First attempt of building new fw from source code.
Changes (from original V4.13)
1. Hanging note bug fixed
2. ADPCM problem fixed
3. Overall speedup - removed all debug x-bus trace port access
4. All unused junk in upper 2k is removed - flash in MCU must be programmed faster

Main question - is this build working ? ))

Maelgrum wrote on 2023-10-07, 18:01:

First attempt of building new fw from source code. Changes (from original V4.13) 1. Hanging note bug fixed 2. ADPCM problem f […]
Show full quote

First attempt of building new fw from source code.
Changes (from original V4.13)
1. Hanging note bug fixed
2. ADPCM problem fixed
3. Overall speedup - removed all debug x-bus trace port access
4. All unused junk in upper 2k is removed - flash in MCU must be programmed faster

Main question - is this build working ? ))

Will test in a second.
Edit: Hangs on boot aswell, same as the 4.16 one. Dont know what part of the code is fixed/changed but its doing something.

Does this one work in your assembler aswell, its one that works with as31.
(Havent done the names yet which were done in the other file.)

S95Sedan wrote on 2023-10-07, 18:19:

Will test in a second.
Edit: Hangs on boot aswell, same as the 4.16 one. Dont know what part of the code is fixed/changed but its doing something.

And original V4.13 works on this card ?
If original V4.13 wont work, this version will not work too ))

S95Sedan wrote on 2023-10-07, 18:19:

Does this one work in your assembler aswell, its one that works with as31.
(Havent done the names yet which were done in the other file.)

you want me to test compilation of this .asm, in keil ?

AT89C52-24JI/C is not OTP. You can reprogram it many times

Maelgrum wrote on 2023-10-07, 18:01:

First attempt of building new fw from source code. Changes (from original V4.13) 1. Hanging note bug fixed 2. ADPCM problem f […]
Show full quote

First attempt of building new fw from source code.
Changes (from original V4.13)
1. Hanging note bug fixed
2. ADPCM problem fixed
3. Overall speedup - removed all debug x-bus trace port access
4. All unused junk in upper 2k is removed - flash in MCU must be programmed faster

Main question - is this build working ? ))

Not working for me, DIAGNOSE does not find the card

maxtherabbit wrote on 2023-10-07, 18:45:

Maelgrum wrote on 2023-10-07, 18:01:

First attempt of building new fw from source code. Changes (from original V4.13) 1. Hanging note bug fixed 2. ADPCM problem f […]
Show full quote

First attempt of building new fw from source code.
Changes (from original V4.13)
1. Hanging note bug fixed
2. ADPCM problem fixed
3. Overall speedup - removed all debug x-bus trace port access
4. All unused junk in upper 2k is removed - flash in MCU must be programmed faster

Main question - is this build working ? ))

Not working for me, DIAGNOSE does not find the card

But original V4.13 works ?

maxtherabbit wrote on 2023-10-07, 18:41:

AT89C52-24JI/C is not OTP. You can reprogram it many times

thank you - I copied and pasted your comment to my previous post - just to keep all info at the same place - it seems I confused "AT89C52" with "AT87C52", which is OTP.

Maelgrum wrote on 2023-10-07, 18:41:

And original V4.13 works on this card ? If original V4.13 wont work, this version will not work too )) […]
Show full quote

S95Sedan wrote on 2023-10-07, 18:19:

Will test in a second.
Edit: Hangs on boot aswell, same as the 4.16 one. Dont know what part of the code is fixed/changed but its doing something.

And original V4.13 works on this card ?
If original V4.13 wont work, this version will not work too ))

S95Sedan wrote on 2023-10-07, 18:19:

Does this one work in your assembler aswell, its one that works with as31.
(Havent done the names yet which were done in the other file.)

you want me to test compilation of this .asm, in keil ?

The unmodified .asm one thats a 100% match works after compiling, the one with your changes doesnt. (this one Re: The Soundblaster DSP project )

Yes, so we known it works in either program.

Maelgrum wrote on 2023-10-07, 18:46:

maxtherabbit wrote on 2023-10-07, 18:45:

Maelgrum wrote on 2023-10-07, 18:01:

First attempt of building new fw from source code. Changes (from original V4.13) 1. Hanging note bug fixed 2. ADPCM problem f […]
Show full quote

First attempt of building new fw from source code.
Changes (from original V4.13)
1. Hanging note bug fixed
2. ADPCM problem fixed
3. Overall speedup - removed all debug x-bus trace port access
4. All unused junk in upper 2k is removed - flash in MCU must be programmed faster

Main question - is this build working ? ))

Not working for me, DIAGNOSE does not find the card

But original V4.13 works ?

Original 4.13 as well as 4.13 patch 3 have both been confirmed working on this card

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

Maelgrum wrote on 2023-10-07, 18:54:

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

yes!

maxtherabbit wrote on 2023-10-07, 19:01:

Maelgrum wrote on 2023-10-07, 18:54:

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

yes!

Excellent !
not only hanging note, but ADPCM bug also fixed here ))

Maelgrum wrote on 2023-10-07, 18:54:

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

Same here, this one works fine.

So i will leave all X-reg 5 stuff in place, and try to optimize code speed

S95Sedan wrote on 2023-10-07, 19:02:

Maelgrum wrote on 2023-10-07, 18:54:

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

Same here, this one works fine.

BTW, you have one card on which you successfully tested V4.16. First build should work on this card.

Maelgrum wrote on 2023-10-07, 19:04:

S95Sedan wrote on 2023-10-07, 19:02:

Maelgrum wrote on 2023-10-07, 18:54:

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

Same here, this one works fine.

BTW, you have one card on which you successfully tested V4.16. First build should work on this card.

Both were on an awe32 ct3900, but can test on either incase more things been changed to ensure compatibility.

Maelgrum wrote on 2023-10-07, 19:04:

S95Sedan wrote on 2023-10-07, 19:02:

Maelgrum wrote on 2023-10-07, 18:54:

Ok, this version leaves all X-bus reg 5 exchange in place
will it work ?

Same here, this one works fine.

BTW, you have one card on which you successfully tested V4.16. First build should work on this card.

That was his CT1740 (no CT1747 bus interface)

If it works, look at single cycle dma

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

If it works, look at single cycle dma

It works, but I still hear the clicking in duke2

Clicking is same as before, or something changed ?