Replies: 11 comments 31 replies
-
|
Hey Hartmut(@hg747), I've made 2 modifications to your entry:
You now have to synchronise your repository with the main one in order to have these changes in your own. Here are the automation results: $ ./test_all.sh hgrosser
******** Test ********
===== Hartmut Grosser ======
4256d19d3e134d79cc6f160d428a1d859ce961167bd01ca528daca8705163910 /home/gcarreno/Programming/1brc-ObjectPascal/results/hgrosser.output
4256d19d3e134d79cc6f160d428a1d859ce961167bd01ca528daca8705163910 Official Output Hash
===========$ ./run_all.sh hgrosser
******** Run ********
===== Hartmut Grosser ======
-- SSD --
Benchmark 1: hgrosser
Time (mean ± σ): 73.390 s ± 0.880 s [User: 71.909 s, System: 1.469 s]
Range (min … max): 71.792 s … 74.810 s 10 runs
===========These are just preliminary results. The official ones will be posted after the weekly Saturday full run of the automation. Good work!! This is an amazing entry, especially being a single threaded one!! Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Thanks a lot for this enjoyable news. |
Beta Was this translation helpful? Give feedback.
-
|
Already done as mentioned in the PR conversation. |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, thank you for merging. Did you see that the 2nd command line parameter has changed and needs some optimizing? Please have a look in file README.md, chapter "Optimizing the 2nd command line parameter". Please try the mentioned values (maybe in a for-loop). If not too much work for you, I would be interested to see the different time measurements. Thanks a lot. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), You're more than welcome!! I did notice the second param while testing. I've quickly used 16, but will do the proper testing with other numbers to check which is quicker. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Happy birthday, Gus!! I wish you the very best for the next year, especially the best for your health and complete satisfaction. Take the day off and have a nice day. 'till tomorrow, Cheers Hartmut |
Beta Was this translation helpful? Give feedback.
-
|
Thank you so, so very much, from way deep in my heart!! 🫂 ❤️ |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, I've seen the new results and am glad about mine. Did you optimize my 2nd command line parameter as I asked you for? Please have a look in file README.md, chapter "Optimizing the 2nd command line parameter". Please try the mentioned values (maybe in a for-loop). If not too much work for you, I would be interested to see the different time measurements. Thanks a lot. And which value did you use for the new official results? Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), I was aware of your asks, but I did this a bit in a rush cuz I was barely awake 😄 I was planning on taking my time to fully run all the possibilities in order to optimise it. I can only apologise for the rush job!! The test was performed with Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Ok, thanks for this infos. I'll be patient... |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, I was patient ... but now we have a new result table and it looks like my 2nd command line parameter was not yet optimized ... I have an idea for an improvement of my hash table, but it makes no sense to compare the new table with the old table, before the old table got optimized. Please could you do the optimization of my 2nd command line parameter? Thanks in advance. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), I can only apologise profusely for not having done your optimization yet. It's now in my top high priority list once I get some proper sleep. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), Results on a not so quiet system: $ ./hgrosser.sh
===== Hartmut Grosser ======
-- 14 --
Benchmark 1: hgrosser-14
Time (mean ± σ): 63.620 s ± 1.023 s [User: 62.311 s, System: 1.295 s]
Range (min … max): 62.628 s … 65.093 s 5 runs
===========
===== Hartmut Grosser ======
-- 15 --
Benchmark 1: hgrosser-15
Time (mean ± σ): 57.649 s ± 0.448 s [User: 56.317 s, System: 1.324 s]
Range (min … max): 57.099 s … 58.290 s 5 runs
===========
===== Hartmut Grosser ======
-- 16 --
Benchmark 1: hgrosser-16
Time (mean ± σ): 52.353 s ± 0.671 s [User: 51.078 s, System: 1.269 s]
Range (min … max): 51.493 s … 52.913 s 5 runs
===========
===== Hartmut Grosser ======
-- 17 --
Benchmark 1: hgrosser-17
Time (mean ± σ): 51.011 s ± 1.317 s [User: 49.722 s, System: 1.286 s]
Range (min … max): 49.731 s … 53.242 s 5 runs
Warning: The first benchmarking run for this command was significantly slower than the rest (53.242 s). This could be caused by (filesystem) caches that were not filled until after the first run. You are already using the '--warmup' option which helps to fill these caches before the actual benchmark. You can either try to increase the warmup count further or re-run this benchmark on a quiet system in case it was a random outlier. Alternatively, consider using the '--prepare' option to clear the caches before each timing run.
