Reply 40 of 112, by kixs
- Rank
- l33t
😲
Thank you very much, this is one step closer to busting the reverse stereo mystery!
So, that leaves us with only Rev.03 untested...
I'm pretty sure Rev.03 which has DSP v3.01 behaves exactly the same as Rev.04 to 06 and all CT1600.
I'm beginning to think that this whole reverse stereo thing is for Rev.01 or 02 with DSP v3.00 but these cards are so rare they are non-existent, or they may be prototypes.
I know that Rev.01 and 02 of the CT1600 also nowhere to be found and can be considered a prototype.
I really wonder where the information about "Creative fixing the reversed stereo issue by guiding programmers to add extra byte before the stereo stream" came from...
Can if be that Creative instructed the programmers of the prototypes and somehow the information leaked to the public as "mandatory"?
Can it be that Creative fixed the issue with DSP v3.01 before releasing the CT1330A to the market..
This means DOSBox has the SBPro emulation reversed from the real card and it behaves like the clones which reverse the stereo with the extra byte, which none of the real SBPro cards does.
Any thoughts?
wrote:I'm beginning to think that this whole reverse stereo thing is for Rev.01 or 02 with DSP v3.00 but these cards are so rare they are non-existent, or they may be prototypes.
I know that Rev.01 and 02 of the CT1600 also nowhere to be found and can be considered a prototype.
An alternative theory I have is that it may be dependent on the speed of the CPU/chipset/bus used.
I mean, apparently the issue stems from the fact that the DSP may or may not miss the first byte of a DMA transfer, resulting in the channels being reversed.
While it could be that some DSPs always miss that byte, while others never miss that byte, I think it's more likely that it depends on the exact configuration of the entire system.
After all, if the stereo was reversed all the time, wouldn't Creative have noticed during initial testing, and just patched the channels on the circuit board?
Also, if they are ALWAYS reversed, why 'un-reverse' them at all? I mean, just make a modification to the spec, that instead of sending the left sample first, you send the right sample first (or vice versa). Then all cards remain compatible.
You may be right Scali, but as of now it is a hard fact that all DSP v3.01 and 3.02 of CT1330A Rev.03-06 and CT1600 Rev.03-08 behave EXACTLY the same, without the possibility of speed or other random system related issues "losing" the first byte.
I think that the reverse switch in games may be used for clone cards or other cards like GUS or PAS16 that emulated the SBPro and may have been reversed to the original.
The only reversed games on the original SBPro are the native GUS games like Jazz Jackrabbit and Epic Pinball, they don't have a reverse stereo switch either, but they are OK with the SB16.
Yes, I've only owned one Sound Blaster Pro (CT1600) in all my life, and it's been in 3 or 4 PCs over the years, but I don't recall it ever reversing stereo.
Question with games like Jazz Jackrabbit and Epic Pinball is if the stereo is actually reversed in hardware, or if they just messed up in software, and send the samples in the wrong order.
@ kixs, Jolaes76, kithylin.
Can you please try this shareware version of Epic Pinball, this native GUS game actually has something suspicious going on that makes the YMF719 card sound correct while all other clones and original not.
http://www.dosgamesarchive.com/download/epic-pinball/
Just enter a game and use the left and right Shift keys to hear the flippers form each side.
I can test this when I get home in about 7 hours.
Requests are also possible... /msg kixs
Okay, the Yamaha YMF719 is absolutely RANDOM in Jazz Jackrabbit, Epic Pinball, and Turrican II, I've tested more than 20 times each game.
All other cards are consistent in all 3 games.
wrote:Okay, the Yamaha YMF719 is absolutely RANDOM in Jazz Jackrabbit, Epic Pinball and Turrican II, I've tested more than 20 times each game.
Probably due to the software game drivers, all other cards are consistent.
Not sure about the other games, but Turrican 2 will randomly swap the channels, I've seen this on an AWE64 Gold as well. So I wouldn't worry about that game 😀
Yes, Turrican II flips the stereo ONCE (also with SB16 and DOSBox) on a consistent card in the music-test menu, but the YMF71x changes channels practically every other try.
To test: run Turrican 2 and press Spacebar (not Enter) to enter the music-test screen, from there press the key '1' to play the first tune, and hit 'esc' to stop the music and return.
Now do this several time like: 1, esc, 1, esc, 1, esc, etc... the YMF719 flip channel whether other cards will not.
wrote:I can test this when I get home in about 7 hours.
Please do, hopefully the CT1330A is consistent and behaves like the CT1600 with these finicky game too.
