VOGONS


First post, by rpocc

User metadata
Rank Newbie
Rank
Newbie

UPD: Issue resolved, it was a bad or wrongly configured drivers
UPD2: Mysteriously, a new issue appeared, see post 7
Maybe this question was asked many times before, but to my surprise I haven't found any discussion about sound differences between CQM sound specifically of AWE64 comparably to other newer cards like Vibra16 or AWE32, SB32, etc.

Recently I got in my hands a bunch of various SB cards to test with the standard utility, DIAGNOSE.EXE.

Apart from PCM and wavetable/sampling synth, I was noticing that FM sound can be divided into 3 primary categories: OPL, CQM, AWE64 CQM. What I have found:

All cards with OPL/Yamaha chips were sounding as expected: an accurately filtered, mild FM synthesis, a bit noisy but pleasant to my ears.
More recent cards, such as CT3630 and generally cards with 3xxx index sound similarly, but with additional high-frequency content, which can be considered either as brighter sound or non-demanded high harmonics dependent on the listener's personal taste.
But what had surprised me, AWE64 Value, CT4500 and CT4520 has the worst CQM sound I heard, having sort of parasite clicking during playback of that Bach piece on all 4 cards I have. And similar artefacts were on the typical game I picked for testing: Supaplex, which sound I know well.

Is this a normal property of AWE64 or am I missing something about it? When I had only one, I was thinking it's a malfunction of my unit and even tried to replace LS245, capacitors, etc, but no, other units sound the same.
Is there other models that have the same property of FM synth to avoid? This is weird because normally AWE64 is fairly valued, obviously for its built-in wavetable bank and easy S/PDIF mod. What about AWE64 Gold?

Last edited by rpocc on 2024-03-04, 16:37. Edited 5 times in total.

Reply 1 of 10, by keenmaster486

User metadata
Rank l33t
Rank
l33t

I haven't noticed the clicking you mention on any of my AWE64 cards, Value or Gold, but they do have the typical harsh CQM sound to them, with some instruments in more complex OPL3 music not sounding correct as well.

World's foremost 486 enjoyer.

Reply 2 of 10, by rpocc

User metadata
Rank Newbie
Rank
Newbie

To be more specific, here's the video with a short demonstration of this effect.

Maybe something's wrong with my config.sys?
(only common and SB-related lines are kept):

DEVICE=C:\DOS\HIMEM.SYS /TESTMEM:OFF

REM SB16
DEVICE=C:\PROGRAMS\CTCM\CTCM.EXE
rem DEVICE=C:\DOS\EMM386.EXE NOEMS
BUFFERS=15,0
FILES=40
DOS=UMB
LASTDRIVE=Z
FCBS=4,0
DOS=HIGH
DEVICEHIGH /L:1,12048 =C:\DOS\SETVER.EXE
STACKS=9,256

REM SB16
DEVICE=C:\PROGRAMS\SB16\DRV\CTSB16.SYS /UNIT=0 /BLASTER=A:220 I:5 D:1 H:5
DEVICE=C:\PROGRAMS\SB16\DRV\CTMMSYS.SYS

The driver files are from AWE32 distributive and the settings work OK with any SB, the only issue I've got is the erratic FM-synth in AWE64.

Reply 3 of 10, by Joseph_Joestar

User metadata
Rank l33t
Rank
l33t

If you're running pure MS-DOS 6.22 try these drivers. They might be newer than what you're currently using.

For Win9x, download this driver CD image. That disc originally shipped with CT4520 cards.

PC#1: Pentium MMX 166 / Soyo SY-5BT / S3 Trio64V+ / Voodoo1 / YMF719 / AWE64 Gold / SC-155
PC#2: AthlonXP 2100+ / ECS K7VTA3 / Voodoo3 / Audigy2 / Vortex2
PC#3: Athlon64 3400+ / Asus K8V-MX / 5900XT / Audigy2
PC#4: i5-3570K / MSI Z77A-G43 / GTX 970 / X-Fi

Reply 4 of 10, by rpocc

User metadata
Rank Newbie
Rank
Newbie
Joseph_Joestar wrote on 2024-03-02, 10:02:

If you're running pure MS-DOS 6.22 try these drivers. They might be newer than what you're currently using.

For Win9x, download this driver CD. That disc originally shipped with CT4520 cards.

Thank you, that helped, case closed!

Reply 5 of 10, by keenmaster486

User metadata
Rank l33t
Rank
l33t

Weird that a driver would fix that problem. The FM functionality is just accessed directly through a port.

Might have something to do with the timing. You're supposed to observe time delays before you write to OPL2 or OPL3 ports, and not all software does that, plus whether or not it causes problems varies from machine to machine.

World's foremost 486 enjoyer.

Reply 6 of 10, by rpocc

User metadata
Rank Newbie
Rank
Newbie
keenmaster486 wrote on 2024-03-04, 06:07:

Weird that a driver would fix that problem. The FM functionality is just accessed directly through a port.

