[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 224.

cayars

full member

Activity: 168

Merit: 100

Update on next release of nvMiner. I'll sum up the rest of this message with one one: FRUSTRATED

I've been working pretty much for the last 24 hours on this.

I have djm34's x17 working. I have tsiv's latest CryptoNight working.

HOWEVER (yea you knew there was a but/however)

I've also been testing/trying/implementing some additional code changes for hash increases. Some are based directly on sp_ and wolf0's mods and some are based on ideas they put forth that I ran with.

Here's the rub. I've increased speed a bit or many algos. Pretty much all of Chris/djm34's are done. Nothing earth shattering by any means but every bit of hash helps long term. tsiv's CryptoNight is another story.

Using pure tsiv code I'm getting slower 750ti hash rates and maybe 10 to 15% faster on kepler. I have another test build of tsiv's code with changes that is getting me 20-25% faster kepler speeds and ever so slightly faster 750ti speeds.

But the rub is that I haven't been able to migrate these changes into nvMiner. It's driving me crazy. I'm doing compares between code and can't find why one EXE runs 20 to 25% faster then the other on kepler.

I now have 3 machines setup doing compiles to figure this out. Main problem is that a full rebuild is around 4 hours. It's a process of elimination and testing. I'll figure it out as I always do but it's time consuming.

So anyway, I didn't want you guys to think I've forgotten about you. I'm working on it as fast as I can but compile times on 5.5 are SLOW regardless of hardware thrown at it.

Carlo

DiggerPhil

newbie

Activity: 18

Merit: 0

Quote from: polanskiman on July 24, 2014, 08:39:35 PM

Quote from: Strannik-74 on July 24, 2014, 07:34:34 PM

The main GPU is not working at 100%, what can be done?

You are clearly using that GPU while mining so the hash-rate drops. A stated above, use the onboard graphics.

As a side note, I am getting lower hash-rates and up to this day no one has been able to help me. Basically I am getting ~300Kh less on every card.

Are you overclocking? I get similar results with stock cards (assuming x11).

cayars

full member

Activity: 168

Merit: 100

Quote from: polanskiman on July 24, 2014, 08:39:35 PM

Quote from: Strannik-74 on July 24, 2014, 07:34:34 PM

The main GPU is not working at 100%, what can be done?

You are clearly using that GPU while mining so the hash-rate drops. A stated above, use the onboard graphics.

As a side note, I am getting lower hash-rates and up to this day no one has been able to help me. Basically I am getting ~300Kh less on every card.

If you wouldn't mind setting up TeamViewer on your side I can try and take a look to see if I can find any problems or identify your slower hashing.

Just PM if you want this help.

Carlo

polanskiman

full member

Activity: 266

Merit: 100

Quote from: Strannik-74 on July 24, 2014, 07:34:34 PM

The main GPU is not working at 100%, what can be done?

You are clearly using that GPU while mining so the hash-rate drops. A stated above, use the onboard graphics.

As a side note, I am getting lower hash-rates and up to this day no one has been able to help me. Basically I am getting ~300Kh less on every card.

bathrobehero

legendary

Activity: 2002

Merit: 1051

ICO? Not even once.

Quote from: Strannik-74 on July 24, 2014, 07:34:34 PM

The main GPU is not working at 100%, what can be done?

Use onboard graphics (as primary, for your monitor) instead of your 750Ti if you want it to work like the rest.

Strannik-74

full member

Activity: 348

Merit: 102

The main GPU is not working at 100%, what can be done?

tc61

hero member

Activity: 494

Merit: 500

wow is anything worth mining anymore??? Seems like a new scam coin is born everyday for a pump and dump...no one holds anymore...

tarzanbigcity

sr. member

Activity: 602

Merit: 250

I just read that Vertcoin is moving to the Lyra2 Proof of work implementation. Who wants to get to work on that? Wink

https://vertcoin.org/pow-algorithm-upgrade-lyra2/

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Good improvements signals but there is a bether way. In the optimal solution, the n threads will always be busy.
Each thread will never read or write to shared memory exept the results (n hashes)

Need to experiment a bit. Probobly have to rewrite some code.

Sigals

member

Activity: 76

Merit: 10

Quote from: sp_ on July 24, 2014, 02:34:53 AM

Quote from: ?? on ??

