Not sure if anyone else experienced this but I had some trouble with Doom 2 SFX using SET BLASTER=A220 I7 D1 P330 T4, changing T4 to T2 worked for me.
That's surprising, given that T2 means SoundBlaster 1.5 and T4 means SoundBlaster Pro (which is what the CS chip is compatible with), AFAIK.
But whatever works for you I guess! 😀 What about other games? No issues with the T4 there, I presume?
Not sure if anyone else experienced this but I had some trouble with Doom 2 SFX using SET BLASTER=A220 I7 D1 P330 T4, changing T4 to T2 worked for me.
That's surprising, given that T2 means SoundBlaster 1.5 and T4 means SoundBlaster Pro (which is what the CS chip is compatible with), AFAIK.
But whatever works for you I guess! 😀 What about other games? No issues with the T4 there, I presume?
So far the only other game I've tried was Monkey Island 1. No issue with T4 with MI1, but I'm aiming to taking some time to do a deeper dive with other games.
@Eivind, I am trying to get the Yamaha Wavetable to work on the TL2 Wavetable adaptor and I need to have the +-12v for it. Any chance you can share the board with me so I can make those changes
@Eivind, I am trying to get the Yamaha Wavetable to work on the TL2 Wavetable adaptor and I need to have the +-12v for it. Any chance you can share the board with me so I can make those changes
Yeah no problem, I added the wavetable kicad files to the github repo.
I got the +-12v working, but still no sound. Anyone got some experience with DB50XG and what is takes? Should I buffer up the midi signal from 3v3 to 5 ("TTL")? I doubt that makes any difference, however I need seeing other people put an optocoupler like 6N137 on that line. Any suggestions?
I got the +-12v working, but still no sound. Anyone got some experience with DB50XG and what is takes? Should I buffer up the midi signal from 3v3 to 5 ("TTL")? I doubt that makes any difference, however I need seeing other people put an optocoupler like 6N137 on that line. Any suggestions?
I don't have experience with the DB50XG, but if you're concerned about the voltage level of the MIDI signal, you could try desoldering R29 and R30 (voltage divider from 5V to 3.3V) and bridging the left-hand side of each footprint together. Note that this will feed any Raspberry Pi you connect with a voltage level it might not like...
You shouldn't need to bother with an optocoupler, that is for electrically isolating two devices running from separate power rails.
Sweet. Got it working! Thank you Eivind
First I tried with removing R30 and R29 and things worked. But then I put the resistors back and added two NPN-Res (Q2 and R28, R29 from TL2 BOM) inverters back to back and things now work for both the Pi and DB50XR.
Conclusion, DB50XR requires the -/+12 (PDME2-S5-D12 and three 10uf 25v Caps from TL2 BOM as C33) and TTL (5v) logic for the MIDI in (with 3v3 level Midi-in the card cannot operate).
Here is what I did on the PCB and I will try to reflect that on the KiCad.
Also, if you wonder why I had to use two NPN (or N-Ch) transistors rather than a single one for 3v3->5v level shift was due to the fact that 3.3v is not available after the switch (switch between WT and RPi) and only 5v is available. Therefore, I found that the easiest way would be the two inverter method.
So far the only other game I've tried was Monkey Island 1. No issue with T4 with MI1, but I'm aiming to taking some time to do a deeper dive with other games.
Could you please try Elder Scrolls: Arena? On MS-DOS 6.22, I can't get the game to launch.
On FreeDOS, it locks after character creation.
I'm booting from MS-DOS 6.22 or PC-DOS, They stops with an EMM386 error.
This is probably EMM386 trying to use a part of memory that's not free.
You didn't provide the relevant config.sys parameters used to start EMM386, so I don't know what it's doing.
Maybe try experimenting with the X=mmmm-nnnn parameter and see if that helps?
Btw, tuning memory managers can be quite the challenge, on any system! 😀 There should be plenty of references here on the forum on how to tackle this.
Eivindwrote on 2023-09-22, 15:20:This is probably EMM386 trying to use a part of memory that's not free.
You didn't provide the relevant config.sys parameters us […] Show full quote
I'm booting from MS-DOS 6.22 or PC-DOS, They stops with an EMM386 error.
This is probably EMM386 trying to use a part of memory that's not free.
You didn't provide the relevant config.sys parameters used to start EMM386, so I don't know what it's doing.
Maybe try experimenting with the X=mmmm-nnnn parameter and see if that helps?
