I have a M559 mobo mounting Cm8330. I'm using this board exclusively in pure msdos. But I noticed pc speaker sounds coming from LINEOUT 3.5" are veeeery quiet. And there is no pc buzzer volume in cmmix. Any way to fix that?

appiah4 wrote on 2024-05-24, 06:58:

All these pages of talk and nobody seems to have made a summary table.

I see your point but I can also see why it doesn't exist. I have stumbled upon this thread because I was checking for alternatives. There is one very important point that can't be easily integrated into such a table. That is taste or personal preference. Do you like the sound or do you not? What really matters to you in a sound card? And that obviously varies widely. I have seen several praises for the CMI8330. I got a specimen here and I repeatedly stuck it in and tried to like it. Damn, I really tried to like it when I only have one ISA slot. But I just don't like the sound of it. So despite being a very sweet aio solution on paper, for me it is not!

As far as I read, there are only two non-SB cards that can be considered SB16 clones: CMI8330 and that ALS100. For the later, I read that it is supposed to be noisy. That instantly kills it for me.

Ultimately, thanks to unisound, I subscribe to your view that it is much easier these days to simply use more than one card and switch between them. As I have found out, one can pretty easily even pick traits from different cards and 'frankenstein' them into a new "sound solution" that has far less compromise than either the CMI or the ALS100. Even with only one ISA port, you can run an ESS-Solo (PCI) with add on wavetable and any SB16-64 card of your choice together.

The CMI8330 went into my sale pile, along with the AWE32 CT2740 R3. Yamaha 719 is still under consideration, might do the filter mod.

ux-3 wrote on 2024-06-20, 05:56:

CMI8330 and that ALS100. For the later, I read that it is supposed to be noisy. That instantly kills it for me.

The ALS100 chip is not inherently noisy. But ALS100-based cards are often built to a very tight budget*, so the layout of the card and possibly the speaker amplifier (if present) are responsible for the noise. By the way, the CMI8330 is also used a lot on budget cards, so I wouldn't expect them to be better.

*: This is kind of the respectful way to say: "Many ALS100-based cards are absolutely cheap crap".

Having built a very capable ALS100 card (The Mega Card for Amstrad MegaPC) I can tell that having flawless external bits won't help the chip much, it is still noisy (very high amount of popcorn/flicker noise) and very high THD (0.1% range). It just sounds plain dirty even in the optimal application.

CMI8330 is also similar but not quite as bad, it is definitely let down by inferior silicon process, the analog parts just are not good...

Tiido wrote on 2024-06-20, 22:20:

Having built a very capable ALS100 card (The Mega Card for Amstrad MegaPC) I can tell that having flawless external bits won't help the chip much, it is still noisy (very high amount of popcorn/flicker noise) and very high THD (0.1% range). It just sounds plain dirty even in the optimal application.

CMI8330 is also similar but not quite as bad, it is definitely let down by inferior silicon process, the analog parts just are not good...

I wonder how bad/good the ALS4000 chip is on that front.

Tiido wrote on 2024-06-20, 22:20:

CMI8330 is also similar but not quite as bad, it is definitely let down by inferior silicon process, the analog parts just are not good...

So, what about that new MK8330 card?
Is it hereditarily tainted?

I'm sure it gets maximum out of the chip, but when the silicon is the limiting factor there's not much that can be done about it unfortunately. Also some CMI8330 flavors are better than others, for example the chips without integrated 3D circuit are less noisy than ones with it (CMI8330A/3D is one such). There are several brandings of the 8330 but I no longer remember if there was any significant performance difference among them, only the integrated 3D function really made a big (and negative) difference in analog performance, being significantly more noisy even with "3D" disabled. On some cards there's a little external 8 pin chip for that, which similarly adds extra noise to the final output.

darry wrote on 2024-06-20, 22:40:

I wonder how bad/good the ALS4000 chip is on that front.

I have one such card and I recall it being quite similar, it definitely didn't stand out with nice clean sound. Similarly all the Avance Logic and early Realtek AC97 codecs (Realtek bought Avance Logic from what I know), they weren't all that good.

Tiido wrote on 2024-06-20, 23:47:

I'm sure it gets maximum out of the chip, but when the silicon is the limiting factor there's not much that can be done about it unfortunately. Also some CMI8330 flavors are better than others, for example the chips without integrated 3D circuit are less noisy than ones with it (CMI8330A/3D is one such). There are several brandings of the 8330 but I no longer remember if there was any significant performance difference among them, only the integrated 3D function really made a big (and negative) difference in analog performance, being significantly more noisy even with "3D" disabled. On some cards there's a little external 8 pin chip for that, which similarly adds extra noise to the final output.

darry wrote on 2024-06-20, 22:40:

I wonder how bad/good the ALS4000 chip is on that front.

I have one such card and I recall it being quite similar, it definitely didn't stand out with nice clean sound. Similarly all the Avance Logic and early Realtek AC97 codecs (Realtek bought Avance Logic from what I know), they weren't all that good.

Too bad about the ALS4000. Even worse is that it does not have I2S or S/PDIF, because it is other wise rather well-rounded on the feature side of things.

When I got some more time, I will experiment with the Trident 4DWAVE instead.

I do have a classic 8330 and while I could deal with the noise level, the distortion/filtering kills the card for me. I just don't like the way it sounds. I just use two cards instead.

In the noughties, I went from an ALS4000 to a SBlive CT4870 and that "upgrade" was very underwhelming as far as sounding much nicer or anything, I know, budget to budget, but probably says the ALS4000 can be "okay". Was only running DX and OGl windows XP and win98 stuff at the time though, so don't know about DOS behaviour... I could get it working for Quake I think, but not super sure.

BitWrangler wrote on 2024-06-21, 16:45:

In the noughties, I went from an ALS4000 to a SBlive CT4870 and that "upgrade" was very underwhelming as far as sounding much nicer or anything, I know, budget to budget, but probably says the ALS4000 can be "okay". Was only running DX and OGl windows XP and win98 stuff at the time though, so don't know about DOS behaviour... I could get it working for Quake I think, but not super sure.

I tend use headphones a lot more than I used to and, while my hearing has surely declined (at least high for frequencies) due to having crossed over into the latter half of my mid 40s, I have become more critical about distortion and noise. Maybe I've weaned myself of my tolerance for 8-bit sound blasters.

Joking aside, sometimes good enough truly is sufficient. The ALS4000 card I've played around most with are truly abysmal on the noise front due to, probably in part, amplifiers that cannot be disabled without changing solder bridges (jumper blocks were too expensive, I guess).

Hi! Does anyone know if the Rockwell WaveArtist RWA010 is compatible with the SB16?

brotherdg2@gmail.com wrote on 2024-08-01, 12:55:

Hi! Does anyone know if the Rockwell WaveArtist RWA010 is compatible with the SB16?

No, SB Pro only - https://www.dosdays.co.uk/media/rockwell/DSA-446553.pdf