Creative's Advanced Signal Processor

[NewRisingSUn]

It hung, never displaying the 0220h line.

I vaguely remember that some SBs supposedly hang when accessing the CMS i/o ports without the chips installed, but thought that was a myth when I learned of your experience with the CT1350B. Unfortunately not.

Can you run the CMSTEST utility with the CMS-less and CMS-having SB 1.5? It'll probably hang at some point, too, but if it only hangs after the first two lines, that would mean I can read the ports and will just have to not write to them to avoid the hanging, and reading should be enough; writing C6 and 39 is probably dispensable.

[5u3]

Ah, now I'm getting it. Tests coming up...

The ASP.SYS driver or anything similar wasn't included with my AWE32 driver CD even after installation in Windows, but I found a driver called CSP.SYS in a forgotten directory on my harddisk. I cannot say where it comes from, but it seems to do the right thing.

First, the "problematic" cards:

CT2910 (the false positive):

SNDID.COM output:

Code:

CT1748 CSP present
DSP command 0x03 (CSP status) returned 80h

CSP.SYS output:

Code:

Error CSPSYS1006: Could not detect the CSP.

DIAGNOSE.EXE output:

Code:

DSP version: 4.13

CT3980 (the false negative):

SNDID.COM output:

Code:

DSP command 0x03 (CSP status) returned FFh

CSP.SYS output:

Code:

Driver Unit Number = 0
SB16 Base I/O Addx = 220 hex

DIAGNOSE.EXE output:

Code:

DSP version: 4.13  Advanced Signal Processor: Detected

And now the good news 
PnP detection works, at least on my cards. 3DSE detection also seems correct. And you got the EMU8K RAM detection working 

[Cloudschatze]

Can you run the CMSTEST utility with the CMS-less and CMS-having SB 1.5?

With:

Code:

            +0 +1 +2 +3
            FF FF FF FF
            FF FF FF FF
writing C6: FF FF FF FF
            FF FF FF FF
writing 39: FF FF FF FF
            FF FF FF FF

Without:

Code:

            +0 +1 +2 +3
            FF FF FF FF
            FF FF FF FF
writing C6:

By the way, your utility identifies Tandy 3-voice + DAC just fine.

[NewRisingSUn]

but I found a driver called CSP.SYS in a forgotten directory on my harddisk.

Well, send it right over here! Possibly one needs to somehow "initialize" it before asking for its status with DSP command 3.

Without:

Code:

            +0 +1 +2 +3
            FF FF FF FF
            FF FF FF FF

Eh? That's not good. That would mean I can detect the CMS chips on the CT1350, but not the CT1320. I could therefore limit the check to 2.x DSPs, but that would also cause problems, since you said you installed the 2.0 DSP in the 1.5 card.

Having seen the photos, how does one exactly install the square CT1351 2.0 DSP into the rectangular CT1320 1.x DSP's socket? Or is there a different 2.0 DSP for the CT1320 upgrade? I notice that the CT1350's DSP is actually version 2.01; does that mean that the CT1320's 2.0 DSP upgrade is not 2.01, but 2.00? Because if so, I could limit the CMS check to 2.01 DSPs.

I think my head hurts now. Ouch.

By the way, your utility identifies Tandy 3-voice + DAC just fine.

Yes, unfortunately it's useless for earlier Tandys because it uses 386 instructions.

[Cloudschatze]

but I found a driver called CSP.SYS in a forgotten directory on my harddisk.

Well, send it right over here! Possibly one needs to somehow "initialize" it before asking for its status with DSP command 3.

Hmm. I'm not loading the CSP driver though. The only thing I did to initialize the card is run "ctcm /t," "diagnose /s," and "aweutil /s."

Having seen the photos, how does one exactly install the square CT1351 2.0 DSP into the rectangular CT1320 1.x DSP's socket? Or is there a different 2.0 DSP for the CT1320 upgrade?

Yep. It's just a rectangular DIP that replaces the existing DSP (but came standard on the later cards, as it were).

