bathrobehero
Legendary
Offline
Activity: 2002
Merit: 1051
ICO? Not even once.
|
|
January 23, 2014, 12:10:02 AM |
|
Great stuff!
Once again though, the x86 compile is slightly faster for scrypt than the x64 version. (231 vs 228 on stock and 260 vs 255 khashes with OC)
|
Not your keys, not your coins!
|
|
|
ktf
Newbie
Offline
Activity: 24
Merit: 0
|
|
January 23, 2014, 12:15:02 AM |
|
Can anyone share the command line used to run cudaminer for the latest version ? I have two GTX 660 and I want to make sure I run it properly. I compiled the latest version from github.
|
|
|
|
Ultimist
|
|
January 23, 2014, 12:24:25 AM |
|
On the newer github builds, is the "-L" switch for scrypt-jane only? It doesn't seem to work with scrypt. Any time I try to use it with scrypt, it gets to the part where it lists the # of hashes & MB per warp, but doesn't seem to reach the autotuning stage and just sits there, telling me about new blocks.
By the way, on all the latest builds in win7x64 I have to use autotuning every time, which is a pain due to the wait being longer. Specifying the -l parameter for my gtx 670 crashes cudaminer every time.
|
|
|
|
cbuchner1 (OP)
|
|
January 23, 2014, 12:25:28 AM |
|
Can anyone share the command line used to run cudaminer for the latest version ? I have two GTX 660 and I want to make sure I run it properly. I compiled the latest version from github.
that's probably an -l Y5x32 then, possibly with an added -C 1
|
|
|
|
cbuchner1 (OP)
|
|
January 23, 2014, 12:38:19 AM |
|
On the newer github builds, is the "-L" switch for scrypt-jane only? It doesn't seem to work with scrypt. Any time I try to use it with scrypt, it gets to the part where it lists the # of hashes & MB per warp, but doesn't seem to reach the autotuning stage and just sits there, telling me about new blocks.
By the way, on all the latest builds in win7x64 I have to use autotuning every time, which is a pain due to the wait being longer. Specifying the -l parameter for my gtx 670 crashes cudaminer every time.
one of the memory allocation routines probably runs into an endless loop. Try playing with -m 0 or -m 1 to see if it goes away. lookup-gap is most likely not helping with scrypt and it is currently not implemented in these kernells: L Y Z the issue with the -l parameter not working is unknown to me. I use it all the time, also on Windows. Hmm..
|
|
|
|
Ultimist
|
|
January 23, 2014, 01:16:07 AM |
|
Following suit, then, is it safe to assume that the -b parameter is also for scrypt-jane only?
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 01:37:35 AM |
|
How does a noob get a current build with the Z kernel?
|
|
|
|
Morgahl
Member
Offline
Activity: 70
Merit: 10
|
|
January 23, 2014, 01:44:38 AM |
|
1) Use a 64 bit Linux, then you should be able to use more than 4 GB at once. I have had someone autotune scrypt-jane (N=32768) with -L 1 and -L 2 on a Tesla M2090 recently He sent me his results via Skype, e.g. with -L 2: GPU #0: maximum total warps (BxW): 324 1.83 khash/s with configuration X71x2
You might check to see what s/he achieves with -L 3 and -L4, the Titan continues to see gains for me until -L 4, might be worth checking if they still have access to the Tesla. 2) Vista, Windows 7, Windows 8 tend to have huge issues with large memory allocations on the card (WDDM driver model restrictions).
Noted, I had hoped this may not be the case I will look into building for Linux. Question on this. Would you see any disadvantage to a USB bootable Distro for testing? 3) I don't have access to such hardware myself, so my testing opportunities are zero, essentially.
No worries here, like I said, curiosity and speculation. I'm pretty satisfied with the improvements despite OS & Driver limitations you have no control over. Still seeing nearly a 45% increase from Commit 92 to Commit 111 Why is cudaminer crashing when I try to mine off of middlecoin.com ?
