VOGONS


Reply 20 of 106, by Jolaes76

User metadata
Rank Oldbie
Rank
Oldbie

EDIT:
deleted because I apparently did the test with wrong cabling.

Last edited by Jolaes76 on 2016-10-26, 09:24. Edited 1 time in total.

"Ita in vita ut in lusu alae pessima iactura arte corrigenda est."

Reply 21 of 106, by kithylin

User metadata
Rank l33t
Rank
l33t

Okay! Well finally I got my "fastest" 486 hooked up. Fired it up and connected it up with the CT1330A in it, and verified quickly in Duke3D setup sound tests that it indeed is playing both speakers correctly with the stereo -correct- and is not reversed.

I can try your program up above you posted and run it on my thing but it'll take some time to get network on it and my laptop setup in there. But I can do it.

Any other games you want me to test? I might take my card out and photo it to show you which one I have.

EDIT: This is my card I own.
CT1330A_Small.jpg
Larger: http://www.outfoxed.net/CT1330A_Large.jpg

EDIT EDIT: Also Confirmed I was wrong, my CT1330A is fully functional and outputs sound in both speakers.

Reply 22 of 106, by kithylin

User metadata
Rank l33t
Rank
l33t

Okay doing some testing for you. I grabbed Descent I demo for DOS and loaded it in. I can confirm Descent I (in setup) -DOES- have reversed stereo! It's playing in the wrong speakers. Duke3D however does -NOT- and plays -CORRECTLY-. I'll continue testing further.

And yes I'm sure my speakers are oriented correctly, they're a set of smallish Harman Kardon powered desktop speakers used for this test that have (L) and (R) on the back of them.
EDIT: And yes, StereoReverse=0 is set in descent.cfg

Reply 24 of 106, by kithylin

User metadata
Rank l33t
Rank
l33t
Jolaes76 wrote:

Duke3d atomic edition 1.4?

No. Retail non-atomic edition 1.3D I don't legally own a copy of Duke3D atomic edition yet, not officially.. I think I have the GoG version but I've had trouble getting that working under real mode MS-DOS. It keeps asking for the CD (Which I don't have) and I'm not using a no-cd crack, haven't found one that works. Maybe some day.. it's not a high priority on my buy list at the moment. I do own the retail disc for the non-atomic edition for DOS though. Bought it retail years ago and still have it.

Reply 25 of 106, by Jolaes76

User metadata
Rank Oldbie
Rank
Oldbie

My atomic edition copy produces reverse stereo on rev 5 CT1330A cards.
EDIT: because I am an idiot and still used swapped cabling for testing. Sorry. With Correct cabling, CT1330A REV5 behaves exactly like REV6.

Last edited by Jolaes76 on 2016-10-26, 09:26. Edited 1 time in total.

"Ita in vita ut in lusu alae pessima iactura arte corrigenda est."

Reply 26 of 106, by kithylin

User metadata
Rank l33t
Rank
l33t

So Here's my testing:
DOOM - Sounds correct, not sure, setup doesn't have anything to test left/right
DESCENT: -INCORRECT-, stereo is reversed by default
Tie Fighter: Correct, sounds pan left to right correctly.
DUKE3D: Correct, sound is correct in setup with reverse stereo off / unticked. (NOT Atomic, version 1.3D Retail)

EDIT: If you see the image mine is Revision 6 CT1330A, I don't know what that means.

EDIT #2: SBTEST.EXE plays the test tone in both speakers at the same time through all of the tests, first one and second one constantly.

EDIT #3: Here's your sbcheck results. And that's my contribution to the thread. If you need anything else tested tell me but it'll be tomorrow now.

sbcheck.jpg

And yes I tested twice on Decent.

Reply 27 of 106, by James-F

User metadata
Rank Oldbie
Rank
Oldbie
kithylin wrote:
So Here's my testing: DOOM - Sounds correct, not sure, setup doesn't have anything to test left/right DESCENT: -INCORRECT-, ste […]
Show full quote

So Here's my testing:
DOOM - Sounds correct, not sure, setup doesn't have anything to test left/right
DESCENT: -INCORRECT-, stereo is reversed by default
Tie Fighter: Correct, sounds pan left to right correctly.
DUKE3D: Correct, sound is correct in setup with reverse stereo off / unticked. (NOT Atomic, version 1.3D Retail)

Fantastic, thank you very much!
It is exactly like the SBPro2 CT1600. 😲

kithylin wrote:

EDIT #2: SBTEST.EXE plays the test tone in both speakers at the same time through all of the tests, first one and second one constantly.

