mattw wrote on 2023-10-15, 09:47:

can you say how much faster STC12C5A60S2 compared to regular 8052 MCU in "6T" is in that case? I just wonder what are the speed limits when those problems start to emerge. Also, thank you for giving so much good details about STC chips and how and with what programmers they can be used.

EDIT: For the original question, I think STC12C5A60S2 would be around 2x-3x of a regular MCU in 6T in overall, with MOVX execution speed controllable depending on the external hardware.

STC12 family is 1T, but execution speed of individual instructions vary and not necessarily correspond to being 12x the speed.
- While NOP is said to be only 1 cycle (12x), when I was using C51 with this MCU, __nop__() runs at 6x in practice. That is, any place requiring one __nop__() now needs six to maintain timing.
- Different instructions have varying cycles, most instructions are about 4x-8x the speed of a standard 8051 MCU. For most routines the overall speed would be about 6x, though routines consisting of mostly heavier instructions could be about only 3x-4x.
- Due to varying execution speed of instructions, the behavior of complex delays using for loops and such can be unpredictable and requires manual calibration.
- MOVX execution time for internal XRAM is 3-4 cycles. For external XRAM the speed depends on ALES (P0 address setup and ALE negative edge cycles) and RWS (MOVX r/w pulse cycles), making the overall execution time 7-20 cycles. It defaults to 14 cycles on startup (ALES=2, RWS=3, which means 3 cycles for P0 address setup and ALE negative edge, plus 4 cycles for MOVX r/w pulse).
- Timers are fully compatible with standard 8051 (12T) by default, so with this MCU there's no need to change any timer-related code. You can set the timer to operate at 1T instead via AUXR bits if desired, but in most cases not necessary. I think a good use case for 1T timers would be to reach insane UART baud rates like 921600 bps or even more for specialized purposes.

I haven't tested 6T mode of STC89/90 family. That mode appears to be a simple clock divider toggle, that doubles the execution speed of everything if set, so timers as well as other timing-sensitive code need to be adjusted accordingly, which means changes in several places if you want working MIDI also. For STC12, only one line of code (that is to disable access to internal XRAM) is needed to get everything, including MIDI, work in most cases. It's just some games and software (including Windows) do not seem to like the difference in the execution speed.

EDIT 2: With STC12 MCU, a MOVX instruction with BUS_SPEED maxed out plus two NOPs would be around the same speed as a standard MCU in 6T at around 24 cycles.

LSS10999 wrote on 2023-10-15, 11:08:

I haven't tested 6T mode of STC89/90 family.

I believe such test worth to be done, because as I understand it - the test you've done is on average higher performance of the MCU than just 2 times. So, it's good we see what happens in 6T mode, i.e. with just 2 times the speed, especially with the clicking problem. Also, @Maelgrum stated many times fixing/patching the MIDI UART timings in the DSP code to make them work with 6T is no problem. last, but not least, I believe MCU with support for 6T are much more common than STC high-performance modes. In fact, on fast search I was not able to find where to buy such STC MCU in PLCC-44 package.

[EDIT] and just an weird idea - what do you all think on slowing down the MCU, i.e. install 12MHz crystal? I mean what are the predictions of the possible behavior other than of course the wrong MIDI UART timings. In any way, I can do such test, because initially I wrongly order 12MHz 8052MCU and also 12MHz Crystal is very common - every broken USB device I have, has one such crystal and so I can borrow it as a donor from there.

mattw wrote on 2023-10-15, 12:00:

I believe such test worth to be done, because as I understand it - the test you've done is on average higher performance of the MCU than just 2 times. So, it's good we see what happens in 6T mode, i.e. with just 2 times the speed, especially with the clicking problem. Also, @Maelgrum stated many times fixing/patching the MIDI UART timings in the DSP code to make them work with 6T is no problem. last, but not least, I believe MCU with support for 6T are much more common than STC high-performance modes. In fact, on fast search I was not able to find where to buy such STC MCU in PLCC-44 package.

