| |
Bamu®
Account closed
Registered: May 2005
Posts: 1332 |
Noise Waveform: real SID vs. Emulation
Hmm, somehow I'm a bit disappointed about the noise emulation (reSID).
At lower frequencies it is really ok, but if sounds are pitched very high, the real has a "shaper" touch....
Does anyone has an idea why this happens and when this will be fixed in emulation?
|
|
| |
Steppe
Registered: Jan 2002
Posts: 1510 |
Dag Lem is presently working on reSID 0.17 to implement the noise register tweaks discovered by SounDemoN (in his Jojojo.sid and also Tamaking.sid tunes IIRC). Maybe this will also fix the problem you mentioned.
Can you give a few examples where the high pitched noises sound noticeably different to a real SID, Nata? |
| |
zdzisek
Account closed
Registered: Apr 2002
Posts: 33 |
The quality of high frequency sounds in reSID highly depends on the clocking mode you use. Clocking at output sample frequency (fast clocking) leads to severe sampling noise caused by that samples are skipped. Bear this in mind when using Sidplay2/w, which is limited to the fast clocking. In VICE (or VSID - a built-in sid player), on the other hand, you can also choose single cycle clocking modes - interpolation and resampling. They greatly help in the process of reducing distortions at high frequencies, at the cost of a higher CPU usage. Interpolation already yields an acceptable sound quality, but you might want to consider using resampling (or, alternatively, fast resampling, which gives the very same result as resampling and is lighter on CPU, yet consumes more memory and it usually takes a couple of seconds required to build sample tables each time it is fired up) if you want to get the smoothest sounds with high frequency content. And should you not be satisfied with sound quality of the audio resampling mode, I'm afraid there's nothing much that can be done here. :-) |
| |
Bamu®
Account closed
Registered: May 2005
Posts: 1332 |
Hmmmmm....
When is the new reSID ready? :-O |
| |
Steppe
Registered: Jan 2002
Posts: 1510 |
I have no clue. And don't even dare bugging me about it on a weekly basis! ;-)
Honestly, I've got no influence on it, and I'm not really a close friend of Dag either. Just heard there's something in the works. May take 4 weeks or 6 months till it's ready. |
| |
Bamu®
Account closed
Registered: May 2005
Posts: 1332 |
Ok, here my example:
It happens only, if you use wavetables...
for example if I play:
81 ; frame 1, noise, A#-7
41 ; frame 2, pulse, C-2
it's even audible in snare-sounds, where there are rapid pitch changes...
I'm really wondering why not so-called "SID experts" recognize this effect. :-O
However, for composing (with emulators) it isn't essential, but could be a nice feature. |
| |
zdzisek
Account closed
Registered: Apr 2002
Posts: 33 |
Please post a sid file with that effect. Also, some examples of those snare sounds you're talking about would be appreciated. |
| |
Bamu®
Account closed
Registered: May 2005
Posts: 1332 |
Every SID file is affected...
Here my latest covers:
http://noname.c64.org/csdb/release/download.php?id=29791
http://noname.c64.org/csdb/release/download.php?id=29700
|
| |
Bamu®
Account closed
Registered: May 2005
Posts: 1332 |
Every SID file is affected...
Here my latest covers:
http://noname.c64.org/csdb/release/download.php?id=29791
http://noname.c64.org/csdb/release/download.php?id=29700
snare:
ADSR: 00F8
Pulse: 800
table:
81 c3 ;g-5
41 aa ;f-5
40 a6 ;d-3
80 ca ;d-6
ff 00 |
