--blocksize: info on its impact on strength and how to use correctly?, Was: --blocksize (TOS #6) |
--blocksize: info on its impact on strength and how to use correctly?, Was: --blocksize (TOS #6) |
 Feb 6 2012, 21:02
Post
#1
|
|
|
Group: Members Posts: 231 Joined: 6-April 09 Member No.: 68706  |
Recently I discovered --blocksize.
Bryant, could you tell some more about its impact on strength and how to use it correctly? I saw some topics mentioning that setting it very low can have positive impact on some sources. I tried the opposite, setting it to maximum on my quick and dirty high-res, multichannel dataset and it turned out to be slightly better than default, 0.05% on average, winning on 22 of 29 samples. Still, it was 0.71% worse than flake. Taking the best result from default / --blocksize reduced the flake's advantage to 0.63%. |
 |
 
|
bryant Increasing the block size from the default will no... Feb 7 2012, 21:01
_mē_ Thanks for the answer. All the numbers that I gave... Feb 7 2012, 21:40 bryant QUOTE (_mē_ @ Feb 7 2012, 12:40) All the ... Feb 8 2012, 20:50 DARcode -hhx6 rather fast?! Are you using a supercompu... Feb 8 2012, 10:16 _mē_ QUOTE (DARcode @ Feb 8 2012, 10:16) -hhx6... Feb 8 2012, 20:24 Porcus QUOTE (_mē_ @ Feb 8 2012, 20:24) I don... Feb 8 2012, 20:53 _mē_ I used SVN version of flake and it's significa... Feb 8 2012, 22:48  |
|
Lo-Fi Version | Time is now: 18th May 2013 - 13:03 |