I notice that the CT1350's DSP is actually version 2.01; does that mean that the CT1320's 2.0 DSP upgrade is not 2.01, but 2.00?

I believe so. "2.00" is etched on the chip.

By the way, your utility identifies Tandy 3-voice + DAC just fine.

Yes, unfortunately it's useless for earlier Tandys because it uses 386 instructions.

Does it detect WSS compatibility too? And on, and on, and on... 

[5u3]

but I found a driver called CSP.SYS in a forgotten directory on my harddisk.

Well, send it right over here! Possibly one needs to somehow "initialize" it before asking for its status with DSP command 3.

Hmm. I'm not loading the CSP driver though. The only thing I did to initilize the card is run "ctcm /t," "diagnose /s," and "aweutil /s."

Well, nobody loads this kind of stuff, it just eats up precious UMBs for nothing but detecting a useless chip...

Here it is: Creative's CSP Physical-Level driver (v1.14) for Sound Blaster 16

[NewRisingSUn]

Well, that CSP.SYS indeed has some crazy detection routine. I have incorporated it, probably not correctly, but try it nonetheless. As you can maybe tell, I actually don't have a card with CSP installed at the moment, so I'm basically working blind-folded here.

I believe so. "2.00" is etched on the chip.

What else is etched on it? Is it also a CT1351, or something else? Also, does the SB 1.5 use a systematically higher DSP version than SB 1.0? I see your other SB 1.5 has DSP 1.05, so have SB 1.0 all 1.00 to 1.04? Because if so, I can say:

DSP < 1.05: SB 1.0 => CMS chips always installed
DSP from 1.05 to 2.00: SB 1.5 => CMS chips optional, but no way of checking
DSP 2.1: SB 2.0 => CMS chips optional, can check
DSP >= 3.0: SBPro/SB16, CMS chips not possible.

Does it detect WSS compatibility too?

No, but Adlib Gold. I can add WSS, do you have one to check?

[Cloudschatze]

I believe so. "2.00" is etched on the chip.

What else is etched on it?

MHS
© CREATIVE LABS
INC CT1321V200
© INTEL 80,82
9109•

Also, does the SB 1.5 use a systematically higher DSP version than SB 1.0? I see your other SB 1.5 has DSP 1.05, so have SB 1.0 all 1.00 to 1.04?

No idea, unfortunately.

Does it detect WSS compatibility too?

No, but Adlib Gold. I can add WSS, do you have one to check?

I was being facetious, but yes, and several Ensoniq cards that are also WSS-compatible.

[Cloudschatze]

DSP from 1.05 to 2.00: SB 1.5 => CMS chips optional, but no way of checking

Would a timeout routine work? Perhaps using the "hang" as a detection method for not having the CMS chips installed?

[NewRisingSUn]

Would a timeout routine work? Perhaps using the "hang" as a detection method for not having the CMS chips installed?

Huh? The hang is not because *I* wait indefinitely, it's because when writing to the CMS ports, the SB asserts an endless ISA bus wait which forces the system to freeze all by itself.

By the way, you're welcome to try the new ASP detection routine.

[Cloudschatze]

Would a timeout routine work? Perhaps using the "hang" as a detection method for not having the CMS chips installed?

Huh? The hang is not because *I* wait indefinitely, it's because when writing to the CMS ports, the SB asserts an endless ISA bus wait which forces the system to freeze all by itself.

Ah, sorry.

Wait... On second thought, you can totally still make it work. Just add the disclaimer, "If you have an SB1.x, and this hangs the system, the SAA1099 chips are not installed." 

By the way, you're welcome to try the new ASP detection routine.

It's going to be a while yet before I have a chance. I'll post the results as soon as I am able.

[5u3]

Yay! 