===========
===== Hartmut Grosser ======
-- 18 --
Benchmark 1: hgrosser-18
Time (mean ± σ): 49.669 s ± 0.698 s [User: 48.353 s, System: 1.306 s]
Range (min … max): 48.948 s … 50.763 s 5 runs
===========
===== Hartmut Grosser ======
-- 19 --
Benchmark 1: hgrosser-19
Time (mean ± σ): 51.537 s ± 0.813 s [User: 50.209 s, System: 1.313 s]
Range (min … max): 50.496 s … 52.471 s 5 runs
===========
===== Hartmut Grosser ======
-- 20 --
Benchmark 1: hgrosser-20
Time (mean ± σ): 53.155 s ± 0.739 s [User: 51.852 s, System: 1.302 s]
Range (min … max): 52.336 s … 54.081 s 5 runs
===========
===== Hartmut Grosser ======
-- 21 --
Benchmark 1: hgrosser-21
Time (mean ± σ): 55.467 s ± 0.753 s [User: 54.056 s, System: 1.399 s]
Range (min … max): 54.522 s … 56.185 s 5 runs
===========
===== Hartmut Grosser ======
-- 22 --
Benchmark 1: hgrosser-22
Time (mean ± σ): 56.560 s ± 0.530 s [User: 55.232 s, System: 1.321 s]
Range (min … max): 55.919 s … 57.316 s 5 runs
===========
===== Hartmut Grosser ======
-- 23 --
Benchmark 1: hgrosser-23
Time (mean ± σ): 58.742 s ± 0.490 s [User: 57.362 s, System: 1.378 s]
Range (min … max): 57.949 s … 59.261 s 5 runs
===========
===== Hartmut Grosser ======
-- 24 --
Benchmark 1: hgrosser-24
Time (mean ± σ): 60.038 s ± 0.680 s [User: 58.602 s, System: 1.434 s]
Range (min … max): 59.066 s … 60.974 s 5 runs
===========
===== Hartmut Grosser ======
-- 25 --
Benchmark 1: hgrosser-25
Time (mean ± σ): 61.331 s ± 1.518 s [User: 59.726 s, System: 1.602 s]
Range (min … max): 59.546 s … 62.752 s 5 runs
===========
===== Hartmut Grosser ======
-- 26 --
Benchmark 1: hgrosser-26
Time (mean ± σ): 62.185 s ± 1.472 s [User: 60.292 s, System: 1.888 s]
Range (min … max): 60.404 s … 64.012 s 5 runs
===========
===== Hartmut Grosser ======
-- 27 --
Benchmark 1: hgrosser-27
Time (mean ± σ): 64.932 s ± 0.614 s [User: 62.414 s, System: 2.505 s]
Range (min … max): 64.357 s … 65.966 s 5 runs
===========
===== Hartmut Grosser ======
-- 28 --
Benchmark 1: hgrosser-28
Time (mean ± σ): 68.924 s ± 1.050 s [User: 65.100 s, System: 3.815 s]
Range (min … max): 67.564 s … 70.230 s 5 runs
===========And the From this preliminary run, I'm guessing that 18 hits the sweet spot, right? Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, thank you very much for doing this optimization run and passing the results. Yes, 18 is the best value and saves about 2.5 seconds. Because the winnings is not more, I will try an other modification to my hash list which could save more seconds. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, I made a new PR for a new version (1.60), which (hopefully) is a little faster. I'm sorry that this version needs a new optimization run for the 2nd command line parameter (but not so many values as before, only from 16 to 22 please). It would be great, if you could do this optimization run with the old input file (with CR's), because then the timing is comparable (the new input file is smaller = faster reading). I hope, I'm asking not too much. Thanks a lot. Please pay attention, that IF the input file has no more CR's, that then a new Conditional "noCR" must be set (see chapter "How to compile" in readme.md). Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Harmut(@hg747), We are never getting back to Please make the necessary changes to your code to make that a reality. All entries must be expecting Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), Once you make that change permanent, I'll run 16 to 22. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, I made a new PR for a new version (1.61) which permanently switches input files to have no CR's. Please do the optimization run for the 2nd command line switch with values from 16 to 22. If possible, I would be curious to see the different timing values. Thanks. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), That's good, I'll merge that and go through the tests from 16 to 22. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), Fine tuning: $ ./hgrosser.sh
===== Hartmut Grosser ======
-- 16 --
Benchmark 1: hgrosser-16
Time (mean ± σ): 50.980 s ± 0.222 s [User: 49.551 s, System: 1.426 s]
Range (min … max): 50.675 s … 51.292 s 5 runs
===========
===== Hartmut Grosser ======
-- 17 --
Benchmark 1: hgrosser-17
Time (mean ± σ): 49.199 s ± 0.394 s [User: 47.710 s, System: 1.486 s]
Range (min … max): 48.602 s … 49.559 s 5 runs
===========
===== Hartmut Grosser ======
-- 18 --
Benchmark 1: hgrosser-18
Time (mean ± σ): 48.047 s ± 0.722 s [User: 46.565 s, System: 1.479 s]
Range (min … max): 47.027 s … 48.792 s 5 runs
===========
===== Hartmut Grosser ======
-- 19 --
Benchmark 1: hgrosser-19
Time (mean ± σ): 49.529 s ± 0.192 s [User: 48.012 s, System: 1.514 s]
Range (min … max): 49.314 s … 49.784 s 5 runs
===========
===== Hartmut Grosser ======
-- 20 --
Benchmark 1: hgrosser-20
Time (mean ± σ): 51.348 s ± 0.615 s [User: 49.790 s, System: 1.550 s]
Range (min … max): 50.291 s … 51.782 s 5 runs
===========
===== Hartmut Grosser ======
-- 21 --
Benchmark 1: hgrosser-21
Time (mean ± σ): 53.108 s ± 0.513 s [User: 51.619 s, System: 1.485 s]
Range (min … max): 52.207 s … 53.473 s 5 runs
===========
===== Hartmut Grosser ======
-- 22 --
Benchmark 1: hgrosser-22
Time (mean ± σ): 54.413 s ± 0.952 s [User: 52.891 s, System: 1.518 s]
Range (min … max): 53.471 s … 55.762 s 5 runs
===========hgrosser-22.json I think that we've proven that 18 is still the soft spot. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, thanks a lot for this optimization run. Yes, 18 is again the best value, although I altered the hash list. Please use 18 for Saturday's. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), Yeaps, already on the automation!! Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Don't we have a new results table for last saturday? |
Beta Was this translation helpful? Give feedback.