[EDIT] and just an weird idea - what do you all think on slowing down the MCU, i.e. install 12MHz crystal? I mean what are the predictions of the possible behavior other than of course the wrong MIDI UART timings. In any way, I can do such test, because initially I wrongly order 12MHz 8052MCU and also 12MHz Crystal is very common - every broken USB device I have, has one such crystal and so I can borrow it as a donor from there.

I just did some tests to be sure.

1. STC90C58RD+ in 6T mode - Other than MIDI (which is expected and this confirms the MCU is indeed running in 6T), everything still works as expected, including Duke Nukem 3D and in Windows.
2. STC12C5A60S2 with BUS_SPEED register maxed out - Still the same with Duke Nukem 3D and Windows (not working). Additionally, MIDI stopped working completely -- games that prefer using MIDI to FM no longer detect MPU401 and use FM instead.

So it seems the issue with STC12C5A60S2 is more than just timing. Rather, the default BUS_SPEED setting is the correct value to be used here, but some programs and drivers are written in a way uncomfortable with the faster MCU.

PS: No noticeable difference regarding clicking between 12T and 6T. I think there may be other factors than DSP code execution time causing it.

LSS10999 wrote on 2023-10-15, 13:44:

PS: No noticeable difference regarding clicking between 12T and 6T. I think there may be other factors than DSP code execution time causing it.

I start to think the same. maybe only capture in the 'digital domain', i.e. see the 'click' case and 'no-click' case as PCM bytes sequence of the same sound (i.e. we see the corruption and its nature in the data-stream), can give real answers, but that's not easy to be tested and probably the modification I came up with here:

Re: The Soundblaster DSP project

is not even entirely proper and sufficient...

Still waiting for last 2 tests to be checked)))

Maelgrum wrote on 2023-10-15, 14:23:

Still waiting for last 2 tests to be checked)))

they are here:

Re: The Soundblaster DSP project

in case people already with PLCC-44 sockets installed missed them. BTW, maybe end of the next week my PLCC-44 sockets and MCUs will arrive and I will be able to do such tests too.

I've been out of town a couple days, I should be able to test tonight

Maelgrum wrote on 2023-10-15, 14:23:

Still waiting for last 2 tests to be checked)))

Been a little bit busy but this is for both thests on my card.
There seems to be very little difference between them aswell.
(If you need any specific tests let me know)

Test 1: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Test 2: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

S95Sedan wrote on 2023-10-15, 17:53:

Been a little bit busy but this is for both thests on my card. There seems to be very little difference between them aswell. (If […]
Show full quote

Maelgrum wrote on 2023-10-15, 14:23:

Still waiting for last 2 tests to be checked)))

Been a little bit busy but this is for both thests on my card.
There seems to be very little difference between them aswell.
(If you need any specific tests let me know)

Test 1: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Test 2: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Thank you S95Sedan for testing!
I will go to path of completely new code for playing 0x14 cmd then.

Maelgrum wrote on 2023-10-15, 18:21:

S95Sedan wrote on 2023-10-15, 17:53:

Been a little bit busy but this is for both thests on my card. There seems to be very little difference between them aswell. (If […]
Show full quote

Maelgrum wrote on 2023-10-15, 14:23:

Still waiting for last 2 tests to be checked)))

Been a little bit busy but this is for both thests on my card.
There seems to be very little difference between them aswell.
(If you need any specific tests let me know)

Test 1: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Test 2: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Thank you S95Sedan for testing!
I will go to path of completely new code for playing 0x14 cmd then.

First make sure the game uses that command.

georgel wrote on 2023-10-15, 18:25:

Maelgrum wrote on 2023-10-15, 18:21:

S95Sedan wrote on 2023-10-15, 17:53:

Been a little bit busy but this is for both thests on my card. There seems to be very little difference between them aswell. (If […]
Show full quote

Been a little bit busy but this is for both thests on my card.
There seems to be very little difference between them aswell.
(If you need any specific tests let me know)

Test 1: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Test 2: (CT3900 AWE32)
- Card is detected and boots normally.
- Tested audio in Doom, Rott, Duke3d: Seems to work and sound fine.
- Audio works fine in Duke2 shooting still has clicking.

Thank you S95Sedan for testing!
I will go to path of completely new code for playing 0x14 cmd then.

