Tech archeology: The frustrating experience of trying to identify mystery cards
ฝัง
- เผยแพร่เมื่อ 30 พ.ค. 2024
- In part 1, we explored a box filled with PC parts to discover what cool treasures lurked within. In this video, let's try to test out some of the card, both the normal and the mysterious ones. What can I actually get working?
Part 1: • Tech archeology: Explo...
Part 2: This part!
00:00 Into
01:08 Cleaning up dirty parts
02:20 Drying the freshly washed parts
04:11 Testing Sound Blaster cards
06:55 Results of testing the "normal" cards
11:30 Flexus 486F55 motherboard
23:27 Kingston 133Mhz 486 upgrade
28:12 Heath 150-307-3 video card
43:16 Compaq Video Controller GA2
46:14 Orchid Turbo Graphics Controller
57:45 Vectrix EX1280
1:04:01 Periscope IV Rev 2
1:06:13 MQX-16 by Music Quest
1:07:06 TOPS AppleTalk card by Centram Systems
1:07:47 Scan Cap by Conversion Technologies
1:08:53 BlackFoot by Arroyo Technologies
1:10:49 So much of this is lost to time
1:15:10 Outro
-- Info
MQX-16 Music Quest Inc.
Scan Cap by Conversion Technology Copyright 1991
BlackFoot by Arroyo Technologies for the Macintosh II NuBus
Patron stynx found this about the Orchid Turbo Graphics card: "Orchid Technology. The Turbo Graphics Controller (TGC), which is functionally compatible with the IBM Professional Graphics Controller, runs 4 to 25 times faster than IBM’s product. The TGC offers 640-by-480 pixel display resolution and can run on a less expensive display than IBM’s. It writes directly to screen memory (bit-map), and has RAM designed specially for video applications. Approximately $2,000."
Fond in PC_Tech_Journal_vol03_n08 page 30
Also from Patron stynx: Arroyo Technologies "Blackfoot JPEG": Macworld May 1992 page 211
Claims to allow 1MB/s JPEG decompression.
-- Links
Patron stynx found this about that Vectrix video card:
bitsavers.org/pdf/vectrix/isa/ex1280/
The card was featured in SMMC 007:
• 0007 A very unprofessi...
Nick Griebel found that a Computer chronicles episode makes a reference to TOP technology, and kind of describes how you can transfer IBM files to an Apple via Appletalk and TOPS. • The Computer Chronicle...
Freetech/Flexus 486F55
theretroweb.com/motherboards/...
Link to dumped ROMs for Vectrix EX1280
Link to dump ROM from BlackFoot card
Heath EGA 150-307-3 Rev 1.33 and P/N 150-307-2 Rev 1.32
www.vogons.org/viewtopic.php?...
AppleTalk TOPS card:
archive.retro.co.za/mirrors/68...
Adrian's Digital Basement Merch store:
my-store-c82bd2-2.creator-spr...
Adrian's Digital Basement ][ (Second Channel)
/ @adriansdigitalbasement2
Support the channel on Patreon:
/ adriansdigitalbasement
-- Tools
Deoxit D5:
amzn.to/2VvOKy1
store.caig.com/s.nl/it.A/id.16...
O-Ring Pick Set: (I use these to lift chips off boards)
amzn.to/3a9x54J
Elenco Electronics LP-560 Logic Probe:
amzn.to/2VrT5lW
Hakko FR301 Desoldering Iron:
amzn.to/2ye6xC0
Rigol DS1054Z Four Channel Oscilloscope:
www.rigolna.com/products/digi...
Head Worn Magnifying Goggles / Dual Lens Flip-In Head Magnifier:
amzn.to/3adRbuy
TL866II Plus Chip Tester and EPROM programmer: (The MiniPro)
amzn.to/2wG4tlP
www.aliexpress.com/item/33000...
TS100 Soldering Iron:
amzn.to/2K36dJ5
www.ebay.com/itm/TS100-65W-MI...
EEVBlog 121GW Multimeter:
www.eevblog.com/product/121gw/
DSLogic Basic Logic Analyzer:
amzn.to/2RDSDQw
www.ebay.com/itm/USB-Logic-DS...
Magnetic Screw Holder:
amzn.to/3b8LOhG
www.harborfreight.com/4-inch-...
Universal ZIP sockets: (clones, used on my ZIF-64 test machine)
www.ebay.com/itm/14-16-18-20-...
RetroTink 2X Upconverter: (to hook up something like a C64 to HDMI)
www.retrotink.com/
Plato (Clone) Side Cutters: (order five)
www.ebay.com/itm/1-2-5-10PCS-...
Heat Sinks:
www.aliexpress.com/item/32537...
Little squeezy bottles: (available elsewhere too)
amzn.to/3b8LOOI
--- Links
My GitHub repository:
github.com/misterblack1?tab=r...
Commodore Computer Club / Vancouver, WA - Portland, OR - PDX Commodore Users Group
www.commodorecomputerclub.com/
--- Instructional videos
My video on damage-free chip removal:
• How to remove chips wi...
--- Music
Intro music and other tracks by:
Nathan Divino
@itsnathandivino - วิทยาศาสตร์และเทคโนโลยี
I commented on Part 1 that many of the boards in your collection came from me, including several that I designed. Now that the holidays are over, I watched Part 2 and can answer some questions you raised: The Orchid TGC (officially called the TurboPGA) will not do anything because it needs initializing software to load the firmware for the onboard 80186 processor's RAM (there is no ROM). Once initialized, I believe it is fully hardware compatible with the IBM PGC, but my memory is not entirely clear on that. The IBM product had a CGA emulator daughter board that produced video at the same scan rates as the main graphics system. The TurboPGA also had a daughter board that you seem to be missing. It was a standard EGA video system, and relied on using a multisync monitor to display either "Pro Graphics" or EGA. Both EGA chips and multisync monitors were fairly new at the time. IBM used a proprietary, single rate monitor. The TurboPGA was considerably faster than the IBM board, and cost about half as much. The "strange" memory chips you see are "Video RAM". Not the generic term for memory used as a frame buffer, but dual port DRAM with a standard random access port for the processor and a serial output port for display refresh. This improved performance by providing nearly 100% bandwidth to the processor. I designed the hardware and most of the firmware under contract to Orchid. There was a paint program available for this board.
