BitWrangler wrote on 2024-03-03, 15:35:

I had been "under a rock" for 3 years as far as PC hardware was concerned when I got given a 9600GT when it wasn't all that old, and running that I was like "Holy shit, why would you need anything faster than this???" still seems to me they do a decent job in "random used office monitor" resolutions. Somehow got three of the things now.

I was testing various graphics cards a few weeks ago in XP to try and see how far back I can go with budget cards to get an experience that I was satisfied with for most use cases. The biggest hurdle was that I wanted it to run Oblivion at a decent frame rate. The 9400 GT came very close, but the 9500 GT did a much better job. I had a similar feeling with that card, asking myself if I would need anything faster than this for XP. I would imagine a 9600 GT to be a pretty amazing upgrade for the time.

In case anyone was under the impression that I just have to turn around and lay my hand on an awesome piece of hardware, today as part of the investigation into "The Soundcard Situation" I turned around and laid my hand on this... Opti 82C931.. maybe the worst SB clone ever produced according to legend... that's if you discount some versions of the SB16 I guess 🤣 ... can it be alllll that bad? I am getting the feeling it might end up in something anyway as soundcards run away when they see me coming or something.

Is there even an ideal spec to curse with this? 96 card so youdathunk the basics would be getting covered by then, might have to stick it in a Win9x build that doesn't DOS around much.

BitWrangler wrote on 2024-03-03, 23:21:

In case anyone was under the impression that I just have to turn around and lay my hand on an awesome piece of hardware, today as part of the investigation into "The Soundcard Situation" I turned around and laid my hand on this... Opti 82C931.. maybe the worst SB clone ever produced according to legend... that's if you discount some versions of the SB16 I guess 🤣 ... can it be alllll that bad? I am getting the feeling it might end up in something anyway as soundcards run away when they see me coming or something.

Is there even an ideal spec to curse with this? 96 card so youdathunk the basics would be getting covered by then, might have to stick it in a Win9x build that doesn't DOS around much.

I've used those in some builds. It's not bad imo. Be warned I had problems using unisound with Opti cards. The card works but sound is distorted. Real weird issue.

The DOS drivers are packed into the Win 3.1 installer. I extracted them and put them on Archive.

kingcake wrote on 2024-03-04, 01:23:

BitWrangler wrote on 2024-03-03, 23:21:

In case anyone was under the impression that I just have to turn around and lay my hand on an awesome piece of hardware, today as part of the investigation into "The Soundcard Situation" I turned around and laid my hand on this... Opti 82C931.. maybe the worst SB clone ever produced according to legend... that's if you discount some versions of the SB16 I guess 🤣 ... can it be alllll that bad? I am getting the feeling it might end up in something anyway as soundcards run away when they see me coming or something.

Is there even an ideal spec to curse with this? 96 card so youdathunk the basics would be getting covered by then, might have to stick it in a Win9x build that doesn't DOS around much.

I've used those in some builds. It's not bad imo. Be warned I had problems using unisound with Opti cards. The card works but sound is distorted. Real weird issue.

The DOS drivers are packed into the Win 3.1 installer. I extracted them and put them on Archive.

Awesome thanks, will know where to look when I'm setting it up.

I read elsewhere they use a crappy opamp for audio amplification, and at full volume it's overdriven, leading to the distortion, so turning it down a bit is meant to help.

BitWrangler wrote on 2024-03-04, 01:37:

kingcake wrote on 2024-03-04, 01:23:

BitWrangler wrote on 2024-03-03, 23:21:

In case anyone was under the impression that I just have to turn around and lay my hand on an awesome piece of hardware, today as part of the investigation into "The Soundcard Situation" I turned around and laid my hand on this... Opti 82C931.. maybe the worst SB clone ever produced according to legend... that's if you discount some versions of the SB16 I guess 🤣 ... can it be alllll that bad? I am getting the feeling it might end up in something anyway as soundcards run away when they see me coming or something.

Is there even an ideal spec to curse with this? 96 card so youdathunk the basics would be getting covered by then, might have to stick it in a Win9x build that doesn't DOS around much.

I've used those in some builds. It's not bad imo. Be warned I had problems using unisound with Opti cards. The card works but sound is distorted. Real weird issue.

