Bitcoin Forum
July 21, 2018, 06:11:33 PM *
News: Latest stable version of Bitcoin Core: 0.16.1  [Torrent]. (New!)
 
   Home   Help Search Donate Login Register  
Pages: « 1 ... 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 [112] 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 ... 1148 »
  Print  
Author Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX]  (Read 3395682 times)
cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 08:54:54 AM
 #2221

Hi-ya Christian,

My 660Ti registers at ~270 khash/s peak (scrypt) for a single instance of cudaminer (which appears to start a single mining thread) - By starting two instances, I'd have expected that rate to halve.  Instead, each instance reports ~200khash/s peak.

This is surprising. What is your launch configuration?
What does GPU-z show for GPU utilization when running just a single instance?
How's GPU utilization and memory usage with 1 and 2 instances?
1532196693
Hero Member
*
Offline Offline

Posts: 1532196693

View Profile Personal Message (Offline)

Ignore
1532196693
Reply with quote  #2

1532196693
Report to moderator
1532196693
Hero Member
*
Offline Offline

Posts: 1532196693

View Profile Personal Message (Offline)

Ignore
1532196693
Reply with quote  #2

1532196693
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1532196693
Hero Member
*
Offline Offline

Posts: 1532196693

View Profile Personal Message (Offline)

Ignore
1532196693
Reply with quote  #2

1532196693
Report to moderator
1532196693
Hero Member
*
Offline Offline

Posts: 1532196693

View Profile Personal Message (Offline)

Ignore
1532196693
Reply with quote  #2

1532196693
Report to moderator
cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 08:58:58 AM
 #2222

How do you set the difficulty factor manually with cudaminer like you can with cgminer?

I ask because I am solo mining directly against my wallet app and the difficulty dropped significantly today for the coin i was on, but yet it still took nearly twice as long as it should have to find a block.  But I have no way of knowing at what difficulty cudaminer is running at.

with -D it prints the stratum difficulty, whenever it changes.  I am not aware of a print feature for getwork. When solo mining it should ask for new work from the server like every 5 seconds. Wouldn't that always include a difficulty number?

Christian
CaptainBeck
Full Member
***
Offline Offline

Activity: 168
Merit: 100


View Profile
January 09, 2014, 09:51:11 AM
 #2223

Quote

holy crap...you just gave me a link to the best most sarcastic site ever.
I WILL use this daily...

PS. All this talk about scrypt-jane is making my windows machines jealous...

Why windows machines???

I've got jane running on my windows machine. My 660ti is about 2.5khs and i still running the rest of my atis on scrypt because they hate jane.

cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 10:29:48 AM
 #2224

I've got jane running on my windows machine. My 660ti is about 2.5khs and i still running the rest of my atis on scrypt because they hate jane.

On Linux I've been getting 3.2 kHash on a 660Ti and I am heading for 3.6 kH once I get the -C 2 option going again.

I am running  K7x3 -i 0 -m 1 and strangely this setting is not liked much by Windows (much slower than e.g. K4x4)
CaptainBeck
Full Member
***
Offline Offline

Activity: 168
Merit: 100


View Profile
January 09, 2014, 10:38:54 AM
 #2225

I've got jane running on my windows machine. My 660ti is about 2.5khs and i still running the rest of my atis on scrypt because they hate jane.

On Linux I've been getting 3.2 kHash on a 660Ti and I am heading for 3.6 kH once I get the -C 2 option going again.

I am running  K7x3 -i 0 -m 1 and strangely this setting is not liked much by Windows (much slower than e.g. K4x4)

Anything more than K 13 for me claims it requires to much memory and gives invalid cpu.

So K13x1 is the best for me it seems.
Ultimist
Full Member
***
Offline Offline

Activity: 350
Merit: 102


"The future of advertisement intelligence"


View Profile
January 09, 2014, 12:12:57 PM
 #2226

Something is definitely not working right for me with the 12/18 version posted in the OP.