I mine there and it isn't crashing. The FAQ for middlecoin seems to indicate that Stales that are submitted count toward proof of work. Does Cudaminer submit stales currently?
|
|
|
|
Morgahl
Member
Offline
Activity: 70
Merit: 10
|
|
January 23, 2014, 01:54:33 AM |
|
Following suit, then, is it safe to assume that the -b parameter is also for scrypt-jane only?
You are correct. How does a noob get a current build with the Z kernel?
Christian has said that there won't be official binaries released until he is comfortable with the feature implementations he is working on. Currently you'll need to wait until some kind soul has the free time to help us out. A few have done recent builds, with Patoberli providing the most recent at commit 111, unfortunate the Z kernel is commit 112 so it might be sometime until another kind soul provides a build.
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 01:58:07 AM |
|
Following suit, then, is it safe to assume that the -b parameter is also for scrypt-jane only?
You are correct. How does a noob get a current build with the Z kernel?
Christian has said that there won't be official binaries released until he is comfortable with the feature implementations he is working on. Currently you'll need to wait until some kind soul has the free time to help us out. A few have done recent builds, with Patoberli providing the most recent at commit 111, unfortunate the Z kernel is commit 112 so it might be sometime until another kind soul provides a build. I'm installing VS2008 as we speak, we'll see how far I get
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 02:15:23 AM |
|
Following suit, then, is it safe to assume that the -b parameter is also for scrypt-jane only?
You are correct. How does a noob get a current build with the Z kernel?
Christian has said that there won't be official binaries released until he is comfortable with the feature implementations he is working on. Currently you'll need to wait until some kind soul has the free time to help us out. A few have done recent builds, with Patoberli providing the most recent at commit 111, unfortunate the Z kernel is commit 112 so it might be sometime until another kind soul provides a build. I'm installing VS2008 as we speak, we'll see how far I get Ok not very far. I only have VS2008 and I don't think I have the environment set up properly. I'm not a very good developer Would someone be willing to compile 112 so I can run it through some testing on a GTX 780 and a TITAN?
|
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 02:47:10 AM |
|
If anyone is good with visual studio and could help me set up to compile, I can compile future releases upon request. I'll keep plugging at this, but I'm getting: d:\users\eric\documents\github\cudaminer\compat\getopt\getopt_long.c(61) : fatal error C1083: Cannot open include file: 'getopt.h': No such file or directory dump.c d:\users\eric\documents\github\cudaminer\compat\jansson\dump.c(14) : fatal error C1083: Cannot open include file: 'jansson.h': No such file or directory hashtable.c d:\users\eric\documents\github\cudaminer\compat\jansson\hashtable.c( : fatal error C1083: Cannot open include file: 'config.h': No such file or directory load.c d:\users\eric\documents\github\cudaminer\compat\jansson\load.c(18) : fatal error C1083: Cannot open include file: 'jansson.h': No such file or directory utf.c