comparative timings (ncplayer -k vs timg)

[dankamongmen]

timg was recently added to Debian. it certainly wasn't for its performance relative to ncplayer:

for i in seq 1 10 ; do /usr/bin/time timg notcurses.png 2>> timg-timings ; done

[schwarzgerat](0) $ cat timg-timings 
1.13user 3.56system 0:00.14elapsed 3266%CPU (0avgtext+0avgdata 70832maxresident)k
757inputs+0outputs (0major+6073minor)pagefaults 0swaps
1.14user 3.55system 0:00.14elapsed 3154%CPU (0avgtext+0avgdata 60132maxresident)k
757inputs+0outputs (0major+6096minor)pagefaults 0swaps
1.14user 3.55system 0:00.14elapsed 3219%CPU (0avgtext+0avgdata 59932maxresident)k
757inputs+0outputs (0major+6409minor)pagefaults 0swaps
1.07user 3.62system 0:00.14elapsed 3242%CPU (0avgtext+0avgdata 59556maxresident)k
757inputs+0outputs (0major+6427minor)pagefaults 0swaps
1.12user 3.58system 0:00.15elapsed 3138%CPU (0avgtext+0avgdata 60712maxresident)k
757inputs+0outputs (0major+5873minor)pagefaults 0swaps
1.07user 3.63system 0:00.14elapsed 3242%CPU (0avgtext+0avgdata 60244maxresident)k
757inputs+0outputs (0major+5782minor)pagefaults 0swaps
1.19user 3.50system 0:00.14elapsed 3198%CPU (0avgtext+0avgdata 59556maxresident)k
757inputs+0outputs (0major+6406minor)pagefaults 0swaps
1.08user 3.61system 0:00.14elapsed 3242%CPU (0avgtext+0avgdata 61304maxresident)k
757inputs+0outputs (0major+6376minor)pagefaults 0swaps
1.22user 3.46system 0:00.14elapsed 3231%CPU (0avgtext+0avgdata 59936maxresident)k
757inputs+0outputs (0major+6404minor)pagefaults 0swaps
1.13user 3.60system 0:00.15elapsed 3139%CPU (0avgtext+0avgdata 94312maxresident)k
757inputs+0outputs (0major+5848minor)pagefaults 0swaps

for i in seq 1 10 ; do /usr/bin/time ncplayer -k -bpixel -sscale notcurses.png 2>> ncplayer-timings ; done

[schwarzgerat](0) $ cat ncplayer-timings 
0.04user 0.01system 0:00.08elapsed 74%CPU (0avgtext+0avgdata 40092maxresident)k
685inputs+0outputs (0major+4540minor)pagefaults 0swaps
0.05user 0.00system 0:00.08elapsed 76%CPU (0avgtext+0avgdata 40196maxresident)k
685inputs+0outputs (0major+4538minor)pagefaults 0swaps
0.04user 0.01system 0:00.08elapsed 72%CPU (0avgtext+0avgdata 40112maxresident)k
685inputs+0outputs (0major+4542minor)pagefaults 0swaps
0.03user 0.02system 0:00.07elapsed 75%CPU (0avgtext+0avgdata 40136maxresident)k
685inputs+0outputs (0major+4537minor)pagefaults 0swaps
0.05user 0.01system 0:00.08elapsed 72%CPU (0avgtext+0avgdata 40068maxresident)k
685inputs+0outputs (0major+4534minor)pagefaults 0swaps
0.04user 0.01system 0:00.08elapsed 74%CPU (0avgtext+0avgdata 40156maxresident)k
685inputs+0outputs (0major+4535minor)pagefaults 0swaps
0.04user 0.02system 0:00.08elapsed 71%CPU (0avgtext+0avgdata 39984maxresident)k
685inputs+0outputs (0major+4537minor)pagefaults 0swaps
0.04user 0.01system 0:00.08elapsed 72%CPU (0avgtext+0avgdata 39740maxresident)k
685inputs+0outputs (0major+4529minor)pagefaults 0swaps
0.03user 0.02system 0:00.08elapsed 75%CPU (0avgtext+0avgdata 40112maxresident)k
685inputs+0outputs (0major+4534minor)pagefaults 0swaps
0.03user 0.02system 0:00.08elapsed 75%CPU (0avgtext+0avgdata 40192maxresident)k
685inputs+0outputs (0major+4540minor)pagefaults 0swaps
[schwarzgerat](0) $ 

=] =] =]

[dankamongmen]

that was on kitty. on xterm, timg doesn't appear to draw sixel. nonetheless, our timings are reasonable compared to it coughing up what look to be halfblocks:

