Maelgrum wrote on 2023-10-07, 21:31:

Something like AT89C51ID2-SLSUM or AT89C51ED2-SLSUM, but pins compatibility shold be checked.

then I think 'AT89C51RD2' is another candidate - easier to find and cheaper to buy.
[EDIT] 'AT89C51RC2' is even cheaper...

S95sedan, do you think the clicking is improved in this test build?

maxtherabbit wrote on 2023-10-07, 21:46:

S95sedan, do you think the clicking is improved in this test build?

Not really to be honest, if it is very very marginal.

what if the dma clicking is not related to the DSP?
aren´t capacitors used to filter audio too? Creative could messed up the caps values from factory?
just thinking...If there is a click on an audio that is expected to have a high pitch.

This website provides audios from different cards, showing all the noises, clicks, Pops, and they even talk about the ct2940 a card that has with DSP 4.13 as being one of the best ones with less issues.

http://nerdlypleasures.blogspot.com/2015/05/1 … t-playback.html

S95Sedan wrote on 2023-10-07, 21:47:

maxtherabbit wrote on 2023-10-07, 21:46:

S95sedan, do you think the clicking is improved in this test build?

Not really to be honest, if it is very very marginal.

I agree. It's either a small improvement or the same, not major

Maelgrum wrote on 2023-10-07, 20:15:

If it works, look at single cycle dma

did this build also fix the thing with saving the flags register? I didn't see you mention it lately

I was under the impression that all of the Vibra chips had an integrated DSP. When examining a CT3930, I see what looks like a small ROM chip. Is it possible that the DSP can be swapped out from this chip? What else could it be?

Kahenraz wrote on 2023-10-08, 04:44:

I was under the impression that all of the Vibra chips had an integrated DSP. When examining a CT3930, I see what looks like a small ROM chip. Is it possible that the DSP can be swapped out from this chip? What else could it be?

I also have a 3930 and was wondering the same, on my card that chip is covered by a label that says "(C) CREATIVE 0148393000" which I didn't want to remove. Unfortunately I don't think it's the DSP, it would be kind of pointless since it's supposedly integrated in the CT2501. Which is a shame as the CT3930 is an interesting and peculiar card, the non-PnP + Vibra + Real OPL + EMU8000 combo makes it really unique... a bug-free DSP would make it the ideal SB16 IMO.

By the way, the DSP chip is actually a microcontroller, not a ROM.

TheMobRules wrote on 2023-10-08, 05:17:

"(C) CREATIVE 0148393000"

that's not the DSP, you can find almost the exact same chip marked as "0148291000" on the older non-PnP cards (like CT2910, which has discrete 8052 MCU on-board) and located on the same place - as that chip is directly 'wired' to the ISA connector gold-fingers my best guess is that's "bus address decoder" that gives access to OPL, EMU8K respective IO port via the ISA bus.

[EDIT] you can even spot a pattern how that chip marking is constructed based on the model number:

0148 XYZT 00:
CT2910 --> 0148 2910 00
CT3930 --> 0148 3930 00

That chip is a PAL/GAL type thing for address decode purposes, and yeah, most probably for EMU8000.

mattw wrote on 2023-10-08, 09:39:

that's not the DSP, you can find almost the exact same chip marked as "0148291000" on the older non-PnP cards (like CT2910, whic […]
Show full quote

TheMobRules wrote on 2023-10-08, 05:17:

"(C) CREATIVE 0148393000"

that's not the DSP, you can find almost the exact same chip marked as "0148291000" on the older non-PnP cards (like CT2910, which has discrete 8052 MCU on-board) and located on the same place - as that chip is directly 'wired' to the ISA connector gold-fingers my best guess is that's "bus address decoder" that gives access to OPL, EMU8K respective IO port via the ISA bus.

[EDIT] you can even spot a pattern how that chip marking is constructed based on the model number:

0148 XYZT 00:
CT2910 --> 0148 2910 00
CT3930 --> 0148 3930 00

on my ct3600 the chip is just covered with a sticker.
the sticker covers up that it is a ct2290 chip.

maybe thats of interest.