As you guessed, ScanCap is a specialized video capture card. It has only a "line buffer", and could not capture a whole frame at once. Thus it was only useful for capturing still images. However, it could capture a wide variety of video signals, including some relatively high resolution (for the time period). It was designed for capturing images from various medical imaging systems so they could be transmitted to a remote doctor. It had multiple optional oscillators (left edge) so that a single board could work with various systems, but only 2 or 3 dozen of these were ever built, and most had just 1 or 2 oscillators. I see that there was an optional 2nd A to D converter, but don't recall why. Perhaps this allowed capturing higher resolution images by alternating between the 2 ADCs. I designed this for a client who was a radiologist. He wanted to be able to study medical images from home, and sold a system based on this board to hospitals all over the U.S.
Blackfoot was indeed a JPEG accelerator for NuBus (Mac) computers. The motivation for it, as you guessed, was that CPUs of that era could not process JPEG fast enough for real-time video, and were painfully slow at compressing/decompressing large JPEG files. Even when software-based JPEG became practical, they generally took shortcuts to improve the speed, at the expense of image quality. Blackfoot, on the other hand, produced about the highest quality JPEG compression/decompression available at the time. I designed the hardware for this, while 2 partners did the firmware and Mac software.
You had one or more boards from Portacom, and I recall that I did some work for them, but don't specifically recognize it. The Vectrix board was, if I remember correctly, given to my by a friend who worked for Vectrix, but I don't really know anything about it.
Do you still have disks flying around from that era? :)
I was an AT in the Navy in the late 80s and they really drilled corrosion control. Water is highly corrosive, so they taught us to use isopropyl alcohol of at least 91% for cleaning. I was a PC tech in the mid 90s and we would get one-off boards all the time. If we couldn't find documentation, we had to hope we could find a known good to compare with. This was pre-internet days, so you had to go with what manuals you had on hand. There was common circuitry on most cards like the circuit for the ISA interface, but sometimes you had to guess the signals based on board and components. Also 2 things I HATE to deal with (and lucky me that's what I mostly worked on in the Navy) - audio and RF. I've had RF problems only show up if you happen to be touching the gear, for example, and audio is always a pain in the ass. :)
I have worked in a small company manufacturing custom PCBs.
Your method of cleaning is exactly what we did right after manufacturing PCBs to make them look neat. Just soap and water, then rinse, rinse with distilled water and then put them into an oven (70°C) to dry them out.
On a large scale, we used a glorified converted dishwasher which was really expensive and basically did the very same job. I’d bet a modern dish washer using a (low temp) ECO program would work just as well.
All you need to do is to cover anything which has a vent hole. Dip switches, beepers and the like usually come with a protective sticker you remove after manufacturing the PCB.
If soap and water isn’t enough, especially when removing residue from leaked batteries and capacitors, you can use PCC, a “Printed Circuitboard Cleaner”. In Germany, you can source it from “Kontakt Chemie” but I am sure that there is something similar in the US. It also works with TFT where ordinary TFT cleaner fails. I have deep cleaned the TFT of my milling machine numerous times without any damage. I can’t be sure if it is safe to use on any LCD/TFT but when everything else fails, why not?
To quick dry, you can use pure isopropanol alcohol. It “dilutes” the water and evaporates without residue. You can buy it in bottles and also as spray cans. For home usage, I prefer the spray cans since you can blast through the cracks between chips and PCBs.
For drying, you can also use compressed air from an ordinary compressor. The extra dry air really helps. Make sure there is no oil in the compressed air. There is no ESD issue using compressed air, all the static electricity is blown away safely.
A surprisingly good DOS program for testing obscure, proprietary video modes supported by extremely few cards is actually the fractal rendering program FRACTINT. You may want to consider adding that to your arsenal of testing software if you don't have it already! ;)
I dimly remember that one! It had modes for cards I’d never even heard of.
Wow fractint is a blast from the past indeed! 😎👍👍🤯🤯
Heh yes, if a cart made it into the wild, fractint supports it including some really obscure modes.
Absolutely, this program runs a lot of extremely obscure video modes, even people that worked in the industry haven't heard of. I first heard about this program from a person that worked for the company Matrox in Canada.
It supported Paradise VGA and other modes that most software didn't. It's a pity that FRACTINT has been forgotten entirely. It was a huge hit back in the day. Real fractals! On your PC! Rendered at lightning speed with integer math! (hence the name) You could zoom in as much as you wanted, and sure enough those were real fractals, infinite. It even supported color cycling for your own hallucinogenic light show. But today? It's gone and nobody remembers it, even on these retro channels.
According to Wiki, Orchid's 1st product was a LAN adapter card and "Lan" is the Vietnamese word for an orchid, hence the name of the company.
cool to know!
Huh.
兰，a chinese charater for orchid too😁
I had this orchid card in 1988, it cost a fortune - $2000 sounds right
It was installed in the fastest PC at that time (386?)
And hooked upto it was a 25-27" 'super' color monitor, requiring at least 3 men to carry - had to install an extra leg under my desk
Used it for autocad and some rudimentary 3d drawings
This sounds like my first bosses setup. He used tango PCB, and AutoCAD, and the monitor was like 5000 lbs. Lmao. It literally bent the middle of his table over time, me ended up getting a new desk lol.
@@jkramerks my desk bend over, I think we used a car jack to get it somewhat looking like a desk again and then build a custom leg
I remember Orchid gear, I was in elementary school. The high school up the street was also the central Pre-Engineering lab for the district, and it was equipped with a fancy CAD and CNC station. Not sure how big the monitor was, but it was huge and colorful compared to the 9" monochrome and 12" green/orange Apple II monitors I had access to at the time.