The DOS drivers are packed into the Win 3.1 installer. I extracted them and put them on Archive.

Awesome thanks, will know where to look when I'm setting it up.

I read elsewhere they use a crappy opamp for audio amplification, and at full volume it's overdriven, leading to the distortion, so turning it down a bit is meant to help.

Oh so maybe unisound just maxxed out the software mixer or something. I was just doing a quick test and didn't dive into the problem.

Had my mITX Socket 7 on the bench today, fixing a failed D: partition. I knew it was a bad idea using a 160GB HD, but it was all I had. Used SeaTools to lock it down to 128GB total, so hopefully that helps. Also swapped out the passive-cooled Pentium 90 for a Pentium 200. I keep flip-flopping on what CPU I want in this thing - I'm playing through Descent 1 on it right now, and the P90 gives period-correct framerates, but the P200 will be "nicer". It also allows it handle some later-90's demoscene stuff. Would really like to be able to cram a full-sized HD in there, but I just don't see that's there's any more room to be found. Of course, I could dump the (600W, hah) SFX PSU for a pico, but at that point I'm really mucking with this build which I quite like in its current state. Everything fits *just* nicely. For once, I think I'll leave well enough alone! 🤓

ubiq wrote on 2024-03-05, 03:58:

Had my mITX Socket 7 on the bench today, fixing a failed D: partition. I knew it was a bad idea using a 160GB HD, but it was all I had. Used SeaTools to lock it down to 128GB total, so hopefully that helps. Also swapped out the passive-cooled Pentium 90 for a Pentium 200. I keep flip-flopping on what CPU I want in this thing - I'm playing through Descent 1 on it right now, and the P90 gives period-correct framerates, but the P200 will be "nicer". It also allows it handle some later-90's demoscene stuff. Would really like to be able to cram a full-sized HD in there, but I just don't see that's there's any more room to be found. Of course, I could dump the (600W, hah) SFX PSU for a pico, but at that point I'm really mucking with this build which I quite like in its current state. Everything fits *just* nicely. For once, I think I'll leave well enough alone! 🤓

Sounds awfully familiar!

Trying to figure out the cause of data corruption in Windows 9x

What a lovely machine. More angles, please!

Prepared an Dell OEM Audigy 2 ZS to be converted to retail 😀 Off to download the required stuff (EEPROM dump, AsProgrammer and CH341 drivers)

ASUS RoG G71GX is the machine of choice 😁 - Core 2 Duo T9600 and GTX260M.

Is this as simple as swapping out the EEROM or is it more involved? Is there any particular reason to do this other then to repurpose the board?

The Dell OEM A2ZS (SB0358) won't detect when trying to install the standard Audigy 2 ZS drivers.
I got a EEPROM dump (thanks to @LoStSOul) and it should be as much as just writing the EEPROM that fixes it and turns it into a retail card.

I wonder if it would be easier to pop off the chip and use a TL866II to reprogram it. Are those little micro probes you have there attached to it?

I've never tried to reprogram a flash chip in-place before.

PcBytes wrote on 2024-03-05, 12:02:

The Dell OEM A2ZS (SB0358) won't detect when trying to install the standard Audigy 2 ZS drivers.
I got a EEPROM dump (thanks to @LoStSOul) and it should be as much as just writing the EEPROM that fixes it and turns it into a retail card.

I was reading about this lately and came across an old post on Anandtech that talks about adding device ID for OEM Audigy 2 ZS cards: https://forums.anandtech.com/threads/audigy-2 … /#post-13810671

Allegedly this can be done to make the normal Creative drivers work with OEM cards.

edited to add:

Here's a snapshot of the techpc site they linked: https://web.archive.org/web/20040609082510/ww … uk/audigy-2.php

Relevant section :

Manual Method […]
Show full quote

Manual Method

If something went wrong using the patch, you can manually modify the drivers yourself. The file you need to modify is buried inside the data1.cab file, located in the \Audio\Setup directory of the software suite, and also in the main folder of the driver update, if you downloaded and extracted that. To extract files from these .cab files, you'll need a program called i6comp.

Specifically, you'll need to download this version, extract it, and put just the i6comp.exe file (located in the Release folder) in the appropriate directory with the data1.cab file. Then download this version, extract it, and put only the ZD51145.DLL file in the same directory. The rest of the files are not important.

