After some trial and error I found out that it was a staggered header, not sequential. Just leaving this here in case someone needs the info later on.

I bought a CT4520 recently, the card shipped with SPK/LINE header jumpered like this.

What does the jumpers do here?

BTW, I also found that the card came with an spdif header without soldered pins. Does this mean the card support digital output potentially? What if I solder the pins to it?

SPDIF one gives TTL signal (not usable for coax as is) of only the EMU8000+CQM, it will not have any of the SB sound in it so the usefulness is a bit limited.

SPK/Line enables (SPK) or disables (Line level) the onboard amplifier chip.

EDIT: This is wrong.

Tiido wrote on 2023-08-16, 04:48:

SPDIF one gives TTL signal (not usable for coax as is) of only the EMU8000+CQM, it will not have any of the SB sound in it so the usefulness is a bit limited.

Not sure if there's any difference between the AWE64 Gold and Value versions with regards to the SPDIF header, but SPDIF output on my CT4390 works just fine with a passive RCA bracket that I cobbled together myself. Here's a picture and some music recordings.

As you say, it's mostly useful for recording AWE and CQM music, but it is possible to also output PCM if the source is 44.1 KHz 16-bit stereo. In this post I showcase Deus Ex running at those settings and correctly outputting via SPDIF.

appiah4 wrote on 2023-08-16, 06:42:

SPK/Line enables (SPK) or disables (Line level) the onboard amplifier chip.

So what is the point of having two separate external connectors(Line Out [green] / SKP Out [black]) when there is a on/off switch of the onboard amplifier chip? I'd assume that both Line Out and SPK Out output amplified signal if the onboard amplifier is turned on by jumpers and vice versa.

What's the current state of the amplifier? Is it turned on? If that is the case how can I turn it off?

Will this work:
5 [3 1]
6 [4 2]

Joseph_Joestar wrote on 2023-08-16, 07:00:

Not sure if there's any difference between the AWE64 Gold and Value versions with regards to the SPDIF header, but SPDIF output on my CT4390 works just fine with a passive RCA bracket that I cobbled together myself. Here's a picture and some music recordings.

As you say, it's mostly useful for recording AWE and CQM music, but it is possible to also output PCM if the source is 44.1 KHz 16-bit stereo. In this post I showcase Deus Ex running at those settings and correctly outputting via SPDIF.

Just straight through adapter like that is not a good way. The chip output is not at all meant to drive a 75ohm load of SPDIF coax connection and will have to dissipate excessive current, there's also no ESD protection so you may kill the output when connecting/disconnecting the cable. The very least there should be a resistor divider which will get the signal level right and dramatically reduce load on the logic level output and with it reduce any effects of ESD issues as the current of those pulses gets reduced aswell and the internal ESD protection diodes of the chip have bigger chance of surviving. Proper way has 374 ohm resistor in series with the incoming signal and 90.9ohm one to ground, forming a divider, and ideally there's a 1:1 pulse transformer on this output to create galvanic separation to break ground loops etc. that may form.

I forgot about the 16bit 44100Hz Stereo PCM output capability. From what I see it only worked in Windows, at least I could never get any DOS game to output anything over the SPDIF output.

OK I just checked and not a single one of the 6 CT4520s I own have that jumper.

Actually, I have used that HEADER to route Line Out from the card to another card internally, it would seem.

So no, I am WRONG, that header is not an amplifier on/off jumper at all. It is, apparently an internal SPK/Line Out header.

I THINK the pinout is something similar to below:

LL NC SL
LR GN SR

Or some such. Remove the jumpers for safe operation.

Tiido wrote on 2023-08-16, 15:05:

Just straight through adapter like that is not a good way. The chip output is not at all meant to drive a 75ohm load of SPDIF coax connection and will have to dissipate excessive current, there's also no ESD protection so you may kill the output when connecting/disconnecting the cable.

Cheers, I'll try to find a better solution then. Didn't realize it could cause damage to the card.

I forgot about the 16bit 44100Hz Stereo PCM output capability. From what I see it only worked in Windows, at least I could never get any DOS game to output anything over the SPDIF output.

Me neither. I tried Duke3D and Crusader: No Remorse in pure DOS, both of which have support for 44.1 KHz 16-bit stereo and got no sound from either of them. Under Windows, playing a 44.1 KHz WAV or MP3 file will work fine too.