The CT2770 is a rather noisy card from the get-go. There's two 16v 47uF caps around the amp chip that need to be swapped for 100uF capacitors, and ideally the output filter caps need to be swapped from 16v 470 to 16v 680-1000uF, whatever fits. That last part is a mod, it really helps the low end cutoff but all of this should be done to any SB16 that comes through. As the caps age, given they're using chinese manufacturers such as Wincap and G-Luxon and even Jamicon which they were using until recently, they're all going to go bad from age. Especially in the realm around the cluster to the right of the CT1745A, as that's where a bunch of stuff goes in and out of that chip, and where the hiss mostly originates. the amp just amplifies that hiss and that's what you're hearing, but replacing the two 47uF caps with 100s helps a ton as that then brings it in line with the datasheet.
I think you're doing excellent work by simply showing these cards on your channel. Who knows when someone'll see a video and go "Oh, I know that card!" and they might have a driver disk.
So yeah, keep the awareness running.
Kinda ironic that the most expensive cards back in the days are sort of worthless, while some very basic things go for a lot of money these days. 😅
It's fairly simple, those high end parts required special software, which nobody in the retro computing space wants. We want stuff that works with ALL the software, not some specialty application.
I remember back when i worked for computerland, we sold PC XT's and PC AT's with autocad, and special monitors that had 4 coaxial connectors (RGB+synch, i think). That and very expensive HP plotters. But you had to have a special monitor for those high res cards, special drivers, and they only worked with autocad, basically, because nobody else wrote software that could use that extra high resolution, because nobody owned the cards and monitors. Some of these systems sold for 10K, and that's in 1980's dollars.
So while that hardware was amazing for the time, it never got widespread adoption, and thus is relegated to the scrap heap of history.
That and after a year or two, better cards and monitors came out anyway. Back in the day it was almost always more cost effective to wait a year for the features you wanted, than to be an early adopter.
@@jeromethiel4323 Yeah, from a logical perspective it all makes sense. But that's why I said it's ironic :) ;)
Great story btw!
@@p_mouse8676 Yeah. It is also sad, because some of these products were extremely good, just too expensive for the mainstream. If only the manuals and software had been archived.
But history is littered with "we should have saved that," but it wasn't because it was not considered important at the time.
Just look at food. What did the ancient Romans eat? How was the food cooked? We really don't know. We have snippets here and there, but no real recipes.
That information is likely lost to us forever. Same as this computer history.
Those cards were for business use. I don't know anyone who does business retro computing. The IBM PC was not a games console, it was first and foremost a business machine. Most computing of the 80s was. It's why computers were such a hit and sold like hotcakes, despite the high (to retail consumers) price tags. When it's a business expense that doubles productivity, it's an easy choice. $10k for a workstation? Pocket change.
@@jeromethiel4323 You know what the ancient Romans ate a lot? Insects! No kidding! Lol.
But yeah I completely get you! :)
It's fascinating how many details get lost in time. Some of them are actually extremely important for the things we take for granted these days!
I don't miss the days of putting a new card into my PC and then not even getting to the BIOS... then putting the card in a friend's nearly identical motherboard and it works perfectly.
That MQX-16 is MPU-401 compatible, if memory serves. It used the same breakout dongle/cable as their later straight up clones. So D-sub 1,6,2 (the rightmost pins) -> Midi Out 2,3,4 and D-sub 5,9 (leftmost) -> Midi In 2,4 respectively should do you good. Pins counted solder-side right to left, top to bottom.
As my comment on the first video of Adrian's investigation of the MQX-16 noted, the 3 RCA connectors are almost definitely duplicating the Roland MPU-401 analog audio connectors, for tape synchronisation in and out, and for a metronome audio signal. The jumpers near the ISA board are almost definitely for the address base and/or the IRQ. From memory, I think 0x330 is the default I/O address for the MPU-401.
Re: washing PC parts with tap water and plain soap, I have done that myself, and even though the water in my area is considered very "hard" I have never had any ill effects, despite just letting the water evaporate naturally from the parts. I've even washed PC fans using this method, and they've all worked perfectly afterwards.
Oh and Adrian, I really like these long videos where you delve deep down into the nature and history of the things you review. I've always felt that I missed a lot on other retro channels when they just show an item for a brief moment with one or two sentences to explain what it is, but rarely what it does, or why.
In other words, please keep making these in-depth videos, TY.
Greetings from Denmark.
Me too. Motherboards, cards, fans, power supplies (out of their metal chassis, of course) .. anything not mechanical, really. :-) Works fine!
I worked in a computer store in a town near a Nuclear Reservation and heard MANY stories of how, when a MILLION DOLLAR piece of electronic equipment got "Crapped Up" a real term with radiation, how they would just take it outside and wash it with water and let it dry.
I bet it is safer from ESD issues too because everything is grounded to each other due to the water. Dry air is where static becomes a danger.
@@noname-gp6hk Yep, you're absolutely right about the dry air.
I used to work for a big mfg company, and we wash pcb's with pure (pure H20) as to not leave residues on board/ics and cause the board to fail during FVT.
The Heath 150-307 board is a Zenith Data Systems Z-449 video card from their Z-248 era systems. Probably designed to allow CGA/EGA/VGA to run on the fixed frequency Flat Tension Mask monitor ZCM-1490
It's a shame that so much information of those rare and obscure cards and motherboards are lost forever and as the title says much history behind those cards and stuff are lost to time. That's why people like you and others in the retro computing community are there to preserve information about these old hardware and software before it lost time for the new young generation for many years to come.
Like tears in rain.
If you ever get stuck only having EDO RAM but need FPM RAM, lift the OE pin of each chip from the board and wire it to the CS pin and the RAM will behave just like FPM.
Bits und bolts actually made a ram stick that has a switch to go between edo and fpm
Every standalone TIGA card I've used required a primary video card in the PC.
The #9 card I have used had on-board paradise vga chip. It wasn't very good, so a separate vga card was better.
The “8bit testing” and “16bit testing” SoundBlaster came back to me like a “Blast from the past” when I’ve heard them! I remember when I’ve first got ModPlay on my 386sx/16MHz, with its buzzer on board instead of a speaker. I never thought I’d ever hear sampled audio coming out of it! Also, after I realized that I could make an 8bit sound card with 18 resistors and a capacitor attached to the parallel port as a DAC, this became my first sound card, which I even built 20-30 more times for friends at uni! Great video Adrian!
In all my years of tinkering with electronics and watching how-to-guides, I had no idea you could just wash a circuit board in the sink.
Depends on your water quality. some sources are full of iron and other bad particles that can be flushed into contacts.
yes you can but you have to be damned sure it's fully dried out before you power it back on or put the battery back into it or it's ruined
I would still use at least de ionized water and final rinse would be with IPA or something similar and imnediate drying off with air and high temp. I dont believe that just water rinsing wouldn't hurt anything at all.
Quite literally every electronic part and board made is washed in some fashion during it's build cycle. The wafers that the microchips themselves are made from (prior to packaging) are washed many times, as they go through various etch stages. Of course, this is with very pure DI water, and rinsed with 100% pure alcohol (or other solvents depending on the stages). Used to work on a wet-etch line at ST Microelectronics, we had very fancy "spin washers/dryers" that a basket of wafers (literally 5-6 figures worth of microchips in a small "basket" measuring maybe 8x8x10 or something like that) would insert into, close the door just like a front-load washer, hit the button and "off they go " rinsing.
PCBs do the same, again with appropriate solvents to dissolve/wash off fluxes and such, then a pure water wash to get the solvents out.
As others have mentioned, the key of course is making sure the water is gone before power is ever applied (although 100% pure DI water isn't a good conductor at all actually, but just a tiny percent of contamination can change that). Technically, you could run a computer under water if it was pure enough, and some have built fully mineral-oil bathed computers in fact (easier to avoid contamination with oil as contaminants don't react as easily and they tend to just "fall out" unlike water where they dissolve into new conductive compounds). Pure water is too hard to "keep pure" and also evaporates much faster than something like mineral oil, but it could be done as a lark.
I've often told people over the years when they say "my keyboard is really dirty (or "I ruined it by spilling a soda on it") to put it in the dishwasher and run a cycle with soap, then another without as an extra rinse, then rinse it with the purest alcohol they can find (typically 99% is avaiable at drug stores if you look hard enough)... the alcohol combines with the water to rapidly evaporate it out, then let it dry really well. (Set on a windowsill, or outside on a sunny breezy day, for several hours (or longer won't hurt obviously). It may or may not work, but if you were going to trash it anyhow, it's worth a shot. (Turn off the heated drying cycle in the d-washer, and top-rack only, LOL). If you can safely remove the keycaps, even better, and if you can fully disassemble it to housing and PCB, almost certainly can get it in great shape again.
I worked in IT in many facettes for more than 20 years. I can see the masses of people who would have had a stroke or heart attack at 1:18...... Washing boards/cards/etc. I get it - it works and doesn't harm them.
Also, both your work and the work of many others who remove/replace capacitors and do other soldering work on motherboards.... That to would have absoltely killed thousands of PC repair techs back in those days.
I love your work and always enjoy it. It often surprises me at the great things you and others are able to do today that would have been forbidden in the past.
Provided you get it fully dry before turning it on it's fine. I agree with not doing that with mechanical drives. It's usually the only effective way of getting rid of dust and grime from so many years. The physical repair is astounding; now and back then it wasn't usually prudent to do that (other than the quick chip repair when you didn't need to solder/desolder). Obviously it's required now, as replacements are dwindling. Cool stuff brings back a ton of memories working with some of those cards/boards and cpus. 3com card especially.
Its perfectly fine as long they dont have any electricity in them. For motherboards you need first remove the battery and drain the caps, after that they are too safe to clean under water. If you dont remove the battery, then corrosion begins to form and the board can become water damaged.
This is why One of my hobbies was collecting the Data Sheets of EVERY board that came across my bench when I was a Tech in a 80ies-90ies Computer Store as well as making a log in a book when I repaired something as to what was wrong, how I fixed it and any info I might have got from a phone call to the manufacture/vender! We were even a Commodore Authorized Repair Center and I photo copied ALL the Repair/Service Manuals they sent us, for my own use of course! Somewhere around the house I have manuals for the Vic 20, C-64, Pet 2010(?) Commodore's Business System with a separate IEEE 5 1/4 Inch Floppy Drive. As well as the Commodore Dot Matrix Printers of the day.
The main differences between the AWE32 and 64 families is the change in ram expansion, some slight changes to the chips - the 64s are meant to have a better noise floor (ESP the gold), and in windows, you get 32 software voices to go with the 32 hardware voices (hence the AWE64 name). But the basic design of the chips/audio interface is identical between the 2 families, hence the 64s work perfectly with software for the 32s.
Too bad Windows stopped supporting hardware voices when Windows Vista came out. Now it is al done in software. CPU's are fast enough to handle that.
Hardware audio hasn't made any significant advancements since Creative killed A3D. Hate that company with a passion.@@G3ld3r3n
@@G3ld3r3n yeah it really sucks that everything is easy and better now
That EDO thing with the Pentium bus width explains why my dad told me you had to buy them in pairs. Now I get it! No one else has explained that in their videos, and I was always confused.
Back in the early days of CAD when we needed better graphics we had two cards in the system connected to two monitors. The base card was typically a cheap CGA card as it was only used to execute drivers and software to enable the more enhanced graphics card. Usually, the enhanced graphics card requires a special type of monitor. I am familiar with the Nth Engine card for example.
The other one I am familiar with was the proprietary “BaileyCAD” card that had an EGA card switched through an ultra high resolution (modern HD in 1985 on an IBM XT/ Compaq Portable) video card.
I remember Orchid making high end video cards. But we sold very few of them, because they were crazy expensive.
I had an Orchid Fahrenheit Pro VLB I bought in the 90s. I don't remember it being that expensive at that time, so they definitely made some more affordable cards. Maybe that was not true earlier on.
