EMU-APS misidentified as 4670

monadity · Jul 3, 2003, 02:14 AM

Hi ,

I just discovered the kxdriver - haven't been using my APS much. . .

When I installed it, kxmixer identified the card as a 4670.

I checked the Registry Pnp Dynamic entries as well as System Info (I'm a former driver geek)
and discovered that the configmgr code return is SUBSYS_00201102.

Being more inclined to hack than ask, I tried changing the inf file.
The install went as APS, but kxmixer behaved as 4670.

I changed the kxmixer string in the registry to try force aps (--startup --aps), which changes the appearance of the input/output page of kxmixer, but not the device in the lower right hand corner.

I would be ok with whatever the panel says, but the IoStatus shows no other inputs besides 3 sp/difs, no outputs except 2 sp/difs and no matter what I patch, I can't seem to get an output from the analog outs on the APS.

I tried what was suggested in some threads - switch F+B, mess with codec params, etc.
Also went into DSP and patched a known good signal on the digital out (known good because it dies if I unpatch) to each output and nothing out. . .

I tried reading this forum and couldn't find anything on this. . sorry if it is a repeat.

Info:

kx ver5.10 - 3533 debug
Win98SE, DX 9.0a
EMU/APS
PII-233
Intel (not Via) PCI bridge 82443LX
info from panel:
CT4670 10K1 [E800]
(Chip Rev 3 - unknown AC codec, but says that card has one)

Mitch

Max M. · Jul 3, 2003, 02:43 AM

It looks like EEPROM (where card id is stored) content is damaged...
Hmm.. interesting - it happens very often with Lives and Audigies but APS... i see that at first time... hmm... it's a problem... you need to flash it with correct data...

monadity · Jul 3, 2003, 03:46 AM

Interesting - it had me thinking that maybe EMU/Creative kicked out a few with the wrong Ids.
Doesn't cause a problem with APS drivers 1.0 and 1.5.

Any Idea how to flash the card?

I haven't seen anything like that on Creative/EMU site, but will check again.

I also will re-verify the ID without any software installed.

Thanks,

Mitch

monadity · Jul 3, 2003, 04:56 AM

According to this article on upgrading a Live! card, it's an EEPROM - pretty standard
little 93C46 chip. Have to pull it and tweak the device ID back.

I re-verified that this card is misreporting. I think I had to directly select the EMU driver for a while, because their INF also does not match this card. Their driver is working again, though!

Thanks for the help! (and I will see if it works when I dig out the prom programmer.)

Mitch

http://www.digit-life.com/articles/l.../livetolive51/

**Eugene Gavrilov** · Jul 5, 2003, 03:50 AM

I'm afraid the driver won't be able to distinguish between your 'fake' APS and the 'real' SBLive Value without re-programming the EEPROM

the kernel-level part relies on the PCI ID when configuring the hardware
actually, the biggest difference (from the driver point of view) is between APS, 10k1, 10k2 (audigy), 10k2.2 (audigy2) and 10k2.2ex (audigy 2 ex) models. the particular card 'flavours' aren't so critical.

the 10k1 and 10k2 are easily recognized even if the EEPROM content is reset (because the lowest part of the ID is read-only and cannot be 'flashed')
but there's no way to distinguish APS from 10k1, as well as audigy, audigy2 and audigy2ex without the PCI Subsystem ID and chip revision information, so, tweaking the INF file won't help at all

/Eugene

monadity · Jul 9, 2003, 01:20 AM

Thanks - I am surprised that the analog outs don't work even if the model is confused- something special that must be set on the card? I tried all of the outputs in the DSP patching page.

Been too busy to deal with it the past week, but checked my programmer and it will burn a 93C46 - so I figure I'll just change the card ID by jumpering some wires from the E2 to the programmer - might have to lift pin 8 so the programmer doesn't cook trying to power the whole card.

Maybe tonight. . .

Mitch

monadity · Jul 13, 2003, 01:26 AM

I pulled the prom and broke it, so had to burn a new one from the info at the link I posted
above. (using 4001 as the subsys ident) Kxdriver works and I am already using it to do things that the EMU driver would only do lamely using Aux buses.

