[ANN] [QRK] Quark | Core 0.10 upgrade

[bitspill]

Finally put my code up on github.

https://github.com/uncle-bob/quarkcoin-cpuminer

I'm looking for suggestions or ideas. Groestl-Hash seem to be the deal breaker right now. It's taking up as much time as all other 5 hashes combined. Any speed up for Groestl would be a win. Currently the V1 release confuses the compiler slightly less and is a smidgen faster. However I think this code is currently the best option for speeding up beyond that. It's still around twice as fast as the default miner.

Added bonus, almost everything for quark now fits a single function. While it won't fit on a GPU until the code size is reduced and memory is shrunk, it's just yet. Consider this my last update for cpuminer.

I wrote this cause I'm greedy, and so I leave my tipjar.
BTC: 1DaSUSqcr13G4QJYamsecW6yHh95QTT39R
QRK: QQRNpg7aGHcvgEFvXEk1DoCtLydboVsAzf

I cannot get it to compile. I'm on Ubuntu 13.04 and I get stuck at:

"checking for pthread_create in -lpthread... yes
./configure: line 5534: syntax error near unexpected token `,'
./configure: line 5534: `LIBCURL_CHECK_CONFIG(, 7.15.2, ,'"

Any help would be appreciated.

I had the same error when compiling quark itself it is due to curl not being properly installed
A fix can be found here https://bitcointalk.org/index.php?topic=4823.0

Edit: Or install libcurl-dev

[1715013466]

1715013466

[1715013466]

1715013466

[1715013466]

1715013466

[mrmanager]

Finally put my code up on github.

https://github.com/uncle-bob/quarkcoin-cpuminer

I'm looking for suggestions or ideas. Groestl-Hash seem to be the deal breaker right now. It's taking up as much time as all other 5 hashes combined. Any speed up for Groestl would be a win. Currently the V1 release confuses the compiler slightly less and is a smidgen faster. However I think this code is currently the best option for speeding up beyond that. It's still around twice as fast as the default miner.

Added bonus, almost everything for quark now fits a single function. While it won't fit on a GPU until the code size is reduced and memory is shrunk, it's just yet. Consider this my last update for cpuminer.

I wrote this cause I'm greedy, and so I leave my tipjar.
BTC: 1DaSUSqcr13G4QJYamsecW6yHh95QTT39R
QRK: QQRNpg7aGHcvgEFvXEk1DoCtLydboVsAzf

I cannot get it to compile. I'm on Ubuntu 13.04 and I get stuck at:

"checking for pthread_create in -lpthread... yes
./configure: line 5534: syntax error near unexpected token `,'
./configure: line 5534: `LIBCURL_CHECK_CONFIG(, 7.15.2, ,'"

Any help would be appreciated.

I had the same error when compiling quark itself it is due to curl not being properly installed
A fix can be found here https://bitcointalk.org/index.php?topic=4823.0

Edit: Or install libcurl-dev

I'm surprised I didn't find this post. I had libcurl-dev installed, but it wasn't working. This has fixed it. Is there a guide anywhere to build Windows binaries? The readme isn't helpful at all.

[cryptohunter]

hold on a min, are people saying the uncle bob miner is  85% faster than the http://stonefoz.myfastmail.com/cpuminer-quark.zip one?

are there any detailed install instructions for linux or follow the same install guide for the uncle bob one as the stonefoz one?

[stonefoz]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github.
Compared to the original miner though, it is about 85% faster, uses SSE2 along with ASM for two of the rounds.

[stonefoz]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github

I can upload a binary as well if there's interest.

That would be great, I would like to see it tested on something not AMD first. There is a "makeprofile.sh" script. It now has enough warnings that I worry it may have issues if not tested. Looks good right now, but the goal is fast, not necessarily good. Anyway, if the middle two lines match anything else the script prints out, at least two hashes where done correctly....


Oh, and there is a new option.
./configure CFLAGS="-msse2 -O3 -DCHEAT"
needs testing but the idea is that it's faster to just try a new nonce that run groestl twice. Anytime that section is found to be needed ran, it just give up. Half of the hashes end up being garbage. But, it now does more than twice the speed, for a net gain. EXPERIMENTAL, need to be ran to see if actually improves payouts as it may be losing shares. Still runs that section when required, but never twice.

[altsay]

So how to mine solo with this new miner?

[glongsword]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github

I can upload a binary as well if there's interest.

That would be great, I would like to see it tested on something not AMD first. There is a "makeprofile.sh" script. It now has enough warnings that I worry it may have issues if not tested. Looks good right now, but the goal is fast, not necessarily good. Anyway, if the middle two lines match anything else the script prints out, at least two hashes where done correctly....


Oh, and there is a new option.
./configure CFLAGS="-msse2 -O3 -DCHEAT"
needs testing but the idea is that it's faster to just try a new nonce that run groestl twice. Anytime that section is found to be needed ran, it just give up. Half of the hashes end up being garbage. But, it now does more than twice the speed, for a net gain. EXPERIMENTAL, need to be ran to see if actually improves payouts as it may be losing shares. Still runs that section when required, but never twice.

cloned the git repo, and got an error message:

user@computer:~/quarkcoin-cpuminer# ./autogen.sh
configure.ac:117: `PTHREAD_FLAGS' includes configure substitution `@PTHREAD_FLAGS@'
configure.ac:117: and is referred to from `minerd_SOURCES';
configure.ac:117: configure substitutions are not allowed in _SOURCES variables

Thoughts?

Thanks!

[trinsic]

Any idea what's going on with this?

Code:

root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 10000
error: {"code":-4,"message":"Transaction too large"}
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 5000
error: {"code":-4,"message":"Transaction too large"}
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 2000
error: {"code":-4,"message":"Transaction too large"}
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 1000
20bfb1a939064973f7e8d08dee352d34f757b625b3e8bac14da98e457675f675
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 1000
54b405e8d532842a7ef1386d3cdeea972ec7ffd4fac965ca89308e6865f395dd
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 1000
error: {"code":-4,"message":"Transaction too large"}

[shazbits]

Logging-in as root to do crypto shenanigans? 

Might need to settxfee 0.01 or more based on how many inputs you get on the wallet.

[trinsic]

It's in a VM, I'm sure it's fine

Thanks I'll try that.

[ig0tik3d]

cloned the git repo, and got an error message:

user@computer:~/quarkcoin-cpuminer# ./autogen.sh
configure.ac:117: `PTHREAD_FLAGS' includes configure substitution `@PTHREAD_FLAGS@'
configure.ac:117: and is referred to from `minerd_SOURCES';
configure.ac:117: configure substitutions are not allowed in _SOURCES variables

Thoughts?

Thanks!

in Makefile.am line 20
change

Code:

quark.c \

to

Code:

quark.c

enjoy))

[Majormax]

Can anyone share their experience on yield and profitability (relative to BTC) over time ?

Is this even relevant to miners at present ?  With the rapid decease in block reward, it would not seem too prudent to sell at present.

As Coinchoose is at present unable to list Coins with a hash that is not Scrypt or SHA256 , that limits the exposure of the coin.


QRK forum is starting to gain some membership : please go and post at : http://forum.quarkcoin.org/index.php

[cryptohunter]

I think this coin has been one of the most stable in price and hash since i have been watching alt coins. I wouldn't sell any of mine for fiat until the reward has halved a few more times and these are starting to get quite scarce. I have traded some for primes although actually i wish i had not now since prime has sunk since i bought them and this has continued to rise.

[eule]

Any idea what's going on with this?

Code:

root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 10000
error: {"code":-4,"message":"Transaction too large"}
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 5000
error: {"code":-4,"message":"Transaction too large"}
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 2000
error: {"code":-4,"message":"Transaction too large"}
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 1000
20bfb1a939064973f7e8d08dee352d34f757b625b3e8bac14da98e457675f675
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 1000
54b405e8d532842a7ef1386d3cdeea972ec7ffd4fac965ca89308e6865f395dd
root@quarky:~/quark/src# ./quarkcoind sendtoaddress QWghDKW59cFJTitJg4zBp7BnjUWM3H4wcF 1000
error: {"code":-4,"message":"Transaction too large"}

Did you recieve many small p2pool transactions?

[Spoetnik]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github.
Compared to the original miner though, it is about 85% faster, uses SSE2 along with ASM for two of the rounds.

uhhhhhhhhhhh 85% faster on AMD Bulldozer CPU's only ?
am i wasting my time ripping my hair out trying to compile this newer version (on windows / for windows) for no speed gain on my Intel CPU ?

[ig0tik3d]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github.
Compared to the original miner though, it is about 85% faster, uses SSE2 along with ASM for two of the rounds.

uhhhhhhhhhhh 85% faster on AMD Bulldozer CPU's only ?
am i wasting my time ripping my hair out trying to compile this newer version (on windows / for windows) for no speed gain on my Intel CPU ?

https://bitcointalk.org/index.php?topic=260031.msg2993417#msg2993417

[Lauda]

Time to buy cheap AMD CPUs 

[Spoetnik]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github.
Compared to the original miner though, it is about 85% faster, uses SSE2 along with ASM for two of the rounds.

uhhhhhhhhhhh 85% faster on AMD Bulldozer CPU's only ?
am i wasting my time ripping my hair out trying to compile this newer version (on windows / for windows) for no speed gain on my Intel CPU ?

https://bitcointalk.org/index.php?topic=260031.msg2993417#msg2993417

dead link = 503 Service Unavailable

also you have made two comments and still not said what it is..

is it optimized for Intel and was the old one ?

can someone shed some light on this please ?
and a windows build would be great too

[ig0tik3d]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github.
Compared to the original miner though, it is about 85% faster, uses SSE2 along with ASM for two of the rounds.

uhhhhhhhhhhh 85% faster on AMD Bulldozer CPU's only ?
am i wasting my time ripping my hair out trying to compile this newer version (on windows / for windows) for no speed gain on my Intel CPU ?

https://bitcointalk.org/index.php?topic=260031.msg2993417#msg2993417

dead link = 503 Service Unavailable

also you have made two comments and still not said what it is..

is it optimized for Intel and was the old one ?

can someone shed some light on this please ?
and a windows build would be great too

is compiled for windows 64 ( from source code http://stonefoz.myfastmail.com/cpuminer-quark.zip)
other link https://docs.google.com/file/d/0B9cvOfoOekSdVzZZcThLZHg4bjA/edit?usp=sharing

[zvs]

it's the same, there is a binary posted about 3 pages back that someone was nice enough to make. If you feel adventures the source has been moved to github.
Compared to the original miner though, it is about 85% faster, uses SSE2 along with ASM for two of the rounds.

uhhhhhhhhhhh 85% faster on AMD Bulldozer CPU's only ?
am i wasting my time ripping my hair out trying to compile this newer version (on windows / for windows) for no speed gain on my Intel CPU ?

don't know how fast/slow original miner is.  the one in the program itself?   quarkcoind got about 100khash, quarkcoinqt got about 200khash, and this minerd64_sse2 gets over 1mhash on an i7 4770K