That mod file you're playing when testing the soundcards, Surfing on a Sine Wave, has an 8-bit AY version that I always use for testing ZX Spectrums and other AY computers. I've always thought of it as my version of "8-bit Dance Party", so that's a really nice coincidence 😊
After washing with water, you should displace it with isopropanol by submerging the cct board for several minutes and then evaporating the alcohol with the blower.
We have repaired million-dollar broadcast video switchers, that had bottled water dumped into the tub, with this method, and it works very well.
I agree, plus dries so much quicker and certain to safely run in a minute
The cache is working correctly on both systems. The 120MHz system even does L1 write-back caching, as observed by "move" line in L1 being so high. The straight line in the write test is expected on any 486 system, as they don't use write allocation. I wrote a detailed post for that on VOGONS. I will post the link in a separate comment to prvent this comment from being "spam filtered" by TH-cam.
In regards to the 486 motherboard: Back in the day, I was working Win95 support and a friend had a call just as we were going for lunch. Guy bought a new US Robotics modem and was having install issues. He was trying to install it the easy way that was suggested which was to run it in PnP mode. To do that he had to remove all jumpers and now he had a bowl full of jumpers and his system was no longer booting. It was one of those earlier motherboards with no real silkscreening and a ton if jumpers, that unless you had the install manual, you were sol on those jumper settings.
You really are a guy who enjoys what you do.
A lot of these more complex video cards were aimed purely at Autocad so didn't even bother supporting standard modes. I have a Pepper SGT somewhere that was superb (expensive) in its day but is basically junk unless you want to use a very early version of autocad :D
Some had drivers for like Lotus123 (spreadsheets) and few other popular msdos programs. And often, Flight Simulator 3 or 4.
The intro gave me goosebumps, haven't listened to chiptunes for years, really took me back than you
Just something to bear in mind. I worked for a company right out of college that made very, very high end systems for crash testing and turbine monitoring. A lot of our own cards and acquisition cards like the other ones in the box were insanely sensitive to bus timing. ISA bus timings could vary a lot once you get to later 386 clones.
4:20 The sound burned into my head fromthis time period was from every time I'd reboot my trusty 386-40: "SBOS INSTALLED!!!" from the SoundBlaster emulation utility I managed to shove into himem for my beloved Gravis Ultrasound. Good times, good times.
Hi Adrian! Regarding the MPU X16(?) MIDI card, the RCA jacks are likely for SMPTE input / output, which is an audible FSK time-code for synchronizing to, or from a multi-track tape recorder. The 3rd RCA jack, could be a "click" output that is used as a metronome, during recording, It looks like there is an opto-isolator on that card (the 6-pin DIP), that would be used for MIDI input. There should be 2 signals on the DB9 that connect to the LED in the isolator through 220 ohm resistors. For MIDI output, positive is typically 220 ohms to 5V, and negative is usually 220 ohms to a couple of inverters in series, as a buffer, that connect to a UART. MIDI UART spec is 31.25Kbaud, no parity, 8-bit, 1 stop bit. It may work as a generic mpu-401, perhaps? Good luck!
Appreciate your commitment to deep diving on some of these obscure pieces of hardware. You put in significantly more effort than some of these things are worth to give them their best shot to shine.
Hi Adrian. It’s likely that Compaq card only works (in MDA mode) on the machine it came out of, which was likely an all-in-one with a monochrome monitor, like the TRS-80 Level 3, but labeled Compaq.
It is sad that information about so many things has been lost purely because "it's old and we're not interested", and now that we're interested in learning about the tech of yesteryear,the documents, specifications and funcions are long lost to history, and unless someone is sitting on a repository (also known as a garage, attic or shed filled with old paperwork), it's not likely that we'll find out about a ot of things that we now no longer know anything about... :(
EDO is supported by 85C496 rev. PR only.
At least no one reported of EDO memory working on SiS 496/497 chipset other than rev. PR.
Hi Adrian. You are hands down my favourite TH-camr. These videos are fascinating. I hope you are well. You deserve every success. All my heart. One Amiga nerd from the UK
The Vectrix card(s) look to be an expansion card for either their VX workstatikn graphics or the Pepe card -- VX workstations ran analog video over 3 bnc connectiins (one per color)
My dad, who was into computing from the eighties onwards, would have loved this stuff. I never inherited his instinctive understanding of what was actually going on inside a machine, but I still enjoy watching someone who really knows their stuff at work!
Good to see working cache on these 486 boards. As it was well known that many 486 boards had fake cache. Where the cache sockets would be would be populated and the board register as having the cache chips installed even though the chips were just blank and did nothing. There are DOS utilities you can use to test if your motherboard has fake cache.
I had one of those Trident video cards. Software that came with it included an image editor that allowed you to create a 3D spinning text object using any font and colors. It could extrude the font by a specified amount, create a linear or radial gradient colored background behind the text, and other effects. It would save the animation as a sequence of images which you would then have to combine to create an animated GIF or other video format.
The TOPS cards! My very first job was in an office that had half macs, half PCs. The PCs had TOPS cards and ran a TSR program that shared out directories as AppleShare servers, and you could print from the PCs to LaserWriters and ImageWriters that were on the AppleTalk network. I still have a TOPS card in a box somewhere, but now I wonder if I still have a floppy somewhere with the software on it.
46:48 The Orchid PGC replacement this is talking about is almost certainly the Orchid TurboPGA, which is pretty well known, and is a totally different card than the one you have there
I thoroughly enjoyed this video. I have de-constructed some old computers from the 1980s, that had some weird cards installed. I regret that I tossed them, because I would have enjoyed sending them to you just for the information you would have been able to glean from just having the card in front of you. I do have a box of old motherboards, and will have a look through them to see if there is anything interesting.
Some of those are pretty rare/expensive now, like IBM 8514 (keep away from 8 bit guy).
I used a Periscope. Very handy for debugging back when debuggers were rather primitive.
rinsing in *distilled* water is a good cleaning method, however beware large surface-mount chips, especially BGAs. Water tends to get sucked in there by capillary (surface tension) forces and may stay there for literal years because it cannot escape, slowly corroding everything. It may seem fine for a few days or weeks of active use and then start failing for seemingly no reason. DIPs don't exhibit this problem, and small non-BGA chips don't seem matter much.