CT2910:

Code:

0330h: MPU-401 (UART mode only)
0388h: YMF289B (OPL3-L)
0220h: DSP 4.13: CT1741 (SB16), Mixer: CT1745 (SB16), FM: YMF289B (OPL3-L)

CT3980:

Code:

0330h: MPU-401 (UART mode only)
0388h: YMF262/CT1747 (OPL3)
0220h: DSP 4.13: CT1741 (SB16), Mixer: CT1745 (SB16), FM: YMF262/CT1747 (OPL3)
       CT1748 CSP present, version 0
       Binaura 3D stereo enhancement supported
       Plug-and-play card
0620h: AWE32 EMU8000 (8192 KiB memory)

CT4390:

Code:

0330h: MPU-401 (UART mode only)
0388h: CT25xx/89xx (CQM)
0220h: DSP 4.16: CT1741 (SB16), Mixer: CT1745 (SB16), FM: CT25xx/89xx (CQM)
       Binaura 3D stereo enhancement supported
       Plug-and-play card
0620h: AWE32 EMU8000 (4096 KiB memory)

[NewRisingSUn]

Just add the disclaimer, "If you have an SB1.x, and this hangs the system, the SAA1099 chips are not installed."

I wonder if an external interrupt could unfreeze the machine. Try this. Does it still hang after you press ENTER? If it hangs, does the counter in the top left counter stop counting as well?

Yay!

Whee.

Code:

CT1748 CSP present, version 0

"Version 0"? Bah, who cares.

[Cloudschatze]

Just add the disclaimer, "If you have an SB1.x, and this hangs the system, the SAA1099 chips are not installed."

I wonder if an external interrupt could unfreeze the machine. Try this. Does it still hang after you press ENTER? If it hangs, does the counter in the top left counter stop counting as well?

I'll try it later as well. I appreciate the effort you're putting into all of this, because I'm actually learning quite a bit.

If ever you want to add the WSS check, I found this, which might be helpful:

Debug Script to Test for Windows Sound System Presence

The Ensoniq (and several other) WSS-compatible cards will fail this check, since they don't implement the 530h-533h configuration and status ports.

[Cloudschatze]

With the CMS1 utility, the counter runs until I press enter, at which point it stops, and is hung with with the frozen counter and "Press ENTER" still displayed. Ah well...

Here is the relevant CSP output using your latest utility:

CT2760

Code:

0220h: DSP 4.12: CT1741 (SB16), Mixer: CT1745 (SB16), FM: YMF262/CT1747 (OPL3)
       CT1748 CSP present, version 0

CT3980

Code:

0220h: DSP 4.13: CT1741 (SB16), Mixer: CT1745 (SB16), FM: YMF262/CT1747 (OPL3)
       CT1748 CSP present, version 18 (← Alternates between this and 237)
       Binaura 3D stereo enhancement supported
       Plug-and-play card

CT3990

Code:

0220h: DSP 4.13: CT1741 (SB16), Mixer: CT1745 (SB16), FM: CT1978 (CQM)
       CT1748 CSP present, version 0 (← Alternates between this and 255)
       Binaura 3D stereo enhancement supported
       Plug-and-play card

How are you differentiating between the CT1978 and CT25xx/CT89xx CQM, by the way?

[NewRisingSUn]

Bah, I'm so dumb. Redownload and try it again if the CSP version is displayed correctly now.

How are you differentiating between the CT1978 and CT25xx/CT89xx CQM, by the way?

CT1978 CQM behaves like OPL3, VLSI CQM behaves like OPL3-L.

With the CMS1 utility, the counter runs until I press enter, at which point it stops, and is hung with with the frozen counter and "Press ENTER" still displayed. Ah well..

I've made one change. Redownload CMS1 and try again. If it still hangs now, I'll give up.

[Cloudschatze]

Bah, I'm so dumb. Redownload and try it again if the CSP version is displayed correctly now.

The results are the same as before. (I've edited my previous post with the correct results.)

With the CMS1 utility, the counter runs until I press enter, at which point it stops, and is hung with with the frozen counter and "Press ENTER" still displayed. Ah well..

I've made one change. Redownload CMS1 and try again. If it still hangs now, I'll give up.

Same thing. Thank-you for the attempt though.

[NewRisingSUn]

The results are the same as before.

Well, it's not really important. I'll just get rid of the version thing entirely.

[Sambor]

Could someone place this ID utility on forum again?
Thanks.