On my auto-tuned GTX 670 sometimes in Nvidia Inspector, GPU usage is shown as 25 to 35% only, and the hashrate is about 85 Kh/s. Other times GPU usage goes to 95% and above. I don't understand why it isn't fully using the GPU. This version is very buggy and never seems to work the same way twice. The autotune never comes up with the same value twice, either, even if run 2 minutes after it was run before. Is it just picking things at random?

▄█▀▀▀▀▀ ● ◯ SaTT - The Revolution of Advertising Transactions ◯ ● ▀▀▀▀▀█▄
███    Website | Ann Thread | Whitepaper | Facebook | Twitter | Telegram    ███
▀█▄▄▄▄▄▄▄▄   PRE-SALE COMPLETED IN LESS THAN 24 HOURS!   ▄▄▄▄▄▄▄▄█▀
patoberli
Member
**
Offline Offline

Activity: 106
Merit: 10


View Profile
January 09, 2014, 12:45:45 PM
 #2227

You can try a to add --benchmark -D to see the results it's getting (leave the -l parameter away). Sometimes the results are very close and thus it takes a different one.

YAC: YA86YiWSvWEGSSSerPTMy4kwndabRUNftf
BTC: 16NqvkYbKMnonVEf7jHbuWURFsLeuTRidX
LTC: LTKCoiDwqEjaRCoNXfFhDm9EeWbGWouZjE
bathrobehero
Legendary
*
Offline Offline

Activity: 1540
Merit: 1026


ICO? Not even once.


View Profile
January 09, 2014, 12:53:56 PM
 #2228

You can try a to add --benchmark -D to see the results it's getting (leave the -l parameter away). Sometimes the results are very close and thus it takes a different one.

That, and also background apps stressing the card (even just a little bit) can affect the results and it's also worht noting that overclocking seem to confuse autotune fairly often as well.

I think K7x32 should be best for your card.

RIP Bittrex
RIP Poloniex
bigjme
Sr. Member
****
Offline Offline

Activity: 350
Merit: 250


View Profile
January 09, 2014, 02:09:50 PM
 #2229

Ok then so my linux system is mining now. Using the old settings of 16x1 I am getting 3.1khash/s and my cpu is hashing at 0.64khash/s

And one thing to note. No driver crashes in linux! So I may be able to get a higher hash rate then I am now

Owner of: cudamining.co.uk
eduncan911
Member
**
Offline Offline

Activity: 101
Merit: 10

Miner / Engineer


View Profile WWW
January 09, 2014, 02:18:39 PM
 #2230

How do you set the difficulty factor manually with cudaminer like you can with cgminer?

I ask because I am solo mining directly against my wallet app and the difficulty dropped significantly today for the coin i was on, but yet it still took nearly twice as long as it should have to find a block.  But I have no way of knowing at what difficulty cudaminer is running at.

with -D it prints the stratum difficulty, whenever it changes.  I am not aware of a print feature for getwork. When solo mining it should ask for new work from the server like every 5 seconds. Wouldn't that always include a difficulty number?

Christian


Excellent answer, thanks!  I will use -D from now on to see the changes.

As for requesting every 5 seconds, that sounds perfect.  But again, I'd want to see these changes.  -D sounds like the way to go.

BTC: 131Zt92zoA7XUfkLhm1p2FwSP3tAxE43vf
cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 02:37:10 PM
 #2231

Something is definitely not working right for me with the 12/18 version posted in the OP.

On my auto-tuned GTX 670 sometimes in Nvidia Inspector, GPU usage is shown as 25 to 35% only, and the hashrate is about 85 Kh/s. Other times GPU usage goes to 95% and above. I don't understand why it isn't fully using the GPU. This version is very buggy and never seems to work the same way twice. The autotune never comes up with the same value twice, either, even if run 2 minutes after it was run before. Is it just picking things at random?


Hmm, too bad that a this software beta version has bugs.  Here, have your money back. I award you 0 LTC.

The thing about autotune is that mid range and high end Kepler GPUs dynamically adjust clock rates "as they see fit" to meet thermal and power requirements. And hence there is a certain randomness to the autotuning.