You’ll need to extract the file CTComp.dat from the data1.cab file using the tool you just downloaded. Using Command Prompt (Go to Run and type cmd), navigate to the directory where the data1.cab file is located. Once you're in that directory from the Command Prompt, type i6comp e data1.cab ctcomp.dat to extract the file.

Now, you can either edit the CTComp.dat file yourself in Notepad to get your card recognized, or you can download an already modified version here (thanks to kode54) to replace the one you just extracted. If you opted to edit the file yourself, see below. If not, just skip to the next step.

Open CTComp.dat using Notepad, and scroll down until you see the only line that starts with the letters “ID”. This line shows the IDs of the cards that the driver will recognize. Right now the IDs are set for the Audigy 2 cards, so you'll want to replace the entire line with the IDs of your specific card. These IDs are:

Audigy: ID = 02000051, 02000052, 02000053, 02000054, 02000055, 02000056, 02000057, 02000058

SBLive! 5.1: ID = 00801061, 00801062, 00801064, 00801069, 0080106A, 0080106B

SBLive!: ID = 00000020, 00000021, 0000002F, 00801022, 00801025, 00801026, 00801027, 00801028, 00801029, 00801031, 00801032, 00801040, 00801050

Replace the ID line with the line of your card, and save the file.

To put the modified CTComp.dat file back into data1.cab, go back into Command Prompt again and type i6comp r data1.cab ctcomp.dat to replace the original. The file was just copied in there, so CTComp.dat will still be showing in the directory. When you are done, delete the leftover CTComp.dat, i6comp.exe, and ZD51145.DLL files. Everything should now be ready to install.

Been playing around with weird old stuff.

Firstly, got a Siemens HiGraph II TIGA card working. It's pretty 'low end' as TIGA goes, an all-in-one design with just 1MB 'dual-ported DRAM' (sounds like VRAM to me) and 2MB of FP DRAM for the TMS34010, and 512kB for the WDC90C11 VGA chip.

Pretty miraculously the drivers are still on the Siemens site, although it took a bit of digging and signing up for support for industrial systems to get them. It's a package for HiGraph I, HiGraph II and CPGraf cards with TIGA 2.2 driver installer for DOS and Win3.1, and additionally contains a second Win3.1 driver (v 1.7 vs 1.6) and a Win3.0 driver. The installer script is a brain-dead mess though so had to do a lot manually. I uploaded the drivers here.

Then I got my hands on a Logitech Soundman Wave, an awful Mediavision Jazz derivative - but one with an OPL4 wavetable synth that works under DOS. This was a bit easier to install, and works as well as any Jazz-based card with severely borked volume levels can work.

Finally messing around with a Unitron EX3135 fake-cache VLB IDE controller that I've had for over a year but not actually done anything with. It's very similar to the [url=Re: Drivers for Hornet Technology VL-200 cached IDE VLB controller?]Hornet Technologyies VL-230[/]. No luck so far - I can make settings do what I want, but so far have not been able to actually get BIOS to detect any drive attached to it.

Edit:
Fixed it - turns out my 2GB DoM is the problem, or rather, it is in combination with this controller. Works fine with less 'intelligent' controllers on this motherboard, it's not a size issue (4.3GB HDD works fine), but for whatever reason it just refuses it. Oh well, another black mark against the card even though it does actually do what I could have expected.

ubiq wrote on 2024-03-05, 03:58:

Had my mITX Socket 7 on the bench today, fixing a failed D: partition. I knew it was a bad idea using a 160GB HD, but it was all I had. Used SeaTools to lock it down to 128GB total, so hopefully that helps. Also swapped out the passive-cooled Pentium 90 for a Pentium 200. I keep flip-flopping on what CPU I want in this thing - I'm playing through Descent 1 on it right now, and the P90 gives period-correct framerates, but the P200 will be "nicer". It also allows it handle some later-90's demoscene stuff. Would really like to be able to cram a full-sized HD in there, but I just don't see that's there's any more room to be found. Of course, I could dump the (600W, hah) SFX PSU for a pico, but at that point I'm really mucking with this build which I quite like in its current state. Everything fits *just* nicely. For once, I think I'll leave well enough alone! 🤓

IMG_0817.jpeg

That case is that?