-
|
Hey Arnaud(@synopse), CRAP!!! You're right!! Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
@gcarreno |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, if you don't have the time to do a results table run before next Saturday, than this is so and we will live with it. In question to the end of the challenge, we could say: the last accepted PR will be on 10th of May at e.g. 12:00 p.m. Then the final results table run could be on Saturday, 11th of May. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), That's actually a nice idea!! Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
I suspect I won't change anything at my code now. |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, I want to improve and optimize one of my 2 programs for 400 stations only and full name compares. But I don't know which one, before I know, if my new version with threads works on your computer with 32 threads and how long it needs. As said, it is my very first multi thread program ever. Do you think it's possible that you run the optimization tests I asked you for? For instructions please have a look in README.md. Thanks a lot. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus (@gcarreno), I'm afraid that you did not see my post from yesterday because you normally answered after a short time. Please don't forget me, Thanks. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), I've been paying attention to all the conversations and I did see yours. I've been really busy with a single problem on my new cryptocoin's explorer and it's been driving me crazy and keeping me awake at night. I can assure you that today, Saturday, I will not touch the cryptocoin side of things and will dedicate all my time to Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), $ hyperfine -w 1 -r 5 -N -n 'hgrosser-th' --export-json 'results/hgrosser-th.json' './bin/hgrosser-th /tmp/measurements-1_000_000_000.txt 18'
Benchmark 1: hgrosser-th
Time (mean ± σ): 10.155 s ± 0.514 s [User: 122.707 s, System: 10.017 s]
Range (min … max): 9.863 s … 11.067 s 5 runs$ hyperfine -w 1 -r 5 -N -n 'hgrosser-th-400' --export-json 'results/hgrosser-th-400.json' './bin/hgrosser-th /tmp/measurements-400-1_000_000_000.txt 18'
Benchmark 1: hgrosser-th-400
Time (mean ± σ): 3.083 s ± 0.059 s [User: 45.178 s, System: 2.064 s]
Range (min … max): 3.019 s … 3.145 s 5 runs |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus, I don't know what to do with this 2 runs. This are not the optimization tests I am waiting for since Tuesday. Now is Sunday. I'm a little disappointed about that. Please have a look in the README.md and do this optimization tests soon, and before you create the next "weekly" results table. Please don't use 400 stations currently, because this creates wrong results. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), I'm sorry for any mistake I've made. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), Just ran a 2 deep loop in a not so quiet machine. Please have a look at these results and inform me what should be the defaults. $ ./hgrosser.sh
===== Hartmut Grosser ======
-- 32 16 64 --
Benchmark 1: hgrosser-th-32-16-64
Time (mean ± σ): 4.883 s ± 0.081 s [User: 136.897 s, System: 4.060 s]
Range (min … max): 4.792 s … 4.972 s 5 runs
-- 32 16 96 --
Benchmark 1: hgrosser-th-32-16-96
Time (mean ± σ): 5.092 s ± 0.118 s [User: 142.227 s, System: 3.844 s]
Range (min … max): 4.995 s … 5.274 s 5 runs
-- 32 16 128 --
Benchmark 1: hgrosser-th-32-16-128
Time (mean ± σ): 5.143 s ± 0.052 s [User: 144.513 s, System: 3.895 s]
Range (min … max): 5.089 s … 5.199 s 5 runs
-- 32 16 192 --
Benchmark 1: hgrosser-th-32-16-192