There are Windows machines on which we cannot get 100% GPU utilization. This happened e.g. on a machine on which I installed Windows Server 2012 R2 for evaluation purposes. It would never quite go above 80% GPU use.

Christian
bigjme
Sr. Member
****
Offline Offline

Activity: 350
Merit: 250


View Profile
January 09, 2014, 02:53:46 PM
 #2232

cbuchner1 thi is my 780 running at over 3.7khash/s!
it is just a straight screen shot not cropped sorry

http://s29.postimg.org/62ttjbizb/Screenshot_from_2014_01_09_14_49_51.png

./cudaminer --algo=scrypt-jane -H 0 -i 0 -d 0 -l T20x1 -o http://127.0.0.1:3339 -u user -p pass

gpu memory usage is 2.86GB
so my gpu and cpu now mine together at over 4.4khash/s!!!!

Owner of: cudamining.co.uk
aliens
Newbie
*
Offline Offline

Activity: 15
Merit: 0


View Profile
January 09, 2014, 03:02:21 PM
 #2233

Can't seem to compile latest git clone. I run ./autogen.sh (which doesn't output anything), ./configure then make, and I get the following errors:

Code:
nvcc -g -O2 -Xptxas "-abi=no -v" -arch=compute_10 --maxrregcount=64 --ptxas-options=-v -I./compat/jansson -o salsa_kernel.o -c salsa_kernel.cu
salsa_kernel.cu(479): error: too few arguments in function call

salsa_kernel.cu(742): error: more than one instance of overloaded function "cuda_scrypt_core" has "C" linkage

salsa_kernel.cu(760): error: too few arguments in function call

3 errors detected in the compilation of "/tmp/tmpxft_00004701_00000000-6_salsa_kernel.cpp1.ii".

Any ideas on fix?

BTC Donations: 1H2SnDm2VSsQRbEhYgY4xjgCqYGxVTQbLw
braytz
Hero Member
*****
Offline Offline

Activity: 490
Merit: 500



View Profile
January 09, 2014, 03:06:40 PM
 #2234

so my gpu and cpu now mine together at over 4.4khash/s!!!!

are you planning a 51% attack?  Grin
bigjme
Sr. Member
****
Offline Offline

Activity: 350
Merit: 250


View Profile
January 09, 2014, 03:08:37 PM
 #2235

I dont even know what a 51% attack is haha

Owner of: cudamining.co.uk
cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 03:30:17 PM
 #2236

cbuchner1 thi is my 780 running at over 3.7khash/s!
it is just a straight screen shot not cropped sorry

Not bad, but half of that 780's compute capability is sitting unused. A GTX 660Ti or GTX 760 could would probably achieve similar hash rates...

I started to build a cheap (800 Euro) 12 kHash/s machine. Wish me luck.

NOTE: there are GTX 760 and GTX 660 (non Ti) models with 4 Gigs of RAM. These can then run 31 warps simultaneously, however their compute capability is only 3.0.

For scrypt-jane, ideally per SMX you want to run 8 warps (internally this number is multiplied by 4 because 4 threads cooperate to compute 1 hash).

GTX 670Ti: 7 SMX  (not enough memory for optimal config K7x8)
GTX 760: 6 SMX     (not enough memory for optimal config K6x8)
GTX 660: 4 SMX (optimal config probably K4x8, but because of some CUDA memory overhead you have to use K31x1)
GTX 440: 2 SMX (optimal config: K2x8)

Christian


bigjme
Sr. Member
****
Offline Offline

Activity: 350
Merit: 250


View Profile
January 09, 2014, 03:39:05 PM
 #2237

hmm, that sucks but its better then my rubbish 2.77khash/s in windows.
so literally on this all that matters is how much memory it has, not too much computing power. and Linux clearly makes a huge difference to performance

good luck! my gpu cost almost as much as that entire build so will be good to see how it runs. is it 3 or 4 660's your running?

