Author

Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 777. (Read 3426922 times)

member
Activity: 70
Merit: 10
What clocks are you running? Cheesy

I bet you can do better with some more time and of course, after Mr Buchner improves cudaminer for Maxwell cards. (looking forward to that!)

I cannot really promise that this produces actual speed benefits. There's no specific new features that a compilation for sm_50 would provide and which sm_35 (Titan kernel) doesn't already offer.

One can only hope that the compiler does some better code generation when targeting sm_50.

Christian


What about the Barrel Shifter that AMD crows so much about that Kepler was missing? This was added to Maxwell Smiley
member
Activity: 70
Merit: 10
hero member
Activity: 676
Merit: 500
i was just writing you this, cause i had this issue too. Just remove the inspector from the folder and the shortcut. Anyway this inspector is very nice program, i undervolted my 760 and now it runs 285khs at 1.075 v instead of 1.20
member
Activity: 68
Merit: 10
Is cudaminer dependent on processor performance to give better hash rates?

with -H 0 and -H 1 options the CPU can become a bottleneck. Otherwise no.
actually for scrypt and scrypt:2048 you get an extra 20khash/s (scrypt:2048) with H1 compared to H2

interesting. -H1 does 4 times more transfers over the PCI express bus than -H 2. Surprising that it's faster. I intend to get rid of all the transfers and do the hash evaluation on the GPU soon (like I do when hashing Keccak). This will allow us to run nVidia cards with 1x connections through risers without performance impact.

Christian


Maybe for users with only one card in a system? Multi-cards = bad?
legendary
Activity: 1364
Merit: 1000
Can't put the image in the chat...
there http://imgur.com/CHYtmpE

But I'm not using the most recent update... seem to be not working with the overclocking of the card
legendary
Activity: 1364
Merit: 1000
Is anybody using a GTX 465 here ? What's your hashrate ?
Thx

100-120kh, what's your?


180khash/s
Clock: 775mhz
Memory: 1603
PSU: 550w
Parameters: -H 1 -d 0 -i 0 -l F11x16 -C 2
sr. member
Activity: 476
Merit: 250
Is anybody using a GTX 465 here ? What's your hashrate ?
Thx

Having a flash video running does not change anything...

It's something else !

Not figuring out what is making me crazy

ok Gpu-z is helping ! Gpu-core clock is only at 705 Mhz ! Why the f*** is it not running full speed...

Will have to do some reading it seems !
newbie
Activity: 1
Merit: 0
Is anybody using a GTX 465 here ? What's your hashrate ?
Thx

100-120kh, what's your?
https://i.imgur.com/1NTbbxs.png?1
legendary
Activity: 1364
Merit: 1000
Is anybody using a GTX 465 here ? What's your hashrate ?
Thx
member
Activity: 106
Merit: 10
Try to open a website with flash showing, for example a youtube video.
This sounds as if your GPU doesn't clock up to 3D speed clocks while running cudaminer.
You could check that with the free program GPU-Z (if you use Windows).
legendary
Activity: 1400
Merit: 1050
Hi people !

Something very intriguing happened to me a few hours ago. I'm mining Dopecoin an alt that uses scrypt.
I'm using the 02-04-14 version with auto-tune. The most I got in all my tries was around 200kh/s with my gtx 670.

I restarted cudaminer without looking at the output and was surprised to hear the fan of my GPU working a lot faster.
I look at the cudaminer window and was shocked to see it was mining at 340mh/s !
I scrolled back to see what config auto-tune used and it was k14x28. I checked on the pool stats to see if the pool was getting the same hash/sec and it was.

Usually my gpu is at 63°C and this time it was at 85°C. I thought I was a jerk not to have noticed I could mine a lot faster earlier and decided to solo mine instead of pool mining. Setup a new .bat with -l k14x28 and got the usual 200kh/s again.

I could not get the 340mh/s again (solo or pool mining)... Tried a lot of times to have auto-tune figure it out again but it did not :/

The gpu usage was the same when mining at 340 or 200mh/s only temp changed.

Did that already happen to anyone else ? How can I get this mining speed again ??
may-be the infamous "google chrome" trick... ? The boost must have fired up for some reason
sr. member
Activity: 476
Merit: 250
Hi people !

Something very intriguing happened to me a few hours ago. I'm mining Dopecoin an alt that uses scrypt.
I'm using the 02-04-14 version with auto-tune. The most I got in all my tries was around 200kh/s with my gtx 670.

I restarted cudaminer without looking at the output and was surprised to hear the fan of my GPU working a lot faster.
I look at the cudaminer window and was shocked to see it was mining at 340mh/s !
I scrolled back to see what config auto-tune used and it was k14x28. I checked on the pool stats to see if the pool was getting the same hash/sec and it was.

Usually my gpu is at 63°C and this time it was at 85°C. I thought I was a jerk not to have noticed I could mine a lot faster earlier and decided to solo mine instead of pool mining. Setup a new .bat with -l k14x28 and got the usual 200kh/s again.

I could not get the 340mh/s again (solo or pool mining)... Tried a lot of times to have auto-tune figure it out again but it did not :/

The gpu usage was the same when mining at 340 or 200mh/s only temp changed.

Did that already happen to anyone else ? How can I get this mining speed again ??
member
Activity: 84
Merit: 10
Devs threw in the towel and there's now enough panda coin on the market to make sure mining never gets profitable again.

when enough people stop mining, you could still make some profit even with the current exchange price of ~25 Satoshis

I expect some huge volatility down the road (both up and down).