Kahenraz wrote on 2024-03-05, 15:19:

I wonder if it would be easier to pop off the chip and use a TL866II to reprogram it. Are those little micro probes you have there attached to it?

I've never tried to reprogram a flash chip in-place before.

Kynar wire, the same type I'm using to do PS2 and Xbox 360 glitch chips.

I did pull it off the circuit afterwards as I couldn't get it to detect in-circuit.

Kahenraz wrote on 2024-03-05, 07:19:

Sounds awfully familiar! […]
Show full quote

ubiq wrote on 2024-03-05, 03:58:

Had my mITX Socket 7 on the bench today, fixing a failed D: partition. I knew it was a bad idea using a 160GB HD, but it was all I had. Used SeaTools to lock it down to 128GB total, so hopefully that helps. Also swapped out the passive-cooled Pentium 90 for a Pentium 200. I keep flip-flopping on what CPU I want in this thing - I'm playing through Descent 1 on it right now, and the P90 gives period-correct framerates, but the P200 will be "nicer". It also allows it handle some later-90's demoscene stuff. Would really like to be able to cram a full-sized HD in there, but I just don't see that's there's any more room to be found. Of course, I could dump the (600W, hah) SFX PSU for a pico, but at that point I'm really mucking with this build which I quite like in its current state. Everything fits *just* nicely. For once, I think I'll leave well enough alone! 🤓

Sounds awfully familiar!

Trying to figure out the cause of data corruption in Windows 9x

What a lovely machine. More angles, please!

That thread really gets in to it huh... I think my SeaTools fix (as per Phil's guide) should do the trick, as it now appears to be a 128GB drive to anything that matters.

I have a bunch of pics of it in my thread where I post all my builds.

supercordo wrote on 2024-03-05, 16:33:

That case is that?

Fractal Design Node 202

The 'lil feet are custom - I added them when I had an actual mITX system in there and was trying to get some airflow to the GPU. When that failed, I had a spare case to play with - tada!

Now to add the CRT on top... 😁

PcBytes wrote on 2024-03-05, 12:02:

The Dell OEM A2ZS (SB0358) won't detect when trying to install the standard Audigy 2 ZS drivers.
I got a EEPROM dump (thanks to @LoStSOul) and it should be as much as just writing the EEPROM that fixes it and turns it into a retail card.

Isn't installing kx drivers a much simpler and better solution?

one question. Did anyone ever build (MPX-16) the system featured in the "Circuit Cellar" section of Byte magazine, Volume 7 Number 11 , 1982 (and v7#12 , v8#1)?

appiah4 wrote on 2024-03-06, 13:22:

PcBytes wrote on 2024-03-05, 12:02:

The Dell OEM A2ZS (SB0358) won't detect when trying to install the standard Audigy 2 ZS drivers.
I got a EEPROM dump (thanks to @LoStSOul) and it should be as much as just writing the EEPROM that fixes it and turns it into a retail card.

Isn't installing kx drivers a much simpler and better solution?

Since EEPROM writers are more common now than custom drivers, my personal view is that un-restricting via reflash is the better way.

I got a batch of untested 30 pin simms for cheap - half the lot appeared to be 256kb simms but I could see less of a pin gap in the middle on some chips which should be 4MB SIMMs. Also a bunch of 60ns 1MB simms:

Set my 386 back up and had to make a cheat-sheet from TH99 for the Super-IO card to save time the next time I try to use it. Some of the RAM turned out to not work and I found after some visual inspection that the SoJ memory chips on the non-working SIMMs had some loose legs, where a chip got bashed or removed at a bad angle.

Those were re-soldered by putting fresh flux on the affected pins and heating them, then cleaning up. That got all sticks of the RAM working and passing in memtest so now I have too many 30-pin SIMMs - I also found I have one more 4x 4mb SIMM set than I thought I had. So many now that I've used these little wooden dowels to keep the SIMMs organised.
And it got me to look at one of the 'dead' sticks in the 30-pin simm box - reflowing the 1st chip on the SIMM in the picture on my USB-PD hotplate with fresh flux. That SIMM works now too 😀 I think that means I've never come across an actually chip-level faulty SIMM? Just bad solder connections...