wbc wrote on 2023-12-30, 22:22:

oh, by the way, I have built a pair of PicoGUSes a couple of weeks ago and can't stop playing with them, as they could not only work as sound card, but are basically a very capable ISA8 field-programmable devboards!

a certain spoiler here - coming soon 😉

stool not included 😁

Ooh, this is very cool. I can’t wait to try it!

This thread reminds me I am still on v0.7.0

Time to change that tonight..

I have a question or issue… I just received my PicoGUS from the Tindie store. The software compatibility is great until now. Even things like Impulse Tracker with it's weird channel allocation schemes seems to work just fine. But I have one annoying issue:
Whenever I reset my computer - whether it is with CTRL-ALT-DEL, the reset button or software command, the PicoGUS is no longer detected. When I do a reset, I can see the led on the PicoGUS blink as if it performs a reset, but the card does not respond anymore. The only way to get it working again is to shut down the power supply. Sometimes that also doesn't work. Is this a known issue?
My mainboard is a Slot-A board with the AMD 650 chipset. I don't know if it has been tested before. I will try the PicoGUS later with my 486 machine, so I don't know if the problem persists.

edit: sometimes after a reset it does work again. But not reliably.

Sounds like yen another reset line issue?

Yes, I read there were some improvements in the last update to improve the boot process. If I could help troubleshoot in any way I would gladly hear what could be done. I’m running the latest firmware, 1.0.1.

digistorm wrote on 2024-01-04, 18:22:

I have a question or issue… I just received my PicoGUS from the Tindie store. The software compatibility is great until now. Eve […]
Show full quote

I have a question or issue… I just received my PicoGUS from the Tindie store. The software compatibility is great until now. Even things like Impulse Tracker with it's weird channel allocation schemes seems to work just fine. But I have one annoying issue:
Whenever I reset my computer - whether it is with CTRL-ALT-DEL, the reset button or software command, the PicoGUS is no longer detected. When I do a reset, I can see the led on the PicoGUS blink as if it performs a reset, but the card does not respond anymore. The only way to get it working again is to shut down the power supply. Sometimes that also doesn't work. Is this a known issue?
My mainboard is a Slot-A board with the AMD 650 chipset. I don't know if it has been tested before. I will try the PicoGUS later with my 486 machine, so I don't know if the problem persists.

edit: sometimes after a reset it does work again. But not reliably.

I don't think it's an issue with the reset line since you say the issue happens on Ctrl-Alt-Del as I've never seen a motherboard that toggles ISA RESET on Ctrl-Alt-Del, but there may be a chance that this board actually does. Does the LED on the PicoGUS blink when doing Ctrl-Alt-Del? When the PicoGUS 2.0 powers on or is reset, the LED sequence should go On -> Off -> On, and then once the card receives valid commands from the bus the LED will toggle on/off. Is the LED on or off when the card is unresponsive? Also: does this only happen on GUS firmware? If the card gets "stuck" with the LED off and this only happens on GUS firmware, it may be a known issue that affects maybe 5% of boards and I'll need to send you a new one. I test all boards I send out for this issue, but there's a chance I missed this one. I'm currently working on a firmware fix for this issue, but I don't want you to have to wait for that if that's what you're experiencing.

If it's not the above issue, I'm interested to hear how your PicoGUS works in your 486 machine to see if it's any different. I haven't tested with any AMD 650 chipset boards (but getting a Slot A system is on my bucket list). This is a bit of an aside, but one thing I've noticed is that very fast systems sometimes don't play very well with PicoGUS or real GUS. My fastest Super Socket 7 computers get "odd".

