| |
ChristopherJam
Registered: Aug 2004
Posts: 1409 |
Nucrunch 0.1
Continuing from the benchmarks WVL posted in Doynamite 1.x:
I dusted off my unfinished nucrunch in December to pack just enough of the second page of Reutastic to give me some workspace for some precalculations. Pity I didn't schedule enough time to pack the entire demo, else it would have been ~90 blocks instead of 190, but I digress. I've spent bits of the past month cleaning up the code, optimizing the packer (mostly by porting it from python to rust :P), and adding reverse direction support.
It's still no more than a component, with an commandline packer and asm decrunch subroutine, but no tools yet for generating an executable from a single commandline. It does at least now support multiple input segments that are unpacked to their destination addresses, and it's also now useable enough to for me to do some benchmarking.
In short, doynamite's ratio looks pretty unbeatable for anything lz based; my ratio's almost identical despite a somewhat different encoding.
Where I can win though is speed at that ratio; nucrunch is usually ten to twenty percent faster. The one exception in the test corpus is 6.bin, where it's 20% slower; not sure why yet.
I've added the times for pucrunch -ffast below for to complete the comparison. Last two columns are nucrunch, and nucrunch -r (the latter decodes in reverse; should be a more useful component for single filers)
If anyone wants to have a play at this stage, poke me and I'll upload some source. Failing that I'll hold off until I at least have something that can make onefilers without any faffing about with relocating the last couple of pages by hand.
filesizes
# bin rle wvl-f wvl-s tc bb pu-f doyna nucru rnucr
- ----- ----- ----- ----- ----- ----- ----- ----- ----- -----
1 11008 8020 4529 4151 4329 3383 3711 3265 3225 3230
2 4973 4314 3532 3309 3423 2648 3005 2512 2498 2490
3 3949 3498 2991 2617 2972 2187 2530 2108 2091 2093
4 7016 6456 4242 4085 4225 3681 3924 3617 3622 3614
5 34760 27647 25781 24895 25210 21306 21182 20405 20447 20516
6 31605 12511 11283 10923 11614 9194 9203 8904 8915 8894
7 20392 17295 12108 11285 11445 9627 9789 9289 9140 9144
8 5713 5407 4179 3916 3936 3251 3656 3132 3165 3187
9 8960 7986 6914 6896 6572 5586 6000 5430 5502 5486
filesize in %
# bin rle wvl-f wvl-s tc bb pu-f doyna nucru rnucr
- ----- ----- ----- ----- ----- ----- ----- ----- ----- -----
1 100 72.9 41.1 37.7 39.3 30.7 33.7 29.7 29.3 29.3
2 100 86.7 71.0 66.5 68.8 53.2 60.4 50.5 50.2 50.1
3 100 88.6 75.7 66.3 75.3 55.4 64.1 53.4 53.0 53.0
4 100 92.0 60.5 58.2 60.2 52.5 55.9 51.6 51.6 51.5
5 100 79.5 74.2 71.6 72.5 61.3 60.9 58.7 58.8 59.0
6 100 39.6 35.7 34.6 36.7 29.1 29.1 28.2 28.2 28.1
7 100 84.8 59.4 55.3 56.1 47.2 48.0 45.6 44.8 44.8
8 100 94.6 73.1 68.5 68.9 56.9 64.0 54.8 55.4 55.8
9 100 89.1 77.2 77.0 73.3 62.3 67.0 60.6 61.4 61.2