d:\users\eric\documents\github\cudaminer\compat\jansson\utf.h(11) : fatal error C1083: Cannot open include file: 'config.h': No such file or directory value.c d:\users\eric\documents\github\cudaminer\compat\jansson\value.c(10) : fatal error C1083: Cannot open include file: 'config.h': No such file or directory cpu-miner.c d:\users\eric\documents\github\cudaminer\cpu-miner.c(17) : fatal error C1083: Cannot open include file: 'stdbool.h': No such file or directory sha2.c d:\users\eric\documents\github\cudaminer\miner.h(10) : fatal error C1083: Cannot open include file: 'stdbool.h': No such file or directory util.c d:\users\eric\documents\github\cudaminer\util.c(19) : fatal error C1083: Cannot open include file: 'stdbool.h': No such file or directory
|
|
|
|
justafool76
Member
Offline
Activity: 85
Merit: 10
|
|
January 23, 2014, 02:59:37 AM |
|
ok guys I am back for round 2 (been up for the last 12 hrs straight and still and not able to get this right)
OK here is some information that I have from my cuda miner I am using the one pre v111
cudaminer-2013-12-18\cudaminer-2013-12-18\x64>cu daminer.exe -H 1 -d 0 -i 0 -l K1x33 -C 1 -m 1 -o stratum+tcp://asia.middlecoin.c om:3333 -u 1MU4EAB6p5xcRPhZ8gFKZSq9znchJpt2iE -p 123
the above info is what I am using in my bat file
[2014-01-23 12:50:28] 1 miner threads started, using 'scrypt' algorithm. [2014-01-23 12:50:28] Starting Stratum on stratum+tcp://asia.middlecoin.com:3333
[2014-01-23 12:50:29] Stratum detected new block [2014-01-23 12:50:29] GPU #0: GeForce GT 750M with compute capability 3.0 [2014-01-23 12:50:29] GPU #0: the 'K' kernel ignores the texture cache argument [2014-01-23 12:50:29] GPU #0: interactive: 0, tex-cache: 0 , single-alloc: 1 [2014-01-23 12:50:29] GPU #0: using launch configuration K1x33 [2014-01-23 12:50:29] GPU #0: GeForce GT 750M, 4224 hashes, 28.68 khash/s [2014-01-23 12:50:29] GPU #0: GeForce GT 750M result does not validate on CPU (i =503, s=1)! [2014-01-23 12:50:31] GPU #0: GeForce GT 750M, 1721280 hashes, 1096 khash/s [2014-01-23 12:50:54] Stratum detected new block [2014-01-23 12:50:54] GPU #0: GeForce GT 750M, 26224704 hashes, 1098 khash/s [2014-01-23 12:51:24] GPU #0: GeForce GT 750M result does not validate on CPU (i =921, s=0)!
as you can see I can get the hash rate upto about 1000kh/z but my cpu is going to 100%
is there anyway that I can keep my hashrate but lower my cpu to a normal number please. as I don't wont to blow up my cpu
I have tried the same in cuda miner v111 and I get the following
<cudaminer_commit_111_x64>cudaminer.exe -H 1 -d 0 -i 0 -l K1x33 -C 1 -m 1 -o stratum+tcp://asia.middlecoin.com:3333 -u 1MU4EAB6p5 xcRPhZ8gFKZSq9znchJpt2iE -p 123 [2014-01-23 13:09:22] 1 miner threads started, using 'scrypt' algorithm. [2014-01-23 13:09:22] Starting Stratum on stratum+tcp://asia.middlecoin.com:3333
[2014-01-23 13:09:23] Stratum detected new block [2014-01-23 13:09:24] GPU #0: GeForce GT 750M with compute capability 3.0 [2014-01-23 13:09:24] GPU #0: interactive: 0, tex-cache: 1D, single-alloc: 1 [2014-01-23 13:09:24] GPU #0: 8 hashes / 1.0 MB per warp. [2014-01-23 13:09:24] GPU #0: Given launch config 'K1x33' exceeds warp limit for 'K' kernel. [2014-01-23 13:09:24] GPU #0: using launch configuration K1x33 [2014-01-23 13:09:24] GPU #0: GeForce GT 750M, 100.50 khash/s [2014-01-23 13:09:34] Stratum detected new block [2014-01-23 13:09:34] GPU #0: GeForce GT 750M, 120.84 khash/s [2014-01-23 13:09:42] Stratum detected new block [2014-01-23 13:09:42] GPU #0: GeForce GT 750M, 119.42 khash/s [2014-01-23 13:09:53] Stratum detected new block [2014-01-23 13:09:53] GPU #0: GeForce GT 750M, 118.92 khash/s [2014-01-23 13:09:54] Stratum detected new block [2014-01-23 13:09:54] GPU #0: GeForce GT 750M, 117.96 khash/s [2014-01-23 13:10:00] Stratum detected new block [2014-01-23 13:10:00] GPU #0: GeForce GT 750M, 118.47 khash/s [2014-01-23 13:10:00] Stratum detected new block [2014-01-23 13:10:00] GPU #0: GeForce GT 750M, 116.77 khash/s [2014-01-23 13:11:03] GPU #0: GeForce GT 750M, 111.36 khash/s [2014-01-23 13:11:50] Stratum detected new block [2014-01-23 13:11:50] GPU #0: GeForce GT 750M, 108.96 khash/s [2014-01-23 13:12:12] Stratum detected new block [2014-01-23 13:12:12] GPU #0: GeForce GT 750M, 108.79 khash/s [2014-01-23 13:13:12] GPU #0: GeForce GT 750M, 108.41 khash/s [2014-01-23 13:13:26] Stratum detected new block [2014-01-23 13:13:26] GPU #0: GeForce GT 750M, 108.50 khash/s [2014-01-23 13:13:27] Stratum detected new block [2014-01-23 13:13:27] GPU #0: GeForce GT 750M, 107.84 khash/s [2014-01-23 13:13:41] Stratum detected new block [2014-01-23 13:13:41] GPU #0: GeForce GT 750M, 108.18 khash/s [2014-01-23 13:13:49] Stratum detected new block [2014-01-23 13:13:49] GPU #0: GeForce GT 750M, 108.56 khash/s [2014-01-23 13:13:59] Stratum detected new block [2014-01-23 13:13:59] GPU #0: GeForce GT 750M, 108.63 khash/s [2014-01-23 13:14:02] Stratum detected new block [2014-01-23 13:14:02] GPU #0: GeForce GT 750M, 108.63 khash/s [2014-01-23 13:14:47] Stratum detected new block [2014-01-23 13:14:47] GPU #0: GeForce GT 750M, 108.59 khash/s [2014-01-23 13:14:52] Stratum detected new block [2014-01-23 13:14:52] GPU #0: GeForce GT 750M, 108.60 khash/s [2014-01-23 13:14:53] Stratum detected new block [2014-01-23 13:14:53] GPU #0: GeForce GT 750M, 108.27 khash/s [2014-01-23 13:14:56] Stratum detected new block [2014-01-23 13:14:56] GPU #0: GeForce GT 750M, 108.33 khash/s [2014-01-23 13:15:32] Stratum detected new block [2014-01-23 13:15:32] GPU #0: GeForce GT 750M, 107.66 khash/s [2014-01-23 13:15:45] Stratum detected new block [2014-01-23 13:15:45] GPU #0: GeForce GT 750M, 107.79 khash/s [2014-01-23 13:15:46] Stratum detected new block [2014-01-23 13:15:46] GPU #0: GeForce GT 750M, 107.39 khash/s [2014-01-23 13:16:03] Stratum detected new block [2014-01-23 13:16:03] GPU #0: GeForce GT 750M, 107.79 khash/s [2014-01-23 13:16:18] Stratum detected new block [2014-01-23 13:16:18] GPU #0: GeForce GT 750M, 107.84 khash/s [2014-01-23 13:17:18] GPU #0: GeForce GT 750M, 107.63 khash/s [2014-01-23 13:17:22] Stratum detected new block [2014-01-23 13:17:22] GPU #0: GeForce GT 750M, 107.75 khash/s [2014-01-23 13:17:30] Stratum detected new block [2014-01-23 13:17:30] GPU #0: GeForce GT 750M, 106.96 khash/s [2014-01-23 13:17:32] Stratum detected new block [2014-01-23 13:17:32] GPU #0: GeForce GT 750M, 107.00 khash/s [2014-01-23 13:17:38] Stratum detected new block [2014-01-23 13:17:38] GPU #0: GeForce GT 750M, 107.31 khash/s [2014-01-23 13:18:03] Stratum detected new block [2014-01-23 13:18:03] GPU #0: GeForce GT 750M, 107.17 khash/s [2014-01-23 13:18:05] Stratum detected new block [2014-01-23 13:18:05] GPU #0: GeForce GT 750M, 107.76 khash/s [2014-01-23 13:18:13] GPU #0: GeForce GT 750M result does not validate on CPU (i =164, s=1)! [2014-01-23 13:19:05] GPU #0: GeForce GT 750M, 107.88 khash/s [2014-01-23 13:19:59] Stratum detected new block [2014-01-23 13:19:59] GPU #0: GeForce GT 750M, 107.79 khash/s [2014-01-23 13:20:01] Stratum detected new block [2014-01-23 13:20:01] GPU #0: GeForce GT 750M, 107.72 khash/s [2014-01-23 13:20:46] Stratum detected new block [2014-01-23 13:20:46] GPU #0: GeForce GT 750M, 107.92 khash/s