Btw, tuning memory managers can be quite the challenge, on any system! 😀 There should be plenty of references here on the forum on how to tackle this.
Thanks for your advice.
I'll look into emm386 more, I think there's some hope of solving the problem.
Also, one more error is happening on 'monkey island' & 'indy3' -lucasfilm games-.
When I run unisound 0.7 , I get this error message.
monkey island :
run-time error R6003
- integer divide by 0
Indy3:
devide error
If I don't run unisound and let the sound come out of my PC speakers, the game runs fine.
On my other vortex86 PC104 computer (wee486 clone), I have no problems at all.
Also, one more error is happening on 'monkey island' & 'indy3' -lucasfilm games-.
When I run unisound 0.7 , I get this error message.
monkey island :
run-time error R6003
- integer divide by 0
Indy3:
devide error
If I don't run unisound and let the sound come out of my PC speakers, the game runs fine.
On my other vortex86 PC104 computer (wee486 clone), I have no problems at all.
These games are notoriously fickle about speed and sound. You didn't say how you launched them / if you used Adlib or MT-32.
I haven't tried Indiana Jones in ages, but for monkey island 1 I first run softmpu;
1SOFTMPU /MPU:330 /SB:220 /IRQ:7
and then launch the game in VGA mode, with mouse support and MT-32 like this:
1MONKEY V MO R
You could also experiment with turning down the CPU speed and disabling the L1 cache in the BIOS.
Eivindwrote on 2023-09-23, 07:38:These games are notoriously fickle about speed and sound. You didn't say how you launched them / if you used Adlib or MT-32.
I h […] Show full quote
Also, one more error is happening on 'monkey island' & 'indy3' -lucasfilm games-.
When I run unisound 0.7 , I get this error message.
monkey island :
run-time error R6003
- integer divide by 0
Indy3:
devide error
If I don't run unisound and let the sound come out of my PC speakers, the game runs fine.
On my other vortex86 PC104 computer (wee486 clone), I have no problems at all.
These games are notoriously fickle about speed and sound. You didn't say how you launched them / if you used Adlib or MT-32.
I haven't tried Indiana Jones in ages, but for monkey island 1 I first run softmpu;
1SOFTMPU /MPU:330 /SB:220 /IRQ:7
and then launch the game in VGA mode, with mouse support and MT-32 like this:
1MONKEY V MO R
You could also experiment with turning down the CPU speed and disabling the L1 cache in the BIOS.
Thank you for your quick and kind attention and advice.
I finally have my beloved Lucasfilm dos game running well.
Just as you advised...
I turned off the L1 cache, lowered the speed to 100Mhz and it runs fine.
I'm guessing that the different characteristics of my PC104 Vortex86DX and Tinyllama's Vortex86EX might be the cause.
It's yet again the same GD-750 mini pcie video card, but it seems to be in stock and available for a good price. The only issue is that it's alibaba and not aliexpress, making it business only iirc 🙁
Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructions on the GitHub; unplug the HID_POWER jumper, disconnect power from the board, hold PRG until the device shows up in the WCH software but that never worked. I tried several times holding the button for up to 60 seconds but it never showed up.
Just to test things I put the HID_POWER jumper back and turned the board on with the USB A to A cable still plugged in and the chip showed up in the WCH software but when I tried to write the hidman binary file it would fail and then the chip would disappear from the software until I power cycled.
I tried several times but it keeps giving me errors and not writing to the chip.
In my haste, I skipped to the BIOS step and that was successful but now no keyboards I plug in will work since I’m assuming the firmware is looking for keyboards on the CH559 chip and not the onboard USB on the SOM.
I’m wondering if I did something wrong somewhere or even overheated the chip when I soldered it on as I used hot air.
Anyone have any suggestions? Also, is there a way to fail back to the native BIOS in the meantime? You know, so I can use a keyboard again? 😉
Thanks in advance!
EDIT:
I reviewed my board and the schematics. It looks like I had a 3.3k ohm resistor at C32 instead of a 3.3uf CAPACITOR. I ordered the right caps but in the meantime I put a 1uf cap on it to test and it let me flash the binary file. Unfortunately I still don’t have keyboard on either the HID or native USB ports. No lights on num lock or anything at power on. Does anyone know what else I should check? USB boot works on the native USB ports FYI.
snipe3687wrote on 2024-04-05, 14:35:Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructio […] Show full quote
Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructions on the GitHub; unplug the HID_POWER jumper, disconnect power from the board, hold PRG until the device shows up in the WCH software but that never worked. I tried several times holding the button for up to 60 seconds but it never showed up.
