and I'm sure I'll find even more errors and have to make yet another edit, but on the off chance that this is actually the final version, please enjoy.

Files

The Grey Scanner For Computer RC2

Support me on Patreon: https://www.patreon.com/cathoderaydude Tip me: https://ko-fi.com/cathoderaydude

Comments

Sauwercraud

You kept busy during your break?

Mees Boe

Can’t fucking wait to watch this👏

[syntax]

sounds like a movie night tonight!

Triple A He They

I'm gonna watch this instead of getting my work done lmao

Kerne

lol, I'm also often stumbling across some obscure bit of early PC minutiae... like how early CD-ROMs needed their own interface card as well (Sony, Mitsumi, Panasonic), until IDE finally became the sole large drive standard. I was reminded of this after recently acquiring a somewhat uncommon Toptek triple CD-ROM interface ISA card in a Tandy 2500.

Nick R

What could be better than an hour and forty minutes of information about this. I don't even know why they make videos of other lengths

Chas Becht

"I absolutely did not intend for this to be an hour and forty minutes long but once I got to a certain point it just didn't make sense for it to be any shorter." Make peace with this, for it is your life.

Chas Becht

Not just IDE, ATAPI.

Mike Lee

Holy SHIT this was a good vid. It is right up there. Don't get me wrong. I've *loved* the 'little guys' stuff too, and other forms you've explored, because those 'apparently' less impactful things are also in fact super interesting... I was going to make some comment about how when you get your teeth into some bigger narrative, you 'really bring it'. But big or small, you always bring it. I like that, in a very real sense, there isn't a distinction.

Kerne

Oh that's right...ATAPI was also part of that standardization, thanks!

Danny

You could make a two hour video about paint drying and it would still be entertaining.

Avi Drissman

Thank you for correctly calling it a “1.4 meg floppy” and not making the “1.44 meg” mistake.

Strawberry Daquiri

the 3 filter and splitter is how they did the first color plate photographs too :3

SaberTail

There is some very cutting edge research into single pixel spectrometers that would allow color images without a bayer grid or multiple sensors. You can find journal articles about them. But from what I can gather, they're nowhere close to being commercially viable yet. If they do start getting made, they'll show up in scientific and medical applications first.

cathoderaydude

Can you clarify? I'm not really sure what this means.

HotKey

OMG! ♥ ♥ ♥ I am not alone in this world with my fringe interests! ♥ ♥ ♥

Avi Drissman

Nitpicking I’m sure but. The high density DOS floppy contained 1440 KB. If you divide that by 1024, you get 1.4 MB, so it is correct to refer to it as a “1.4 meg floppy” as you did. Far too many people looked at “1440 KB” and called it a “1.44 meg floppy” which is incorrect.

Iceron Ranga

Considering the xerox 914 was already a more or less flatbed construction in 1959 i can see the term flatbed evolving in the copy machine space first and then only being applied to scanners because of their similarity in use.

Alex Cruise

Haven't watched yet; hopefully you already knew that TWAIN stands for Technology Without An Interesting Name 😁

cathoderaydude

The general consensus from what I read is that the name was picked solely because it sounded good, made uppercase so people would notice it, and then at some point somebody made this backronym which the standards body never wanted or approved of.

cathoderaydude

That's my thinking in general, it just seems weird that none of the early references say e.g. "a flatbed design, similar to a copy machine", you know what I mean?

Torbjörn Lindgren

10:54 ISA Bus Speed - IIRC the protocol requires at least 3 clock cycles per transfer (either 8-bit or 16-bits). So an 8-bit bus at 8MHz can do ~2.67MB/s max, a 16-bit bus twice that, but I don't think even a '286 can actually do that in PIO mode. My understand was that neither could the motherboard Intel 8257 DMA controllers - it's not like the floppy or hard disk back then could get anywhere near those numbers anyway. IIRC ~2MB/s is a number that I've seen bandied around (for both) from people who test various "emulated hardware". There WERE faster ISA cards like the AHA-154x bus-master SCSI controller which used a 16-bit ISA slot, it had it's own DMA controller on the card (hence "bus master") to get maximum performance. The original 154x cards ran at a fixed speed of 5.0MB/s, but later B and up cards could be to run at 5.0, 5.7, 6.7 and 8.0 MB/s DMA speeds but the manual warns that not all machines supported this and only use it if the manufacturer said it was OK - I'm guessing these required 8, 10 & 12MHz ISA bus though this isn't stated anywhere. Given that Adaptec had to add dedicated hardware even for the original 5.0MB/s I'm going to assume this confirms this isn't doable via PIO or 8257.