I did quite some testing but I am worried about the results…

My Slot-A motherboard does indeed make the PicoGUS do a reset when I press CTRL-ALT-DEL or use a software reset command (REBOOT.COM)
On my Slot-A motherboard, whenever the PicoGUS resets, it blinks 3 times (consistently)!
I did a lot of tests with the GUS firmware loaded, and the results are, as I said, a bit troubling:
First time I power it on, it almost always works.
Whenever I press the reset button, it flashes 3 times. Most of the times the LED goes off and it doesn't work. Sometimes it stays on and it does work. But the more I do a reset, the less it will work (and the LED stays off). I never got it to work by pressing CTRL-ALT-DEL, but this was later in the testing when the chance of working was already decreasing, and the system only responds to CTRL-ALT-DEL after the POST so it takes a long time to test. Pressing reset can be done much quicker. Power cycling does not seem to make a difference once it refuses to start up, it seems. Or at least the success rate is the same as with pressing the reset button.

Then I decided to flash the CMS firmware (once I got a successful boot), because I never had such a card in the past. The flashing succeeded, but I got this message:
Programming complete. Waiting for the card to reboot...
ERROR: card is not alive after programming firmware

I tried to reset my system 20 times, and it didn't come alive. Now I got really worried! I also discovered that I don't have any functioning USB micro cables (they only seem to be for charging) so I cannot recover the firmware at this moment.

Then I took out the 486 and put the PicoGUS in there. This was after I had some dinner, so some time has passed.
When I turned it on, the LED flashed 1 time (as it should) and the LED stayed on. And what do you know? CMS mode worked.
But then I started pressing reset to try to reboot, and none of the times did I get it to work again! Power cycle also didn't help. CTRL-ALT-DEL does not trigger the reset on this motherboard (SiS 85C471 chipset) and software reboot also does not trigger reset.
Power cycling again did not help.

After writing this text, I powered the 486 on again, and it did not boot (LED stayed off). Then I pushed the reset button, and now it started correctly. The first try (when it worked first time on the 486) the ISA bus was still running at 10 MHz. After it failed, I set it to 8 MHz, but that didn't change anything (for CMS mode). O, yeah, on the Slot-A board, the ISA bus is running at 8.33 MHz and when it worked, GUS software seemed to work just fine.

I then flashed the firmware back to GUS mode, which succeeded but again gave an error as it didn't want to boot. Then after 4 reset presses I saw the LED stay on and it works as a GUS again.

Does this provide enough information?

digistorm wrote on 2024-01-05, 20:14:

I did quite some testing but I am worried about the results… […]
Show full quote

I did quite some testing but I am worried about the results…

My Slot-A motherboard does indeed make the PicoGUS do a reset when I press CTRL-ALT-DEL or use a software reset command (REBOOT.COM)
On my Slot-A motherboard, whenever the PicoGUS resets, it blinks 3 times (consistently)!
I did a lot of tests with the GUS firmware loaded, and the results are, as I said, a bit troubling:
First time I power it on, it almost always works.
Whenever I press the reset button, it flashes 3 times. Most of the times the LED goes off and it doesn't work. Sometimes it stays on and it does work. But the more I do a reset, the less it will work (and the LED stays off). I never got it to work by pressing CTRL-ALT-DEL, but this was later in the testing when the chance of working was already decreasing, and the system only responds to CTRL-ALT-DEL after the POST so it takes a long time to test. Pressing reset can be done much quicker. Power cycling does not seem to make a difference once it refuses to start up, it seems. Or at least the success rate is the same as with pressing the reset button.

Then I decided to flash the CMS firmware (once I got a successful boot), because I never had such a card in the past. The flashing succeeded, but I got this message:
Programming complete. Waiting for the card to reboot...
ERROR: card is not alive after programming firmware

I tried to reset my system 20 times, and it didn't come alive. Now I got really worried! I also discovered that I don't have any functioning USB micro cables (they only seem to be for charging) so I cannot recover the firmware at this moment.

Then I took out the 486 and put the PicoGUS in there. This was after I had some dinner, so some time has passed.
When I turned it on, the LED flashed 1 time (as it should) and the LED stayed on. And what do you know? CMS mode worked.
But then I started pressing reset to try to reboot, and none of the times did I get it to work again! Power cycle also didn't help. CTRL-ALT-DEL does not trigger the reset on this motherboard (SiS 85C471 chipset) and software reboot also does not trigger reset.
Power cycling again did not help.

