Skip to main content

Environment:

- ASUS p4b266, 2.0 ghz, with 1gb memory

- Windows 2000 service pack 2, not really tuned for audio but given my little experiment I don't think this should matter. It was not installed as a standard PC. I have another partition that I had installed as a Standard PC. I don't remember how much tuning I did with it, but the problem I am about to talk about also occurs there.

- ASUS C-Media audio driver: 7/16/2002, 5.12.1.639 (installed two days ago). This came with p4b266.

- old SONY DAT PCM-2600 without a manual. It is connected via COAX(S/PDIF) to the S/PDIF In on the back of the PC. That is internally connected to the p4b266, it dosn't use a PCI slot.

- the cable I am using is NOT an S/PDIF RCA-to-RCA cable, it is just typical analog RCA cable

- I changed a parameter in the C-Media mixer to send S/PDIF directly to the analog out which in turns goes to my Hi Fi.

What works
- ignoring the digial I/O, the analog I/O (line and mic), really high quality (ho ho), works fine.

The problem:
when I switch the C-Media mixer to S/PDIF in as opposed to Line In, I can hear the audio, in this case just someone talking. It is 95% clean. The remainder is crackling/fuzz/distortion that is only audible when someone is talking. When the person stops talking, the crackling goes away.

I have read about clicks and pops that people have because of clocking problems. My impression is that they are hearing intermittent sounds, they do not occur constantly. In my case this is constant noise when the person is talking.

The audio from the analog out of the DAT is fine. The meters on the DAT only go up to about -6.

I haven't tried sending audio from the PC via its S/PDIF out to the DAT because I can't figure out what to do about clocking due to the lack of a manual. With the DAT set to Digital In, it just flashes "UNLOCK". I don't really care about this path.

Question:
- Is it possible that the distortion, or whatever it is, is caused by my cheating and just using an analog RCA cable as opposed to a true S/PDIF cable?

Comments

Tommy P. Sun, 03/16/2003 - 02:47

Question:- Is it possible that the distortion, or whatever it is, is caused by my cheating and just using an analog RCA cable as opposed to a true S/PDIF cable?

Given that you have the sampling rates set to the same 44.1khz(the Sony should be set to output at the same frequency that the C-Media is set to )
It could be the limited bandwidth of the audio RCA cable causing extreme jitter, or a mismatched S/PDIF gain setting.
Sounds like an S/PDIF I/O gain issue, not a cable issue. Although most newer equipment is more standardized now, maybe the Sony has level adjustments neccesary for the S/PDIF data signal.
Although you should be using a higher bandwidth S/PDIF data cable instead of the standard audio one you're using, you would more likely hear pops and clicks caused by the jitter.
First thing, if you can, get the right S/PDIF cables hooked up.
Then see if the Sony has switchable S/PDIF gain settings. I believe there were three settings that were being used when the specification for
S/PDIF was first came out. Sometimes there is a variable setting, but usually its just a high/low, or high/med/low setting. I don't think the C-Media has these settings available, but I havn't checked it, so you'll have more luck adjusting the Sony to match up to the C-Media.

Fozz Tue, 03/18/2003 - 05:36

Tommy.

Thanks for the response.

The Sony and the PC/C-Media are set to the same sampling rate of 48k.

Neither the Sony nor the C-Media software have any gain controls. Now that I say that, I think I will record at various levels and see if I still hear the same sound at the different levels.

I'm thinking the signal flow in the dat looks like this:
- play the tape it self
- send the digital signal out the S/PDIF or AES/EBU
- send the digital signal to the digital to analog converter

I don't know if this makes sense but I'm thinking if there were something wrong with the actual digital signal in the Sony, then it would also show in the analog out. But the analog out is fine. Therefore I'm thinking the problem lies some place after the signal leaves the Sony:
- the cable not being correct
- the C-Media hardware on the motherboard
- the C-Media drivers.

There is just a skinny little ribbon cable inside the PC connecting the S/PDIF IN/OUT ports on the back of the PC to the motherboard. I don't know how safe that is from electro magnetic interference.

I am going to experiment some more and get a real S/PDIF cable, as you suggest.

Thanks again for the info.

Opus2000 Sun, 03/23/2003 - 09:28

Well, first of all you can not adjust input level of a digital signal...digital is digital and can not be altered in terms of level input.
Make sure the C Media is set to take the clock from the incoming signal from the Sony.
Also you may want to check and see if there is a copy protection scheme on the Sony digital out which may possibly cause something of this nature.

I highly doubt that it's a cable issue as any RCA cable will do just fine. The clock is embedded into the signal and not through the cable itself. The reason for digital cables is merely to protect from static discharge or impedance reasons.

I would first look into the clocking scheme for the C Media more so than anything else

Opus :D

Fozz Sat, 03/29/2003 - 03:07

I bought a 75 ohm S/PDIF cable to use instead of the typical analog RCA cable.

I've been experimenting off and on for the last week or so. I don't totally understand everything but I can, so far, get rid of the distortion, when it does appear. This is what I have learned so far:

Power on periferals first:

If you had asked me, should one power on periferals, before powering on the PC, I would have said yes. Did I think about that when I first started testing the DAT and the S/PDIF ... No!.

Now I power on the DAT, etc, before the PC, but that is not a fool proof solution.

What I have been doing every day for the past week or so is power on all periferals, then the PC and then immediately play a DAT tape. If there was distortion the day before, I fixed it (more about that in a moment) so that presumably the previous day ended with "no distortion". In the morning, when I would power everything on in the correct order, half the time it would be fine and half the time there would be distortion, even though I had fixed it the day before.

Switch the S/PDIF Format:

The little C-Media mixer that came with the Asus p4b266 as three buttons on the left hand side: Volume, Record and Advanced. When I click on Advanced, that pops up a window containing an S/PDIF tab and 4 others. This S/PDIF tab has an Output and an Input section.

The Input section has a Format option with two choices: Normal and Reverse.

Whenever I hear distortion all I have to do is switch the Format. If it is set to Normal, I switch it to Reverse, if it is set to Reverse, I change it back to Normal. That switching has always gotten rid of the distortion.

I don't understand why it would be working fine on a given day and not on the next day. Once I get it working the only thing I do for the remainder of the day is maybe a little audio work recording from the DAT into Cakewalk or dial into my ISP and surf the web. That dialing in doesn't screw things up.

I'm not going to worry about why. If there is distortion, I will just switch the format. I hope I don't have this problem when I get a real audio interface, most likely the RME multiface or Digiface.

Use a 75ohm S/PDIF Cable, not an Analog Cable

Once I was comfortable with knowing how to get rid of the distortion I did a few experiments by switching between the 75 ohm S/PDIF cable, that I used in my experimenting above, and the analog RCA cable that I had originally used before I bought the S/PDIF cable.

The analog RCA cable consistently gives static and distortion. The switching of the Format from Normal to Reverse does not eliminate it. Some of the distortion goes away but there is so much static that you can barely hear what is on the tape.

Powering everything off, switching back to the S/PDIF cable, and changing that Format option, if need be, eliminates the static and distortion.

Opus2000 Sun, 03/30/2003 - 08:47

Fozz

Using a 75 Ohm SPDIF cable most likely was not the main fix here.

I think what's happening is that the clock source is not properly being recognized by the built in digital input on the motherboard.

When you have the external digital device on first and power the computer on it sees the embedded clock signal from the SPDIF connection and latches on to it.

I think that was your issue. Clicks, pops or possible distortion may arise from improper clocking scenarios

Opus

Opus2000 Sun, 03/30/2003 - 13:24

Tommy

tsk tsk tsk tsk....shame on you!!
Look at any digital input device and tell me where you can adjust the "digital" input level?!!

DA-88's, ADAT's, Tascam MX2424, ANY Apogee product, DAT Recorders.

You can adjust the analog "output" if the unit is a D/A but NOT the digital input!

HA! Trying to make me look bad here! :D
:s:

Fozz Mon, 03/31/2003 - 01:44

Opus,

For the combination of hardware that I have, the 75 ohm cable is my only choice for getting rid of that distortion and static. When I powered things off and on in the same sequence and only switched cables, the RCA cable consistently gave distortion and static which I could not get rid of by switching that wierd C-Media Foramt option. Maybe I got rid of some of the distortion but it still sounded horrible. With the 75 ohm cable, I didn't get the static but half the time I would get distortion which I could consistently get rid of by switching the Format option.

Regarding your comment: "the clock source is not properly being recognized by the built in digital input on the motherboard.", that is an interesting thought. That could be related to the C-Media mixer's Format/Normal/Reverse option. Maybe, when the clock source is not recognized I get horrible distortion. Switching the format cleans things up perfectly.

And, maybe the real answer is that I should get with the program and get a real audio interface. If you get what you pay for, the audio interface junk on the p4b266 probably cost $10. An Apogee PSX-100 and RME Digiface sound like a nice combination.