oh
interesting design flaw... could certainly help to patch this!
btw are you sure that the pictures are for CT2760? the ones I find on the net do not match?!

I believe there are two different revisions of the CT2760. The one I have, is for the 'older' first revision with the TDA1543T DAC.

Sorry, I should have mentioned that. 😐

oooohhhhh
more creative weirdness 🤣
when I have the time I will remove the opamps from the CT2230 and report back
good catch again Eep386!

If I'm not mistaken, there are several revisions of the CT2230 as well. Some can only take the TDA1543T, while later ones can take either that or the superior TDA1387.
Also, if I'm not mistaken, the CT2230 is one of those cards with floating amps.

Wow, with all these annoying issues it becomes obvious that Creative sure made crappy sound cards. 😕 They dominated the market through dirty tricks and sly marketing, not because of the quality of their products.

Isn't it true that the reason why the SB16 lacked downward compatibility with the SBPro was because they were playing catchup with Media Vision, and rushing the SB16 to market before the PAS16 could have the chance to gain a significant market share?

The good news is, at least this latest issue is fixable. It just looks a little ugly afterward once the wires are fitted. 😒

I couldn't tell you about the PAS16 situation, but it wouldn't surprise me to hear that the SB16 cut a lot of corners to make it to market faster. Though not using half of a quad amp and then doing nothing about the unused portion is just plain bone-headed: they're paying more money for poorer performance. On the first-run CT2760, they could have removed an entire quad amp and just used a 4558 dual amp instead of another quad, and saved a few cents per card in the process. But then maybe they simply got those lousy, slow and noisy 3403 amps at a knock-down drag-out price...

digger wrote on 2021-01-03, 19:58:

Wow, with all these annoying issues it becomes obvious that Creative sure made crappy sound cards. 😕 They dominated the market through dirty tricks and sly marketing, not because of the quality of their products.

Isn't it true that the reason why the SB16 lacked downward compatibility with the SBPro was because they were playing catchup with Media Vision, and rushing the SB16 to market before the PAS16 could have the chance to gain a significant market share?

Exactly my thoughts, as well..

And not to mention all the useless revisions of what essentially was the same card (SB16)..
Ironically, the subsequent releases were all inferior incarnations of the original CT17xx model.

With the exceptions of the different CD-ROM interfaces or a modem (these kind of made sense still).
Someone might wonder what all this confusion was good for. Developing , supporting and making a quadzillion new card PCBs surely wasn't cheap.

Well, darn. It looks like Creative's bad habits started before the Sound Blaster 16. 😐
I found that the upper half of the quad amp at U11 on a CT1600 Sound Blaster Pro 2 was floating too.
download/file.php?mode=view&id=100390

It's hard to tell but pins 8-10 and 12-14 are floating.
Thankfully this is the only quad amp on the board that has unused, floating inputs.

Eep386 wrote on 2021-01-16, 00:16:

Well, darn. It looks like Creative's bad habits started before the Sound Blaster 16. :| I found that the upper half of the quad […]
Show full quote

Well, darn. It looks like Creative's bad habits started before the Sound Blaster 16. 😐
I found that the upper half of the quad amp at U11 on a CT1600 Sound Blaster Pro 2 was floating too.
download/file.php?mode=view&id=100390

It's hard to tell but pins 8-10 and 12-14 are floating.
Thankfully this is the only quad amp on the board that has unused, floating inputs.

Sloppy. Strap those boys down!

* Ringing/Hissing artifacts, Vibra only.
A faint high pitched ringing will be heard after a sample is done playing.
Sometimes the card will hiss instead of ring, but the cause it the same.
In any case, these "artifacts" can be heard only on the Vibra cards.

Hello all, I was trying to reproduce it on several SB with different Vibra chips (2501,2502,2504,2505,2511), for example mentioned Blood - it sounds the same on all the cards - with Vibra chip or with different non Vibra SB card with original SB chips.
Is there way where it can be detected?

The noisiness of the early cards seems to depend on the individual mainboard they are used with

Is this the single cycle DMA bug?

https://youtu.be/gYHvWLrsF-4

I ask because though I assumed it is, various people here have said it isn't that bad or is preferable to X, y, or z, but this is clearly awful.

This is a ct2230 btw.