Not sure about larger chips...
Blowing water from the boards is most definitely NOT enough drying. They need at least a full day of drying before being used. Older DIP boards are more tolerant, newer with 0.1mm traces are much less so, modern laptops and smartphones are probably not washable outside specialized labs.
I love that you exist to show some love and regard to these old cards!
Lost to the ages is why I'm glad you do this TH-cam channel to at least document what you can.
Try to find the following software: Orchid's PGA Palette Paint to see if it allows the card to run.
I was curious about that Vectrix EX1280 and other than the bitsavers pictures you found, it looks like there is support for it in MAME oddly enough. Or at least support was started there, not sure how complete it is. But you might be able to do some testing using MAME to at least see if it works at all. Or maybe even try contacting the dev who committed the code - I think the commit was just from 2020 so relatively recent. He must have some pretty in-depth info to implement the driver.
Even in the late 90's it was difficult to find info on ISA cards. So much is lost since then, it's so sad.
That's one of the benefits of a post-internet age. Nothing will ever be lost to time again.
I'm glad to have at least partially helped in concluding the PGC/PGA type of card from the last video. I too have noticed that old resources are slowly vanishing from the Web. In some cases I printed out some handy sites back in the 90s, and have been able to find them via the Wayback Machine. Whenever possible and relevant, I try to note the sources in descriptions in some vids in hopes to keep old resources alive.
It’s wild how much has been lost, and how soon it was forgotten. One of my first tinkering computers growing up was a Compaq Deskpro 386 that I rescued from somewhere. Those machines required a setup disk to change BIOS parameters, and when I pulled out the battery I lost those settings. I never was able to get them back because the standard BIOS software didn’t work on my machine because it had EISA slots and required a special version. Never did find it. Sucked too cause I lost the configuration for the 16MB of RAM on an expansion card in the machine. After that it would only ever see 2MB.
These days the community would build something to fix it but in the late 90s this workstation class machine that was only 7-8 years old was a doorstop.
The large mystery EGA/VGA card looks familiar.
I have pulled out a bunch of them disassembling old CAD stations back in 1994. Some of them even had 286 processors on board!
Those stations were used for AutoCAD. I have asked the owners some questions and sadly, I can’t remember everything I had learned back then.
I am very sure that they have a compatibility mode for booting purposes only. Once AutoCAD is running, it takes over and uses them in a totally different mode. This might explain why they are so slow in your test.
The machines I had disassembled had 20+ inch color CRTs for the workspace and a small monochrome monitor for selecting the tools and status information. The small monitor was connected to the 9-pin output and the gigantic screen was connected by VGA. Those large monitors were fixed frequency and couldn’t be used with a standard VGA card. Also they didn’t display anything during booting.
I had kept some of the cards as souvenirs because they look cool, but I was not allowed to have any software, the special driver disks had to remain at the site.
I've washed / cleaned components that way with zero issues. Of course only boards never mechanical hard disks, drives ect. I have a high flow blower I use as well for automotive cleaning purposes. One thing I insure is my water in my area was very hard and I used only water out of my reverse osmosis tap. I'd also wait a good day regardless at minimum to power up any card or board as a last safe guard to fully insure any residual water was dried.
As a side note I gave areas where socket or larger IC chips were located extra attention so if any water or moisture made it's way under the chip it was gone or dried.
Love your videos, sadly only stumbled upon it recently 😢. I'm located in Hong Kong now so you got another viewer from there now 😅
Compaq's VDU does support 720x348 MDA text mode but it can only display that mode when it's used with a Compaq machine and monitor of that era, like a Deskpro or Portable which came with this particular card.It's the exact same card that is used in one of your previous projects Adrian, The Deskpro 8086 🙂, which I find super cool content by the way 👌
It uses one of the pins on the video connector to switch modes and when it does both the VSYNC and HSYNC signals are kept low for a certain amount of frames to prevent CRT damage. It needs a Compaq BIOS and a Dual Mode Monitor (one of the very first multisync-ish monitors if you will 😁 ) to do so, just like the CTRL-ALT-< and CTRL-ALT-> key combinations or Compaq DOS' MODE command ;-) So on a regular clone PC it can only act as a CGA. The Dual Mode monitor would act as an RGBI CGA monitor and will not cannot on an MDA or Hercules card because of the different pinout.
Maybe you could make the VDU switch by manually manipulating its registers using debug but that would require a bit of reverse engineering as I think the code in the BIOS manipulates the card's registers directly. The early Compaq video cards, the VDU and especially CEGC (the EGA equivalent, the Compaq Enhanced Graphics Card) are definitely super quirky and interesting.
For instance: Did you know the internal monitor of the Portable uses an analog video signal and no TTL RGBI? That little 9" CRT in my 1984 Compaq Portable can display 640x350 EGA in 16 grey/green scales which I find super cool! It looks amazing!
I bought Arcade game PCBs in bulk from operators, tested, repaired if needed, and sold them for years. Now and then, I would getnone that was close to working , but wouldn't quite kick over. My one little trick before getting out a probe was to stick them in a dish washer! I added a little bit of vinigar and a rinsing agent. 75% of the time, it did the trick! If it did, I would remove all the ICs, clean and coat the legs, then reinstall them like I did all my boards I sold.
I think I remember the SCAN-CAP cards used for science and medical purposes. I had the opportunity to work with an electron microscope a long time ago. The microscope was generating a BAS (monochrome) signal which was displayed on a TV screen. The PC was connected between the microscope and the screen and used to print out the image, zoom into the picture and save the picture to disk. I am not perfectly sure, but I think this was a SCAN-CAP card.
Also I know that such cards were also often used for normal microscopes. Instead of snapping slides, a video camera was used connected to a computer or video printer.
To snap pictures of games, they usually did that using a camera with film. The German 64’er magazine even printed an article how to do that. Which monitor works best, which distance and setting for a 35mm camera, etc.