I'm pretty sure the output of the last algorithm is used as the input of the next one for X11, precisely so you can't do that.

Yes you can. If each thread is working on a different hash.

example
4 threads 4 hashes

HASH1: x1->x2->x3->
HASH2: x4->x5->x6->
HASH3: x7->x8->x9->
HASH4: x10->x11

Swap the 4 hashes

HASH4: x1->x2->x3->
HASH1: x4->x5->x6->
HASH2: x7->x8->x9->
HASH3: x10->x11

Swap the 4 hashes

HASH3: x1->x2->x3->
HASH4: x4->x5->x6->
HASH1: x7->x8->x9->
HASH2: x10->x11

Swap the 4 hashes

HASH2: x1->x2->x3->
HASH3: x4->x5->x6->
HASH4: x7->x8->x9->
HASH1: x10->x11

Complete

Hello, great idea sp_, but shouldn't your models be more like:

Code:

START MINING :
TASK 1:
thread1_hash_id_1: x1->x2->x3->
TASK 2:
thread1_hash_id_2: x1->x2->x3->
thread2_hash_id_1: x4->x5->x6->
TASK 3:
thread1_hash_id_3: x1->x2->x3->
thread2_hash_id_2: x4->x5->x6->
thread3_hash_id_1: x7->x8->x9->
TASK 4:
thread1_hash_id_4: x1->x2->x3->
thread2_hash_id_3: x4->x5->x6->
thread3_hash_id_2: x7->x8->x9->
thread4_hash_id_1: x10->x11->SUBMIT_hash_id_1
TASK 5:
thread1_hash_id_5: x1->x2->x3->
thread2_hash_id_4: x4->x5->x6->
thread3_hash_id_3: x7->x8->x9->
thread4_hash_id_2: x10->x11->SUBMIT_hash_id_2
TASK 6:
thread1_hash_id_6: x1->x2->x3->
thread2_hash_id_5: x4->x5->x6->
thread3_hash_id_4: x7->x8->x9->
thread4_hash_id_3: x10->x11->SUBMIT_hash_id_3
TASK 7:
thread1_hash_id_7: x1->x2->x3->
thread2_hash_id_6: x4->x5->x6->
thread3_hash_id_5: x7->x8->x9->
thread4_hash_id_4: x10->x11->SUBMIT_hash_id_4
TASK 8:
thread1_hash_id_8: x1->x2->x3->
thread2_hash_id_7: x4->x5->x6->
thread3_hash_id_6: x7->x8->x9->
thread4_hash_id_5: x10->x11->SUBMIT_hash_id_5
TASK 9:
thread1_hash_id_9: x1->x2->x3->
thread2_hash_id_8: x4->x5->x6->
thread3_hash_id_7: x7->x8->x9->
thread4_hash_id_6: x10->x11->SUBMIT_hash_id_6
TASK 10: ----- STRATUM DETECTED A NEW BLOCK ---- // REMOVE STALE SHARE (can also force submit for pool like Waffle doe)
thread1_hash_id_10: x1->x2->x3->
thread2_hash_id_7: REMOVE
thread3_hash_id_8: REMOVE
thread4_hash_id_9: REMOVE
TASK 11:
thread1_hash_id_11: x1->x2->x3->
thread2_hash_id_10: x4->x5->x6->
TASK 12:
thread1_hash_id_12: x1->x2->x3->
thread2_hash_id_11: x4->x5->x6->
thread3_hash_id_10: x7->x8->x9->
TASK 13:
thread1_hash_id_13: x1->x2->x3->
thread2_hash_id_12: x4->x5->x6->
thread3_hash_id_11: x7->x8->x9->
thread4_hash_id_10: x10->x11->SUBMIT_hash_id_10
___ ETC ___

isn't it better to submit when each single hash is complete to avoid potential stale shares due to waiting to complete each hash ?

also, we should make sure that each thread take the same time, so better to deal with the slowest hash with only one other hash, for example:

Code:

TASK 4:
thread1_hash_id_4: x1->x2 - 7.5mhs
thread2_hash_id_3: x3->x5->x6-> - 7.5mhs
thread3_hash_id_2: x7->x8->x9-> - 7.5mhs
thread4_hash_id_1: x10->x11->SUBMIT_hash_id_1 - 7.5mhs

unitedminers

full member

Activity: 241

Merit: 100

Quote from: S_tring on July 24, 2014, 08:51:51 AM

Code:

export PATH=$PATH:/usr/local/cuda/bin

Thank you! Didn't think someone answers so quick and copied the backup image to my SSD.