Christian


Agreed.
full member
Activity: 182
Merit: 100
I cannot really promise that this produces actual speed benefits. There's no specific new features that a compilation for sm_50 would provide and which sm_35 (Titan kernel) doesn't already offer.

One can only hope that the compiler does some better code generation when targeting sm_50.

Christian
Aha, this is where nVidia engineers could go ahead and shine then, making their compiler more efficient!
hero member
Activity: 756
Merit: 502
What clocks are you running? Cheesy

I bet you can do better with some more time and of course, after Mr Buchner improves cudaminer for Maxwell cards. (looking forward to that!)

I cannot really promise that this produces actual speed benefits. There's no specific new features that a compilation for sm_50 would provide and which sm_35 (Titan kernel) doesn't already offer.

One can only hope that the compiler does some better code generation when targeting sm_50.

Christian
hero member
Activity: 812
Merit: 1000
What clocks are you running? Cheesy

I bet you can do better with some more time and of course, after Mr Buchner improves cudaminer for Maxwell cards. (looking forward to that!)
hero member
Activity: 756
Merit: 502
Devs threw in the towel and there's now enough panda coin on the market to make sure mining never gets profitable again.

when enough people stop mining, you could still make some profit even with the current exchange price of ~25 Satoshis

I expect some huge volatility down the road (both up and down).

Christian
newbie
Activity: 52
Merit: 0
I tested with driver version 332.17, since I've heard there is reduced performance with the latest drivers.

You might want to try -C 2 for scrypt:2048, I found it helped performance quite a bit.

With -C 2 I get  "the 'T' kernel ignores the texture cache argument"


+1 C 2 for scrypt and please try higher L values. 3 and 4 and maybe higher and auto tuning would net you more than 4khs (I hope) :p

thanks!


Autotune results with different lookup values


-L 3
Scrypt-jane with -L4 [2014-02-20 06:16:09] GPU #1: 2950.30 hash/s with configuration t20x6
[2014-02-20 06:29:35] GPU #0: using launch configuration t5x16
[2014-02-20 06:29:36] GPU #0: GeForce GTX 750 Ti, 2.98 khash/s
[2014-02-20 06:29:36] Total: 2.98 khash/s
[2014-02-20 06:29:41] GPU #0: GeForce GTX 750 Ti, 3.12 khash/s
[2014-02-20 06:29:41] Total: 3.12 khash/s
[2014-02-20 06:29:46] GPU #0: GeForce GTX 750 Ti, 3.12 khash/s
[2014-02-20 06:29:46] Total: 3.12 khash/s
[2014-02-20 06:29:51] GPU #0: GeForce GTX 750 Ti, 3.12 khash/s

-L 4

[2014-02-20 06:16:09] GPU #0: 2950.30 hash/s with configuration t20x6
[2014-02-20 06:16:09] GPU #0: using launch configuration t20x6
[2014-02-20 06:16:11] GPU #0: GeForce GTX 750 Ti, 2.88 khash/s
[2014-02-20 06:16:11] Total: 2.88 khash/s
[2014-02-20 06:16:16] GPU #0: GeForce GTX 750 Ti, 2.76 khash/s
[2014-02-20 06:16:16] Total: 2.76 khash/s

-L 5
[2014-02-20 06:48:03] GPU #1: using launch configuration t20x7
[2014-02-20 06:48:05] GPU #1: GeForce GTX 750 Ti, 2.26 khash/s
[2014-02-20 06:48:05] Total: 2.26 khash/s
[2014-02-20 06:48:10] GPU #1: GeForce GTX 750 Ti, 2.38 khash/s
[2014-02-20 06:48:58] Total: 2.38 khash/s
[2014-02-20 06:49:03] GPU #1: GeForce GTX 750 Ti, 2.46 khash/s
[2014-02-20 06:49:03] Total: 2.46 khash/s

-L6 ends with cudaminer crashing and this scrolling

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaFree(h_V
[thr_id][MAXWARPS[thr_id]]-h_V_extra[thr_id][MAXWARPS[thr_id]])' (D:/Christian/D
ocuments/Visual Studio 2010/Projects/CudaMiner/salsa_kernel.cu line 875)

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaMalloc((
void **) &tmp, mem_size)' (D:/Christian/Documents/Visual Studio 2010/Projects/Cu
daMiner/salsa_kernel.cu line 251)

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaMalloc((
void **) &tmp, mem_size)' (D:/Christian/Documents/Visual Studio 2010/Projects/Cu
daMiner/salsa_kernel.cu line 252)

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaMalloc((
void **) &tmp, mem_size)' (D:/Christian/Documents/Visual Studio 2010/Projects/Cu
daMiner/salsa_kernel.cu line 253)

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaMalloc((
void **) &tmp, mem_size)' (D:/Christian/Documents/Visual Studio 2010/Projects/Cu
daMiner/salsa_kernel.cu line 254)

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaHostAllo
c((void **) &tmp, state_size, cudaHostAllocDefault)' (D:/Christian/Documents/Vis
ual Studio 2010/Projects/CudaMiner/salsa_kernel.cu line 257)

[2014-02-20 06:58:35] GPU #0: cudaError 30 (unknown error) calling 'cudaHostAllo
c((void **) &tmp, state_size, cudaHostAllocDefault)' (D:/Christian/Documents/Vis
ual Studio 2010/Projects/CudaMiner/salsa_kernel.cu line 258)

...snip...

EDIT: -L 5 test results were after I added back in my GTX 780 Ti back in the system so that it could be used for BF4 during the day since I apparently forgot to save the single card results.
Jump to: