Pages:
Author

Topic: Ufasoft Miner - Windows/Linux, x86/x64, SSE2/OpenCL, Open Source - page 54. (Read 631117 times)

newbie
Activity: 34
Merit: 0
I tested it with the same pool. I'd suggest making the time somewhat shorter and/or having the algorithm only check as many hashes as it can do in a certain time similar to jgarzik's miner. If you could post the rest of the code, I would be much obliged. I'll start looking at how it might be ported to linux, since given the 50% performance improvement using your kernel, I'd like to get that being used on all my machines as soon as possible.
sr. member
Activity: 404
Merit: 251
So, how about the rest of the source code? I'd like to build this for linux to see how it runs there.

I know that jgarzik's will only ocmpute a certain number of hashes before returning the "I couldn't find anything" result to be sure that the result is returned in a reasonable amount of time. Does this employ the same tactic?

Source code is not secret, but it uses WinInet API, and cannot be directly recompiled.

Tactic is following:
Now "getwork period" == 30 sec is hardcoded, and each thread takes portions by "4*1024*1024 nonces". After trying aver 3 millions nonces thread chexk timeout (30 sec). and takes another portion of nonces or request new "GETWORK".

Probably 30 sec is very long period and is the reason of "Not accepted" results.
I have tested it with Pool http://mining.bitcoin.cz/.

newbie
Activity: 34
Merit: 0
So, how about the rest of the source code? I'd like to build this for linux to see how it runs there.

EDIT: Seems to give a large number of NOT ACCEPTED results when using this with slush's pool. I know that jgarzik's will only ocmpute a certain number of hashes before returning the "I couldn't find anything" result to be sure that the result is returned in a reasonable amount of time. Does this employ the same tactic?
sr. member
Activity: 493
Merit: 250
IDENA.IO - Proof-Of-Person Blockchain
sr. member
Activity: 404
Merit: 251
Performance: 2.3 MHash/s per core on 2.3 GHz CPU [1000 CPU cycles per hash]

What architecture? It will be excellent for Intel Core, Core2 or Core i3-7 and old Athlons K8 but not impressive for Athlons K10.

Core 2 Quad
hero member
Activity: 696
Merit: 500
bitjet

rpcminer-cpu.exe is usual CPU miner. More interesting what are your results with rpcminer-4way.exe ?
It uses SSE2 too (as bitcoin-miner.exe).



Im not exactly sure how to use the 4 way.
full member
Activity: 238
Merit: 100
Performance: 2.3 MHash/s per core on 2.3 GHz CPU [1000 CPU cycles per hash]

What architecture? It will be excellent for Intel Core, Core2 or Core i3-7 and old Athlons K8 but not impressive for Athlons K10.
sr. member
Activity: 404
Merit: 251
bitjet

rpcminer-cpu.exe is usual CPU miner. More interesting what are your results with rpcminer-4way.exe ?
It uses SSE2 too (as bitcoin-miner.exe).

hero member
Activity: 696
Merit: 500
ok with the RCPMINER-CPU.EXE miner I was showing about 4500khash/s

With this app I am showing 10.8-11mhash/s

Is this DOUBLE what I was getting with the RCP or am I just mathematically challenged??
sr. member
Activity: 404
Merit: 251
Do you use some custom compilation of "rpcminer-4way.exe"?

Precompiled binaries from http://www.mediafire.com/file/e39ygb9hbiesb09/bitcoin-rpcminer-20110213-win32bin.zip
on my Core Quad 2.33 shows with 4 miner threads:
2462 khash/s


It's very strange
sr. member
Activity: 493
Merit: 250
IDENA.IO - Proof-Of-Person Blockchain
hi ufasoft i have tested you version of miners in my xp 64, my result are 11.3 M/hash with 5 threads of my phenon 1055t at 3.1 ghz per core, are poor compared with these results = 14.4 M/hash with the rpcminer-4way of puddinpop. Thanks for this work and continue working i u have future versions i continue testing =)
sr. member
Activity: 404
Merit: 251
Pages:
Jump to: