Besides the BLASTER environment variable, is it possible you have been messing around in the BIOS since it (completely) worked last and now?

jakethompson1 wrote on 2022-08-04, 00:05:

... is it possible you have been messing around in the BIOS since it (completely) worked last and now?

No. Haven't touched bios for the past month at least.

jakethompson1 wrote on 2022-08-04, 00:05:

Besides the BLASTER environment variable...

Sorry, edited the original post. Blaster is set according to jumpers, did not touch them since I got card working in games.

Tried to use different isa slot, did not help.

Basically when things stop working without any reason that we may have caused it can mean:
- cold solder joints
- bad capacitors
- corrosion on connecting elements
- too high voltage from source (in this case it would mean motherboard - but I doubt something went through ISA slot)

So probably it just needs some restoration.

varrol wrote on 2022-08-04, 13:50:

Basically when things stop working without any reason that we may have caused it can mean: - cold solder joints - bad capacitors […]
Show full quote

Basically when things stop working without any reason that we may have caused it can mean:
- cold solder joints
- bad capacitors
- corrosion on connecting elements
- too high voltage from source (in this case it would mean motherboard - but I doubt something went through ISA slot)

So probably it just needs some restoration.

Thanks! Do you have any hints what to look for? I coudn't see or smell anything off. How would one go about checking the capacitors for example?

Not sure, but I think on that card FM synth is handled by an external chip (internet seems to think a DXP44Q). If that's right, then it might be worth pressing on that chip whilst running the Duke3D sound test, in case one of the solder joints has failed. A photo of the card might help.

snufkin wrote on 2022-08-04, 16:10:

Not sure, but I think on that card FM synth is handled by an external chip (internet seems to think a DXP44Q). If that's right, then it might be worth pressing on that chip whilst running the Duke3D sound test, in case one of the solder joints has failed. A photo of the card might help.

I have reverse engineered the DXP44Q it’s a YMF262 and a YAC512 in one chip. 👍 All pins from both Chips can be found on the DXP .

snufkin wrote on 2022-08-04, 16:10:

... A photo of the card might help.

Thanks, here you go. Also updated original post.

https://imgur.com/a/Qn11fHL

Sucho wrote on 2022-08-04, 18:10:

snufkin wrote on 2022-08-04, 16:10:

... A photo of the card might help.

Thanks, here you go. Also updated original post.

https://imgur.com/a/Qn11fHL

This is a clone OPL3 (YMF262) with integrated DAC (YAC512) as i mentioned before.
I really like the slightly janky offbrand clones. 😁

I have started do make a pinout, but i have to make some measurements to perfectly lable the missing pins.

Super interesting: The blue labeled ones on the DXP44Q are the connections between the YAC512 and YMF262 ... they go from one pin to another on the DXP.
I love the weirdness 😁
the "44Q" is most likely saying 44 Pins / Quad flat package chip.

I dont think the DXP is broken. It is controlled by the ESS688F.
(Sidenote The ESS688FZ has an integrated FM "emulation") This information is not relevant for you, i just mentioned it for other people searching for informations. 😀

DerBaum wrote on 2022-08-04, 18:23:

(Sidenote The ESS688FZ has an integrated FM "emulation") This information is not relevant for you, i just mentioned it for other people searching for informations. 😀

and this exact Pinetech PT-201 card came in two variants, with DXP44Q and without it as evidenced by google image search

DerBaum wrote on 2022-08-04, 18:23:

This is a clone OPL3 (YMF262) with integrated DAC (YAC512) as i mentioned before. I really like the slightly janky offbrand clo […]
Show full quote

Sucho wrote on 2022-08-04, 18:10:

snufkin wrote on 2022-08-04, 16:10:

... A photo of the card might help.

Thanks, here you go. Also updated original post.

https://imgur.com/a/Qn11fHL

This is a clone OPL3 (YMF262) with integrated DAC (YAC512) as i mentioned before.
I really like the slightly janky offbrand clones. 😁

I have started do make a pinout, but i have to make some measurements to perfectly lable the missing pins.

dxp44q(NOT READY).JPG

Super interesting: The blue labeled ones on the DXP44Q are the connections between the YAC512 and YMF262 ... they go from one pin to another on the DXP.
I love the weirdness 😁
the "44Q" is most likely saying 44 Pins / Quad flat package chip.

I dont think the DXP is broken. It is controlled by the ESS688F.
(Sidenote The ESS688FZ has an integrated FM "emulation") This information is not relevant for you, i just mentioned it for other people searching for informations. 😀

Is it, the DXP, directly coupled to the ESS688F or is it using the 74F244 visible on the photo ?
edit: nevermind, looking at the photos of the card without DXP, it's obvious the F244 is used for something else.

Sucho wrote on 2022-08-03, 23:58:

It worked just fine during the week, but after starting pc today, sound blaster does not show up in Wolf3D.

Have you changed something in the BIOS? Like PNP settings?

I have tested something:
I have a Toshiba satellite pro 430 Laptop. It also has a ESS688.
I can boot without any drivers or settings (Safe mode command prompt) and run Duke3d with sound and music.
I dont know if it works bacause it is a laptop and you can set Address IRQ and DMA in the BIOS! But it works.

Have you tried Duke3D? It initializes audio hardware.
I also have a Thinkpad wich only produces sound in dos games if i run duke3d first...

Just some ideas...

weedeewee wrote on 2022-08-04, 19:33:

Is it, the DXP, directly coupled to the ESS688F or is it using the 74F244 visible on the photo ?
edit: nevermind, looking at the photos of the card without DXP, it's obvious the F244 is used for something else.

This may be interesting:

Here are some of my reverse enginnering results.
DXP is connected exactly like a YMF262 (directly to the control chip).

The missing chip on top is YAC512.
The missing chip in the center is YMF262.
The missing chip on the bottom is DXP44Q.

DerBaum wrote on 2022-08-04, 19:46:

Sucho wrote on 2022-08-03, 23:58:

It worked just fine during the week, but after starting pc today, sound blaster does not show up in Wolf3D.

Have you changed something in the BIOS? Like PNP settings? ...

I can check if there are any PnP settings in bios, but sound card itself is set up with jumpers. I did no changes in bios after putting it into system. It ran fine for a week. Then suddenly this weird behavior.

Checked bios, PnP is not mentioned anywhere. Also, tried to push on dxp chip while running duke3d music test, but still same no fm chip found message.
Also tried to boot duke3d up with music card set up as both sound blaster and adlib, but it fails on music initialization.
If I choose music card none, game runs fine.

P.S. thanks for all the info, really interesting stuff.
Also, was wondering what am I missing with that ess688fz missing on my chip. Great stuff.

Sucho wrote on 2022-08-04, 21:56:

Checked bios, PnP is not mentioned anywhere. Also, tried to push on dxp chip while running duke3d music test, but still same no fm chip found message.
Also tried to boot duke3d up with music card set up as both sound blaster and adlib, but it fails on music initialization.
If I choose music card none, game runs fine.

The strange thing is that it just stopped working...

You can try to Recap the board... Old 90´s caps sometimes let hardware do weird stuff (when they dry out for example).
This is my last idea.

Try it in another machine?

Otherwise check for damage, ensure the underside of board isn't shorting to another surface, etc. Then recap and / or check traces for continuity or broken solder joints.

https://github.com/hkzlab/ES688_ISA8/blob/mas … 8_Datasheet.pdf says OPL is not going thru ESS, but is parallel to it, only Address decoding/chip select is handled by ESS
if you are electronically inclined you could slap an oscilloscope on FMCSB, FMCLK and FMENB245, on DAC output, and on traces in between identified by DerBaum edit: the crossed ones arent important if adlib is not even detected