as you can see I get the cpu thing again
please note that I am using a asus laptop with a gt750m 4gb gpu
I also have a asus laptop g7vw that has a gtx 670m 3gb gpu but I have not mess with the bat file on that one as of yet.
I will be changing my bat file back to the old settings until I can get some help can any one help me please
below is this
k1x23 141.3 kh/s for my gt750m 4gb card
f14x8 114.87 kh/s for my gtx 670m 3gb card
above 2 line is what the auto thingy did on both of my cards before I starting playing around please can someone please oh please please find it in there heart to help me
I don't know how to right code I don't know how to compile stuff and I really need some help from someone here please. to get this sought before I throw meself and my wheelchair off the balcony. (so to say that but I am desperate.)
plzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 03:07:08 AM |
|
I think you're having trouble because of the -l setting, try to stick closer to what the autotune says to start off with.
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 03:25:52 AM |
|
Can someone send me stdint.h from their VS2010 installation?
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 03:38:35 AM |
|
d:\users\eric\documents\github\cudaminer\cpu-miner.c(37) : fatal error C1083: Cannot open include file: 'curl/curl.h': No such file or directory sha2.c d:\users\eric\documents\github\cudaminer\compat\sys\time.h(2) : error C2059: syntax error : 'string' d:\users\eric\documents\github\cudaminer\compat\sys\time.h(3) : error C2059: syntax error : 'string' d:\users\eric\documents\github\cudaminer\miner.h(13) : fatal error C1083: Cannot open include file: 'pthread.h': No such file or directory util.c d:\users\eric\documents\github\cudaminer\util.c(23) : fatal error C1083: Cannot open include file: 'curl/curl.h': No such file or directory Maddening. I'm a hacker at best, so I'm trying to just plug holes in a dam it seems. These files all exist inside the project and are included, and in the place they are expected to be
|
|
|
|
orrett3
Newbie
Offline
Activity: 33
Merit: 0
|
|
January 23, 2014, 03:50:28 AM |
|
The Dependencies have to be in a directory one up from the source folder.
|
|
|
|
cwizard
Member
Offline
Activity: 98
Merit: 10
|
|
January 23, 2014, 03:55:53 AM |
|
The Dependencies have to be in a directory one up from the source folder.
so if the source is $$user$$/documents/github/cudaminer the dependencies have to be in ../cudaminer ? (/docs/github) I grabbed VS2010 express and I'm not making any better headway
|
|
|
|
tuaris
|
|
January 23, 2014, 04:06:24 AM |
|
The Dependencies have to be in a directory one up from the source folder.
so if the source is $$user$$/documents/github/cudaminer the dependencies have to be in ../cudaminer ? (/docs/github) I grabbed VS2010 express and I'm not making any better headway I have VS2010 Ultimate sitting on my system. I'm downloading SP1 for it, I'll try building when that's done and let you know if I have better luck.
|
|
|
|
|