Owner of: cudamining.co.uk
patoberli
Member
**
Offline Offline

Activity: 106
Merit: 10


View Profile
January 09, 2014, 03:39:53 PM
 #2238

Just wanted to recompile newest version but received this error (downloaded up until commit 81):
Code:
 C:\Users\Patrick\Desktop\cudaminer\cudaminer>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\bin\nvcc.exe" -gencode=arch=compute_10,code=\"sm_10,compute_10\" --use-local-env --cl-version 2012 -ccbin "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\bin"  -I. -Icompat -Icompat\jansson -Icompat\getopt -I"..\pthreads\Pre-built.2\include" -I"..\curl-7.29.0\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include" --opencc-options -LIST:source=on   --keep --keep-dir Release -maxrregcount=64 --ptxas-options=-v --machine 32 --compile -cudart static     -DWIN32 -DNDEBUG -D_CONSOLE -D_CRT_SECURE_NO_WARNINGS -DCURL_STATICLIB -DSCRYPT_KECCAK512 -DSCRYPT_CHACHA -DSCRYPT_CHOOSE_COMPILETIME -D_MBCS -Xcompiler "/EHsc /W3 /nologo /O2 /Zi  /MD  " -o Release\salsa_kernel.cu.obj "C:\Users\Patrick\Desktop\cudaminer\cudaminer\salsa_kernel.cu"
[color=red][b]C:/Users/Patrick/Desktop/cudaminer/cudaminer/salsa_kernel.cu(798): error : identifier "opt_benchmark" is undefined[/b][/color]
C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V110\BuildCustomizations\CUDA 5.5.targets(592,9): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\bin\nvcc.exe" -gencode=arch=compute_10,code=\"sm_10,compute_10\" --use-local-env --cl-version 2012 -ccbin "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\bin"  -I. -Icompat -Icompat\jansson -Icompat\getopt -I"..\pthreads\Pre-built.2\include" -I"..\curl-7.29.0\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v5.5\include" --opencc-options -LIST:source=on   --keep --keep-dir Release -maxrregcount=64 --ptxas-options=-v --machine 32 --compile -cudart static     -DWIN32 -DNDEBUG -D_CONSOLE -D_CRT_SECURE_NO_WARNINGS -DCURL_STATICLIB -DSCRYPT_KECCAK512 -DSCRYPT_CHACHA -DSCRYPT_CHOOSE_COMPILETIME -D_MBCS -Xcompiler "/EHsc /W3 /nologo /O2 /Zi  /MD  " -o Release\salsa_kernel.cu.obj "C:\Users\Patrick\Desktop\cudaminer\cudaminer\salsa_kernel.cu"" exited with code 2.
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========
I guess the second line is the problem?

YAC: YA86YiWSvWEGSSSerPTMy4kwndabRUNftf
BTC: 16NqvkYbKMnonVEf7jHbuWURFsLeuTRidX
LTC: LTKCoiDwqEjaRCoNXfFhDm9EeWbGWouZjE
cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 03:50:01 PM
 #2239

good luck! my gpu cost almost as much as that entire build so will be good to see how it runs. is it 3 or 4 660's your running?

4 x GTX 640  (4x90 Euro)
1 x GTX 660  (150 Euro)

20 GB of Video RAM total, 4 GB of system memory. And a puny AMD Sempron 145 single core as CPU.

The mainboard has 5 PCI-x x16 slots, however with single slot spacing. And I don't want to buy risers.

Mainboard, SSD boot disk, RAM, CPU and power supply and case total at around 300 Euros.
cbuchner1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
January 09, 2014, 03:53:21 PM
 #2240

Just wanted to recompile newest version but received this error (downloaded up until commit 81):

I pushed a fix for that...

NOTE:

1) autotune was improved (with scrypt-jane it can use up to 4 GB memory now on 32 bit systems, 12 GB on 64 bit systems - untested)
2) the -C 2 option now works with Kepler kernels in scrypt-jane mode...
3) automatic upgrade from -C 1 to -C 2 if more than 2 GB of RAM are used (the -C 1 has an upper limit of 2 GB)
Pages: « 1 ... 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 [112] 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 ... 1148 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!