After writing this text, I powered the 486 on again, and it did not boot (LED stayed off). Then I pushed the reset button, and now it started correctly. The first try (when it worked first time on the 486) the ISA bus was still running at 10 MHz. After it failed, I set it to 8 MHz, but that didn't change anything (for CMS mode). O, yeah, on the Slot-A board, the ISA bus is running at 8.33 MHz and when it worked, GUS software seemed to work just fine.

I then flashed the firmware back to GUS mode, which succeeded but again gave an error as it didn't want to boot. Then after 4 reset presses I saw the LED stay on and it works as a GUS again.

Does this provide enough information?

Thank you very much for the detailed report! From what you describe the RP2040 is not stable after overclock even on non-GUS firmware. I'll do a special build with less of an overclock and DM the firmware to you to see if it fixes the issue. Part of what I want to work on this weekend is optimizing code so I don't need to overclock as much. I test all PicoGUSes to ensure they're stable but it seems like yours is not behaving...

Hi polpo im having some odd behavior with the v2 card with some games like doom with GUS emulation. The led flashes so it processing but no audio comes out. Some games work like epic pinball.
I swapped in my v1 board and doom worked fine.
Edit I am using latest firmware

Pickle wrote on 2024-01-06, 02:08:

Hi polpo im having some odd behavior with the v2 card with some games like doom with GUS emulation. The led flashes so it processing but no audio comes out. Some games work like epic pinball.
I swapped in my v1 board and doom worked fine.
Edit I am using latest firmware

Hmm, I haven't heard of something like this. My first hunch is somehow DMA is involved, since Doom uses DMA and Epic Pinball does not. Do you use the same DMA setting on both cards? What chipset do you have?

I might be having the same issue. I'll have to try out Epic Pinball but so far no games (Descent, Doom v1.9, Rise of the Triad) either detect the card or play anything through it. UFO: Enemy Unknown/X-COM: UFO Defense tried to do something but it was more like garbled noise.

Of the demos I tried only Second Reality and Paper had music. Dope and Out of Sight Out Of Mind started but I didn't hear anything.

At the moment the PicoGUS v2 (from the latest Tindie batch) is the only sound card in the system, a 486 with a SiS 85C496/497 chipset (the one that has firmware flashing issues from DOS). I've tried both DMA 1 and 3.

I don't know if you already checked (but it wouldn't hurt)… at what IRQ do you have the PicoGUS set and are you sure it is not used elsewhere?

The IRQ isn't taken (PicoGUS is set to 5) but the 240h I/O port was by the PNP Ethernet card (oops). Unfortunately it doesn't solve all the issues.

Rise of the Triad detects the card now but only plays sound effects (no music) and Doom and the Dope demo are still silent. UFO plays more normal tones now but I might be running the wrong version of the game for full GUS support, so I'll forget about that one for now.

I tried Epic Pinball and it does work, as does the intro music of Pinball Fantasies (then the demo version crashes but the same happens when it's set to "no sound", might be a video card thing.)

Did you install the Ultrasnd files and set the correct environment variables in AUTOEXEC.BAT? Some of those games need ULTRAMID to be working and the correct path to be set.

Yes to both, I took the v4.11 drivers from the wiki page (installed to C:\ULTRASND) and set the environment variables.

Just now I did get sound in Doom (no music) but after I rebooted it was gone again. Nothing had been changed before that.

I have also tried IRQ7 btw but since it didn't improve things I put it back to the recommended IRQ5.

polpo wrote on 2024-01-07, 06:38:

Pickle wrote on 2024-01-06, 02:08:

Hi polpo im having some odd behavior with the v2 card with some games like doom with GUS emulation. The led flashes so it processing but no audio comes out. Some games work like epic pinball.
I swapped in my v1 board and doom worked fine.
Edit I am using latest firmware

Hmm, I haven't heard of something like this. My first hunch is somehow DMA is involved, since Doom uses DMA and Epic Pinball does not. Do you use the same DMA setting on both cards? What chipset do you have?