First make sure the game uses that command.

According to this:
https://github.com/lethal-guitar/Duke2 ... DIGISND.C
Duke2 may have samples in PCM and Adpcm,
so it can be not best game to test 0x14 single cycle dma playback.
[Edit] duke2 by default uses Adpcm compressed sound, so it is good test for Adpcm playing, but not for cmd 0x14 test.

This is what I get when playing DOTT using the first of the two latest test builds. The voice samples are all sped up and sound like the chipmunks. Also it still clicks...

I've done rebuilding DosBox from sources, added my custom sb trace logging code, run Mikey 123 - and got this:

1F2          - IRQ Request, 8-bit
2D1          - Enable Speaker
340 B9       - Set Time Constant 14084 Hz
4E1          - DSP Version
548 FF 03    - Set DMA Block Size
61C          - Auto - Initialize DMA DAC, 8-bit
7DA          - Exit Auto - Initialize DMA Operation, 8-bit
81C          - Auto - Initialize DMA DAC, 8-bit
9DA          - Exit Auto - Initialize DMA Operation, 8-bit
101C          - Auto - Initialize DMA DAC, 8-bit
11DA          - Exit Auto - Initialize DMA Operation, 8-bit
121C          - Auto - Initialize DMA DAC, 8-bit
13DA          - Exit Auto - Initialize DMA Operation, 8-bit

Not using 0x14 single cycle DMA

With this logging, i can test any game what commands it really uses.

From Duke Nukem 2:

1D1          - Enable Speaker
240 38       - Set Time Constant 5000 Hz
375 15 09    - DMA DAC, 4-bit ADPCM Reference
4D0          - Halt DMA Operation, 8-bit
540 38       - Set Time Constant 5000 Hz
675 15 09    - DMA DAC, 4-bit ADPCM Reference
7D0          - Halt DMA Operation, 8-bit
840 38       - Set Time Constant 5000 Hz
975 15 09    - DMA DAC, 4-bit ADPCM Reference
10D0          - Halt DMA Operation, 8-bit
1140 72       - Set Time Constant 7042 Hz
1275 3E 05    - DMA DAC, 4-bit ADPCM Reference
13D0          - Halt DMA Operation, 8-bit
1440 39       - Set Time Constant 5025 Hz
1577 25 05    - DMA DAC, 2.6-bit ADPCM Reference
16D0          - Halt DMA Operation, 8-bit
1740 72       - Set Time Constant 7042 Hz
1875 3E 05    - DMA DAC, 4-bit ADPCM Reference
19D0          - Halt DMA Operation, 8-bit
2040 06       - Set Time Constant 4000 Hz
2175 3A 03    - DMA DAC, 4-bit ADPCM Reference
22D0          - Halt DMA Operation, 8-bit

Uses ADPCM 4 and 2.6 bit

Maelgrum wrote on 2023-10-16, 01:39:

From Duke Nukem 2:

Details

Copy code to clipboard

1D1          - Enable Speaker
240 38       - Set Time Constant 5000 Hz
375 15 09    - DMA DAC, 4-bit ADPCM Reference
4D0          - Halt DMA Operation, 8-bit
540 38       - Set Time Constant 5000 Hz
675 15 09    - DMA DAC, 4-bit ADPCM Reference
7D0          - Halt DMA Operation, 8-bit
840 38       - Set Time Constant 5000 Hz
975 15 09    - DMA DAC, 4-bit ADPCM Reference
10D0          - Halt DMA Operation, 8-bit
1140 72       - Set Time Constant 7042 Hz
1275 3E 05    - DMA DAC, 4-bit ADPCM Reference
13D0          - Halt DMA Operation, 8-bit
1440 39       - Set Time Constant 5025 Hz
1577 25 05    - DMA DAC, 2.6-bit ADPCM Reference
16D0          - Halt DMA Operation, 8-bit
1740 72       - Set Time Constant 7042 Hz
1875 3E 05    - DMA DAC, 4-bit ADPCM Reference
19D0          - Halt DMA Operation, 8-bit
2040 06       - Set Time Constant 4000 Hz
2175 3A 03    - DMA DAC, 4-bit ADPCM Reference
22D0          - Halt DMA Operation, 8-bit

