I could plug in an amp or speaker, but there are two problems with that:

Are you saying that you are currently trying to use it with passive (non-amplified) speakers? That's not possible, as the card has no on-board amplifier to drive passive speakers.

Is there another system that you could test the card in?

What amplified speakers have you tried?

Have you tried with a pair of headphones?

derSammler wrote:

the card has no on-board amplifier to drive passive speakers.

If that's why the output is so quiet, then OK.

Most of my testing was done using headphones. So it's really jarring to hear the difference between the sound card and internal audio (the SiS audio on this thin client has quite an amp).

However, this doesn't address the audio interference that I have been getting out of the sound card.

Have you tried disabling the built-in SIS audio to see if that makes a difference?

It almost sounds like a grounding issue or filtering capacitor issue as it definitely should not be getting interference such as mouse movement noise.

Are you using the same exact model of HP thin client that is on Phil's video?

If not, what is the exact model number?

The thin client is the exact same one Phil had, which I bought from the seller he linked. It has the expansion module for PCI card slots.

And yes, I have tried disabling the on board audio and other devices, except for the USB controller, and no luck.
(I probably should test with the USB controller disabled for good measure)

BenWit wrote:

The thin client is the exact same one Phil had, which I bought from the seller he linked. It has the expansion module for PCI card slots.

And yes, I have tried disabling the on board audio and other devices, except for the USB controller, and no luck.
(I probably should test with the USB controller disabled for good measure)

You've got a PM.

I did some further testing, and found interesting stuff.

Disabling the USB controller didn't help. However, I noticed a proper pattern with the audio interference. The noises line up with the drive access light, and whenever there's a display update on Windows (especially harsh noise when Start menu is opened).

The mouse issue I reported on earlier doesn't happen under Windows, but under DOS with the CuteMouse driver loaded.

cyclone3d wrote:

You've got a PM.

Thanks for the offer.

Ultimately, I think I'm going to keep the sound card for a future project. The interference issue appears to be caused by my particular HP thin client, but be sure to let me know how well it works with yours.

Also, I haven't gotten a clear answer on this. Is it normal for for these sound cards to have a very quiet output and require an external amp?

BenWit wrote:

Thanks for the offer. […]
Show full quote

cyclone3d wrote:

You've got a PM.

Thanks for the offer.

Ultimately, I think I'm going to keep the sound card for a future project. The interference issue appears to be caused by my particular HP thin client, but be sure to let me know how well it works with yours.

Also, I haven't gotten a clear answer on this. Is it normal for for these sound cards to have a very quiet output and require an external amp?

The card itself will not power passive speakers. You need speakers that have an amp in them.

If the output is low with headphones, then there may be something wrong with the card. You will never get the same volume over headphones as you will with the built-in amplified setup, but that is completely normal for a card without a built in amp. With headphones it should be loud enough though.

I'm going to see if I can reproduce the issues you are having once the thin client I ordered arrives.

I do have a few ideas.

Do you have a different PCI sound card you can try in Windows to see if you still get interference?

You could always try to get the seller of the thin client to replace the thin client. Noise over the PCI bus like this is definitely not normal.

I did testing with headphones with fairly low impedance. It was a little quiet with games under pure MS-DOS, but I was able to hear Doom and Epic Pinball just ok. The biggest problem really was the interference that occurred with mouse movement on the CuteMouse driver, along with drive access noises. Under Windows, I could barely make out the startup/system sounds and MIDI playback. I didn't do any testing with DOS games under Windows.

I also had a change of focus on what to run on my thin client. I might pick up a PCI GPU for running Windows games from the late 90s and very early 2000s. However, I am open to picking up another thin client if it works perfectly fine with the sound card, and setting it up as a dedicated DOS machine.

Hmmm... I kinda think something is wrong with that card.

How loud is the music and sound in the DOS test program? Just run SETUPDS from DOS and you should be able to test. It should be pretty loud with regular headphones.

The thin client should be here in a few days and I will test with a card then.

OK, this time around I have audio samples under DOS. These were recorded on another computer through the line in with Audacity (input gain was at 65 percent). There were some interesting problems.

First off, the card outputs a really loud pop sound when the thin client is powered on. Then I ran into a curious issue, a constant popping sound. This happened on the line in recording and a pair of powered computer speakers, regardless of audio cable used. The constant popping doesn't happen on headphones or my unpowered speakers. The popping eventually slowed down when I launched Epic Pinball, and was gone by the time the game got to the auto demo. I'm starting to lean toward that there is something wrong with the sound card itself.

For comparison purposes, I tested the SETUPDS sound files under Windows with the built in SiS audio.

Here are the recordings, just warning that the first 3 files have VERY LOUD popping from the sound card.

Thin client arrived today and I am seeing the same weird issues you are having.

I am going to see if Phil can give us the exact settings and setupds, etc he is using with this thin client.

I even tried DSDMA which ended up only locking up the system when I went to do the sound test.

Something is really weird.

I tried again, this time I updated the BIOS like Phil did in his HP thin client video. It didn't improve anything sound card wise.

For the DOS driver, I have been using the one modified for SiS chipsets which includes an SISINIT.BAT script. The script consists of:

@echo off
setupds.exe /s
ymfirq 5 edge
SET BLASTER=A220 I5 D1 T4

It has worked fine for getting sound under DOS, doesn't even need an expanded memory manager. Though the audio interference happens even if the driver is not running.