I can get this to work fine by using my on mobo ESS Solo card instead of the 2230, so it's not the game or the processor speed (though I have messed with the cpu speed to try to fix it as well).

Yuck, that sounds like it. This is a particularly bad example, it's normally not *this* extreme.

I'd take a ringing ViBRA any day over that popping racket.
No small wonder I've taken to ESS cards and the original Sound Blaster Pro 2 lately.

Yesterday I tested one of the ALS100 based SB16 clones and compared it to a real SB16 CT2290 (with 4.13 DSP and CT1703-A and I already knew this was a very good sounding SB16) and to Labway A151-A00 (Audician with Yamaha YMF71x).

Now the "problem" is that I was not able to hear DMA bugs with real SB16 at all. I tested with the Mickey 123 that was on the yotube video couple of posts above. Also tested with Star Trek Judgement rites. Both were just fine and what little "clicks" the Judgement rites had can be heard with the Labway as well so it was likely the samples itself.

BTW, the ALS100 was just bad. Very noisy and it did not work properly with the Mickey at all.

When I did the comparison, sound was routed through GUS ACE and only later I started to think that maybe it might have filtered away those DMA clicks. I'll test again but what would be the best games to test with? I want to hear the worst case scenario with real SB16.

aitotat wrote on 2021-08-19, 05:15:

When I did the comparison, sound was routed through GUS ACE and only later I started to think that maybe it might have filtered away those DMA clicks. I'll test again but what would be the best games to test with? I want to hear the worst case scenario with real SB16.

Definitively do the tests with the SB16 connected directly to your speakers.

Try with Day of the Tentacle (intro) and Prince of Persia (walking around, opening doors). Here's a video showcasing DMA clicking, in case it helps. Phil also demonstrates it here at around the 3:00 mark.

Of my two SB16-based cards, only the CT2230 (DSP v4.11) that I use on one of my old machines has the single-cycle DMA clicking problem and one of the worst since it occasionally drops out some samples.

Wolfenstein 3-D (especially when opening and closing doors), Llamatron: 2112 and Jill of the Jungle are good examples for testing the single-cycle DMA.

I did some more testing. I played first level of Prince of Persia but no clicks until the door at the end of the level. There were also some clicks with DOTT but Jill of the jungle was the best game to test it. So yes, the DMA bug is (of course) present but otherwise this is very good sounding card. No noise at all. I'd definitely take the clicks over the noise ALS100 produces. But not need to choose since the Labway doesn't have noise or bugs.

I took some pictures of the CT2290 since this is one of the best SB16 models. Or maybe if would be better to say that this individual card has one of the best chip configuration since not all the CT2290 models are the same. Maybe it helps someone to identify models with less clicking.

aitotat wrote on 2021-08-19, 17:12:

I took some pictures of the CT2290 since this is one of the best SB16 models. Or maybe if would be better to say that this individual card has one of the best chip configuration since not all the CT2290 models are the same. Maybe it helps someone to identify models with less clicking.

The same goes for the CT2230 since some have the DSP v4.13 instead of v4.11 and some have the less noisy CT1703-TBS and CT1703-A chips instead of CT1703-T like mine has. Certain variations may lack or have fewer CD-ROM ports.

The SB16 was notorious for the amount of variations and revisions.

Eep386 wrote on 2021-01-03, 18:32:

One more fault of many models of the SoundBlaster 16:
At least two models suffer from a rookie design mistake: unterminated (floating) portions of quad op-amps...

It's an interesting finding, it's a beginners mistake to leave floating opamp or CMOS logic inputs. I wouldn't expect this from a professional HW manufacturer. Maybe for floating-check it would be enough to use ohmmeter to measure resistance between output and input if there present some resistance of feedback network. Did someone checked AWE32 and AWE64 for this? I would look at my SB cards I have but guess AWE64 use only dual opamps.
I also wonder how many other bugs SBs have. In old good DOS days when I played games a lot I had ESS1868 with wavetable MIDI clone card and I didn't noticed any heavy disturbing bugs, also background noise was on acceptable level...

I checked my old AWE32 CT3900 and I have found this 3 opamps:
U26: MC3403 - fully utilized
U27: MC3403 - fully utilized
U31: TL074C - part 2 and 3 are floating!
As the opamps are powered by symmetric +-5V I simply wired them as folowers with + input to GND.