number of frames to depack
# bin rle wvl-f wvl-s tc bb pu-f doyna nucru rnucr
- ----- ----- ----- ----- ----- ----- ----- ----- ----- -----
1 0 11 13 14 15 58 54 27 22 22
2 0 5 7 7 9 38 39 17 14 14
3 0 4 6 6 7 28 31 12 10 10
4 0 8 9 9 10 43 51 20 17 18
5 0 36 39 42 59 300 298 119 104 107
6 0 20 25 25 37 126 152 49 59 59
7 0 22 25 26 32 138 139 60 51 52
8 0 6 8 8 10 43 47 18 16 17
9 0 9 12 12 16 73 81 32 28 29
kilobytes output per second
# bin rle wvl-f wvl-s tc bb pu-f doyna nucru rnucr
- ----- ----- ----- ----- ----- ----- ----- ----- ----- -----
1 49.0 41.4 38.5 35.9 9.3 10.0 20.0 24.5 24.5
2 48.7 34.8 34.8 27.0 6.4 6.2 14.3 17.4 17.4
3 48.3 32.2 32.2 27.6 6.9 6.2 16.1 19.3 19.3
4 42.9 38.2 38.2 34.3 8.0 6.7 17.2 20.2 19.1
5 47.3 43.6 40.5 28.8 5.7 5.7 14.3 16.4 15.9
6 77.4 61.9 61.9 41.8 12.3 10.2 31.6 26.2 26.2
7 45.4 39.9 38.4 31.2 7.2 7.2 16.6 19.6 19.2
8 46.6 35.0 35.0 28.0 6.5 6.0 15.5 17.5 16.5
9 48.7 36.5 36.5 27.4 6.0 5.4 13.7 15.7 15.1
cycles per byte consumed
# bin rle wvl-f wvl-s tc bb pu-f doyna nucru rnucr
- ----- ----- ----- ----- ----- ----- ----- ----- ----- -----
1 0 27 56 66 68 337 286 163 134 134
2 0 23 39 42 52 282 255 133 110 111
3 0 22 39 45 46 252 241 112 94 94
4 0 24 42 43 47 230 255 109 92 98
5 0 26 30 33 46 277 277 115 100 103
6 0 31 44 45 63 269 325 108 130 130
7 0 25 41 45 55 282 279 127 110 112
8 0 22 38 40 50 260 253 113 99 105
9 0 22 34 34 48 257 265 116 100 104
decrunch time for nucrunch/rnucrunch relative to doynamite
1: 81.5% (-18.5%) 81.5% (-18.5%)
2: 82.4% (-17.6%) 82.4% (-17.6%)
3: 83.3% (-16.7%) 83.3% (-16.7%)
4: 85.0% (-15.0%) 90.0% (-10.0%)
5: 87.4% (-12.6%) 89.9% (-10.1%)
6: 120.4% ( 20.4%) 120.4% ( 20.4%)
7: 85.0% (-15.0%) 86.7% (-13.3%)
8: 88.9% (-11.1%) 94.4% ( -5.6%)
9: 87.5% (-12.5%) 90.6% ( -9.4%)
|
|
... 95 posts hidden. Click here to view all posts....
|
| |
MagerValp
Registered: Dec 2001
Posts: 1078 |
I don't have time to code anything from scratch at the moment, but if you share your benchmarking code I could probably do LZMV and exomizer.
And while exomizer produces small files, it really is butt slow. |
| |
ChristopherJam
Registered: Aug 2004
Posts: 1409 |
Here you go:
https://dl.dropboxusercontent.com/u/62734159/csdb/20160210_cycl.. |
| |
Axis/Oxyron
Account closed
Registered: Apr 2007
Posts: 91 |
So any volunteers to optimize the exo-decruncher? ;oP |
| |
Bitbreaker
Registered: Oct 2002
Posts: 508 |
Quote: So any volunteers to optimize the exo-decruncher? ;oP
Doesn't make much sense as long as you also want the decruncher to be tiny, most of all, as it needs a 156 byte scratchpad during decrunching. |
| |
MagerValp
Registered: Dec 2001
Posts: 1078 |
Added timed benchmarks for LZMV, both the original 256 byte sliding window and the 4k version I used in U4. Note that neither is optimized for speed right now, but size and convenience.
