cbuchner1
commented
10 years ago Use -i 0
----- Ursprüngliche Nachricht ----- Von: "ImmortalJ" notifications@github.com Gesendet: 04.02.2014 09:06 An: "cbuchner1/CudaMiner" CudaMiner@noreply.github.com Betreff: [CudaMiner] Fermi autotune issue in 2014-02-02 Windows release...(#87)
I'm not really sure what is going on with autotune in this release, but I did a debug output with the benchmark flag on my GTX 570: C:\Users\Kamiyama\Desktop\cudaminer-2014-02-02\x86>cudaminer.exe --benchmark -D * CudaMiner for nVidia GPUs by Christian Buchner * This is version 2014-02-02 (beta) based on pooler-cpuminer 2.3.2 (c) 2010 Jeff Garzik, 2012 pooler Cuda additions Copyright 2013,2014 Christian Buchner LTC donation address: LKS1WDKGED647msBQfLBHV3Ls8sveGncnm BTC donation address: 16hJF5mceSojnTD3ZTUDqdRhDyPJzoRakM YAC donation address: Y87sptDEcpLkLeAuex6qZioDbvy1qXZEj4 [2014-02-03 16:21:46] 1 miner threads started, using 'scrypt' algorithm. [2014-02-03 16:21:46] GPU #0: GeForce GTX 570 with compute capability 2.0 [2014-02-03 16:21:46] GPU #0: interactive: 1, tex-cache: 0 , single-alloc: 0 [2014-02-03 16:21:46] GPU #0: 32 hashes / 4.0 MB per warp. [2014-02-03 16:21:46] GPU #0: Performing auto-tuning (Patience...) [2014-02-03 16:21:46] GPU #0: maximum total warps (BxW): 260 [2014-02-03 16:21:47] x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15 x16 [2014-02-03 16:21:47] 7: 3.93| 7.86|11.39|14.69|18.36|21.00|23.76|26.35|28.00|30.27|45.63|48.87|47.74|50.58|49.41|50.48| kH/s [2014-02-03 16:21:48] 8: 4.49| 8.98|13.24|17.07|20.65|24.00|27.15|30.57|31.56|34.59|52.15|56.89|54.56|57.81|56.47|58.51| kH/s [2014-02-03 16:21:49] 9: 5.05|10.11|14.90|19.20|23.23|27.00|30.55|33.88|35.51|38.92|58.67|62.84|61.38|65.03|63.53|64.90| kH/s [2014-02-03 16:21:50] 10: 5.71|11.23|16.27|21.33|26.23|29.54|33.94|38.21|39.45|43.24|66.42|68.57|69.33|72.26|70.59|72.11| kH/s [2014-02-03 16:21:51] 11: 6.18|12.35|18.21|23.47|28.39|33.00|37.33|41.41|44.00|47.57|71.70|76.80|75.02|79.48|77.65|79.32| kH/s [2014-02-03 16:21:52] 12: 6.74|13.47|19.86|25.60|30.97|36.00|40.73|45.18|48.00|51.89|78.22|83.78|80.52|86.71|83.48|86.54| kH/s [2014-02-03 16:21:53] 13: 7.17|14.34|20.13|27.28|34.10|38.40|44.12|49.67|51.29|56.22|83.20|90.76|88.66|93.94|90.43|95.09| kH/s [2014-02-03 16:21:54] 14: 7.9| 15.4| 21.7| 26.7| 36.1| 42.0| 47.5| 53.5| 55.2| 59.7| 89.6| 96.0| 95.5|101.2| 97.4|101.0| kH/s [2014-02-03 16:21:56] 15: 8.4| 16.3| 23.2| 28.7| 32.4| 45.0| 50.9| 55.7| 59.2| 64.0| 97.8|102.9|102.3|106.7|104.3|106.7| kH/s [2014-02-03 16:21:57] 16: 8.8| 17.1| 24.8| 30.6| 34.1| 56.9| 57.8| 59.4| 48.5| 52.8|106.3|111.7|110.9|117.5|112.9|118.7| kH/s [2014-02-03 16:21:58] 17: 9.5| 17.8| 25.5| 32.0| 36.8| 58.3| 61.4| 63.1| 51.0| 55.5|112.9|118.7|117.9|122.8|121.8| | kH/s [2014-02-03 16:21:59] 18: 9.9| 18.9| 