Fellow CT1330A owners, as a final test please play the attached WAV file with MPXPlayer (all included).
You simply run mpxplay.exe and select LR.WAV to perform the test, 'S' to Stop, hit Esc to quit.
The audio should be reversed on a SBPro, but correct on a SB16 and later.
As for games that do not sound correct with the SBPro but are correct with SB16 or GUS;
To my best guess the developers simply did not optimize the game for the SBPro which is reversed to the SB16 because it processes stereo PCM differently.
Games that are optimized for the SBPro (thanks to non lazy developers) like Doom, Duke3D, Tie Fighter, actually have their PCM already 'fixed' for the SBPro in software.
Jazz Jackrabbit (1994) and Epic Pinball (1993) both by Epic MagaGames, simply didn't bother to test with a SBPro or include a reverse switch.... 😐
wrote:Jazz Jackrabbit (1994) and Epic Pinball (1993) both by Epic MagaGames, simply didn't bother to test with a SBPro or include a reverse switch.... 😐
As I recall, most of Epic Megagames' titles were "MOD file" driven, meaning they sound much better on a GUS, which is I think where their focus was.
They simply supported Sound Blaster Pro and Sound Blaster 16, due to being the "de facto" standard, but probably didn't bother to test any further, apart from making sure that it did produce sounds 🤣
If I'm not mistaken, then some of those games will also support the Windows Sound System (WSS), and if your Sound Blaster/Sound Blaster Pro clone has WSS support, then you can select that and have a much "cleaner" sound output.
MPXPLAY....
even the latest revision will NOT autodetect a rev5 CT1330A as an SB Pro, it is wrongly picked up as an SB16 and plays jitter only that way.
If I manually specify "SBP" and anything lower than 22 khz in mpxplay.ini then it will start (in two bloody minutes!)
and plays back "RIGHT CHANNEL" "LEFT CHANNEL" swapped as it was supposed to do
Otherwise, it is a great utility 😀
Epic Pinball also produces reverse stereo.
"Ita in vita ut in lusu alae pessima iactura arte corrigenda est."
Thank for checking Jolaes76!
wrote:MPXPLAY....
even the latest revision will NOT autodetect a rev5 CT1330A as an SB Pro, it is wrongly picked up as an SB16 and plays jitter only that way.
If I manually specify "SBP" and anything lower than 22 khz in mpxplay.ini then it will start (in two bloody minutes!)
MPXPlayer searches for BLASTER variable,
For the CT1330A it should be: SET BLASTER=A220 I7 D1 T2
MPXPlayer is indeed a great utility; I use it primarily for adjusting volume and testing a -0.1dbFS sinewave for distortion with a VST oscilloscope.
This way I can maximize signal to noise ratio without distortion even from a primitive card like the SBPro.
@kixs, can you test with your Rev.04 please.
wrote:Epic Pinball also produces reverse stereo.
That means all the reversed games and software without a reverse switch didn't bother to test with the SBPro...
On the other hand, the "Reversed Stereo Bug on the SBPro1 CT1330A" is HISTORY! Rejoice. 😁
At least we came to the conclusion that the SBPro and all its iterations are consistent and exactly the same.
I assume we can call it history... Surprisingly, the echoes of this confusion rang for quite some time. 😲
"Ita in vita ut in lusu alae pessima iactura arte corrigenda est."
The confusion will keep on ringing believe me... especially when DOSBox is still reversed to the real SBPro and also emulates the "extra byte reverse switch" which the original SBPro does not.
I hope elderly Vogoners will point young and confused Vogoners directly to this thread. 😀
I once tried to bust all the SB16 myths and mysteries and gather all the information in one thread, I succeeded for a very short while.
A few weeks later, the information is ignored and vanished in the threads of history like it was never there... oh well, google still point to the right places for those who want to know.
Last tests with Rev 4.
Epic Pinball... reverse stereo
MpxPlay... reverse stereo (detects SBpro and plays fine with only SET BLASTER parameters)
Requests are also possible... /msg kixs
I guess it puts the final nail in the coffin of this myth.
Thanks to all participants of this thread for clarifying 25 years of confusion.
Hope this information will help DOSBox developers improve the emulation and shed some light on future debates on this matter.
Thanks!
I have the final revision of this card. If needed, I would perform any test you want.
I'm talking about this beauty here: http://www.amoretro.de/wp-content/uploads/cre … _sound_card.jpg
www.AmoRetro.de Visit my huge hardware gallery with many historic items from 16MHz 286 to 1000MHz Slot A. Includes more than 80 soundcards and a growing Wavetable Recording section with more than 300 recordings.