Search found 157 matches

Go to advanced search

by sinamas
Fri Mar 03, 2006 9:42 pm
Forum: Development
Topic: Optimal RGB Mix/Clamped Add/Sub
Replies: 52
Views: 40614

clamp = carries - (carries >> 5);
Birlliant. I had a feeling it could be optimized further.
by sinamas
Fri Mar 03, 2006 8:01 pm
Forum: Development
Topic: Optimal RGB Mix/Clamped Add/Sub
Replies: 52
Views: 40614

What do you think of this for an rgb15 clamped add?

Code: Select all

unsigned int sum=pixel1+pixel2;
unsigned int tmp=(sum-((pixel1^pixel2)&0x0420))&0x8420;
sum-=tmp;
tmp|=tmp>>1;
tmp|=tmp>>2;
tmp|=tmp>>1;
tmp>>=1;
sum|=tmp;
by sinamas
Fri Mar 03, 2006 6:13 pm
Forum: Development
Topic: Optimal RGB Mix/Clamped Add/Sub
Replies: 52
Views: 40614

The only way I can turn parts from 32-63 to 31 is with r = (x+y)|(31-(x+y)&32)&31. And all that per channel. Otherwise r = (x+y); r = r >= 32 ? 31 : r; The ternary would probably translate to a cmp and a cmov (no branch) if you compile for x86 >= i686. If you absolutely insist on avoiding a...
by sinamas
Wed Feb 22, 2006 3:21 pm
Forum: Bug Reports/Feature Requests
Topic: More audio output options.
Replies: 24
Views: 8457

Re: More audio output options.

Try FCE Ultra.. it has a 96KHz option... see if there's a real difference between 48KHz and 96KHz. Actually the NES uses a PSG. If it was sample-based it would have a sample rate of about 1.79 MHz, of course the final sound output is limited to lower frequencies by the DAC among other things, but i...
by sinamas
Sun Oct 30, 2005 10:54 pm
Forum: ZSNES Talk
Topic: Calling all Windows users who can use a DOS prompt
Replies: 102
Views: 44667

I guess some failed k6-III+ may have been released as k6-2+ with half of the cache disabled, but wouldn't they've changed the model number then?
by sinamas
Sun Oct 30, 2005 1:09 pm
Forum: ZSNES Talk
Topic: Calling all Windows users who can use a DOS prompt
Replies: 102
Views: 44667

AFAIK a k6-III has 256k of L2 cache, while a k6-2+ has 128k. A k6-2 doesn't have on-chip L2 cache. I guess this makes the k6-2+ closer to a k6-III than a k6-2 anyway, and it seems that binaries compiled with arch set to k6-2 has the same md5sum as when compiled with arch set to k6-3, on gcc 3.4. The...
by sinamas
Fri Oct 21, 2005 4:10 pm
Forum: ZSNES Talk
Topic: Calling all Linux users
Replies: 31
Views: 19614

k6-3: processor : 0 vendor_id : AuthenticAMD cpu family : 5 model : 9 model name : AMD-K6(tm) 3D+ Processor stepping : 1 cpu MHz : 401.078 cache size : 256 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr ...

Go to advanced search