Nexxen wrote on 2026-10-31, 13:07:

Wanted to use a socket A platform for some tests. The PCI video card I used would make it freeze just after entering bios or boo […]
Show full quote

Wanted to use a socket A platform for some tests.
The PCI video card I used would make it freeze just after entering bios or boot screen. Either way I had limited time.

Put an AGP and ok, ISA and ok.
Video card is a S3 Trio64V+, board is https://theretroweb.com/motherboards/s/qdi-k7 … -k7e-kinetiz-7e
Maybe a chipset related issue?
1st time I had issues with this card.

Possibly unrelated, but when I was trying to set up dual head in a KT series board back in the day, the only PCI cards that would work, albeit alongside a GF3 in the AGP slot, were Trident TGUI and GD5434 and the S3 virge and 1st gen Trio I had wouldn't boot. I did not try them independently.

Edit: it's been itching my brain that it might not have been a GD5434 but a GD5436

BitWrangler wrote on 2026-10-31, 13:34:

Nexxen wrote on 2026-10-31, 13:07:

Wanted to use a socket A platform for some tests. The PCI video card I used would make it freeze just after entering bios or boo […]
Show full quote

Wanted to use a socket A platform for some tests.
The PCI video card I used would make it freeze just after entering bios or boot screen. Either way I had limited time.

Put an AGP and ok, ISA and ok.
Video card is a S3 Trio64V+, board is https://theretroweb.com/motherboards/s/qdi-k7 … -k7e-kinetiz-7e
Maybe a chipset related issue?
1st time I had issues with this card.

Possibly unrelated, but when I was trying to set up dual head in a KT series board back in the day, the only PCI cards that would work, albeit alongside a GF3 in the AGP slot, were Trident TGUI and GD5434 and the S3 virge and 1st gen Trio I had wouldn't boot. I did not try them independently.

Ok, this confirms possible incompatibilities.
Better than something is dying 😀
Ty!

About 6 months ago I managed to get my hands on Matrox Parhelia, brand new in box. It is the later "bug fixed" AGP 8X version, with higher clocks. The listing however, did not have picture of the box... just generic picture of the card taken from the internet, so it was a gamble, and the price was, although high, still lower than other two options I had at the time. Once the package arrived, I was a bit underwhelmed. The box was in great condition, but it had a very large sticker (white inventory sticker), right on the front side. Very large... it was about 1/5 of front side actually, so it really bothered me.

Yesterday I decided to take a risk and try to take off that annoying sticker. Took a hairdryer, some precision blades and after about 1.5 hours I finally managed to take of the sticker without damaging the box or the paint. What remained as a problem was the glue, a large quantity of it across the whole area where the sticker was. I googled a bit, and found about three solutions. The first was the scotch tape and lifting the old glue with it, the second was isopropyl pads, and the third was Zippo lighter fluid. At first I though scotch tape is the way to go, however it is sort of mechanical method, and since the area is large, the chance for damage was very high, Isopropyl also seemed like a more aggressive chemical method. So in the end I decided to go for lighter fluid.

I was skeptical at first... but it really proved to be a miracle method. Fast evaporation, no damage to cardboard or paint... it just needed patience and a dozen of cotton swabs. Took about 30 minutes, but the result was amazing... Never really tried something like this before, and I am so happy I managed to restore the box back to decent condition. To protect it from other items in the drawer, I've also put it in a new cellophane wrap, non-stretchable so it won't stick or take paint off in the future. For some unknow reason I have failed to take a picture of that annoying sticker, as I was so focused on getting rid of it. But at least, you can take a look at cleaned and wrapped box, tucked away with other goodies. I fully recommend this lighter fluid method btw... it's literally magic at work.

@Shader_BiH: I admire your work and your patience to preserve the packages. I collect also graphics cards. Matrox Parhelia is really great. Have never known that it's early release had some bugs, thx for information.