But I'm not sure if it would have changed something because my .bashrc file looks already like this at the end.

Code:

# enable programmable completion features (you don't need to enable
# this, if it's already enabled in /etc/bash.bashrc and /etc/profile
# sources /etc/bash.bashrc).
if ! shopt -oq posix; then
  if [ -f /usr/share/bash-completion/bash_completion ]; then
    . /usr/share/bash-completion/bash_completion
  elif [ -f /etc/bash_completion ]; then
    . /etc/bash_completion
  fi
fi

export LD_LIBRARY_PATH=/usr/local/cuda/lib

export PATH=$PATH:/usr/local/cuda/bin

What do you think?

S_tring

full member

Activity: 252

Merit: 102

OPEN Platform - Powering Blockchain Acceptance

Quote from: unitedminers on July 24, 2014, 08:39:37 AM

Quote from: th00ber on July 22, 2014, 04:47:35 PM

you can use the linux nvidia tools to get the infos
/usr/bin/nvidia-smi --query-gpu=index,temperature.gpu,fan.speed --format=csv,noheader,nounits

Thank you

but I can already read the GPU (Gainward GTX 750 Ti GS) info from the Nvidia Settings GUI.

The problem is, I'm running the nvidia-331 / nvidia-340 driver but the "Fan Info" shows for "Speed (RPM)" unsupported, "Speed (%)" 30 and more (=works) but below there should be a bar where I could set the fan speed in percent, so I guess it doesn't show up because of "Speed (RPM)" unsupported. If I let everything at default and start ccMiner with nvidia-331 the heat rises only and the watts for a PC with one SSD and one 750 Ti jumps after a few minutes from about 95-105 watts to about 140 watts.

BTW if I run the nvidia-340 driver I get the following error. Any hints?

Code:

./ccminer: error while loading shared libraries: libcudart.so.6.0: cannot open shared object file: No such file or directory

Code:

export PATH=$PATH:/usr/local/cuda/bin

unitedminers

full member

Activity: 241

Merit: 100

Quote from: th00ber on July 22, 2014, 04:47:35 PM

you can use the linux nvidia tools to get the infos
/usr/bin/nvidia-smi --query-gpu=index,temperature.gpu,fan.speed --format=csv,noheader,nounits

Thank you

but I can already read the GPU (Gainward GTX 750 Ti GS) info from the Nvidia Settings GUI.

The problem is, I'm running the nvidia-331 / nvidia-340 driver but the "Fan Info" shows for "Speed (RPM)" unsupported, "Speed (%)" 30 and more (=works) but below there should be a bar where I could set the fan speed in percent, so I guess it doesn't show up because of "Speed (RPM)" unsupported. If I let everything at default and start ccMiner with nvidia-331 the heat rises only and the watts for a PC with one SSD and one 750 Ti jumps after a few minutes from about 95-105 watts to about 140 watts.

Here the output of nvidia-smi:

Code:

user@miner:~$ nvidia-smi
Thu Jul 24 17:05:46 2014       
+------------------------------------------------------+                       
| NVIDIA-SMI 331.89     Driver Version: 331.89         |                       
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 750 Ti  Off  | 0000:01:00.0     N/A |                  N/A |
| 31%   39C  N/A     N/A /  N/A |     55MiB /  2047MiB |     N/A      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Compute processes:                                               GPU Memory |
|  GPU       PID  Process name                                     Usage      |
|=============================================================================|
|    0            Not Supported                                               |
+-----------------------------------------------------------------------------+

Update 1: I thought I have found a solution but didn't work.

nvidia-settings -a [gpu:0]/GPUFanControlState=1
nvidia-settings -a [fan:0]/GPUCurrentFanSpeed=80

Code:

user@miner:~$ nvidia-settings -a "[fan:0]/GPUCurrentFanSpeed=80"


ERROR: Error assigning value 80 to attribute 'GPUCurrentFanSpeed'
       (miner:0[fan:0]) as specified in assignment
       '[fan:0]/GPUCurrentFanSpeed=80' (Unknown Error).