Uses ADPCM 4 and 2.6 bit

So any "click build" testing we did with duke2 should be invalidated

DOTT:

1E0 C6       - DSP Identification
2D3          - Disable Speaker
3E0 AA       - DSP Identification
4E1          - DSP Version
5E2 06       - Creative protection check
6E2 6B       - Creative protection check
7E4 AA       - Write Test Register
8E8          - Read Test Register
940 64       - Set Time Constant 6410 Hz
1014 00 00    - DMA DAC, 8-bit
11------------- IRQ8
12D1          - Enable Speaker
13D1          - Enable Speaker
1440 A5       - Set Time Constant 10989 Hz
1514 DF 2F    - DMA DAC, 8-bit
16------------- IRQ8
1714 5F 03    - DMA DAC, 8-bit
18------------- IRQ8
1914 5A 02    - DMA DAC, 8-bit
20------------- IRQ8
2140 A5       - Set Time Constant 10989 Hz
2214 80 0A    - DMA DAC, 8-bit
23------------- IRQ8
2440 A5       - Set Time Constant 10989 Hz
2514 80 0A    - DMA DAC, 8-bit
26------------- IRQ8
2740 A5       - Set Time Constant 10989 Hz
2814 80 0A    - DMA DAC, 8-bit
29------------- IRQ8
3040 A5       - Set Time Constant 10989 Hz
3114 80 0A    - DMA DAC, 8-bit
32------------- IRQ8
3340 A5       - Set Time Constant 10989 Hz
3414 B7 18    - DMA DAC, 8-bit
35D0          - Halt DMA Operation, 8-bit
3640 A5       - Set Time Constant 10989 Hz
3714 B7 18    - DMA DAC, 8-bit
38D0          - Halt DMA Operation, 8-bit
3940 A5       - Set Time Constant 10989 Hz
4014 B7 18    - DMA DAC, 8-bit
41------------- IRQ8
4240 A5       - Set Time Constant 10989 Hz
4314 FF 0A    - DMA DAC, 8-bit
44D0          - Halt DMA Operation, 8-bit
4540 A5       - Set Time Constant 10989 Hz
4614 FF 0A    - DMA DAC, 8-bit
47D0          - Halt DMA Operation, 8-bit
4840 A5       - Set Time Constant 10989 Hz
4914 53 06    - DMA DAC, 8-bit
50------------- IRQ8
5140 A5       - Set Time Constant 10989 Hz
5214 53 06    - DMA DAC, 8-bit
53------------- IRQ8
5440 A5       - Set Time Constant 10989 Hz
5514 FF 0A    - DMA DAC, 8-bit
56D0          - Halt DMA Operation, 8-bit
5740 A5       - Set Time Constant 10989 Hz
5814 FF 0A    - DMA DAC, 8-bit
59------------- IRQ8
6040 A5       - Set Time Constant 10989 Hz

