 |
Cheat Engine
The Official Site of Cheat Engine
|
| View previous topic :: View next topic |
| Author |
Message |
hcavolsdsadgadsg
I'm a spammer
![]() Reputation: 26
Joined: 11 Jun 2007
Posts: 5801
|
 Posted: Sun Apr 05, 2009 9:07 pm Post subject: [Snippet / ASM] Fast MMX alpha blending |
 |
|
Last night I sat down and tried my hand at some MMX. After some reading and experimentation, I got to work.
dest = alpha * (source - dest) / 255 + dest;
Previously the best I could do was work on each channel of the pixel, so doing the math for the entire pixel at once is probably a bit quicker I imagine. 
I use a DIB Section for drawing, so I get a pointer to the bits. (which is dst)
color is a dword, and serves as the source.
Basically this is how I tried to lay it out...
| Code: | 00 XX 00 RR 00 GG 00 BB
-
00 XX 00 RR 00 GG 00 BB
*
00 AA 00 AA 00 AA 00 AA
/
00 FF 00 FF 00 FF 00 FF
+
00 XX 00 RR 00 GG 00 BB |
and the code behind it.
| Code: | pxor mm3, mm
mov eax, dword ptr [dst]
movd mm0, dword ptr [eax]
movd mm1, dword ptr [color] //src
movd mm2, dword ptr [alpha]
punpcklbw mm0, mm3 //unpack dst to words
punpcklbw mm1, mm3 //unpack color
punpcklbw mm2, mm2 //unpack alpha
punpcklbw mm2, mm2
punpcklbw mm2, mm3
psubusb mm1, mm0 //(color - dest)
pmullw mm1, mm2 //alpha * (color - dest)
psrlw mm1, 8 //alpha * (color - dest) / 256
paddusw mm1, mm0 //alpha * (color - dest) / 256 + dest
packuswb mm1, mm3
movd dword ptr [eax], mm1 |
Have fun. |
|
| Back to top |
|
 |
Fallen`
Expert Cheater
 Reputation: 0
Joined: 24 Aug 2008
Posts: 224
Location: United States
|
| Posted: Sun Apr 05, 2009 9:29 pm Post subject: |
|
|
Thanks for the Information, Ill be sure to check it out later.
_________________
Maplestory's Fun, But It waste's your Time. Your hard work Get's demolished Because You're better than everyone else. Nexon, You truly. Suck.
.::Acomplishments::.
Level 114 Arch Mage (i/l)
Level 100 Sniper (Hacked)
Level 98 Cheif Bandit(Hacked) |
|
| Back to top |
|
|
hcavolsdsadgadsg
I'm a spammer
![]() Reputation: 26
Joined: 11 Jun 2007
Posts: 5801
|
|
| Back to top |
|
|
hcavolsdsadgadsg
I'm a spammer
![]() Reputation: 26
Joined: 11 Jun 2007
Posts: 5801
|
| Posted: Fri Apr 10, 2009 2:15 am Post subject: |
|
|
| Moose wrote: | This is slower than using plain x86 instructions.
The idea of max speed with mmx comes from utilizing all instruction pipelines in parallel. Essentially you can process at least 2 pixels at once, or at least perform 2 instructions in 1 cc.
Good thing you are learning though. |
Of course, but drawing can't always be done 2 pixels at a time. The function this is for basically mimics SetPixel.
Later on, drawing for sprites, pictures, etc will probably all be handled like that. Coding seems to come in inspired bursts recently though, so like always, it's done when it's done.
Anyway, there's not really that much chance for instruction pairing here, since it's so short. The emms is expensive, which sucks, but this still seems to be faster than the straight C version, since you have to do it per channel in the pixel.
I'll also probably try threading the drawing eventually, or maybe something like dirty rectangles.
Really, once I get a bit more experience, I think I'll try my hand at something a little more modern, like DX. |
|
| Back to top |
|
|
pkedpker
Master Cheater
![]() Reputation: 1
Joined: 11 Oct 2006
Posts: 412
|
| Posted: Sun Apr 12, 2009 11:45 am Post subject: |
|
|
one person created a new array for C++ using all registers + MMX (xmm0-xmm7) and SSE (mm0-mm7) registers and hes array works 9-12 times speedup faster then the normal array C++ that uses just normal registers. So its not slow
_________________
|
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