Tiido wrote on 2023-08-16, 15:05:

Just straight through adapter like that is not a good way. The chip output is not at all meant to drive a 75ohm load of SPDIF coax connection and will have to dissipate excessive current, there's also no ESD protection so you may kill the output when connecting/disconnecting the cable. The very least there should be a resistor divider which will get the signal level right and dramatically reduce load on the logic level output and with it reduce any effects of ESD issues as the current of those pulses gets reduced aswell and the internal ESD protection diodes of the chip have bigger chance of surviving. Proper way has 374 ohm resistor in series with the incoming signal and 90.9ohm one to ground, forming a divider, and ideally there's a 1:1 pulse transformer on this output to create galvanic separation to break ground loops etc. that may form.

I forgot about the 16bit 44100Hz Stereo PCM output capability. From what I see it only worked in Windows, at least I could never get any DOS game to output anything over the SPDIF output.

Very helpful information! There was a thread about adding SPDIF to a CT4520. I think your precious info should be linked to that one in case others will find it in the future.

appiah4 wrote on 2023-08-16, 15:09:

OK I just checked and not a single one of the 6 CT4520s I own have that jumper. […]
Show full quote

OK I just checked and not a single one of the 6 CT4520s I own have that jumper.

Actually, I have used that HEADER to route Line Out from the card to another card internally, it would seem.

So no, I am WRONG, that header is not an amplifier on/off jumper at all. It is, apparently an internal SPK/Line Out header.

I THINK the pinout is something similar to below:

LL NC SL
LR GN SR

Or some such. Remove the jumpers for safe operation.

I think it's the previous owner who misunderstood the header for a switch, hence the jumpers setting.

BTW, I'm wondering why you got yourself 6 CT4520s.

GunKneeNeon wrote on 2023-08-16, 15:45:

appiah4 wrote on 2023-08-16, 15:09:

OK I just checked and not a single one of the 6 CT4520s I own have that jumper. […]
Show full quote

OK I just checked and not a single one of the 6 CT4520s I own have that jumper.

Actually, I have used that HEADER to route Line Out from the card to another card internally, it would seem.

So no, I am WRONG, that header is not an amplifier on/off jumper at all. It is, apparently an internal SPK/Line Out header.

I THINK the pinout is something similar to below:

LL NC SL
LR GN SR

Or some such. Remove the jumpers for safe operation.

I think it's the previous owner who misunderstood the header for a switch, hence the jumpers setting.

BTW, I'm wondering why you got yourself 6 CT4520s.

You never know when you might need one. Or six.

Tiido wrote on 2023-08-16, 15:05:

Just straight through adapter like that is not a good way. The chip output is not at all meant to drive a 75ohm load of SPDIF coax connection and will have to dissipate excessive current, there's also no ESD protection so you may kill the output when connecting/disconnecting the cable. The very least there should be a resistor divider which will get the signal level right and dramatically reduce load on the logic level output and with it reduce any effects of ESD issues as the current of those pulses gets reduced aswell and the internal ESD protection diodes of the chip have bigger chance of surviving. Proper way has 374 ohm resistor in series with the incoming signal and 90.9ohm one to ground, forming a divider, and ideally there's a 1:1 pulse transformer on this output to create galvanic separation to break ground loops etc. that may form.

To follow up on this, I recently saw a picture of an unboxed AWE64 Gold, and the official SPDIF bracket doesn't appear to have any additional circuitry. It's seems to be completely passive, unless Creative stowed away some components inside the actual cable.

This looks fairly similar to my homemade bracket, albeit with better cable shielding. Not sure what to make of that.

That is definitely gonna hide at least one resistor in the heatshrink. Other factor is that the Coax device receiving a TTL signal is not required to survive being subjected to the possibly much higher voltage level (depending on how much current the source can push into the 75ohm termination), it is most unlikely that it is just a straight cable.

Tiido wrote on 2023-09-25, 12:02:

That is definitely gonna hide at least one resistor in the heatshrink. Other factor is that the Coax device receiving a TTL signal is not required to survive being subjected to the possibly much higher voltage level (depending on how much current the source can push into the 75ohm termination), it is most unlikely that it is just a straight cable.

I think you may be right about resistor(s) being hidden beneath the heatshrink at the base of the bracket. Here's a slightly better picture which shows a bit more detail:

Maybe someone who has one of these original brackets would be willing to take off the heatshrink and examine it more thoroughly, for research and replication purposes.

I was under the impression that the gold contained all the passives to condition the signal on the card, as opposed to the other AWE cards which did not

You're right, I had totally forgotten about it. Gold indeed has an actual line-driver chip on it for the SDPIF output. AWE has a straight signal from EMU8000 and that definitely needs additional effort.

Tiido wrote on 2023-09-25, 13:39:

You're right, I had totally forgotten about it. Gold indeed has an actual line-driver chip on it for the SDPIF output.

So a completely passive coaxial SPDIF bracket would be fine for the AWE64 Gold after all?

Yeah, only for Gold but not any of the others.