Stephen Brooks 2008-10-20 13:03:05 | I just recompiled Muon1 because I found out something about HTTP proxies that should smooth over a few beginners' problems with it not uploading when you're behind a proxy. The EXE got bigger by 30KB which made me wonder what exactly changed, I don't think any debugging or weird code got added, maybe my libraries just got bigger.. |
Stephen Brooks 2008-10-20 22:59:00 | I've heard that it's a lot slower than the -d version so don't go upgrading all at once. I'm going to have a look at diffs and see what's been changed. I left the previous version up for this sort of reason. |
[TA]Assimilator1 2008-11-09 10:47:03 | On the subject of changes, would you be able to make the new client flush more regularly? I think some people with even just slightly slower PCs are getting dis-heartened by the client apparently doing 'nothing' for several days. TA has certainly lost quite a few crunchers since that client was released (whichever 1 that was?). |
waffleironhead 2008-11-12 16:58:36 | It's been a long time since I ran dpad(about a year?). I am getting a file saved over and over again in my muon1 v444e folder. It is named "Simulation_scorerampdistance_XXXXXXX.log" where the xxxxxxxx is different for each file IE 0013F243. Is this something new that is saved to the disk or is this something to be concerned about? |
Stephen Brooks 2008-11-12 17:31:27 | That's harmless, just some debug code I hadn't turned off (I've commented it out in the development code now). |
[TA]Assimilator1 2008-11-15 11:14:46 | What sort of slow down are you seeing with version e? |
[TN]opyrt 2008-11-29 12:35:41 | Stephen: Could you please write on the download section of the main page that users should not download version 4.44e unless they're behind a proxy? A friend of mine downloaded it becaus it was the newest, and when I told him to download the old one, he got a 20% increase in performance. |
shauge 2008-12-01 21:47:23 | I experienced 66% increase after switching to the 4.44d from the 4.44e version and had it not been for another team member posting his muonbench results on our team web page, I would probably still only "help the project" with much less than I do now. So why is the 4.44e still the recommended version on the front page, so long since the "a lot slower" post was posted? |
[DPC]white_panther 2008-12-02 01:18:50 | oke here mine bench first the D version Uptime (secs),Mpts in file,Estimate kpts/sec 8296,13139887.6,0.00 11019,13141139.9,459.84 14047,13142718.7,492.24 18597,13144396.4,437.68 23452,13146037.3,405.74 28302,13147710.1,390.99 32542,13149385.5,391.73 54656,13157719.8,384.64 57985,13158857.9,381.78 61616,13160532.7,387.19 65855,13162164.1,387.02 87964,13170498.4,384.23 108859,13178829.1,387.23 113401,13180473.3,386.14 116727,13181458.0,383.38 117029,13181485.1,382.56 121572,13183148.7,381.91 123992,13184002.9,381.30 124596,13184091.7,380.09 126105,13184283.1,376.84 130040,13185814.1,377.24 134582,13187494.6,376.98 138216,13189071.7,378.57 139124,13189382.2,378.32 139425,13189442.5,377.91 143056,13191043.0,379.60 163644,13199407.6,383.14 163946,13199483.9,382.89 168489,13201160.0,382.49 173032,13202797.1,381.88 and now the E version Uptime (secs),Mpts in file,Estimate kpts/sec 8870,13136163.7,0.00 10984,13136681.3,244.94 12796,13136943.4,198.61 16430,13138185.0,267.37 21277,13139861.8,298.09 26122,13141536.2,311.42 30953,13143023.8,310.66 57219,13151356.2,314.23 59935,13152026.0,310.63 86503,13160344.3,311.47 90729,13161432.6,308.69 95862,13163109.5,309.75 100993,13164717.5,309.95 101295,13164749.5,309.29 125744,13173080.6,315.87 129969,13174361.2,315.42 135101,13176045.3,315.94 139327,13177086.0,313.69 142949,13178150.1,313.15 145664,13178616.1,310.34 151098,13180243.2,309.92 177363,13188575.6,311.06 177665,13188625.8,310.80 so a big differens |
Stephen Brooks 2008-12-02 12:03:11 | Yes, removed it for now. There might be an 'f' version when I work out what is making it slow (most people seemed OK with the manual proxy configuration anyway). |
shauge 2008-12-02 17:18:50 | Just a guess, but you might check if some code is compiled in debug mode. Compiler optimizations is typically disabled when compiled in debug mode. |
Zerberus 2008-12-11 21:41:15 | Will the color but be fixed too? In Vista, 256 color mode is unreliable and the palette shifts all the time. It looks very funny . This is caused by the Aero desktop. |
Zerberus 2008-12-11 21:43:05 | Ah yes, and the -txttobin bug, of course... |
Stephen Brooks 2008-12-15 16:00:15 | I've made a note of the 16M colour mode suggestion. I can probably do this. By the way, the reason not much has happened around here for months is that I've got 98 pages of my thesis written so far, probably more than half-way through now, but I can't really allow myself to get distracted off it at all, or it'll never get done. |
Zerberus 2008-12-15 23:14:50 | No problem, real life has always priority. |
[TA]Assimilator1 2008-12-19 18:16:55 | Fair enough Stephen, when will you of finished it by? |
Stephen Brooks 2008-12-20 02:03:30 | In a couple of months, can't be more specific though. I'm really slow at writing up. |
[TA]Assimilator1 2008-12-20 07:10:40 | lol |
[TA]Assimilator1 2008-12-23 00:20:19 | [DPC]white_panther What spec PC was that on btw? |
[DPC]white_panther 2008-12-23 01:23:56 | see [url]http://tweakers.net/gallery/20176/sys[/url] running 2 at the same time so 2 cores for version D and 2 for version E |
[TA]Assimilator1 2008-12-23 13:32:58 | Ah a Q6600, nice ,not overclocked? Btw I take it you set each client to just 2 threads? (not that it should make any difference). |
[DPC]white_panther 2009-01-02 22:02:43 | Q6600 running now @ 3384 (376x9) at the time of banchmark running @ 3348 (372x9) still with the stock cooler each client set to 2 threads in the config.txt |