Previous Thread
Next Thread
Print Thread
Page 20 of 22 1 2 18 19 20 21 22
Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
I was under the impression that digital output of SNESAPU matched the SNES when set to 32khz/stereo/16-bit and gaussian resampling. That it was all of the features (echo feedback crosstalk, analog simulation, other resampling methods, etc.) that attempted to improve the quality of the sound at the expense of accuracy. So, you've piqued my (boundless) curiosity, can you elaborate?

SNeESe is the same as libopenspc?


-nZero
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Latest SNeESe is in a class by itself for accuracy - it passes all of the diagnostic cartridge tests with zero hacks, among other things (neither of the 2 big-name emulators even pass all the tests, hacks or not). It's audio was originally based on OpenSPC but has undergone 3 years of modifications based on test programs run on real hardware since then. OpenSPC has not backported their improvements.

Joined: Jul 2001
Posts: 105
Likes: 1
R
Senior Member
Senior Member
R Offline
Joined: Jul 2001
Posts: 105
Likes: 1
According to Antiresonance, he built the SNES SPDIF converter and made SNESAPU act exactly like his SNES. The high quality options do decrease accuracy for quality, but if you select "Like the SNES" it should be pretty close(turning off analog simulation may help too).

Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Ooh, people get snippy when I attack their golden calf. SNESAPU was accurate several years ago (and still is in the common case, if you don't mind that the sound driver program is running slightly wrongly to produce the "correct" results). The current state of knowledge of the SPC700 and DSP has advanced a great deal since then (particularly in the various corner cases that e.g. Squaresoft liked to get out and push), and it is no longer considered accurate at any setting.

Joined: Oct 2004
Posts: 296
S
Senior Member
Senior Member
S Offline
Joined: Oct 2004
Posts: 296
Just tried Blargg's Game Music Box, and it's some darn good sound emulation. In particular, it plays Road Runner's Death Valley Rally properly. With Audio Overload's current spc system, there are certain notes that hang much longer than they should, resulting in eventual dissonance.

Joined: Jul 2001
Posts: 105
Likes: 1
R
Senior Member
Senior Member
R Offline
Joined: Jul 2001
Posts: 105
Likes: 1
Not snippy, just that he claimed it to be like the SNES and that it output the same as a real SNES. If it turns out to be wrong I'd gladly accept a more accurate one.

His comment
Quote
SNESAmp 3.1 is out. This version contains many improvements to the DSP emulation. The ADPCM decompression, gaussian interpolation, mixing, and echo FIR filtering now work as they do in the Super NES hardware. The syncronization between the DSP and SPC700 has improved to provide more consistent timing between KON writes.

Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Boris: exactly. And the next AO will sound exactly that good too =) Feel free to try some of the infamous Squaresoft "wind" tunes if you're into that kinda thing.

Reznor: That's 18 months old. The really in-depth research has taken place in the last 6 months, with one important discovery happening last week.

Besides, next year when consumers all run 64-bit Windows 100% assembly software will die the painful death it richly deserves. (The ZSNES authors are currently working on a ground-up rewrite in C/C++ to avoid that sort of thing).

Joined: Jul 2001
Posts: 105
Likes: 1
R
Senior Member
Senior Member
R Offline
Joined: Jul 2001
Posts: 105
Likes: 1
Do you have any specific information? I'd really like to read about it since SNES is probably my favorite console.

I read the ZSNES dev forum alot since there are quite a few people that do tests on real hardware posting there.

I also read that emulators have been using the wrong IPL ROM, what's up with that?

Not meaning to be annoying, just curious.

Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
I don't have specific written examples (I believe those are known as "the anomie document"), but the ZSNES dev forum is definitely where a lot of good stuff gets hashed out.

I don't know what's up with the IPL ROM - it seems like it's trivial to dump if you can run test programs, and certainly the one we're using in MAME looks right.

Joined: May 2002
Posts: 67
N
Member
Member
N Offline
Joined: May 2002
Posts: 67
Interesting stuff. I haven't followed the SNES emulation scene very closely for probably a year and a half, so thanks for the update. Always nice to see some sort of progress going on.


-nZero
Page 20 of 22 1 2 18 19 20 21 22

Moderated by  R. Belmont, Richard Bannister 

Link Copied to Clipboard
Who's Online Now
1 members (1 invisible), 57 guests, and 2 robots.
Key: Admin, Global Mod, Mod
ShoutChat
Comment Guidelines: Do post respectful and insightful comments. Don't flame, hate, spam.
Forum Statistics
Forums9
Topics9,331
Posts122,197
Members5,077
Most Online1,283
Dec 21st, 2022
Our Sponsor
These forums are sponsored by Superior Solitaire, an ad-free card game collection for macOS and iOS. Download it today!

Superior Solitaire
Powered by UBB.threads™ PHP Forum Software 8.0.0