Might have something to do with the timing. You're supposed to observe time delays before you write to OPL2 or OPL3 ports, and not all software does that, plus whether or not it causes problems varies from machine to machine.

Well, I was testing the card using just Diagnose.exe, so there can be no more native and compatible program than Creative standard diagnostic utility...

However, I have new questions and thinking of continue this topic instead of making more.

Right after that I took all of my untested sound cards and made a large OPL playback test, using the last version of ADLIB Tracker II, using particular song called whatslov.a2m by encore.
The test setup is pretty simple: Iwill P55XB2 (Intel 430TX/PCIset TX Triton II) running Pentium MMX-166 with installed 3COM Etherlink III (No IRQ/DMA conflict with Sound blaster), a random ISA videocard and the soundcard. MS-DOS 6.22.

So, half of those were playing the song without any issues:
Reference sound example

Creative CT2940 Sound Blaster 16
Creative CT2940 SB Vibra16 FPNP
Creative CT3620 AWE32
Creative CT3630 AWE32 Value
Creative CT3600 AWE32
Creative CT4180 Vibra16 PNP?
Creative CT4500 AWE64 (Reference card)
Creative CT4520 AWE64

There were also another half, which was completely unable to play these songs, performing a complete mess instead of that:
Example of wrong sound. HARSH NOISE AND GLITCH INSIDE!
Creative CT2230 Sound Blaster 16
Creative CT2230 Sound Blaster 16 CSM
Creative CT2290 Sound Blaster 16 IDE
Creative CT2810 SB Vibra16 IDE
HP MM PRO 16V-A (HP 5064-2620)
HP MM PRO 16Y-A (HP 5064-2620)
Reveal SC400 Rev. 4
SoundPro
Asound Asonic-NET

I was trying to play it with loaded CTCM from AWE64 package and SB16 package (completely reinstalled) and also without any CTCM or any other creative-related driver and no difference: neither of these cards (all with Yamaha YMF262-M, CT1747 or OPL3 clones by Aztech and others) were able to play the song properly.

Also, one of AWE64 units were displaying stable corruption of sound as like one or couple of FM channels were failing. What could cause such behaviour?
Defective AWE64 OPL malfunction example

The listed true-OPL cards are OK because I was able to play any mods from older version of Adlib Tracker II, designed for OPL2 (Adlib Tracker II (OPL2) ver. 2.0.27). I quite can't get that: it's specifically stated on the website that the tracker is compatible with OPL3 but neither of original cards worked well, and only CQM cards were OK.

Example of OPL2-compatible song

The last question is: how could I fix the 10 kHz digital hiss on my CT2810 VIbra16-based card? Looks like it has something to do with voltage regulator, power rail recapping but maybe somebody knows something more specific to look at?
Hissing CT2810 example with OPL2-compatible song

Reply 7 of 10, by auron

User metadata
Rank Oldbie
Rank
Oldbie

just to clarify the initial problem, diagnose.exe shouldn't matter here, the clicking issue was discussed in the past and was said to be fixed by upgrading aweutil from 1.20 to 1.35: AWE64 clicking on FM

regarding adlib tracker ii, it seems that for the most part it's older cards that gave you trouble. did you try running the program with l1+l2 cache disabled in BIOS? and the vibra issue as far as i know is inherent to the chipset, i really wouldn't recommend those cards.

Reply 8 of 10, by sfryers

User metadata
Rank Newbie
Rank
Newbie
rpocc wrote on 2024-03-04, 16:22:

There were also another half, which was completely unable to play these songs, performing a complete mess instead of that:
Example of wrong sound. HARSH NOISE AND GLITCH INSIDE!

That's exactly how my Aztech AZT2320 (integrated OPL3) sounded until I found this thread: How-to: Resolving ISA sound and OPL2/3 issues. Using the suggested BIOS settings fixed the problem immediately.

MT-32 Editor- a timbre editor and patch librarian for Roland MT-32 compatible devices: https://github.com/sfryers/MT32Editor

Reply 9 of 10, by rpocc

User metadata
Rank Newbie
Rank
Newbie
sfryers wrote on 2024-03-04, 18:04:
rpocc wrote on 2024-03-04, 16:22:

There were also another half, which was completely unable to play these songs, performing a complete mess instead of that:
Example of wrong sound. HARSH NOISE AND GLITCH INSIDE!

That's exactly how my Aztech AZT2320 (integrated OPL3) sounded until I found this thread: How-to: Resolving ISA sound and OPL2/3 issues. Using the suggested BIOS settings fixed the problem immediately.

Oh, thank you a lot, setting 8-bit I/O Recovery to 8 immediately solved all problems except for 4520. And problem with 4520 was stupidly solved by returning the original 74LS245 in place of the new one which I had replaced priorly thinking that it might be causing the clicking issue. Now I'm happy and preparing a kind of test with 10 Creative and 8 other cards which I recorded last night in a long and boring run. Will post as soon as I consolidate the measurement data.