My board is called MV035F and has an OPTI chipset
https://theretroweb.com/motherboards/s/edom-wintech-mv035

My v2 jumpers are set the same as the V1 with IRQ 5 and DMA 3. The other sound card is set to IRQ 7 and DMA 1. PGUSINIT does not report any issues and its quite odd i can switch them and the V1 works.
With the V2 led flashing id think the I/O is working.
Anything else you like me to run or try?

I upgraded my v1.x board to firmware 1.0.1 last night and everything is working smoothly..

Pickle wrote on 2024-01-07, 22:33:

My board is called MV035F and has an OPTI chipset https://theretroweb.com/motherboards/s/edom-wintech-mv035 […]
Show full quote

polpo wrote on 2024-01-07, 06:38:

Pickle wrote on 2024-01-06, 02:08:

Hi polpo im having some odd behavior with the v2 card with some games like doom with GUS emulation. The led flashes so it processing but no audio comes out. Some games work like epic pinball.
I swapped in my v1 board and doom worked fine.
Edit I am using latest firmware

Hmm, I haven't heard of something like this. My first hunch is somehow DMA is involved, since Doom uses DMA and Epic Pinball does not. Do you use the same DMA setting on both cards? What chipset do you have?

My board is called MV035F and has an OPTI chipset
https://theretroweb.com/motherboards/s/edom-wintech-mv035

My v2 jumpers are set the same as the V1 with IRQ 5 and DMA 3. The other sound card is set to IRQ 7 and DMA 1. PGUSINIT does not report any issues and its quite odd i can switch them and the V1 works.
With the V2 led flashing id think the I/O is working.
Anything else you like me to run or try?

I was just checking to see if the LED on my board was flashing as well, to see if our cases are similar, and what do you know? I suddenly got sound and music in all games that didn't work yesterday (Doom, Doom 2, Rise of the Triad). I forgot to try the demos.

I did not change a single thing since my tests yesterday, I swear. It also kept working after rebooting and power cycling the machine.

PicoGUS settings: DMA 1, IRQ 5

The only thing I can think of that is different, and apologies if it's a silly thought, is that today it's icily cold in the room where that computer is.

This might be (my latest) dumb question, but does there exist large-scale testing of Picos being overclocked?

E.g. 100+ random Picos, all overclocked and analysed, for statistical results?

I was just idly wondering if perhaps 1 out of 100, or whatever, might not like the overclock. This seems less likely, given that I understand the PicoGUS's overclock of the RP2040 is mild, and anecdotally, people say the RP2040 is usually capable of rather large overclocks - but just thought I'd ask the broad question, on the basis that in my experience of PC overclocking, whenever we experience weird behaviour, we always look first to the overclock, in case it is exacerbating something.

A test firmware without the overclock might be another angle from which to check that speculation, in troublesome instances.

So I had full sound on everything (including in the "Dope" demo by Complex that was silent before), then I lost it after trying the "Out Of Sight Out Of Mind" demo, and got it back after cycling the power for the second time.

Then while playing Doom the sound and music disappeared once the game loaded E1M2, after playing through E1M1 and the listening to the intermission screen for a bit.

As with Pickle, the LED on the board was flickering happily and Epic Pinball still has music and sound.

I'm completely stumped. Polpo, do you have any idea what it could be?

Summary of my specs / situation:

• SiS 85C496/497 chipset (Asus 486SP3 motherboard)
• Am5x86 (486) at 133 MHz (33MHz bus speed)
• PicoGUS 2.0 from Tindie, firmware 1.0.1, PGUSINIT 2.0.0
• UltraSound drivers v4.1.1 as linked in the wiki
• DMA 1, IRQ 5
• Running PGUSINIT without any flags but also tried "/d 8" with Doom, as suggested in the wiki

It's also still damn cold in that room so it's certainly not a temperature issue 😜