Make using astra ai the Prey 1998 maim menu boot like as the knowen "Prey OS" that made in the 90s by 3D Realms from the Unreal Engine 1 tech sapertlly from the game PreyEngine becuase the industry know better in that peroid the Unreal 1 tech instad of private new engine, and save expenses on this project also. Now for the UT99 as a new mod:
https://x.com/duke_unreal/status/211678804658 … CRoiqcqKKs&s=19

eliot_new wrote on 2026-11-01, 10:44:

@Shader_BiH: I admire your work and your patience to preserve the packages. I collect also graphics cards. Matrox Parhelia is really great. Have never known that it's early release had some bugs, thx for information.

Hey thanks 😀 From what I understood, it is more of a architectural issues of the first revision. The newer one should be about 10-20% faster, depending on the game. Can't wait to get my hand on the original version, because I would really like to compare the two. Also, I am determined to find a workaround on FarCry blue textures bug on these cards, that will be a challenge.

The trash bag box build is getting an upgrade to it's case downgrade. Was running around yesterday and picked up some small totes, polypropylene. I had intended to look at small blue box recycling bins, as the ones I am using for such seem the right size. However, the ones on sale for reasonable money that I came across had reduced dimensions, more highly rounded corners and more slope on the sides, so had a lot less "floor space" than the ones I have in use. So, these totes seemed next best thing, also have a lid for "putting away" though I intend to run them open ATM. The dimensions on the label are a bit of a lie, being I think external at the rim of the box. I was looking for 10x13 like the trash bag box, and these come out at about 10.5 x 13.5 at the base.

The point is/was, to keep several mATX boards that aren't great, but still interesting enough and have slots and sockets to run interesting spare cards and CPUs, from hogging "real" cases when I might not want to use them that often. But leave them somewhat assembled and ready to run to play around with when the urge strikes. Shown in pics here is a "stunt board" not the AM2+ board that will be moving in shortly. Though that one might get put in one of these later.

Further members of this "family" might include socket A and 370 boards.

Things I don't quite like.... maybe not quite as sturdy as the recycling bins I was thinking of, with a rim not as wide for racking one sided like those do, and maybe a little narrow for supporting on runners. But I may attempt that, looks like two strips of 1/8 hardboard or ply, screwed down offset, so it has a recess and a step up would probably do okay for hanging. Or maybe just strips of 3/8 for runners in a drawer type rack.

IDK if I should go at this with a hot knife rather than trying to saw it though, might be easier and less prone to crack to put some holes in like that.

Label of tote included for the "if your friend jumped off a cliff would you too?" types 🤣

Got a switch for free and used it to run Blood in coop.

Ah... does anyone have the Aztech driver disks?

I've been pulling together software/drivers that will hopefully enable an Aztech PCI 288 Q3DII from 1999, but each component I found came from a different source so I'm not confident I have the complete/accurate file set.

AFAIK, the PCI 288 represents the last sound card from Aztech and the last audio processor from Trident. It is powered by Trident 4Dwave-NX connected to a SigmaTel STAC9708/11 and these were configured with two oscillating crystals to somehow boast ~100dB SNR (according to the marketing blurb). Other brands using the NX fitted only one oscillating crystal and did not achieve the same result - I find that intriguing.

I found magazine reviews showing the Aztech's SNR score completely thrashed everything Creative Labs (and everyone else) was offering (bearing in mind Creative acquired Ensoniq hence Creative's two very different product lines). Either way, to appreciate Aztec's ~100dB SNR any listener would need to build the kind of silent PC that did not exist in homes due to keyboard clicks, force feedback joysticks, buzzing CRTs, vibrations in steel cases, 80mm PSU fans, HDDs, etc., etc. and that makes the engineering strategy perplexing.

It also lacks the typical musician I/O connectors that you'd expect to see in a sound proof studio so it wasn't designed for audio production; I found evidence of 4DWave-NX cards being used in audio-acquisition systems though. How do I explain this card? It may just be a byproduct of ordinary corporate control because when product lines are cancelled the budget owners redirect their resources from cancelled activities (next generation research, marketing, etc.) to remaining activities (testing, tuning, etc.) rather than accept their own reduced importance - that is a known empire building behaviour that can produce an accidental masterpiece.

EDIT: An original disk has not materialised but I've recreated the Aztec Multimedia Studio CD bundle, the last release of the Trident reference drivers, and the Sound Blaster 16/Pro patch for MS-DOS. I'll upload the set to archive.org at some point.