Can you please listen with headphones, the separation may be less obvious but it is definitely there.
One channel would produce a sinus-wave, the other will be kind of noise.
Please test again, because this test is the most revealing one.

Also, SBCHECK reports SBPro2 instead of SBPro1, so obviously we can ignore the type of card and look at the DSP version only.

Last edited by James-F on 2016-10-26, 12:50. Edited 2 times in total.


my important / useful posts are here

Reply 28 of 106, by Jolaes76

User metadata
Rank Oldbie
Rank
Oldbie

Both my cards are rev5.
When I get home I will try the Shareware versions of Duke3D and Descent yet I suspect I will get the same results as with the last retail versions. I am using headphones as well.

"Ita in vita ut in lusu alae pessima iactura arte corrigenda est."

Reply 29 of 106, by James-F

User metadata
Rank Oldbie
Rank
Oldbie

I want to emphasize the importance of the SBTEST utility for this test, it inserts an extra byte before the second test.
This extra byte is actually Creatives instruction to reverse the stereo on the first revision of the CT1330A cards which ignored the first byte of the PCM stream and as a result reversed the stereo where they should not.

This issue known to be fixed with the CT1600 cards and now apparently also the Rev.06 of CT1330A.
This is strange that there is a difference between Rev.05 and Rev.06 stereo behavior...

EDIT:
Quake is one of these games that will insert the extra byte when the game starts, while Doom, Duke3D, Tie-Fighter and Descent don't insert the extra byte.
On the clone SBPro2 cards the stereo reverses in Quake and SBTEST (second test) with the extra byte, but not on the original SBPro2 CT1600.
Have a look at the chart I've posted in the SBTEST column, SBPro2 stays L twice, this is directly affects Quake.

Last edited by James-F on 2016-10-25, 08:08. Edited 3 times in total.


my important / useful posts are here

Reply 30 of 106, by kithylin

User metadata
Rank l33t
Rank
l33t

I'll try again tomorrow or in a few days. Thankfully my CT1330A has a volume knob.. my ear buds do not. Too tired to mess with it tonight.

Reply 31 of 106, by James-F

User metadata
Rank Oldbie
Rank
Oldbie

Thank you, waiting for results, the SBTEST is the most important test actually.
You can plug the headphones in the speaker headphones-out if they have one.


my important / useful posts are here

Reply 32 of 106, by kixs

User metadata
Rank l33t
Rank
l33t

I didn't have the time to do any retro stuff this morning ... so my tests are still pending...

I only have Dooms installed on CF card, but I guess SBTEST should suffice. I could install shareware versions of Descent and Duke3D Atomic from above link.

My Amibay: SALE | BUY - Updated on 2020-05-25

Reply 33 of 106, by James-F

User metadata
Rank Oldbie
Rank
Oldbie

kixs, what revision is your CT1330A?
You can also test Quake shareware from this site which adds the extra byte and flips the stereo on some SBPro cards, start a game and walk towards a fire torch.
But yes, Doom, Duke3D, Descent, Quake (all shareware) and most importantly SBTEST should reveal the stereo behavior of any SBPro card.

Waiting for your results.


my important / useful posts are here

Reply 34 of 106, by Jolaes76

User metadata
Rank Oldbie
Rank
Oldbie

I think I messed up something previously either with the cabling or not using the shareware game versions recommended earlier. Or I was just too tired and started Xwing's setup instead of Tie Fighters'. Sorry for the confusion.
Now that I sorted everything out, even with Tie Fighter floppy version I get the same results as others with the rev6 card. TF indeed sounds corrects without swapping channels on a rev5 CT1330A as well.

For me, harekiet's test utility plays the sine wave on the left and a little noise (cracks, faint ring) on the right all the time.

So the rev5 CT1330A behaves the same as a rev6. The only test I have not done yet is with Duke3D shareware.