Oh man yes, great to see all these old but not crazy old components on here- bringing me back in time. Made me think of the AdLib sound cards. I remember my friend showing me one way back, and how his computer could make 'music' - I was amazed. It was an ISA with volume knob on the card- I'm amazed to see some selling for $300+ on eBay these days!
Thank you for your amazing content. I've watched it all and look forward to the updates.... From the UK 🇬🇧
enjoyed the adapter sluething. I remember PGA/PGC from when I worked at IBM and the Orchid cards as well. Matrix also had a PGA card back in '87. PC Mag explained PGA in their May 26, 1987 edition (Petzold n Rosch) if you're tired of reading old IBM manuals.
I think that 1991 Scan-Cap board can not only capture video but probably supported Video in a Window. Hauppauge had a adapter board in 1992 with almost the same external jacks. Was called Win/TV or something.
18:25 Seeing that startup screen was like a gut punch of nostalgia. My first computer was (IIRC) a pentium running at 133MHz with Win 98SE. Can't recall how much RAM it had, probably 32 or 64 MB.
With that Vectrix Card. The site your Patreon linked has a PDF in the parent directory about Pepe. I would expect that the card is related to this. My understanding from that brochure is that the card is essentially used by things like CAD as a supported device, but not as a graphics card. The Wikipedia page for Vectrix Corporation also explicitly says that the card wasn't supported by the bios as a video card and you needed to have another graphics card for system operation.
Guessing here, but the cards you have look to be newer that what I can see with Pepe. Pepe looks like it was two cards across two ISA slots that were connected to each other. I wonder if the VX1280 was an improved version that didn't need the second ISA slot.
I wonder if you could find an old copy of Autocad and see if it might recognize that card.
From 1985 to 1990, computers were advancing at an extremely fast rate and the standards kept changing. Monochrome DOS to EGA than VGA, 386 32 bit and the 486 in less than 5 years. DOS to Windows 3 Optional HDDs, to everyone having one, etc.
Really interesting to see the various cards, especially the Creative SoundBlaster 16 which reminded me of the sound cards I used to have & then the Orchid card which reminded me of the Orchid Righteous 3D card I had which was (I think) followed by the 3Dfx Rage 32 (???), then the VooDoo2 & then the VooDoo 3 graphics cards !!!
Talk about bringing back old memories.
Scary to think that the Orchid was 1985 .... 39 years ago !!!!!!!!!!!!!!!!!!!!!!!
God I'm getting old !!!
memories wow!!! this really takes me back .. i did alot of hardware /software instalation in 1990-2005 you had to be good at mystery jumper and dos
Great stuff! I love tech archeology videos.
Just started watching your stuff, didn't know you were in my hometown. Cheers!
The TI based cards had stuff on them that VGA didn't have, particularly hardware facilities like Bounded Block Transfers, ('BitBlit) and greater colour depth. If you see a TMS34010 or TMS34020, then it supports the TIGA standard. They did find use in arcade machines, like 'Mortal Kombat'. In the UK, there was Pluto graphics TIGA card that could be run 'outboard' from a PC, with an 8 bit interface card in the PC.
I have the TOPS DOS driver (Version 3.0) disks if you are interested. The drivers came on four 720K floppies.
I have a vague memory that the old Cad cards sometimes used a jumper cable between themselves and regular video cards capable of it, to pass video (similar to the newer SLI). The Cad cards were more for special processing, and didn't work like a regular video card, even thought they had the external connector. They were some what hardware/software specific. I never encountered one, so I'm not for certain, as I said, it's vague memory. I also remember using FCC ID's to try to get some idea what blank cards were, with mixed success.
Blast from the past. I've used many of these things back in the day. I use isoprop alcohol instead of water to clean them...
The VTEC that's been scratched out is a QA reject. Who knows what it failed.
FPM = Fast Page Mode
19:00 Get the PC-Doctor POST card. Or PM me and I'll get you one.
23:00 Write through cache...
25:50 as to tools for checking speeds, I wrote PC Bench for just that purpose in the late 1980's.
31:30 if I recall right, the jumper is for setting base address for multi-card systems. I.e. won't fix a problem if it's not happy.
47:30 Orchid Graphics card -- it was used for video editing systems (among others) in parallel with a CGA or Hercules card. I wrote diagnostics for it back in the stone age. Similar to the Targa card back in the day. It needed custom software to set it up. The BIOS didn't know what to do with it.
1:00:00 I always add a bit of epoxy to jumper wires if I want the card to survive storage...
I have the same blower unit in my shop at work and it's great. ESD safe as well which can be critical repairing VFDs and other industrial boards.
If I remember correctly there were a number of non-standard GFX cards made specifically in the early days of Autocad to try an give better than standard resolutions and these always had specific drivers for those cards. Some had standard outputs others needed a default GFX card.
Owing to the cost of cache RAM, back in the day, many generic motherboards were fitted with fake cache, or sometimes low speed cache that would work with a 486 SX /25, but would fail with a 33mHZ DX or DX2. The reall sneaky ones would fake up the action of on board cache, by turning the on-chip cache on and off from the BIOS.
Great video: gave me an interesting mix of happy memories and awful flashbacks.
Wow. Feature length Digital Basement. I like it.
Just had to say Nathan's chiptunes still killing it. 🤘🖤🏴☠️
26:20 Those "move" throughput differences are extreme -- is that possibly a write-back vs. write-through cache setting difference between the CPU tests? I would expect if the cache strategy is write-through, the CPU might be delayed on making its next read until the write is complete, which would constrain the move speed to be no more than the base memory write speed, which is kind of what we see in the 2nd graph.
Several years ago I cleaned a keyboard that had soup poured into it. Similar to your method, I washed the circuit board in a sink with dish soap. I then wrapped the board in a towel, and baked it in my oven set to the lowest "Warm" temperature for an hour or two. Just another way to gently remove moisture.
Your board washing method is very much industry standard, so much so it was in many electronics manufacturing handbooks I had in my life.
Though, for speed, what we do at work is also give the circuits a purge of water by using IPA.
Stuff with SMPS transformers and batteries soldered to them I use only IPA for safety and corrosion prevention.