6114 53 06    - DMA DAC, 8-bit
62------------- IRQ8
6340 A5       - Set Time Constant 10989 Hz
6414 53 06    - DMA DAC, 8-bit
65------------- IRQ8
6640 A5       - Set Time Constant 10989 Hz
6714 FF 0A    - DMA DAC, 8-bit
68D0          - Halt DMA Operation, 8-bit
6940 A5       - Set Time Constant 10989 Hz
7014 FF 0A    - DMA DAC, 8-bit
71------------- IRQ8
7240 A5       - Set Time Constant 10989 Hz
7314 53 06    - DMA DAC, 8-bit
74------------- IRQ8
7540 A5       - Set Time Constant 10989 Hz
7614 53 06    - DMA DAC, 8-bit
77------------- IRQ8
7840 A5       - Set Time Constant 10989 Hz
7914 FF 0A    - DMA DAC, 8-bit
80------------- IRQ8
8140 A5       - Set Time Constant 10989 Hz
8214 53 06    - DMA DAC, 8-bit
83------------- IRQ8
8440 A5       - Set Time Constant 10989 Hz
8514 FF 0A    - DMA DAC, 8-bit
86------------- IRQ8
8740 A5       - Set Time Constant 10989 Hz
8814 53 06    - DMA DAC, 8-bit
89------------- IRQ8
9040 A5       - Set Time Constant 10989 Hz
9114 FF 0A    - DMA DAC, 8-bit
92------------- IRQ8
9340 A5       - Set Time Constant 10989 Hz
9414 53 06    - DMA DAC, 8-bit
95------------- IRQ8
9640 D3       - Set Time Constant 22222 Hz
9714 DF 2F    - DMA DAC, 8-bit
98------------- IRQ8
9914 5F 03    - DMA DAC, 8-bit
100------------- IRQ8
10114 9F 2C    - DMA DAC, 8-bit
102------------- IRQ8
10314 88 2F    - DMA DAC, 8-bit
104------------- IRQ8
10540 A5       - Set Time Constant 10989 Hz
10614 FF 0A    - DMA DAC, 8-bit
107D0          - Halt DMA Operation, 8-bit
10840 A5       - Set Time Constant 10989 Hz
10914 FF 0A    - DMA DAC, 8-bit
110------------- IRQ8
11140 A5       - Set Time Constant 10989 Hz
11214 53 06    - DMA DAC, 8-bit
113------------- IRQ8
11440 A5       - Set Time Constant 10989 Hz
11514 53 06    - DMA DAC, 8-bit
116------------- IRQ8
11740 A5       - Set Time Constant 10989 Hz
11814 FF 0A    - DMA DAC, 8-bit
119D0          - Halt DMA Operation, 8-bit
12040 A5       - Set Time Constant 10989 Hz
12114 FF 0A    - DMA DAC, 8-bit
122------------- IRQ8
12340 A5       - Set Time Constant 10989 Hz
12414 53 06    - DMA DAC, 8-bit
125------------- IRQ8
12640 A5       - Set Time Constant 10989 Hz
12714 53 06    - DMA DAC, 8-bit
128------------- IRQ8
12940 A5       - Set Time Constant 10989 Hz
13014 FF 0A    - DMA DAC, 8-bit
131------------- IRQ8
13240 A5       - Set Time Constant 10989 Hz
13314 53 06    - DMA DAC, 8-bit
134------------- IRQ8
13540 D3       - Set Time Constant 22222 Hz
13614 DF 2F    - DMA DAC, 8-bit
137------------- IRQ8
13814 5F 03    - DMA DAC, 8-bit
139------------- IRQ8
14014 9F 2C    - DMA DAC, 8-bit
141------------- IRQ8
14214 88 2F    - DMA DAC, 8-bit
143------------- IRQ8
14440 A5       - Set Time Constant 10989 Hz
14514 BF 0B    - DMA DAC, 8-bit
146------------- IRQ8
14740 D3       - Set Time Constant 22222 Hz
14814 DF 2F    - DMA DAC, 8-bit
149------------- IRQ8
15014 5F 03    - DMA DAC, 8-bit
151------------- IRQ8
15214 9F 2C    - DMA DAC, 8-bit
153------------- IRQ8
15414 5A 1C    - DMA DAC, 8-bit
155------------- IRQ8
15640 A5       - Set Time Constant 10989 Hz
15714 D1 0F    - DMA DAC, 8-bit
158------------- IRQ8
15940 D3       - Set Time Constant 22222 Hz
16014 DF 2F    - DMA DAC, 8-bit
161------------- IRQ8
16214 5F 03    - DMA DAC, 8-bit
163------------- IRQ8
16414 9F 2C    - DMA DAC, 8-bit
165------------- IRQ8
16614 56 1A    - DMA DAC, 8-bit
167------------- IRQ8
16840 D3       - Set Time Constant 22222 Hz
16914 DF 2F    - DMA DAC, 8-bit
170------------- IRQ8
17114 5F 03    - DMA DAC, 8-bit
172------------- IRQ8
17314 9F 2C    - DMA DAC, 8-bit
174------------- IRQ8
17514 FF 2F    - DMA DAC, 8-bit
176------------- IRQ8
17714 5F 03    - DMA DAC, 8-bit
178------------- IRQ8
17914 9F 2C    - DMA DAC, 8-bit
180------------- IRQ8
18114 FF 2F    - DMA DAC, 8-bit
182------------- IRQ8
18314 5F 03    - DMA DAC, 8-bit
184------------- IRQ8
18514 9F 2C    - DMA DAC, 8-bit
186------------- IRQ8
18714 FF 2F    - DMA DAC, 8-bit
188------------- IRQ8
18914 5F 03    - DMA DAC, 8-bit
190------------- IRQ8
19114 1B 10    - DMA DAC, 8-bit
192------------- IRQ8
19340 A5       - Set Time Constant 10989 Hz