Got decent dos support sorted for sk41g using sq2500 aureal card instead of turtle beach. Screamer, shadow warrior, redneck rampage, doom, duke 3d all work in real dos mode. Yamaha ymf744 was more stable and less difficult to get going but ruined by single speaker output in real dos mode. Solo-1 doesn't like the sk41g. I'm happy with setup so will be backing up SD card and autoexec and config files

Thermalwrong wrote on 2023-03-16, 03:12:

I bought 2x very cheap 64GB CF cards and one is working great in my iPod 3rd gen, the other I got curious about and took apart, […]
Show full quote

I bought 2x very cheap 64GB CF cards and one is working great in my iPod 3rd gen, the other I got curious about and took apart, that no longer works sadly. Oh well, it was still cheap and now I know the flash chips are garbage but functional with an SM2234H controller. Still a good CF card at 2x the price actually since 64GB is almost never in that price range. Got a pair of 16GB SLC industrial CF cards today as well and wow those are fast, don't think I've seen 100MB read on a CF card before.

Beyond that I'm trying to get things working so that I can pack them away and get onto moving since I'm doing that soon (procrastinating just a bit). So far I've got 3x of the 486 boards I purchased as not working into a working state, in each case it was stilly stuff like jumpers or cache chips in the wrong places that I couldn't spot at the time. Seriously, 486s are such an experience since there's 3+ CPU types to support and the board will just not function in most cases until the jumpers are just right.
With today's one it was the Gemlight GMB-486SG-V2.2 board which has a different jumper configuration than that board on theretroweb: https://theretroweb.com/motherboards/s/dtk-pkm-0038s The JP24 jumper being the most significant difference.
Of course, thanks to this forum and the person maintaining this awesome page: https://www.iwriteiam.nl/GMB-486SG.html
I now have the manual for the version of my board and it's working 😀 the repair work I did on it months ago to just get a post code of "--" was successful and it was just jumper config. Another one down.

Before now, I've never come across a motherboard with the same silkscreened name but two different PCB revisions.

Thanks a lot.

The 2.2 version i have is indeed different. At the other manual, JP24 is at a different location, and has a resistor over that position. Under the resistor i can see a rectangle that indicates there could be a Jumper position also.

I have a 133MHz 486 CPU, that i want to try to install on this mobo.

Pull up/down 0ohm resistor likely, though normally they just short it with a small bit of wire.

DOS Retro wrote on 2026-11-04, 05:19:

Thanks a lot. […]
Show full quote

Thanks a lot.

The 2.2 version i have is indeed different. At the other manual, JP24 is at a different location, and has a resistor over that position. Under the resistor i can see a rectangle that indicates there could be a Jumper position also.

The attachment photo_2026-11-04_06-09-00.jpg is no longer available

I have a 133MHz 486 CPU, that i want to try to install on this mobo.

Actually i dont see any settings for 33x4 and setting up the 3.45Volts in this manual

Trashbytes wrote on 2026-11-04, 05:23:

Pull up/down 0ohm resistor likely, though normally they just short it with a small bit of wire.

Let's not forget that sometimes designers would add those just to jump tracks when there weren't enough layers.
Doesn't look like it in this case, but it definitely happened.

myne wrote on 2026-11-04, 05:48:

Trashbytes wrote on 2026-11-04, 05:23:

Pull up/down 0ohm resistor likely, though normally they just short it with a small bit of wire.

Let's not forget that sometimes designers would add those just to jump tracks when there weren't enough layers.
Doesn't look like it in this case, but it definitely happened.

Oh yea, Ive seen that before along with jumper wire rather than an actual jumper.

I love the older dual sided PCBs that you can tell they were done by hand with the beautiful curved hand drawn tracks and all the bodge wiring, truly amazing to look at and reverse engineer on paper as a way to understand what you are looking at. It being dual sided or single sided also helped with understanding the circuit.

I'm thinking 1970 to about mid to late 80s were its hey day for hand made PCBs.

By any means necessary...