Worth knowing considering people have experience with water and powered up electronics (Macbook users apparently like to give their lappie a coffee)
Onto the rest of the video, I'd like to know what the bespoke cards were, hopefully by the time 1H is over.
Confirm you could fix a motherboard that had been given sugary coffee to drink by washing in water. For best rapid results dry with blade type hand dryer, immediately after washing using small paint brush. Didn't always save the board, but warranty was invalid due to coffee so nothing lost trying.
It would be great if you could scan in both sides of the unpopulated PCB's as well as take high res photos of the chips on the boards as it may help people reverse engineer or replicate the PCB's in the future. Also if you could dump the GAL's and PAL's on the cards along with the ROM's that would be great.
I have a board similar to that SCAN CAP, the Truevision Targa+. Yes, it's a full frame grabber for video. I got mine for a few bucks when a local news station offloaded a bunch of the things back in the 90s. I had the software for it at one point.
I love this. Memories... oh wow.
I never had access to all these rare birds back in the day, and was reduced to scrapping out old systems for parts to build "new" ones. I feel ya. One place some very old hardware can be found is in CNC machine tools. I've found CNC systems still running MS-DOS, WinNT 3.51, WinNT 4.0, and Win2K. Don't even get me started on HP's old HP-UX systems - I still support those at my current job. (Fortunately, I have software for those, and good docs.)
Oh yeah, I had lots of fun to get a CNC milling machine back up and running in 2006 after the Mitsubishi 286 decided to forget everything about its hardware when the CMOS battery went flat. That beast was so old it didn't have a setup ROM, you had to have a BIOS setup floppy disc. I eventually managed to get that, figure out the correct hard drive type and XCOPY everything to a SCSI drive that I could then stuff into a used Pentium 90 with a fresh DOS 6.22 install. I tried getting the hard drive (possibly RLL, absolutely no info available, it used the classic Shugart connectors and control and data cable setup but didn't work with an ST506 controller, so I think it has to have been RLL, ESDI would have configured automatically and worked with any random BIOS hard drive type) and its controller card to work in a different 286, to no avail.
10:51 I’ve encountered this in automotive parts as well! I first noticed it when I got a MAF sensor from the company that supplies the part to the OEM. It was identical save for the (rather crude) removal of the automaker specific part number and logo.
hey Adrian, just watching you with that dryer, it made me think - what about a dehydrator? I use one in my kitchen for drying fruit and vegetables, but it strikes me you could easily put boards in there in racks, set a sensible temperature and let them dry. With a decent one you could probably dry quite a few boards at once.
While that would work, it leaves all the minerals that were dissolved in the water on the board, so you would get spots. I guess it's better to blow most of the water away with compressed air before.
Well, or maybe use a bit of dishwasher rinse agent in the water for the final wash?
@@kpanic23 yes, he mentioned that in his region the water is quite soft so doesn't cause issues. I guess ideally you would want to use distilled water
I watched this video while wearing my "Ron's Computer Videos" shirt. Literally.
The SoundBlaster with the noise is because it's picking up RFI, likely from components on the motherboard. The design or some component on the SoundBlaster might be susceptible to this. The noise will vary depending on the components being utilized, so your simple test isn't triggering the RFI from the PC component, but running a more complex program may.
About that 386 adapter for Periscope. Actally, those pins in 168-pin socket are the same as in cheap, say, 40 by 1, round pin-header-female available an AliEx. I was able to extract them one by one from 40-pin row by forcing them off from the back side using rigid tweezers. And I have used exact same method to extract damaged pins from 168-pin socket, replacing them with salvaged ones. The only thing, "new" pins feel a bit loose in 168-pin socket without soldering them to the PCB that socket placed on. In your case, those replacement pins would be followed by soldered pins, so they could not fall off.
You're doing the tech gods' work, Adrian.
Both videos have been highly enjoyable. As a sise note, I actually try to avoid water and use medical alcoholl or isopropyl. I have used a combination of warm water, small preassure washer and a soft brush for really grimy stuff.
I worked for a company that sold raster image processors. We made several custom cards whose jobs were simply to interface with output devices. One board was for a specific raster accelerator but was eventually replaced by a software function. Even if you had the drivers for these boards, they're worthless without our dongle protected software. If you managed to get the dongle and the software, they were only useful for sending data over thick cables to filmsetters and platesetters, which no hobbyist would ever have. Such is the fate of many small-time cards.
Really brings back memories, if maybe not such good ones, lol. Faffing around for hours on end with jumpers and settings because that new card or motherboard you bought second hand never came with a manual. I got my hands on this old Burroughs 286 my Dad got from his friend who worked there. Of course, no manual. Faffed around with that thing every day after school for a solid week or two. Swapping stuff around with bits of hardware I had laying around, or out of my XT. Every couple of days I'd get it to move a little farther through the POST sequence, only to get stuck again. So I'd wait until my parents fell asleep, pull the line off the house phone and drop it down to the basement through a hidden hole I made in the floor right above my computer desk. I'd be awake all night dialing up every BBS I knew to post questions with the hope someone had an idea of what I needed to do next. After that, carefully sneak the cord back upstairs and get the house phone plugged back in before one of my parents got up.
I do miss the old BBS days, lol...
I remember having trouble getting working drivers for a (new) graphics card in the early 90s. Some stuff was already lost without help from time passing...
@9:41
I could be wrong but there looks to be a surface mount resistor oddly positioned near the dip switches ??
That Heath card reminds me of the cards that Zenith was using in the Z-248 systems I've worked on... they had some sort of odd mixed-mode trinitrons that were EGA/VGA like this, very unusual. Perhaps there's Zenith documentation for a system that has a rebranded version of this card?
5:44 the awe64 was just an awe32. The other 32 channels was only supported in windows with the special software pack-- as far as i could tell back when i was a teenage expert, the other 32 was done in software. I think i asked in a couple tracker forums and they confirmed theres no support for additional voices in hardware, thats why none ofnthe trackers or cubic players had awe64 mode.
The memory modules were outrageous. I thinknit was $100 for 8 megabytes
On the video card that had the dim output, you can see a broken off resistor