19414 FF 0A    - DMA DAC, 8-bit
195D0          - Halt DMA Operation, 8-bit
19640 A5       - Set Time Constant 10989 Hz
19714 FF 0A    - DMA DAC, 8-bit
198------------- IRQ8
19940 A5       - Set Time Constant 10989 Hz
20014 53 06    - DMA DAC, 8-bit
201D0          - Halt DMA Operation, 8-bit
20240 A5       - Set Time Constant 10989 Hz
20314 53 06    - DMA DAC, 8-bit
204------------- IRQ8
20540 A5       - Set Time Constant 10989 Hz
20614 FF 0A    - DMA DAC, 8-bit
207D0          - Halt DMA Operation, 8-bit
20840 A5       - Set Time Constant 10989 Hz
20914 FF 0A    - DMA DAC, 8-bit
210------------- IRQ8
21140 A5       - Set Time Constant 10989 Hz
21214 53 06    - DMA DAC, 8-bit
213------------- IRQ8
21440 A5       - Set Time Constant 10989 Hz
21514 53 06    - DMA DAC, 8-bit
216------------- IRQ8
21740 A5       - Set Time Constant 10989 Hz
21814 FF 0A    - DMA DAC, 8-bit
219------------- IRQ8
22040 A5       - Set Time Constant 10989 Hz
22114 53 06    - DMA DAC, 8-bit
222------------- IRQ8
22340 D3       - Set Time Constant 22222 Hz
22414 DF 2F    - DMA DAC, 8-bit
225------------- IRQ8
22614 5F 03    - DMA DAC, 8-bit
227------------- IRQ8
22814 9F 2C    - DMA DAC, 8-bit
229------------- IRQ8
23014 FF 2F    - DMA DAC, 8-bit
231------------- IRQ8
23214 5F 03    - DMA DAC, 8-bit
233------------- IRQ8
23414 36 02    - DMA DAC, 8-bit
235------------- IRQ8
23640 A5       - Set Time Constant 10989 Hz
23714 DF 2F    - DMA DAC, 8-bit
238D0          - Halt DMA Operation, 8-bit
23940 D3       - Set Time Constant 22222 Hz
24014 DF 2F    - DMA DAC, 8-bit
241------------- IRQ8
24214 5F 03    - DMA DAC, 8-bit
243------------- IRQ8
24414 9F 2C    - DMA DAC, 8-bit
245------------- IRQ8
24614 34 0C    - DMA DAC, 8-bit
247------------- IRQ8
24840 A5       - Set Time Constant 10989 Hz
24914 F9 0A    - DMA DAC, 8-bit
250D0          - Halt DMA Operation, 8-bit
25140 A5       - Set Time Constant 10989 Hz
25214 80 0A    - DMA DAC, 8-bit
253------------- IRQ8
25440 A5       - Set Time Constant 10989 Hz
25514 23 16    - DMA DAC, 8-bit
256------------- IRQ8
25740 A5       - Set Time Constant 10989 Hz
25814 40 0D    - DMA DAC, 8-bit
259D0          - Halt DMA Operation, 8-bit
26040 A5       - Set Time Constant 10989 Hz
26114 92 17    - DMA DAC, 8-bit
262D0          - Halt DMA Operation, 8-bit
26340 A5       - Set Time Constant 10989 Hz
26414 01 2D    - DMA DAC, 8-bit
265D0          - Halt DMA Operation, 8-bit
26640 A5       - Set Time Constant 10989 Hz
26714 DF 2F    - DMA DAC, 8-bit
268D0          - Halt DMA Operation, 8-bit
26940 A5       - Set Time Constant 10989 Hz
27014 2B 22    - DMA DAC, 8-bit
271------------- IRQ8
27240 A5       - Set Time Constant 10989 Hz
27314 DF 2F    - DMA DAC, 8-bit
274------------- IRQ8
27514 2A 00    - DMA DAC, 8-bit
276------------- IRQ8
27740 A5       - Set Time Constant 10989 Hz
27814 25 23    - DMA DAC, 8-bit
279------------- IRQ8
28040 A5       - Set Time Constant 10989 Hz
28114 DF 2F    - DMA DAC, 8-bit
282D0          - Halt DMA Operation, 8-bit
28340 D3       - Set Time Constant 22222 Hz
28414 DF 2F    - DMA DAC, 8-bit
285------------- IRQ8
28614 5F 03    - DMA DAC, 8-bit
287------------- IRQ8
28814 9F 2C    - DMA DAC, 8-bit
289------------- IRQ8
29014 3E 10    - DMA DAC, 8-bit
291------------- IRQ8
29240 A5       - Set Time Constant 10989 Hz
29314 DF 2F    - DMA DAC, 8-bit
294------------- IRQ8
29514 5F 03    - DMA DAC, 8-bit
296------------- IRQ8
29714 D9 19    - DMA DAC, 8-bit
298------------- IRQ8
29940 D3       - Set Time Constant 22222 Hz
30014 DF 2F    - DMA DAC, 8-bit
301------------- IRQ8
30214 5F 03    - DMA DAC, 8-bit
303------------- IRQ8
30414 9F 2C    - DMA DAC, 8-bit
305------------- IRQ8
30614 FF 2F    - DMA DAC, 8-bit
307------------- IRQ8
30814 5F 03    - DMA DAC, 8-bit
309------------- IRQ8
31014 9F 2C    - DMA DAC, 8-bit
311------------- IRQ8
31214 08 06    - DMA DAC, 8-bit
313------------- IRQ8
31440 A5       - Set Time Constant 10989 Hz
31514 DF 2F    - DMA DAC, 8-bit
316------------- IRQ8
31714 5F 03    - DMA DAC, 8-bit
318------------- IRQ8
31914 F4 1A    - DMA DAC, 8-bit
320------------- IRQ8