Robert Yates

I'm certainly not an electronics expert, more of an enthusiastic amateur. But regarding your segment about the fluorescent tube power supply starting at 57:36; It looks like they are generating the supply for the fluorescent tube, from the low voltage DC supplied to the motherboard using those right most components as seen at 58:09. If you do an image search for: fluorescent tube driver battery, you will find many examples of circuits to drive a fluorescent tube off of a battery that look basically the same as this. Two power transistors driving the transformer in a half-bridge configuration at high frequency, producing a stepped-up enough voltage to ionize the gas in the tube, and often having other provisions to give enough current to heat the tube's filaments. It seems like a very common circuit that was for a time constructed by hobbyist for a cheap and efficient low voltage light source. I even remember during the late 80's early 90's having a cheap flashlight that came out of a supermarket which used a similar circuit to power a short fluorescent tube from 4 AA batteries. As to why they decided to do this? Perhaps it was better than trying to find somewhere for a traditional magnetic ballast and starter, the components for an electronic ballast would have been about this complicated anyway, so just starting with low voltage saves the effort of running a high voltage line over in the first place. It is also possible that the main board is controlling the intensity of the lamp? If so, starting from a lower voltage that has a guaranteed common ground makes that control easier ( assuming optoisolators were uncommon when this was designed ). And it has the benefit of being inherently agnostic to input voltage, powering the tube from the mains would have necessitated changes between 120v versus 240v versions. As to why they didn't use CCFL's, I'm assuming as well that they had not become a broadly commercial product at this time. But while looking for the history of cold cathode fluorescent tubes I found a document titled "COLD CATHODE FLUORESCENT LAMPS (CCFL’s) A HISTORY AND OVERVIEW" that didn't actually give that much history, but did relate on how the CCFL tubes used in LCD's did not aim for good color rendering. Rather they were tuned to only emit light in the wavelengths that were most suitable to the LCD pixels and their associated filters. Of course it would be reasonable to assume that this would be the case for any other application where the final product is going to be filtered in some way such as scanners. I suppose that wouldn't apply to this scanner of course since it's only capable of grayscale, but I found it interesting none the less.

cathoderaydude

Fascinating! That had been bothering me, wondering exactly how they derived the HV from the DC. I wonder then what controls the relay; I believe I noticed a smaller TO92 component below those, perhaps that provides enough current to trigger the coil.

Robert Yates

I had spied that little transistor left of the relay but wasn't really sure what it was doing, it's definitely plausible that it is for driving the relay coil. Most of the schematics I've seen for this style of fluorescent inverter seemed to be setup to cause the driving transistors to self oscillate by feeding their outlets into their compatriots input. This generally works well enough as the transistors will be slightly different from each other due to manufacturing variability, and so one will will tend to turn on faster during initial power on thus suppressing the other and beginning the oscillation. I had thought perhaps that extra transistor could be a modification to guarantee good and stable oscillation since I'm sure they would like this to be a bit more reliable than a cheap flashlight ( at least one would hope ). But I do better with digital rather than analog, and I can just about understand oscillator circuits if someone explains them to me slowly and carefully, and the effect wears off after a few minutes.

Bill Florio

This is great! I did a zine in the 90s and the half tone option worked for off set printing whereas dithering did not. The half tone didn't work as well as the arcane plastic hole thing you referred to, but the only affordable option in nyc who did that process seemed like a pedifile serial killer type dude who found excuses to make you wait and talk while the tones developed.

Mitchell Foreacre

54:00 folding the image: Additionally the image needs the be focused down to the CCD, the fold could be adding more image runway to shorten the whole scanner package length. I still work for HP in the Print engineering/support. We just retired our last (to my knowledge) CCD scanner moving to the CIS (scan bar)

Mitchell Foreacre

58:00 AC/DC Motherboard. This is semi common, usually there are large insulation holes in the Fiberglass between the HV/LV circuits

Mat Fordon

I think they could stuff 3 image sensors into a smartphone.... they're already rotating the cameras to run a longer lens stack along the body of the phone in some of them.

Ryan Butler

Also, quickdraw actually has loads of hardware independence, and a variety of video cards were produced for mac's as early as 1986. Software from as early as 1985 scales to the size of the screen you have. So mac's also had hardware independent video, with 32bit color introduced in 1988 or so.

Ryan Butler

The radius pivot had the first widely available Mac video card, in 1986, which was also the first widely available dual raster bitmap monitor solution on any platform.