filesizes
# bin rle wvl-f wvl-s tc bb2.0 pu-f doyna nucru rnucr LZMPi LZMV4k LZMV256 exoraw
- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
1 11008 8020 4529 4151 4329 3322 3711 3265 3225 3230 3184 4205 4539 2988
2 4973 4314 3532 3309 3423 2513 3005 2512 2498 2490 2410 3183 3575 2241
3 3949 3498 2991 2617 2972 2098 2530 2108 2091 2093 1931 2551 3018 1817
4 7016 6456 4242 4085 4225 3682 3924 3617 3622 3614 3571 4343 4314 3454
5 34760 27647 25781 24895 25210 20530 21182 20405 20447 20516 20362 23845 26116 19631
6 31605 12511 11283 10923 11614 8998 9203 8904 8915 8896 8719 10619 11352 8337
7 20392 17295 12108 11285 11445 9241 9789 9289 9140 9145 9256 11154 12188 8751
8 5713 5407 4179 3916 3936 3165 3656 3132 3166 3187 3048 3959 3987 3059
9 8960 7986 6914 6896 6572 5491 6000 5430 5502 5487 5563 6505 6943 5295
filesize in %
# bin rle wvl-f wvl-s tc bb2.0 pu-f doyna nucru rnucr LZMPi LZMV4k LZMV256 exoraw
- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------- ------
1 100.0 72.9 41.1 37.7 39.3 30.2 33.7 29.7 29.3 29.3 28.9 38.2 41,2 27.1
2 100.0 86.7 71.0 66.5 68.8 50.5 60.4 50.5 50.2 50.1 48.5 64.0 71,9 45.1
3 100.0 88.6 75.7 66.3 75.3 53.1 64.1 53.4 53.0 53.0 48.9 64.6 76,4 46.0
4 100.0 92.0 60.5 58.2 60.2 52.5 55.9 51.6 51.6 51.5 50.9 61.9 61,5 49.2
5 100.0 79.5 74.2 71.6 72.5 59.1 60.9 58.7 58.8 59.0 58.6 68.6 75,1 56.5
6 100.0 39.6 35.7 34.6 36.7 28.5 29.1 28.2 28.2 28.1 27.6 33.6 35,9 26.4
7 100.0 84.8 59.4 55.3 56.1 45.3 48.0 45.6 44.8 44.8 45.4 54.7 59,8 42.9
8 100.0 94.6 73.1 68.5 68.9 55.4 64.0 54.8 55.4 55.8 53.4 69.3 69,8 53.5
9 100.0 89.1 77.2 77.0 73.3 61.3 67.0 60.6 61.4 61.2 62.1 72.6 77,5 59.1
number of frames to depack
# bin rle wvl-f wvl-s tc bb2.0 pu-f doyna nucru rnucr LZMPi LZMV4k LZMV256 exoraw
- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------- ------
1 0.0 11.5 13.5 14.5 15.5 46.1 54.7 27.5 22.0 21.3 16.9 13.7
2 0.0 5.5 7.5 7.5 9.5 31.2 39.6 17.5 13.7 13.6 9.6 8.4
3 0.0 4.5 6.5 6.5 7.5 22.7 31.9 12.5 10.0 10.1 7.1 6.2
4 0.0 8.5 9.5 9.5 10.5 37.5 52.0 20.5 16.6 17.1 11.5 9.4
5 0.0 36.5 39.5 42.5 59.5 222.2 298.6 119.5 101.4 100.6 58.4 46.6
6 0.0 20.5 25.5 25.5 37.5 112.3 152.8 49.5 57.9 56.6 35.8 38.2
7 0.0 22.5 25.5 26.5 32.5 109.6 139.8 60.5 50.3 50.0 35.0 29.1
8 0.0 6.5 8.5 8.5 10.5 35.4 47.7 18.5 15.6 15.8 10.3 8.9
9 0.0 9.5 12.5 12.5 16.5 62.3 81.7 32.5 27.1 27.2 16.6 14.2
kilobytes output per second
# bin rle wvl-f wvl-s tc bb2.0 pu-f doyna nucru rnucr LZMPi LZMV4k LZMV256 exoraw
- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------- ------
1 46.9 39.9 37.2 34.8 11.7 9.9 19.6 24.5 25.3 33.1 41.0
2 44.3 32.5 32.5 25.6 7.8 6.1 13.9 17.7 17.9 26.4 30.1