So DOTT really uses 0x14 cmd,
and does creative protection check ))

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

If it works, look at single cycle dma

So I need to retest this build with DOTT in addition to the most recent one you've done

I dont hear clicking in that chipmunk sample 🙁. We really need a standalone test for clicking, small program doing nothing else but playing single cycle DMA with samples exuberating the problem.

rasz_pl wrote on 2023-10-16, 02:20:

I dont hear clicking in that chipmunk sample 🙁. We really need a standalone test for clicking, small program doing nothing else but playing single cycle DMA with samples exuberating the problem.

It's there I promise. Was noticable in person, just not my cell phone recording of the sound coming through the speakers

strange thing i see in log:

1------------- IRQ8
240 A5       - Set Time Constant 10989 Hz
314 B7 18    - DMA DAC, 8-bit
4D0          - Halt DMA Operation, 8-bit
540 A5       - Set Time Constant 10989 Hz
614 B7 18    - DMA DAC, 8-bit
7D0          - Halt DMA Operation, 8-bit
840 A5       - Set Time Constant 10989 Hz
914 B7 18    - DMA DAC, 8-bit
10------------- IRQ8

1. Playing buffer is stopped during a play. But stopped by some game timings, so it can end at any sample value. And another playback is started.
If last output sample of first buffer differs significantly from first sample of next buffer, it will click.
2. Halt DMA operation is strange:

1    mov r0,#8
2    movx a,@r0
3    anl a,#0e7h
4    orl a,#42h
5    movx @r0,a
6    mov 2eh,#64h
7    djnz 2eh,$
8    anl a,#0a7h
9    movx @r0,a
10    clr ex1

Some embedded delay is here - djnz 100 times.
May be lowering this value can affect clicks ?.... interesting