Is this slow for an i7 10700k?

[ecdeveloper]

Hey all, great project, thanks for putting the effort!

I have an i7 10700k with 32GB ram, 1TB+2TB nvme (3200 write speed) running on Ubuntu 20. The speed looks much higher from what I see others have with similar setup. Is it normal for my hardware?

Running with:

./build/chia_plot -n -1 -r 16 -t /mnt/ssd2/madmax/ -2 /mnt/ssd1/madmax/ -d /mnt/ssd1/madmax_final/ -p -f

Plot Name: plot-k32-2021-06-12-12-19- [P1] Table 1 took 25.2295 sec [P1] Table 2 took 182.435 sec, found 4294958453 matches [P1] Table 3 took 249.296 sec, found 4294933525 matches [P1] Table 4 took 332.004 sec, found 4294886325 matches [P1] Table 5 took 311.728 sec, found 4294754161 matches [P1] Table 6 took 277.429 sec, found 4294584613 matches [P1] Table 7 took 194.314 sec, found 4294284281 matches Phase 1 took 1572.45 sec [P2] max_table_size = 4294967296 [P2] Table 7 scan took 23.9863 sec [P2] Table 7 rewrite took 114.161 sec, dropped 0 entries (0 %) [P2] Table 6 scan took 47.1398 sec [P2] Table 6 rewrite took 98.6587 sec, dropped 581276769 entries (13.5351 %) [P2] Table 5 scan took 46.2642 sec [P2] Table 5 rewrite took 91.9907 sec, dropped 761965435 entries (17.7418 %) [P2] Table 4 scan took 45.0176 sec [P2] Table 4 rewrite took 87.8892 sec, dropped 828844959 entries (19.2984 %) [P2] Table 3 scan took 44.7446 sec [P2] Table 3 rewrite took 92.3199 sec, dropped 855077879 entries (19.909 %) [P2] Table 2 scan took 46.7557 sec [P2] Table 2 rewrite took 98.6613 sec, dropped 865577861 entries (20.1533 %) Phase 2 took 844.253 sec Wrote plot header with 268 bytes [P3-1] Table 2 took 48.9049 sec, wrote 3429380592 right entries [P3-2] Table 2 took 95.9241 sec, wrote 3429380592 left entries, 3429380592 final [P3-1] Table 3 took 109.8 sec, wrote 3439855646 right entries [P3-2] Table 3 took 77.3934 sec, wrote 3439855646 left entries, 3439855646 final [P3-1] Table 4 took 104.738 sec, wrote 3466041366 right entries [P3-2] Table 4 took 81.709 sec, wrote 3466041366 left entries, 3466041366 final [P3-1] Table 5 took 113.22 sec, wrote 3532788726 right entries [P3-2] Table 5 took 85.2263 sec, wrote 3532788726 left entries, 3532788726 final [P3-1] Table 6 took 117.379 sec, wrote 3713307844 right entries [P3-2] Table 6 took 86.5728 sec, wrote 3713307844 left entries, 3713307844 final [P3-1] Table 7 took 177.422 sec, wrote 4294284281 right entries [P3-2] Table 7 took 122.459 sec, wrote 4294284281 left entries, 4294284281 final Phase 3 took 1222.15 sec, wrote 21875658455 entries to final plot [P4] Starting to write C1 and C3 tables [P4] Finished writing C1 and C3 tables [P4] Writing C2 table [P4] Finished writing C2 table Phase 4 took 52.971 sec, final plot size is 108826444537 bytes Total plot creation time was 3691.86 sec Started copy to /mnt/ssd1/madmax_final/plot-k32-2021-06-12-12-19-hash.plot Process ID: 3219220 Number of Threads: 16 Number of Buckets: 2^7 (128)

As a follow up question, once the next plot creation starts - the speed of each phase goes even higher. May that be due to IO operations that are related to moving the final plot around? Any way to mitigate that?

[dEMonaRE]

I have exact same setup with aio cooling i think its normal, xeons' time less than this in windows whole creation time vary between 3000 sec -3900 sec with different number of threads

[Smallthing]

too slow my Phase 1 took 1063.97 sec with sas ssd and 950sec with optane 900p 280g and my cpu is 10700(without K)