3 43.0 29.7 29.7 25.8 8.5 6.1 15.5 19.4 19.2 28.4 32.6
4 40.4 36.2 36.2 32.7 9.2 6.6 16.8 20.6 20.1 31.1 38.1
5 46.6 43.1 40.0 28.6 7.7 5.7 14.2 16.8 16.9 30.3 38.0
6 75.5 60.7 60.7 41.3 13.8 10.1 31.3 26.7 27.3 44.9 42.1
7 44.4 39.1 37.7 30.7 9.1 7.1 16.5 19.8 20.0 29.7 35.6
8 43.0 32.9 32.9 26.6 7.9 5.9 15.1 18.0 17.7 28.1 32.6
9 46.2 35.1 35.1 26.6 7.0 5.4 13.5 16.2 16.1 27.4 32.1
cycles per byte consumed
# bin rle wvl-f wvl-s tc bb2.0 pu-f doyna nucru rnucr LZMPi LZMV4k LZMV256 exoraw
- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------- ------
1 0.0 28.2 58.6 68.7 70.4 272.9 289.7 165.6 134.2 129.4 30.2 24.4
2 0.0 25.1 41.7 44.6 54.6 244.4 259.3 136.9 108.0 107.5 37.8 33.2
3 0.0 25.3 42.7 48.8 49.6 212.3 247.9 116.6 93.8 94.6 35.2 30.6
4 0.0 25.9 44.0 45.7 48.8 200.3 260.4 111.4 90.3 93.1 32.2 26.2
5 0.0 26.0 30.1 33.6 46.4 212.8 277.1 115.1 97.5 96.4 33.0 26.3
6 0.0 32.2 44.4 45.9 63.5 245.3 326.3 109.3 127.7 125.1 22.3 23.8
7 0.0 25.6 41.4 46.2 55.8 233.2 280.7 128.0 108.2 107.4 33.7 28.1
8 0.0 23.6 40.0 42.7 52.4 219.9 256.4 116.1 96.6 97.2 35.5 30.6
9 0.0 23.4 35.5 35.6 49.3 223.0 267.7 117.6 96.9 97.4 36.5 31.1
|
| |
mankeli
Registered: Oct 2010
Posts: 146 |
just for comparison:
cycles per byte consumed (bin): 4+4 = 8
kilobytes output per second (bin): (312*63*50)/8 -> 122.85kB/s
etc.
:) |
| |
MagerValp
Registered: Dec 2001
Posts: 1078 |
That requires an unrolled loop and hardcoded addresses, so it's not a realistic comparison. |
| |
Bitbreaker
Registered: Oct 2002
Posts: 508 |
also with direct loading into registers and presorting to maximize register reuse upon same values, we get below 6 cycles ;-) For further brainfuck we can combine registers with SAX, SHA, ... on some stores :-D |
| |
ChristopherJam
Registered: Aug 2004
Posts: 1409 |
Quoting Bitbreakeralso with direct loading into registers and presorting to maximize register reuse upon same values, we get below 6 cycles ;-)
Exactly what's required for half-cell Koalas, or there wouldn't be enough CPU time to update 37 bytes of d800 values every four lines :)
But yeah, I was assuming the .bin was already in place, so this is the excess cycles over time taken to transfer from external storage, be it 1541 or something more exotic
|
| |
ChristopherJam
Registered: Aug 2004
Posts: 1409 |
Quoting MagerValpAdded timed benchmarks for LZMV, both the original 256 byte sliding window and the 4k version I used in U4. Note that neither is optimized for speed right now, but size and convenience.
Thank you! Looking very similar to TinyCrunch there, maybe I should be dusting that off too
|
| Previous - 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 - Next |