What the hell happened here? A junk Zalman aero, banged around a bit, original fan was badly broken. I had these two Arctic cooling fans that came in with other fans, guess they were part of a HSF at some point. Cooler was a CNPS9700 LED originally I think. Original fan sucked ~0.3A and the arctic cooling ones were only 0.15A, didn't have airflow numbers for anything, but I figured half the oomph is half the oomph so I stuck both on there in push, pull. Had to have the back one inverted of course so they were both working in the same direction.

They both had these little rubber plug feet mounting things which had little eyehole tags on, so I just looped through those with twist ties and sandwiched the cooler between them as it were. Possibly it could have looked cleaner with half unfolded paper clips.

Figure it should be okay for a 65W CPU I'm gonna use it on. Didn't want to "waste" my other nicer Zalman Aero, Hyper 212 or Gamma400 on a piddly lower watt CPU when I've got some furnaces to cover.

florian3 wrote on 2026-10-28, 01:05:

Recently got a prototype FPGA MIDI synth board made and have been experimenting with porting classic MIDI synthesizer emulators […]
Show full quote

Recently got a prototype FPGA MIDI synth board made and have been experimenting with porting classic MIDI synthesizer emulators to it.
It’s based on a Lattice ECP5-45F with 32 MiB parallel NOR flash and 1–2 MiB SRAM. No MCU, just FPGA gateware, which so far has been developed using LLMs based on the work of Nuked, gearmulator, and MAME.
The board fits a Wave Blaster header but can also run standalone.
Working so far: Yamaha FB-01, Roland MT-32/CM-32L, CM-64 (including U-110 cards), SC-55/SC-55mkII, SC-88/SC-88 Pro, and Yamaha DB50XG.
Still experimenting with adding more synths.

This is super interesting and an incredible achivement! Anymore info about this project?

Baleog wrote on 2026-11-05, 06:37:

florian3 wrote on 2026-10-28, 01:05:

Recently got a prototype FPGA MIDI synth board made and have been experimenting with porting classic MIDI synthesizer emulators […]
Show full quote

Recently got a prototype FPGA MIDI synth board made and have been experimenting with porting classic MIDI synthesizer emulators to it.
It’s based on a Lattice ECP5-45F with 32 MiB parallel NOR flash and 1–2 MiB SRAM. No MCU, just FPGA gateware, which so far has been developed using LLMs based on the work of Nuked, gearmulator, and MAME.
The board fits a Wave Blaster header but can also run standalone.
Working so far: Yamaha FB-01, Roland MT-32/CM-32L, CM-64 (including U-110 cards), SC-55/SC-55mkII, SC-88/SC-88 Pro, and Yamaha DB50XG.
Still experimenting with adding more synths.

This is super interesting and an incredible achivement! Anymore info about this project?

Since then I have been working on adding Ensoniq Soundscape (synth only). Unfortunately I don't have a real Soundscape Elite card (and also no sound ROM for it) but interestingly the newer firmware for the basic S-2000 that is uploaded to the card every time it's initialized includes the program for the ES5510 effects chip that is only on the Elite. My ES5510 is based on MAME and I do not know how accurate it is, but basically I now have a classic Soundscape core with the original samples with effects support.

Another synth that I am working on is the ICS WaveFront on the Turtle Beach Tropez Plus and others. The firmware ROMs on the Tropez Plus are socketed and based on the firmware it was possible to dump the sound ROMs from software. The ICS WaveFront implementation is also based on MAME, but I ran into a couple issues. There is not much on the YSS225 effects chip of the Tropez Plus so I am working on that now.

The YSS225 is also used on an optional daughterboard for the MediaTrix AudioTrix Pro. I do already have a working OPL4 synth core based on openMSX, so I am hoping to support OPL4+YSS225.

I did also look into larger Yamaha synths but from what it looks like it won't fit on my current FPGA.

There are still a couple more synths I am looking to add.

I am FINALLY getting around to fixing the G2 white level issue on this Dell P1110 monitor this weekend. My first time using WinDAS so I'm curious how it will go. I made space on my computer desk to set up a Win98 PC right next to my main desktop, so once I get the monitor calibrated, I'll pick a tower to use and get down with some 90s gaming goodness.

Tried fixing the 2180CDT. Unfortunately unable to fix. CPU remains cold no matter what I use - K6 or Pentium MMX.