3% faster mining with phoenix+phatk for everyone

[Opsamk]

The more efficient you make miners, the higher the difficulty rises.

[1714778315]

1714778315

[1714778315]

1714778315

[1714778315]

1714778315

[1714778315]

1714778315

[PcChip]

Wow, really awesome find. You're gonna be a minor celebrity around here soon no doubt.

I salute your ingenuity, and am firing up all the VNC sessions to edit my miners kernel files

edit: You added ten megahashes per second to one of my 5850's.  I'm now hitting 400 MH/s on only 1.162 volts.  I love you.


That was a very selfless act giving it away on the forums like that.  Unfortunately well over 60% of the community would have wanted to charge a fee for their optimization.

[eusor]

Nice!

On a 6950 : 349 MH/s -> 366 MH/s (3.75%)

Thanks!

[fitty]

Everybody, thanks for trying this out. I guess it really does work  for other people. Keep donating, I like it

Now that we know it works, it would be great to check this change into the source code repository somewhere so that everyone gets this change by default, without having to manually edit any files. How is this done? Also, how do I post into the non-newbie thread? I feel that this post it doesn't belong here.

I believe after 5 posts you can post in the miner section. Don't worry this will get around quick and it will I'm sure get added within a week or two at most. Great job! I hope your donation made it. =)

[PcChip]

I hope it doesn't get added and we keep it a secret.  Delete this post

 

Edit: To the OP - if you want to see the results of your kernel edit to my 5770:

210 MH/s -> 217 MH/s , check the graph (dip is when I swapped kernels)
One degree cooler and more stable usage

[JDog]

Excellent advice, thanks!!!

[godofal]

got a massive increase on openCL miner, from 159 to 166 max
well, okay, not massive, but in terms of % its above what the others got

and it'l come in handy for my mining rig

thanks for the tip and effort, gonna send you a thank you

[bitclown]

2.6% increase on Phoenix' poclbm kernel. Nice work!

[PcChip]

2.6% increase on Phoenix' poclbm kernel. Nice work!

I wasn't aware that poclbm supported BFI_INT ?

Why not use PhatK with the 2.4 Stream SDK ?

[icolsuineg]

Thanks, donation send, I got a total of about 100 MH/s from all my rigs, works as advertised.

[naukop]

Went from 193.7Mh/s to 195.1Mh/s on my 5770 (900/600) after I edited BitcoinMiner.cl and restarted guiminer.

 running guiminer phoenix.exe VECTORS BFI_INT FASTLOOP=false AGGRESSION=12 -k phatk WORKSIZE=128

Hmm, but now has rate has actually dropped into 186-188 range, not sure what is going on.

[Opsamk]

So bitless, how far are you in Computer Science?

[bitless]

So bitless, how far are you in Computer Science?

Naukop - Could the drop in the hashrate be due to overheating? I've noticed that my card heats up more as a result of the change (and then either it hangs or I have to overclock it less ), so could the same be happening to you?

Opsamk - m.sc., then asm programming for 10 years or so; i'm not looking to be-rich myself on bitcoins, just playing around with these hashes and ATI's hardware

[coblee]

Worked for me. 380 -> 392 (3.16%)
Thanks. This is worth donating for!

[jprockbelly]

Thanks so much for sharing!

I only got about 2% increase, but hey, thats pretty good for 30 seconds work

[msb8r]

From 339.7MH -> 351.7MH (+3.5%) on my 6950@860 using Guiminer (poclbm).

Very nice find, bitless. Donation on it's way 

[Crs]

Works for me too.
I have two 5850s:

bitcoin0
[27/06/2011 12:38:36] Result: 09c2f420 accepted
[330.86 Mhash/sec] [38843 Accepted] [992 Rejected] [RPC]nmc@nmcpc:~$
new bitcoin0
[27/06/2011 12:43:09] Result: 805929c4 accepted
[340.86 Mhash/sec] [4 Accepted] [0 Rejected] [RPC]nmc@nmcpc:~/phoenix$

bitcoin1
[27/06/2011 12:38:45] Result: d593d475 accepted
[27/06/2011 12:38:49] Result: f23c0c10 accepted
[332.14 Mhash/sec] [39162 Accepted] [970 Rejected] [RPC]nmc@nmcpc:~$
new bitcoin1
[27/06/2011 12:43:45] Result: 7b29a04b accepted
[342.27 Mhash/sec] [11 Accepted] [0 Rejected] [RPC]nmc@nmcpc:~/phoenix$

Thanks for sharing this!

[-ck]

Thanks for your code. It's nice to see people get rewarded for their efforts (though it doesn't always happen). Well, I've tried it on my own miner (minerd) and have some very strange findings.

https://forum.bitcoin.org/index.php?topic=21275.0

First of all, minerd supports up to 4 vectors, and when I add this change to my kernel, it actually _slows down_ the 4 vector version. But when I override it to set 2 vectors, it speeds it up. However, once it's sped up, I then get runs of rejected shares. I tried it multiple times with and without and it does appear to be just this change that causes it, so I'm not sure what's going on.

[being]

Thank you! Works great!

[t3h]

Wasn't the claimed improvement with hdminer 5% anyway? I wonder if Phoenix is now faster than hdminer... or was it before anyway?

5770 from ~200 to ~210