Time (mean ± σ): 5.239 s ± 0.055 s [User: 148.639 s, System: 3.977 s]
Range (min … max): 5.173 s … 5.322 s 5 runs
-- 32 16 256 --
Benchmark 1: hgrosser-th-32-16-256
Time (mean ± σ): 5.405 s ± 0.102 s [User: 153.700 s, System: 4.027 s]
Range (min … max): 5.305 s … 5.570 s 5 runs
-- 32 17 64 --
Benchmark 1: hgrosser-th-32-17-64
Time (mean ± σ): 6.052 s ± 0.135 s [User: 168.629 s, System: 4.103 s]
Range (min … max): 5.811 s … 6.117 s 5 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
-- 32 17 96 --
Benchmark 1: hgrosser-th-32-17-96
Time (mean ± σ): 6.106 s ± 0.110 s [User: 170.842 s, System: 3.972 s]
Range (min … max): 5.959 s … 6.220 s 5 runs
-- 32 17 128 --
Benchmark 1: hgrosser-th-32-17-128
Time (mean ± σ): 6.079 s ± 0.102 s [User: 168.348 s, System: 3.724 s]
Range (min … max): 5.983 s … 6.239 s 5 runs
-- 32 17 192 --
Benchmark 1: hgrosser-th-32-17-192
Time (mean ± σ): 6.315 s ± 0.113 s [User: 178.316 s, System: 3.860 s]
Range (min … max): 6.180 s … 6.429 s 5 runs
-- 32 17 256 --
Benchmark 1: hgrosser-th-32-17-256
Time (mean ± σ): 6.471 s ± 0.095 s [User: 182.470 s, System: 3.852 s]
Range (min … max): 6.333 s … 6.592 s 5 runs
-- 32 18 64 --
Benchmark 1: hgrosser-th-32-18-64
Time (mean ± σ): 6.934 s ± 0.126 s [User: 192.952 s, System: 4.145 s]
Range (min … max): 6.776 s … 7.122 s 5 runs
-- 32 18 96 --
Benchmark 1: hgrosser-th-32-18-96
Time (mean ± σ): 7.197 s ± 0.183 s [User: 204.000 s, System: 4.022 s]
Range (min … max): 6.974 s … 7.343 s 5 runs
-- 32 18 128 --
Benchmark 1: hgrosser-th-32-18-128
Time (mean ± σ): 7.198 s ± 0.127 s [User: 200.920 s, System: 3.881 s]
Range (min … max): 6.991 s … 7.314 s 5 runs
-- 32 18 192 --
Benchmark 1: hgrosser-th-32-18-192
Time (mean ± σ): 7.290 s ± 0.077 s [User: 206.812 s, System: 3.786 s]
Range (min … max): 7.167 s … 7.349 s 5 runs
-- 32 18 256 --
Benchmark 1: hgrosser-th-32-18-256
Time (mean ± σ): 7.605 s ± 0.168 s [User: 214.971 s, System: 3.915 s]
Range (min … max): 7.332 s … 7.784 s 5 runs
-- 32 19 64 --
Benchmark 1: hgrosser-th-32-19-64
Time (mean ± σ): 7.825 s ± 0.119 s [User: 223.646 s, System: 4.332 s]
Range (min … max): 7.638 s … 7.950 s 5 runs
-- 32 19 96 --
Benchmark 1: hgrosser-th-32-19-96
Time (mean ± σ): 7.906 s ± 0.138 s [User: 225.110 s, System: 4.156 s]
Range (min … max): 7.663 s … 7.996 s 5 runs
-- 32 19 128 --
Benchmark 1: hgrosser-th-32-19-128
Time (mean ± σ): 7.950 s ± 0.134 s [User: 227.847 s, System: 4.099 s]
Range (min … max): 7.786 s … 8.077 s 5 runs
-- 32 19 192 --
Benchmark 1: hgrosser-th-32-19-192
Time (mean ± σ): 8.079 s ± 0.122 s [User: 230.096 s, System: 3.966 s]
Range (min … max): 7.890 s … 8.216 s 5 runs
-- 32 19 256 --
Benchmark 1: hgrosser-th-32-19-256
Time (mean ± σ): 8.336 s ± 0.127 s [User: 239.489 s, System: 3.975 s]
Range (min … max): 8.181 s … 8.500 s 5 runs
-- 32 20 64 --
Benchmark 1: hgrosser-th-32-20-64
Time (mean ± σ): 8.352 s ± 0.060 s [User: 240.793 s, System: 4.584 s]
Range (min … max): 8.251 s … 8.395 s 5 runs
-- 32 20 96 --
Benchmark 1: hgrosser-th-32-20-96
Time (mean ± σ): 8.421 s ± 0.117 s [User: 243.560 s, System: 4.509 s]
Range (min … max): 8.295 s … 8.561 s 5 runs
-- 32 20 128 --
Benchmark 1: hgrosser-th-32-20-128
Time (mean ± σ): 8.460 s ± 0.199 s [User: 241.307 s, System: 4.391 s]
Range (min … max): 8.136 s … 8.672 s 5 runs
-- 32 20 192 --
Benchmark 1: hgrosser-th-32-20-192
Time (mean ± σ): 8.485 s ± 0.139 s [User: 242.975 s, System: 4.414 s]
Range (min … max): 8.271 s … 8.605 s 5 runs
-- 32 20 256 --
Benchmark 1: hgrosser-th-32-20-256
Time (mean ± σ): 8.787 s ± 0.201 s [User: 253.167 s, System: 4.345 s]
Range (min … max): 8.498 s … 8.987 s 5 runs
-- 32 21 64 --
Benchmark 1: hgrosser-th-32-21-64
Time (mean ± σ): 8.850 s ± 0.126 s [User: 251.338 s, System: 5.494 s]
