|
| |
AKA :
Doynax LZ
Credits :
Download :
Look for downloads on external sites:
Pokefinder.org
User Comment
Submitted by Mr Wegi on 8 May 2013
| Well, finally! How could so long underestimated this diamond! All I did this guy is perfect. When I saw "bitbucket" my mind exploded. Doynax you're great - congratulations! Well, thanks to separate Bitbreaker now Bit(s)breaker? :-) |
User Comment
Submitted by Oswald on 7 May 2013
| thanks, that clears it up for me. |
User Comment
Submitted by Krill on 7 May 2013
User Comment
Submitted by Bitbreaker on 7 May 2013
| @Oswald: I tried out saving cycles in the copy loops, and indeed it turned out to perform slower due to too much overhead. But feel free to try on your own :-) Also it will bloat the code further, but usually one wants to waste as few as possible bytes for a depacker and loader. |
User Comment
Submitted by doynax on 7 May 2013
@ Oswald:
Certainly. The match/literal innerloops may even be unrolled.
Unfortunately the average match length/literal run is rather short. Typically 3-6 bytes though YMMV, especially for highly compressible files (the --statistics flag comes in handy here.) Basically I judged it wouldn't be worth it in terms of set-up cycles and bytes.
It's the bit fiddling, e.g. decoding run lengths and match offsets, which eats the bulk of the cycles. |
User Comment
Submitted by Oswald on 7 May 2013
there are many aspects I'm not looking at when saying this but cant copy loops be constructed like:
lda src,y
sta dst,y
dey
bne -
if the loop have long enough runs it's probably worth the lengthyer setup. probably not the case just wondering. |
User Comment
Submitted by The Human Code Machine on 7 May 2013
| Finally! Thx for the good work. |
User Comment
Submitted by Yogibear on 6 May 2013
User Comment
Submitted by WVL on 6 May 2013
|
|
|
 | Search CSDb |
|
| Navigate | |
|
| Detailed Info | |
|
| Fun Stuff | |
· Goofs
· Hidden Parts
· Trivia
|
|
| Forum | |
|
| Support CSDb | |
|
 |  |
|
