lekt wrote:weird, yesterday 0x...6240, today 0x...6250, system changed address. but v2.97 still works for you? think doesn't work today. will review code.
i suspect that in your case wasapi give different pointers depend on code of each version. could you take pointer again for v2.97.xx 160 code.
lekt wrote:weird, yesterday 0x...6240, today 0x...6250, system changed address. but v2.97 still works for you? think doesn't work today. will review code.
i suspect that in your case wasapi give different pointers depend on code of each version. could you take pointer again for v2.97.xx 160 code.
jesuscheung wrote:i have strong feeling that the sound of 2.93 should be used on youporn... hehe
v2.93 xx use shift left 1 (unsigned char <<=1;) as v2.40.1 256. vocal very good emotion. as i said bitwise operation is best, fastest is shift unsigned byte to 1 digit. and/or/xor/not not good as shift. +/- is bad, * and / very bad.
v2.93 160 have especially vocal emotion, hear female singer whisper to your ears.
jesuscheung wrote:i have strong feeling that the sound of 2.93 should be used on youporn... hehe
v2.93 xx use shift left 1 (unsigned char <<=1;) as v2.40.1 256. vocal very good emotion. as i said bitwise operation is best, fastest is shift unsigned byte to 1 digit. and/or/xor/not not good as shift. +/- is bad, * and / very bad.
v2.93 160 have especially vocal emotion, hear female singer whisper to your ears.
lekt wrote:...i tried Stud_PE setting ...:
- Headers/Characteristics/Relocations stripped: check it
- Headers/DllCharacteristics/Dynamic Base: uncheck it
do you tried this MOD? try it.
...
maybe v2.97.1 160 will work if you make MOD by Stud_PE for testing, think if /DINAMICBASE:NO then system will give more stability wasapi addresses (ASLR...)
jesuscheung wrote:i have strong feeling that the sound of 2.93 should be used on youporn... hehe
v2.93 xx use shift left 1 (unsigned char <<=1;) as v2.40.1 256. vocal very good emotion. as i said bitwise operation is best, fastest is shift unsigned byte to 1 digit. and/or/xor/not not good as shift. +/- is bad, * and / very bad.
v2.93 160 have especially vocal emotion, hear female singer whisper to your ears.
how does bitshift work ?
my buffer size is 1792, can bitshift be used ?
xor sounded better than neg.
cheers
use shift for counter variables in loop. i don't know why system give very pretty sound when CPU works with shift.
can make this for your case:
1792 x 4 = 7168 bytes = 7 x 8 x 128bytes
unsigned char i=2,j=4;
a:
w(h,0xFFFFFFFF);
gb(n,160,k);
b:
*d++=*r++;*d++=*r++;*d++=*r++;...// 128 statments, i use different methods for this copy 128 byes
if(i>0){i<<=1;goto b;}
if(j>0){j<<=1;i=2;goto b;}
rb(n,160,0);
i=2;j=4;
goto a;
i think need small buffer size, so then not 128 bytes, ie. 10 bytes for buffer size 160. big render loop code not good, code jump to far address. as nearer as better.
lekt wrote:...i tried Stud_PE setting ...:
- Headers/Characteristics/Relocations stripped: check it
- Headers/DllCharacteristics/Dynamic Base: uncheck it
do you tried this MOD? try it.
...
maybe v2.97.1 160 will work if you make MOD by Stud_PE for testing, think if /DINAMICBASE:NO then system will give more stability wasapi addresses (ASLR...)
lekt wrote:
v2.93 xx use shift left 1 (unsigned char <<=1;) as v2.40.1 256. vocal very good emotion. as i said bitwise operation is best, fastest is shift unsigned byte to 1 digit. and/or/xor/not not good as shift. +/- is bad, * and / very bad.
v2.93 160 have especially vocal emotion, hear female singer whisper to your ears.
how does bitshift work ?
my buffer size is 1792, can bitshift be used ?
xor sounded better than neg.
cheers
use shift for counter variables in loop. i don't know why system give very pretty sound when CPU works with shift.
can make this for your case:
1792 x 4 = 7168 bytes = 7 x 8 x 128bytes
unsigned char i=2,j=4;
a:
w(h,0xFFFFFFFF);
gb(n,160,k);
b:
*d++=*r++;*d++=*r++;*d++=*r++;...// 128 statments, i use different methods for this copy 128 byes
if(i>0){i<<=1;goto b;}
if(j>0){j<<=1;i=2;goto b;}
rb(n,160,0);
i=2;j=4;
goto a;
i think need small buffer size, so then not 128 bytes, ie. 10 bytes for buffer size 160. big render loop code not good, code jump to far address. as nearer as better.
wonder if that's why you get glitches, copying byte by byte is not fast enough, not scalable to hirez.
the d++ and r++ will be doing an addition for each byte or an lea so not removing additions.