Range (min … max): 8.643 s … 8.985 s 5 runs
-- 32 21 96 --
Benchmark 1: hgrosser-th-32-21-96
Time (mean ± σ): 8.839 s ± 0.094 s [User: 254.251 s, System: 5.404 s]
Range (min … max): 8.733 s … 8.927 s 5 runs
-- 32 21 128 --
Benchmark 1: hgrosser-th-32-21-128
Time (mean ± σ): 8.897 s ± 0.118 s [User: 254.693 s, System: 5.263 s]
Range (min … max): 8.790 s … 9.029 s 5 runs
-- 32 21 192 --
Benchmark 1: hgrosser-th-32-21-192
Time (mean ± σ): 9.077 s ± 0.111 s [User: 259.670 s, System: 5.230 s]
Range (min … max): 8.965 s … 9.243 s 5 runs
-- 32 21 256 --
Benchmark 1: hgrosser-th-32-21-256
Time (mean ± σ): 9.185 s ± 0.114 s [User: 263.032 s, System: 5.226 s]
Range (min … max): 9.053 s … 9.274 s 5 runs
-- 32 22 64 --
Benchmark 1: hgrosser-th-32-22-64
Time (mean ± σ): 9.277 s ± 0.185 s [User: 260.456 s, System: 7.713 s]
Range (min … max): 9.108 s … 9.563 s 5 runs
-- 32 22 96 --
Benchmark 1: hgrosser-th-32-22-96
Time (mean ± σ): 9.275 s ± 0.112 s [User: 260.175 s, System: 7.744 s]
Range (min … max): 9.125 s … 9.423 s 5 runs
-- 32 22 128 --
Benchmark 1: hgrosser-th-32-22-128
Time (mean ± σ): 9.226 s ± 0.098 s [User: 260.527 s, System: 7.499 s]
Range (min … max): 9.125 s … 9.353 s 5 runs
-- 32 22 192 --
Benchmark 1: hgrosser-th-32-22-192
Time (mean ± σ): 9.520 s ± 0.124 s [User: 270.825 s, System: 7.602 s]
Range (min … max): 9.395 s … 9.716 s 5 runs
-- 32 22 256 --
Benchmark 1: hgrosser-th-32-22-256
Time (mean ± σ): 9.570 s ± 0.100 s [User: 267.501 s, System: 7.585 s]
Range (min … max): 9.482 s … 9.719 s 5 runs
===========hgrosser-th-32-22-256.json Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus (@gcarreno), thank you very much for those optimization tests. But you did much more then I asked for. I wanted only 5 tests for 3 parameters each = 15 tests and you made 35 tests. No problem, thanks for your effort, but not neccessary next time. Sorry in case my README.md was not clear enough. The best results are for hgrosser-th-32-16-64 = 4.883 seconds. After looking in the new results table I see there only my old 1-thread program with 52 seconds and not my new 32-thread-program with 4.883 seconds. I'm very unlucky about that. Do you think it's possible to add my 32-thread-program with 4.883 seconds retroactive to this table? I would be very happy about that. And please make sure that for next Saturday my 32-thread-program will be part of the game. I'm working on an improved version which will need no further optimization tests :-)) Cheers and many thanks for your help, Hartmut |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), Once you shift the code from the threaded to the official one, or at least rename one to old and rename the threaded one to the official, I'll re-run your entry on a quiet machine and replace the values on the results table. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hello Gus (@gcarreno), thank you Gus for that offer. I'm not sure if I understood correctly, what I should do. In a new PR I renamed the single-threader to *_old in *.pas and *.lpi and the executable. Hope this was correct. Otherwise please feel free to rename the single-threader as you need it. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
Hey Hartmut(@hg747), I'll have a look at it and I'll run it in a quiet system right before I go to sleep and will update the table after I wake up. Cheers, |
Beta Was this translation helpful? Give feedback.
-
|
@gcarreno Great. Thanks. |
Beta Was this translation helpful? Give feedback.
-
Hey Hartmut(@hg747),
This will be your very own discussion so we can discuss it further.
Welcome to the family 😄
Cheers,
Gus
Beta Was this translation helpful? Give feedback.
All reactions