Newest at the top
2024-11-01 22:14:37 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 252 seconds) |
2024-11-01 22:10:08 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
2024-11-01 22:08:00 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 265 seconds) |
2024-11-01 22:06:56 +0100 | hgolden_ | (~hgolden@static-198-44-129-51.cust.tzulo.com) (Ping timeout: 252 seconds) |
2024-11-01 22:06:43 +0100 | Tuplanolla | (~Tuplanoll@91-159-69-59.elisa-laajakaista.fi) Tuplanolla |
2024-11-01 22:04:20 +0100 | hgolden__ | (~hgolden@static-198-44-129-83.cust.tzulo.com) hgolden |
2024-11-01 22:03:10 +0100 | morb | (~morb@pool-108-41-100-120.nycmny.fios.verizon.net) (Ping timeout: 265 seconds) |
2024-11-01 21:59:57 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
2024-11-01 21:58:49 +0100 | Feuermagier | (~Feuermagi@user/feuermagier) Feuermagier |
2024-11-01 21:58:38 +0100 | morb | (~morb@pool-108-41-100-120.nycmny.fios.verizon.net) |
2024-11-01 21:58:27 +0100 | Feuermagier | (~Feuermagi@user/feuermagier) (Remote host closed the connection) |
2024-11-01 21:56:06 +0100 | Feuermagier | (~Feuermagi@user/feuermagier) Feuermagier |
2024-11-01 21:52:56 +0100 | euandreh | (~Thunderbi@2804:d59:891b:2c00:896a:156a:7a25:3f91) euandreh |
2024-11-01 21:49:08 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 272 seconds) |
2024-11-01 21:45:46 +0100 | <tomsmeding> | the measurements are quite reliable here, individual executions differ maybe 5%; but the differences we're looking at are 100% or more, so hyperfine is a luxury :p |
2024-11-01 21:45:00 +0100 | <zzz> | i need to go grab something to eat. in the meantime i will leave hyperfine running |
2024-11-01 21:44:10 +0100 | <tomsmeding> | (my CPU is 3 generations newer (10th gen) and runs at ~160% the frequency (it maxes at 5GHz); together that neatly explains mine being 2x as fast as yours) |
2024-11-01 21:43:52 +0100 | <zzz> | ok now i'm getting similar results |
2024-11-01 21:42:19 +0100 | <tomsmeding> | they are within 5% of each other on my machine |
2024-11-01 21:41:56 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
2024-11-01 21:41:51 +0100 | <tomsmeding> | zzz: what is also confusing is how this one takes 1.3 seconds (which is about 2x as slow as my machine, which is quite to be expected given the cpu freq + age difference), but the previous one with the \case was 7x as slow as my machine |
2024-11-01 21:41:27 +0100 | <zzz> | ok this is weird. let me doublecheck something |
2024-11-01 21:40:50 +0100 | sord937 | (~sord937@gateway/tor-sasl/sord937) (Quit: sord937) |
2024-11-01 21:40:04 +0100 | <tomsmeding> | perhaps ghc is less eager to inline across modules, and in one stuff gets inlined and in the other it doesn't? |
2024-11-01 21:39:41 +0100 | <zzz> | ok wait. trying one thing |
2024-11-01 21:39:31 +0100 | <zzz> | yes |
2024-11-01 21:39:22 +0100 | <tomsmeding> | can you reproduce your original measurements :p |
2024-11-01 21:38:03 +0100 | zzz | confused |
2024-11-01 21:37:36 +0100 | <zzz> | https://paste.jrvieira.com/1730493448663 |
2024-11-01 21:32:48 +0100 | GdeVolpiano | (~GdeVolpia@user/GdeVolpiano) GdeVolpiano |
2024-11-01 21:31:02 +0100 | <zzz> | y |
2024-11-01 21:30:48 +0100 | todi | (~todi@p57803331.dip0.t-ipconnect.de) todi |
2024-11-01 21:30:41 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 248 seconds) |
2024-11-01 21:30:17 +0100 | <tomsmeding> | can you try my last paste? |
2024-11-01 21:29:56 +0100 | <zzz> | i consistently get better results for compiled rnf |
2024-11-01 21:29:29 +0100 | <zzz> | one file for each version |
2024-11-01 21:29:28 +0100 | <tomsmeding> | zzz: https://paste.tomsmeding.com/WGmJJRuD |
2024-11-01 21:29:12 +0100 | <zzz> | tomsmeding: oh, i did that so i could easily share it. i ran separate files |
2024-11-01 21:28:52 +0100 | <monochrom> | That is strange. |
2024-11-01 21:28:26 +0100 | <tomsmeding> | are you perhaps testing one version with it all in one file and one version split over multiple files? |
2024-11-01 21:28:08 +0100 | <tomsmeding> | zzz: if I compile your code as-is (but do put it all in one file), I get similar timings too |
2024-11-01 21:27:47 +0100 | <zzz> | monochrom: 8GB Ram :) |
2024-11-01 21:26:57 +0100 | <monochrom> | OK I'll stop until I think up a better joke! |
2024-11-01 21:26:18 +0100 | <monochrom> | What if the web browser already eats up 0.4999GB and the OS eats up the rest :) |
2024-11-01 21:25:47 +0100 | <zzz> | tomsmeding: i don't understand why in your test the difference between deepseq and rnf disappears |
2024-11-01 21:25:10 +0100 | <tomsmeding> | case 1 and 2 run lean |
2024-11-01 21:25:06 +0100 | <tomsmeding> | monochrom: that would be for case 3 only |
2024-11-01 21:24:54 +0100 | <EvanR> | intel celery |
2024-11-01 21:24:48 +0100 | <monochrom> | OK now I'm hoping "oh it only has 0.5GB RAM so yeah it's thrashing" |
2024-11-01 21:24:26 +0100 | <tomsmeding> | but yeah with a CPU of that class I would not expect a 7x difference, but I guess it could be |