27.4| 33.9| 38.4| 62.8| 64.0| 67.8| 54.6| 58.8|119.5|125.7|124.8|130.1| | | kH/s [2014-02-03 16:21:59] 19: 10.7| 20.3| 29.0| 36.3| 40.5| 65.1| 68.6| 71.5| 57.0| 62.7|123.9|132.7|131.7| | | | kH/s [2014-02-03 16:22:00] 20: 11.2| 21.0| 30.5| 38.2| 42.7| 68.6| 72.3| 74.2| 60.0| 65.3|132.8|139.6|138.7| | | | kH/s [2014-02-03 16:22:01] 21: 11.6| 22.0| 32.5| 39.5| 67.2| 72.0| 74.7| 79.1| 63.0| 68.6|136.9|149.3| | | | | kH/s [2014-02-03 16:22:02] 22: 12.4| 23.1| 34.1| 41.4| 47.6| 75.4| 79.5| 80.5| 66.0| 72.6|143.4| | | | | | kH/s [2014-02-03 16:22:03] 23: 12.9| 23.7| 35.6| 43.3| 49.7| 78.9| 81.8| 85.3| 69.0| 75.1|149.9| | | | | | kH/s [2014-02-03 16:22:03] 24:13.24|25.18|36.57|45.18|51.20|82.29|85.33|87.77|72.00|78.37| | | | | | | kH/s [2014-02-03 16:22:04] 25:13.79|26.23|38.10|47.06|53.33|85.71|90.32|92.75|75.00|80.81| | | | | | | kH/s [2014-02-03 16:22:05] 26:14.34|26.41|39.62|48.94|81.57|89.14|92.44|96.46|78.00|84.90| | | | | | | kH/s [2014-02-03 16:22:05] 27: 14.9| 27.9| 41.1| 50.1| 57.6| 92.6| 96.0|100.2| 80.2| | | | | | | | kH/s [2014-02-03 16:22:06] 28: 15.2| 28.4| 42.7| 52.7| 87.8| 96.0| 99.6|102.4| 83.1| | | | | | | | kH/s [2014-02-03 16:22:06] 29: 15.7| 29.9| 38.1| 54.6| 61.9|101.2|103.1|104.6| | | | | | | | | kH/s [2014-02-03 16:22:07] 30: 16.3| 30.0| 39.5| 68.6| 94.1|102.9|105.0|106.7| | | | | | | | | kH/s [2014-02-03 16:22:07] GPU #0: 149925.93 hash/s with configuration F23x11 [2014-02-03 16:22:07] GPU #0: using launch configuration F23x11 [2014-02-03 16:22:07] GPU #0: GeForce GTX 570, 147.20 khash/s [2014-02-03 16:22:07] Total: 147.20 khash/s [2014-02-03 16:22:12] GPU #0: GeForce GTX 570, 147.76 khash/s [2014-02-03 16:22:12] Total: 147.76 khash/s Ctrl-C [2014-02-03 16:22:13] workio thread dead, waiting for workers... [2014-02-03 16:22:14] Total: 144.19 khash/s [2014-02-03 16:22:14] worker threads all shut down, exiting. It looks like it skipped over several different possible settings, and it's using a much slower setting than it should be. In the previous release it was testing settings from 1x1-16 all the way through 120x1. In the meantime I'm using my previous best setting of 15x16, and it seems to be working okay. But I still wanted to bring this to your attention. Thanks :) — Reply to this email directly or view it on GitHub.
ImmortalJ
I'm not really sure what is going on with autotune in this release, but I did a debug output with the benchmark flag on my GTX 570:
It looks like it skipped over several different possible settings, and it's using a much slower setting than it should be. In the previous release it was testing settings from 1x1-16 all the way through 120x1. In the meantime I'm using my previous best setting of 15x16, and it seems to be working okay. But I still wanted to bring this to your attention.
Thanks :)