Just to test things I put the HID_POWER jumper back and turned the board on with the USB A to A cable still plugged in and the chip showed up in the WCH software but when I tried to write the hidman binary file it would fail and then the chip would disappear from the software until I power cycled.
I tried several times but it keeps giving me errors and not writing to the chip.
In my haste, I skipped to the BIOS step and that was successful but now no keyboards I plug in will work since I’m assuming the firmware is looking for keyboards on the CH559 chip and not the onboard USB on the SOM.
I’m wondering if I did something wrong somewhere or even overheated the chip when I soldered it on as I used hot air.
Anyone have any suggestions? Also, is there a way to fail back to the native BIOS in the meantime? You know, so I can use a keyboard again? 😉
Thanks in advance!
EDIT:
I reviewed my board and the schematics. It looks like I had a 3.3k ohm resistor at C32 instead of a 3.3uf CAPACITOR. I ordered the right caps but in the meantime I put a 1uf cap on it to test and it let me flash the binary file. Unfortunately I still don’t have keyboard on either the HID or native USB ports. No lights on num lock or anything at power on. Does anyone know what else I should check? USB boot works on the native USB ports FYI.
OK edit number two I was able to get everything working by adding that capacitor and checking the circuit on the USB I found I wasn’t getting 5 V due to one of the fuses, not being soldered all the way, which would explain why I needed to provide power to get the chip to flash also had the same issue with the top port so both of them were not getting power for various reasons now I’ve moved onto other issues the one I’ve spent several hours on at this point is the MT 32 PI I’m not getting sound playback on the I 2S bus. it was working on PWM but after fiddling with it some more now that’s not working either so I’m guessing it’s a software issue, but I can’t seem to figure out what it is. Is there anyone who knows how to troubleshoot that further?
snipe3687wrote on 2024-04-05, 14:35:Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructio […] Show full quote
Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructions on the GitHub; unplug the HID_POWER jumper, disconnect power from the board, hold PRG until the device shows up in the WCH software but that never worked. I tried several times holding the button for up to 60 seconds but it never showed up.
Just to test things I put the HID_POWER jumper back and turned the board on with the USB A to A cable still plugged in and the chip showed up in the WCH software but when I tried to write the hidman binary file it would fail and then the chip would disappear from the software until I power cycled.
I tried several times but it keeps giving me errors and not writing to the chip.
In my haste, I skipped to the BIOS step and that was successful but now no keyboards I plug in will work since I’m assuming the firmware is looking for keyboards on the CH559 chip and not the onboard USB on the SOM.
I’m wondering if I did something wrong somewhere or even overheated the chip when I soldered it on as I used hot air.
Anyone have any suggestions? Also, is there a way to fail back to the native BIOS in the meantime? You know, so I can use a keyboard again? 😉
Thanks in advance!
EDIT:
I reviewed my board and the schematics. It looks like I had a 3.3k ohm resistor at C32 instead of a 3.3uf CAPACITOR. I ordered the right caps but in the meantime I put a 1uf cap on it to test and it let me flash the binary file. Unfortunately I still don’t have keyboard on either the HID or native USB ports. No lights on num lock or anything at power on. Does anyone know what else I should check? USB boot works on the native USB ports FYI.
OK edit number two I was able to get everything working by adding that capacitor and checking the circuit on the USB I found I wasn’t getting 5 V due to one of the fuses, not being soldered all the way, which would explain why I needed to provide power to get the chip to flash also had the same issue with the top port so both of them were not getting power for various reasons now I’ve moved onto other issues the one I’ve spent several hours on at this point is the MT 32 PI I’m not getting sound playback on the I 2S bus. it was working on PWM but after fiddling with it some more now that’s not working either so I’m guessing it’s a software issue, but I can’t seem to figure out what it is. Is there anyone who knows how to troubleshoot that further?
Make sure you have the gain for the Line In on the Autoexec.bat file.
snipe3687wrote on 2024-04-05, 14:35:Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructio […] Show full quote
Hey guys, just finished building my TinyLlama v2 board but I’m having problems with the CH559.
I tried following the instructions on the GitHub; unplug the HID_POWER jumper, disconnect power from the board, hold PRG until the device shows up in the WCH software but that never worked. I tried several times holding the button for up to 60 seconds but it never showed up.