Update 2: WORKS

Code:

http://www.gpugrid.net/forum_thread.php?id=2925

Fan control in Linux, the easy way:

Open a Terminal and type,

    cd /etc/X11
    sudo nvidia-xconfig --cool-bits=4


This will take effect after a restart and open NVIDIA Server Settings / Thermal Settings

BTW if I run the nvidia-340 driver I get the following error. Any hints?

Code:

./ccminer: error while loading shared libraries: libcudart.so.6.0: cannot open shared object file: No such file or directory

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

I come from the Demoscene. My school is old school

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: Amph on July 24, 2014, 05:06:22 AM

someone tried to build a rig with 7+ vga?

Quote from: yellowduck2 on July 24, 2014, 04:18:02 AM

Quote from: sp_ on July 24, 2014, 03:38:46 AM

I haven't looked at TSIV's code. Isn't Cryptonite just a variation of x11 + scryptn? 20% gain is a good job. Now do another 20%

Anyway, I will start implementing some code soon. I will start with the 11 x'es. One by One.

Do u mind me asking if u have a degree / master / phd in computer science ? U spot something that no one here understand at first. I see u have to explain in so many post before people gets it.

Are you related to Satoshi ?

his nick remind me of sp h miner, perhaps retired coder from amd lol

actually sph(lib) has nothing to do with amd... (it is rather an optimized code for cpu...)

yellowduck2

hero member

Activity: 868

Merit: 1000

Quote from: Amph on July 24, 2014, 05:06:22 AM

someone tried to build a rig with 7+ vga?

Quote from: yellowduck2 on July 24, 2014, 04:18:02 AM

Quote from: sp_ on July 24, 2014, 03:38:46 AM

I haven't looked at TSIV's code. Isn't Cryptonite just a variation of x11 + scryptn? 20% gain is a good job. Now do another 20%

Anyway, I will start implementing some code soon. I will start with the 11 x'es. One by One.

Do u mind me asking if u have a degree / master / phd in computer science ? U spot something that no one here understand at first. I see u have to explain in so many post before people gets it.

Are you related to Satoshi ?

his nick remind me of sp h miner, perhaps retired coder from amd lol

or christ friend trolling us

hmmmmm

Amph

legendary

Activity: 3248

Merit: 1070

someone tried to build a rig with 7+ vga?

Quote from: yellowduck2 on July 24, 2014, 04:18:02 AM

Quote from: sp_ on July 24, 2014, 03:38:46 AM

I haven't looked at TSIV's code. Isn't Cryptonite just a variation of x11 + scryptn? 20% gain is a good job. Now do another 20%

Anyway, I will start implementing some code soon. I will start with the 11 x'es. One by One.

Do u mind me asking if u have a degree / master / phd in computer science ? U spot something that no one here understand at first. I see u have to explain in so many post before people gets it.

Are you related to Satoshi ?

his nick remind me of sp h miner, perhaps retired coder from amd lol

or christ friend trolling us

yellowduck2

hero member

Activity: 868

Merit: 1000

Quote from: sp_ on July 24, 2014, 04:44:21 AM

Quote from: yellowduck2 on July 24, 2014, 04:18:02 AM

Do u mind me asking if u have a degree / master / phd in computer science ? U spot something that no one here understand at first. I see u have to explain in so many post before people gets it.
Are you related to Satoshi ?

Ask your boss.

yes boss.

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Quote from: yellowduck2 on July 24, 2014, 04:18:02 AM

Do u mind me asking if u have a degree / master / phd in computer science ? U spot something that no one here understand at first. I see u have to explain in so many post before people gets it.
Are you related to Satoshi ?

Ask your boss.

yellowduck2

hero member

Activity: 868

Merit: 1000

Quote from: sp_ on July 24, 2014, 03:38:46 AM

I haven't looked at TSIV's code. Isn't Cryptonite just a variation of x11 + scryptn? 20% gain is a good job. Now do another 20%

Anyway, I will start implementing some code soon. I will start with the 11 x'es. One by One.

Do u mind me asking if u have a degree / master / phd in computer science ? U spot something that no one here understand at first. I see u have to explain in so many post before people gets it.

Are you related to Satoshi ?

Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 224. (Read 3426936 times)