DerBaum wrote on 2023-10-08, 11:11:

on my ct3600 the chip is just covered with a sticker.
the sticker covers up that it is a ct2290 chip.

it looks like Creative totally cheap-out, since its PAL (i.e. programmable) chip, it seems they had old re-marked chips left-over from CT2290 cards and reprogrammed them.

mattw wrote on 2023-10-08, 11:34:

it looks like Creative totally cheap-out, since its PAL (i.e. programmable) chip, it seems they had old re-marked chips left-over from CT2290 cards and reprogrammed them.

The custom marked PALs usually are OTP or mask programmed, so they can't be re-programmed. To me, it looks like the chip is not for EMU8000 decoding, but for IDE port decoding. And the IDE interface on the CT2290 and the CT3600 might be identical. The bus interface chip of the CT3600 is PnP capable, so it needs to provide the I/O address base selection signals to the PAL, while on the CT2290, the IDE base port is configured using jumpers. I suppose the sticker is just to "clean up" the number so it matches the card, not to indicate that it has a different program.

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

maxtherabbit wrote on 2023-10-09, 16:34:

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

Don't despair, i have some ideas about single cycle DMA, but large scale fw rewrite is needed, so it takes some time.

maxtherabbit wrote on 2023-10-09, 16:34:

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

third release:
1. Hanging note bug fixed
2. Some ADPCM decoding bug fixed
3. Small speed-up of fw code, but not solving Single DMA bug

Maelgrum wrote on 2023-10-09, 17:18:

third release: 1. Hanging note bug fixed 2. Some ADPCM decoding bug fixed 3. Small speed-up of fw code, but not solving Single D […]
Show full quote

maxtherabbit wrote on 2023-10-09, 16:34:

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

third release:
1. Hanging note bug fixed
2. Some ADPCM decoding bug fixed
3. Small speed-up of fw code, but not solving Single DMA bug

So no fix for the flags register not being pushed to stack before interrupt?

maxtherabbit wrote on 2023-10-09, 16:34:

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

These are the fixes that are out there?

1v413-8k_4ca3dca8_patch3
2- Fixed hanging note bug
3
4v413-8k_16a46526_patch4
5- Fixed hanging note bug
6- Fixed PSW bug in ExtInt0/ExtInt1 interrupt handlers

Id just go with the last one for now.
Theres a couple things which can be improved but it will probably take time.

Here im mostly cleaning up the code before doing anything, making it a bit easier to work with.

S95Sedan wrote on 2023-10-09, 17:21:

These are the fixes that are out there? […]
Show full quote

maxtherabbit wrote on 2023-10-09, 16:34:

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

These are the fixes that are out there?

Copy code to clipboard

1v413-8k_4ca3dca8_patch3
2- Fixed hanging note bug
3
4v413-8k_16a46526_patch4
5- Fixed hanging note bug
6- Fixed PSW bug in ExtInt0/ExtInt1 interrupt handlers

Id just go with the last one for now.
Theres a couple things which can be improved but it will probably take time.

Here im mostly cleaning up the code before doing anything, making it a bit easier to work with.

When I said "third maelgrum release" I was referring to the ones made by assembling modified disassembly from the dumps, not the hex edited versions. Also you forgot the ADPCM fix

maxtherabbit wrote on 2023-10-09, 17:25:

S95Sedan wrote on 2023-10-09, 17:21:

These are the fixes that are out there? […]
Show full quote

maxtherabbit wrote on 2023-10-09, 16:34:

Does the third "4.13 maelgrum" release fix all known bugs so far in 4.13? I'm gonna burn it to all my cards and call it done if yes. I've given up on seeing single cycle DMA get fixed in firmware, don't think it's possible

These are the fixes that are out there?

Copy code to clipboard

1v413-8k_4ca3dca8_patch3
2- Fixed hanging note bug
3
4v413-8k_16a46526_patch4
5- Fixed hanging note bug
6- Fixed PSW bug in ExtInt0/ExtInt1 interrupt handlers

Id just go with the last one for now.
Theres a couple things which can be improved but it will probably take time.

Here im mostly cleaning up the code before doing anything, making it a bit easier to work with.

When I said "third maelgrum release" I was referring to the ones made by assembling modified disassembly from the dumps, not the hex edited versions. Also you forgot the ADPCM fix

You mean the typo in X1231? Thats all i can see as changed in the later versions.

1mov	a,0ffh
2to
3mov	a,#0ffh