"Ita in vita ut in lusu alae pessima iactura arte corrigenda est."

Reply 35 of 106, by James-F

User metadata
Rank Oldbie
Rank
Oldbie
Jolaes76 wrote:

For me, harekiet's test utility plays the sine wave on the left and a little noise (cracks, faint ring) on the right all the time.
So the rev5 CT1330A behaves the same as a rev6. The only test I have not done yet is with Duke3D shareware.

Excellent, thank you for checking another time, this is gold right here!

So up till now we have CT1330A Rev.5 and Rev.6 that behave EXACTLY like the SBPro2 CT1600, even SBTEST plays twice on the Left channel without changing channels.
Now we have to test Rev 3 and 4 (both DSP v3.01) to make absolutely sure that this whole reverse stereo issue is all a bad misunderstanding that lasted for 25 years. 😀
Note that Rev.1 and 2 are non-existent or are so rare no one seen one in person, anyone can prove otherwise?

Up till now the only cards that flip the stereo image with the SBTEST utility are the clone cards.
The only clone card that actually has reversed PCM is the Aztech NX Pro (but its FM and CD are still correct).

Fellows we are making history here, and you are part of it. 😁


my important / useful posts are here

Reply 36 of 106, by kixs

User metadata
Rank l33t
Rank
l33t
James-F wrote:
kixs, what revision is your CT1330A? You can also test Quake shareware from this site which adds the extra byte and flips the st […]
Show full quote

kixs, what revision is your CT1330A?
You can also test Quake shareware from this site which adds the extra byte and flips the stereo on some SBPro cards, start a game and walk towards a fire torch.
But yes, Doom, Duke3D, Descent, Quake (all shareware) and most importantly SBTEST should reveal the stereo behavior of any SBPro card.

Waiting for your results.

I'll check the rev when I get home from work... if it's rev 5 or 6 the tests aren't necessary any more, right?

My Amibay: SALE | BUY - Updated on 2020-05-25

Reply 37 of 106, by Scali

User metadata
Rank l33t
Rank
l33t
Jolaes76 wrote:

For me, harekiet's test utility plays the sine wave on the left and a little noise (cracks, faint ring) on the right all the time.

The noise happens on my CT1600 as well. I think it's a result of the stereo solution they chose. They 'pingpong' samples to left and right on a 44.1 kHz DAC, which has very poor channel separation and lots of aliasing.
So yes, all digital sounds panned hard left can be heard softly and crackling on the right channel as well, and vice versa.
It's a joke really, such poor quality, considering the steep cost of the SB Pro at the time.
That is something the emulation generation will never understand, I'm afraid. Emulated SB Pros sound crisp and hi-fi, and people will say: "What's the point of a GUS?"
Well, the point is that a GUS actually sounds crisp and hi-fi in real life. SB's did not. The SB16 and AWE models still left quite a bit to be desired in the quality department, although they did get better at every iteration.
A PAS16 on the other hand did deliver crisp audio.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 38 of 106, by James-F

User metadata
Rank Oldbie
Rank
Oldbie
kixs wrote:

I'll check the rev when I get home from work... if it's rev 5 or 6 the tests aren't necessary any more, right?

It is still worth running SBTEST just to be double sure.

Scali wrote:

The noise happens on my CT1600 as well.
It's a joke really, such poor quality, considering the steep cost of the SB Pro at the time.

Maybe then, but now it is all part of the magic no emulator can emulate. 😀
Poor channel separation and unique analog low pass filter, all part of the SBPro sound.
Nowadays it is still steep cost + magic. 🤣

BTW, SBTEST played from the left channel twice on your CT1600?


my important / useful posts are here

Reply 39 of 106, by NJRoadfan

User metadata
Rank Oldbie
Rank
Oldbie

Creative isn't the only ones who outputted a mono signal and de-multiplexed it with filters. A similar setup was used on Apple IIgs stereo cards to demux the Ensoniq 5503DOC's audio to left and right channels. Stereo separation varied by the quality of the components of the card.... which Creative and most Apple IIgs soundcard vendors cheaped out on.

The reversed stereo audio problem is present on the IIgs too, but for a different reason. Applied Engineering flipped the right and left channel output on their sound cards vs. the industry standard definition.