Topic: DiabloMiner GPU Miner - page 65. (Read 866596 times)

Oh, and last time I noticed, they didn't do JNI yet, so you can't run my miner with it.

I call bullshit: http://gcc.gnu.org/java/

I really like Diablo Miner but there is few problems with it:

1. When I run the new Diablo miner versions with .exe instead of .bat in it, it does not work. I get black console screen for maybe 25ms and it exits. Not enough time to even hit Pause button to see what's wrong. I need tu use .bat file instead.

2. No speed improvements on new version with BFI_INT. I even get speed decrease. I get 260 Mh/s with 2011-04-23 version and with 2011-05-19 I get 250 Mh/s

I use ATI HD5850 with Catalyst 10.11 and ATI SDK 2.1

Java does not work that way.

is there any way I can use the miner without installing java?
can you put it in a warper and compile the whole thing including the java?

Could you do me a favor and test with Diablo's most recent version too.. .I dunno why but I get varying results with my card here at work.  His latest release is outperforming mine by a longshot here at work, but on my NVidia card at home my version outperforms his -shrug-

-w 64 on nvidia really cuts down on the desktop interactivity abuse (and may even increase speed), however this is due to nvidia driver bugs.

As for -z, I'm surprised -z helps anywhere. Nvidia driver design must be extremely bad. -z is usually slower on AMD.

Just do a

Since classpath can be a pain, and Maven is quite good at handling dependencies just tell mvn to build the cp string:
and it will print the classpath.

Edit: intended as general tip, not just for this case

Hey just a quick note on how much "Power" you need to finish one share from one block. This is only an "average" on how long it takes to solve one. I did a quick summary that I run at 345MHases/s it took me 57s to do 5 shares, will give you 3933MHashes/Share or on an avg 1Share/11.4s. This is just ball parking small numbers..

In your case.. CPU is very slow compared to an ATI card.. same goes for that GeForce.. Looking at https://en.bitcoin.it/wiki/Mining_hardware_comparison

I'd say your system puts about maybe for fun 10MHashes/s that would make one share that's 3933MHashes / 10 = 393.3s or 6min 33sec for one share. In that time my system would have computed about 34,5 shares aka 34,5 times faster..

Is it worth it?? I donno.. the system just jumped up another difficulty bump yesterday, from 152274 to 244139. So it will just take longer for full blocks to finish now. That means it take more shares on avg find the solution == less bitcoins for us all..

Cheers!

As far as OSX having poor OpenCL implementation, that sure seems to be the case.  DiabloMiner running in OSX gives me about 80 ~ 90 MHash / sec and switching over to Windows to run DiabloMiner gives me 185 ~ 205 MHash on the same gear.  So it is certainly better to be running Windows.

However, your improvements will be marginal at best since you're running NVidia gear... if you really, really want to give it a try, you should definitely install Windows or Linux for maximum performance.

I have a Mac Pro (single quadcore 2.66 Xeon 3500 series "Bloomfield", 6gb ram, Geforce GT120, Snow Leopard) that I am using DiabloMiner on. I've got experience using pyrit (tool for cracking WPA wireless keys) on it, utilising the GPU to speed things up, so I thought I'd give this a go.

I have the latest OSX Nvidia CUDA drivers installed (4.0 RC) and I confirmed that pyrit was still using the GPU cores so hopefully everything's all ok. I downloaded DiabloMiner, got it working with the public pool of my choice (Eligius) and I'm watching the output...

...ugh.

First of all, I'm behind a firewall at work, so despite the fact that it can connect to the public pool, I'm getting errors saying that it can't connect to Bitcoin. Is this a problem? A google search suggests it's not a major problem.

Secondly, my slow results are really surprising me. Reading the display is a little difficult, I have two khash numbers so I'm not sure which to look at. Is one GPU and one CPU? And how come sometimes one will drop to 0 for ages and then start ticking again. But right now it says 1429. A lot of the time it's under 1000. Sometimes it jumps quite high (3, 4, even 5000). I also turned on debug and I can see a ghash value of 0.5. What is that?

Lastly, I tried changing -f to 1000 on a whim. Seemed to speed things up a touch. But not sure if it's a good idea?

I'm under no illusions that this is going to be a quick process but the output is worrying me that it's not actually working. How can I tell?

I've heard that OSX is shit for this because of shitty OpenCL implementation. That being the case, would I get better results if I stuck XP on the MacPro and used that?

Could you do me a favor and test with Diablo's most recent version too.. .I dunno why but I get varying results with my card here at work.  His latest release is outperforming mine by a longshot here at work, but on my NVidia card at home my version outperforms his -shrug-

Also, I've noticed that lowering that work unit to 64 and raising z to some multiple of 2 gives me much better performance.  Seems like it fills the shader processors better that way (but what do I know, really? )

I´m running this: start /DC:\blablah\Diablo-D3-DiabloMiner-b975ca7\Diablo-D3-DiabloMiner-b975ca7 DiabloMiner-Windows.exe -u myuser -p mypw -o deepbit.net -r 8332 -f1, and tried all -v and all -w values.
I can´t reach nothing near my 430 Mhash/sec with my 5870 on the phatk, what am i doing wrong?

Hey Diablo, I'm glad to hear it

I do however, have a bit of an issue... I'm not familiar at all with Maven, so I don't know how to set up a profiler for use with it, and google seems to be no help 

Google provides a profiler that works great for me at work...  I'm in the business of creating interactive ebooks built on top of a game engine for Android, so I use Google's Java Profiler to see which processes are taking the longest time to execute.

If you know how to set up a profiler with Maven I would very much like to know how to do so, after all it's MUCH MUCH MUCH better to macro-optimize the slowest pieces of code than to sprawl through the code line by line making minimal micro-optimizations like I did.

If you could help me to set up a profiler with Maven then I would love to continue doing what I can to help you with your project