Ok, here is a vid on how Phil set up the card on his thin client.

https://youtu.be/F2O0FolLDwc

It also sounds like there is some background noise/interference until the driver in Windows is loaded up.

Looks like he is using DSDMA instead of the modified for SiS chipsets setupds.

I tried it myself in pure DOS but it didn't seem to help.

What I did notice is that Phil was booting into Windows 98 SE first and then rebooting into DOS. I wonder if that has anything to do with it.

Also check out the file name of the BIOS file in the thinclient vid. It is different then the name of the one that the boot floppy flasher maker gave me.

I'll do some more testing on my end tomorrow if I get the chance.

I am working with only a 512MB DOM right now so it is kinda tricky to get everything set up. I think I am going to resort to using a parallel CD drive until the cables I ordered come in.

I'll double check this card in my KT7A build which already has the same YMF744 card in it. I'll also test that card in the thin client since I know it is fine.

Ok, so finally got around to pulling out a Pentium 233 MMX system that I have one of these cards installed in to do some testing.

1. The card works fine in the 233 MMX system. So loud with my Senheiser Momentum headphones that I had to turn the main volume down to -12db in DOS to make it not distorted.

2. The card that was in the 233 MMX system shows the same exact super low volume symptoms when installed in the thin client.

My thoughts so far are this:
1. Maybe there are different versions of the riser card. the one that doesn't work properly is not feeding some voltage line through to the card. I am going to ask Phil to get me some close up pictures of the riser from his thin client so I can compare. It really looks to me like the one I have doesn't have all the traces connected. It is not a straight through riser. There are blank places for jumpers, caps, plugs, etc.

2. The card must be initialized in Windows first on this platform. I am waiting for a 44-pin IDE cable to arrive so I can actually hook a drive up that is big enough for me to install Windows 98SE from it.

A bit more progress to report. I have tested all the different YMF7x4 cards I have.

I have a couple other AOPEN ones, but the model is AW724. If the jumpers are in the "Ear Out" (1-2) position, I get the same low volume as with the 744L II cards. Setting the jumpers to "Line Out" (2-3) position, they work just fine.

All other cards worked fine except an early version of the Yamaha WF-192XG. It has a resistor at R70 as well as a YMF724C-V chip compared to the later ones which have a YMF724E-V.

The C-V based card will not work with DDMA. I am probably going to remove the R70 resistor at some point to see if that makes a difference.

At this point, I am pretty sure that the problem with the 744L II cards is that the HP thin client is missing voltage on one of the PCI pins that the card needs for the power for the op-amps.

I tested with a PCI riser cable instead of the riser card and it didn't make a difference so I think it is a design flaw of the client itself. I am hoping that I can just jumper power from a different PCI pin to the one that is missing power if my theory is correct.

I'm in the same boat- issues with the same card on the same thin client. With headphones, super quiet and fairly distorted audio. When I hook up an RCA cable, I was able to get loud but distorted volume for a little bit by amplifying through my cheapo behringer audio interface but I haven't been able to replicate that since, and hooking up to powered speakers or a TV I can't hear anything.

Have you guys seen this thread?
Strange sound issue on YMF744, connectivity problem?

Same sound card on an unnamed thin client. I'm able to replicate that experience as well- random volume surges when plugging and unplugging. Fooling around a bit, I figured out that if you have one RCA cable plugged in and ground the other, you can reliably get that volume surge, but only for a second- there doesn't seem to be a way to maintain that volume, and granted it's still distorted. Sorry if that's a bit of a non-sequitur, just figured might as well as share in case it meant something- if it does or doesn't I sure as heck wouldn't be able to tell.

Bringing this thread back from the dead.

I have figured out what the problem is with the AW744-L2 card and the HP T5720 thin client.

It is PCI bus noise.

Fully populating the capacitor places on the PCI riser card completely fixes the issue. I am guessing this will also fix issues with "compatibility" with other cards as well. It should also significantly clear up the audio output if you are using a different sound card.

For the smd ones, I just grabbed some that were the correct physical size off of a dead video card. I tinned the pads with my iron and then used a hot air solder gun to melt the solder and each cap just dropped into the proper place though I did hold them from blowing away by using a pick on one side to keep them from moving from the air current from the hot air gun. Temp was at 350F.. just depends on what temp your solder melts at.

For the through the hole electrolytic capacitors, I used these (Left to right when looking at the Riser)

TC4 (12v line) - 16v 220uf - this one was a bugger to do since there was crappy solder in the holes from the factory. Had to set my soldering iron to 425F for this one.

TC2, TC1, TC3 - 6.3v 1000uf - these were super easy but I did have to carefully scrape the solder pads to expose the copper as it seemed like there was some sort of coating on them. Soldering iron temp was set at 375F for these.

I also had to remove 4 of the fins from the chipset/GPU heatsink for the caps to clear. After looking at it, it looks like that is not necessary if you mount TC4 and TC2 on the back of the riser.

On a side note, if you want to upgrade the CPU to an NX1750, the CN4 header outputs 5v so you can add a 5v fan to cool the CPU. I will be adding a 50x15mm blower fan to suck hot air from the CPU heatsink and exhaust it out the case. Not sure if I will need to trim part of the case or not yet (to allow it to exhaust the hot air out of the case). This should not only cool the CPU better, but should help keep the whole inside of the case cooler as well.