[schwarzgerat](0) $ cat ncplayer-timings 
invalid string stashed 0
0.06user 0.01system 0:00.32elapsed 24%CPU (0avgtext+0avgdata 44248maxresident)k
678inputs+0outputs (0major+4089minor)pagefaults 0swaps
invalid string stashed 0
0.06user 0.01system 0:00.24elapsed 30%CPU (0avgtext+0avgdata 44104maxresident)k
678inputs+0outputs (0major+4088minor)pagefaults 0swaps
invalid string stashed 0
0.07user 0.00system 0:00.24elapsed 32%CPU (0avgtext+0avgdata 44816maxresident)k
678inputs+0outputs (0major+4103minor)pagefaults 0swaps
invalid string stashed 0
0.06user 0.01system 0:00.24elapsed 33%CPU (0avgtext+0avgdata 44424maxresident)k
678inputs+0outputs (0major+4088minor)pagefaults 0swaps
invalid string stashed 0
0.07user 0.00system 0:00.24elapsed 33%CPU (0avgtext+0avgdata 44392maxresident)k
678inputs+0outputs (0major+4093minor)pagefaults 0swaps
invalid string stashed 0
0.06user 0.01system 0:00.24elapsed 33%CPU (0avgtext+0avgdata 44096maxresident)k
678inputs+0outputs (0major+4086minor)pagefaults 0swaps
invalid string stashed 0
0.07user 0.00system 0:00.25elapsed 31%CPU (0avgtext+0avgdata 44152maxresident)k
678inputs+0outputs (0major+4114minor)pagefaults 0swaps
invalid string stashed 0
0.05user 0.03system 0:00.24elapsed 32%CPU (0avgtext+0avgdata 44140maxresident)k
678inputs+0outputs (0major+4086minor)pagefaults 0swaps
invalid string stashed 0
0.07user 0.00system 0:00.29elapsed 26%CPU (0avgtext+0avgdata 44324maxresident)k
678inputs+0outputs (0major+4152minor)pagefaults 0swaps
invalid string stashed 0
0.06user 0.01system 0:00.24elapsed 31%CPU (0avgtext+0avgdata 44504maxresident)k
678inputs+0outputs (0major+4097minor)pagefaults 0swaps

[schwarzgerat](0) $ cat timg-timings 
0.75user 2.39system 0:00.11elapsed 2763%CPU (0avgtext+0avgdata 64240maxresident)k
757inputs+0outputs (0major+5103minor)pagefaults 0swaps
0.53user 1.65system 0:00.09elapsed 2257%CPU (0avgtext+0avgdata 57752maxresident)k
757inputs+0outputs (0major+5217minor)pagefaults 0swaps
0.55user 1.75system 0:00.09elapsed 2356%CPU (0avgtext+0avgdata 57008maxresident)k
757inputs+0outputs (0major+5236minor)pagefaults 0swaps
0.61user 1.71system 0:00.09elapsed 2357%CPU (0avgtext+0avgdata 58456maxresident)k
757inputs+0outputs (0major+5117minor)pagefaults 0swaps
0.59user 1.84system 0:00.10elapsed 2346%CPU (0avgtext+0avgdata 90832maxresident)k
757inputs+0outputs (0major+4959minor)pagefaults 0swaps
0.68user 1.85system 0:00.10elapsed 2387%CPU (0avgtext+0avgdata 90672maxresident)k
757inputs+0outputs (0major+4980minor)pagefaults 0swaps
0.64user 1.72system 0:00.10elapsed 2300%CPU (0avgtext+0avgdata 57444maxresident)k
757inputs+0outputs (0major+5137minor)pagefaults 0swaps
0.59user 1.85system 0:00.10elapsed 2421%CPU (0avgtext+0avgdata 58532maxresident)k
757inputs+0outputs (0major+5118minor)pagefaults 0swaps
0.55user 1.81system 0:00.09elapsed 2416%CPU (0avgtext+0avgdata 57112maxresident)k
757inputs+0outputs (0major+5134minor)pagefaults 0swaps
0.67user 1.79system 0:00.10elapsed 2288%CPU (0avgtext+0avgdata 57172maxresident)k
757inputs+0outputs (0major+5276minor)pagefaults 0swaps
[schwarzgerat](0) $ 

[dankamongmen]

note that we've grown a bit slower since this time in kitty, due to our use of deflate. this speeds us up on a remote machine, generally (due to reduced bandwidth), but typically slows us down on a local machine.

[dankamongmen]

i reran this today and since these data, we've held the line pretty much exactly on kitty, and cut our linux console times by about 50%