Thanks,

Mitch

**Eugene Gavrilov** · Jul 13, 2003, 02:14 AM

good

I'm glad your card is working now

a small note for the other kX users: don't repeat re-programming unless you know what you are doing

/Eugene

Jul 3, 2003, 02:14 AM	#1
monadity DriverHeaven Newbie Join Date: Jul 2003 Location: US - Georgia Posts: 5 Rep Power: 0	EMU-APS misidentified as 4670 Hi , I just discovered the kxdriver - haven't been using my APS much. . . When I installed it, kxmixer identified the card as a 4670. I checked the Registry Pnp Dynamic entries as well as System Info (I'm a former driver geek) and discovered that the configmgr code return is SUBSYS_00201102. Being more inclined to hack than ask, I tried changing the inf file. The install went as APS, but kxmixer behaved as 4670. I changed the kxmixer string in the registry to try force aps (--startup --aps), which changes the appearance of the input/output page of kxmixer, but not the device in the lower right hand corner. I would be ok with whatever the panel says, but the IoStatus shows no other inputs besides 3 sp/difs, no outputs except 2 sp/difs and no matter what I patch, I can't seem to get an output from the analog outs on the APS. I tried what was suggested in some threads - switch F+B, mess with codec params, etc. Also went into DSP and patched a known good signal on the digital out (known good because it dies if I unpatch) to each output and nothing out. . . I tried reading this forum and couldn't find anything on this. . sorry if it is a repeat. Info: kx ver5.10 - 3533 debug Win98SE, DX 9.0a EMU/APS PII-233 Intel (not Via) PCI bridge 82443LX info from panel: CT4670 10K1 [E800] (Chip Rev 3 - unknown AC codec, but says that card has one) Mitch

Jul 3, 2003, 02:43 AM	#2
Max M. h/h member-shmember Join Date: Dec 2002 Location: Evil Empire Posts: 2,639 Rep Power: 69	It looks like EEPROM (where card id is stored) content is damaged... Hmm.. interesting - it happens very often with Lives and Audigies but APS... i see that at first time... hmm... it's a problem... you need to flash it with correct data... __________________

Jul 13, 2003, 01:26 AM	Thread Starter #7
monadity DriverHeaven Newbie Join Date: Jul 2003 Location: US - Georgia Posts: 5 Rep Power: 0	Problem solved. . . thanks! I pulled the prom and broke it, so had to burn a new one from the info at the link I posted above. (using 4001 as the subsys ident) Kxdriver works and I am already using it to do things that the EMU driver would only do lamely using Aux buses. Thanks, Mitch

Jul 3, 2003, 03:46 AM	Thread Starter #3
monadity DriverHeaven Newbie Join Date: Jul 2003 Location: US - Georgia Posts: 5 Rep Power: 0	Interesting - it had me thinking that maybe EMU/Creative kicked out a few with the wrong Ids. Doesn't cause a problem with APS drivers 1.0 and 1.5. Any Idea how to flash the card? I haven't seen anything like that on Creative/EMU site, but will check again. I also will re-verify the ID without any software installed. Thanks, Mitch

Jul 3, 2003, 04:56 AM	Thread Starter #4
monadity DriverHeaven Newbie Join Date: Jul 2003 Location: US - Georgia Posts: 5 Rep Power: 0	According to this article on upgrading a Live! card, it's an EEPROM - pretty standard little 93C46 chip. Have to pull it and tweak the device ID back. I re-verified that this card is misreporting. I think I had to directly select the EMU driver for a while, because their INF also does not match this card. Their driver is working again, though! Thanks for the help! (and I will see if it works when I dig out the prom programmer.) Mitch http://www.digit-life.com/articles/l.../livetolive51/

Jul 5, 2003, 03:50 AM	#5
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,119 Rep Power: 75	I'm afraid the driver won't be able to distinguish between your 'fake' APS and the 'real' SBLive Value without re-programming the EEPROM the kernel-level part relies on the PCI ID when configuring the hardware actually, the biggest difference (from the driver point of view) is between APS, 10k1, 10k2 (audigy), 10k2.2 (audigy2) and 10k2.2ex (audigy 2 ex) models. the particular card 'flavours' aren't so critical. the 10k1 and 10k2 are easily recognized even if the EEPROM content is reset (because the lowest part of the ID is read-only and cannot be 'flashed') but there's no way to distinguish APS from 10k1, as well as audigy, audigy2 and audigy2ex without the PCI Subsystem ID and chip revision information, so, tweaking the INF file won't help at all /Eugene

Jul 9, 2003, 01:20 AM	Thread Starter #6
monadity DriverHeaven Newbie Join Date: Jul 2003 Location: US - Georgia Posts: 5 Rep Power: 0	Thanks - I am surprised that the analog outs don't work even if the model is confused- something special that must be set on the card? I tried all of the outputs in the DSP patching page. Been too busy to deal with it the past week, but checked my programmer and it will burn a 93C46 - so I figure I'll just change the card ID by jumpering some wires from the E2 to the programmer - might have to lift pin 8 so the programmer doesn't cook trying to power the whole card. Maybe tonight. . . Mitch

Jul 13, 2003, 02:14 AM	#8
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,119 Rep Power: 75	good I'm glad your card is working now a small note for the other kX users: don't repeat re-programming unless you know what you are doing /Eugene