Just to test things I put the HID_POWER jumper back and turned the board on with the USB A to A cable still plugged in and the chip showed up in the WCH software but when I tried to write the hidman binary file it would fail and then the chip would disappear from the software until I power cycled.
I tried several times but it keeps giving me errors and not writing to the chip.
In my haste, I skipped to the BIOS step and that was successful but now no keyboards I plug in will work since I’m assuming the firmware is looking for keyboards on the CH559 chip and not the onboard USB on the SOM.
I’m wondering if I did something wrong somewhere or even overheated the chip when I soldered it on as I used hot air.
Anyone have any suggestions? Also, is there a way to fail back to the native BIOS in the meantime? You know, so I can use a keyboard again? 😉
Thanks in advance!
EDIT:
I reviewed my board and the schematics. It looks like I had a 3.3k ohm resistor at C32 instead of a 3.3uf CAPACITOR. I ordered the right caps but in the meantime I put a 1uf cap on it to test and it let me flash the binary file. Unfortunately I still don’t have keyboard on either the HID or native USB ports. No lights on num lock or anything at power on. Does anyone know what else I should check? USB boot works on the native USB ports FYI.
OK edit number two I was able to get everything working by adding that capacitor and checking the circuit on the USB I found I wasn’t getting 5 V due to one of the fuses, not being soldered all the way, which would explain why I needed to provide power to get the chip to flash also had the same issue with the top port so both of them were not getting power for various reasons now I’ve moved onto other issues the one I’ve spent several hours on at this point is the MT 32 PI I’m not getting sound playback on the I 2S bus. it was working on PWM but after fiddling with it some more now that’s not working either so I’m guessing it’s a software issue, but I can’t seem to figure out what it is. Is there anyone who knows how to troubleshoot that further?
Make sure you have the gain for the Line In on the Autoexec.bat file.
Thanks for the reply!
It actually ended up being something hardware related after all. I wired up a Pi 4 with jumper wires and everything worked perfectly but no signal from the system on the MIDI line. I took one of the jumper wires I had connected between the Pi 4 and the TinyLlama and poked the MIDI signal line coming directly off the sound chip and I got sound. I didn't do this for long because I believe the raw signal is outputing 5v and the Pi is expecting 3.3v on that line but it was enough for me to figure out that R29 was not making contact despite seemingly being soldered completely. I popped it off, cleaned the pads and resoldered it giving it an extra 30 seconds of heat and a good amount of flux. this seems to have reflowed everything properly and now it's working great!
I was pulling my hair out for a bit on that one! Part of the fun I guess. 😀
snipe3687wrote on 2024-04-15, 19:46:Thanks for the reply!
It actually ended up being something hardware related after all. I wired up a Pi 4 with jumper wires and […] Show full quote
OK edit number two I was able to get everything working by adding that capacitor and checking the circuit on the USB I found I wasn’t getting 5 V due to one of the fuses, not being soldered all the way, which would explain why I needed to provide power to get the chip to flash also had the same issue with the top port so both of them were not getting power for various reasons now I’ve moved onto other issues the one I’ve spent several hours on at this point is the MT 32 PI I’m not getting sound playback on the I 2S bus. it was working on PWM but after fiddling with it some more now that’s not working either so I’m guessing it’s a software issue, but I can’t seem to figure out what it is. Is there anyone who knows how to troubleshoot that further?
Make sure you have the gain for the Line In on the Autoexec.bat file.
Thanks for the reply!
It actually ended up being something hardware related after all. I wired up a Pi 4 with jumper wires and everything worked perfectly but no signal from the system on the MIDI line. I took one of the jumper wires I had connected between the Pi 4 and the TinyLlama and poked the MIDI signal line coming directly off the sound chip and I got sound. I didn't do this for long because I believe the raw signal is outputing 5v and the Pi is expecting 3.3v on that line but it was enough for me to figure out that R29 was not making contact despite seemingly being soldered completely. I popped it off, cleaned the pads and resoldered it giving it an extra 30 seconds of heat and a good amount of flux. this seems to have reflowed everything properly and now it's working great!
I was pulling my hair out for a bit on that one! Part of the fun I guess. 😀
I am glad you got it working. I had a similar issue with the Yamaha wavetable that was expecting a certain level of voltage to register the right values on the digital input side. Anyways, enjoy your TL
