Krypta - it looks like r0() and r1() aren't used; try deleting them if you like.
Note that they aren't used by the phatk kernel that I've got, and even though I can't think of circumstances in which r0() will compile correctly, still, be careful if your kernel is different!
Topic: 3% faster mining with phoenix+phatk for everyone - page 2. (Read 36831 times)
It works really well:: +3MHs on my 5770 at stock frequencies.
Seeing all these success stories makes me wonder if the file i downloaded was actually assembled correctly.
When apply this little 'tweak'
my mh/sec rate increses by seeming 2mhash/s per 6870 core But, the value at what im hashing at jumps all the fuck over the place. like 20-150-270-279-410-4-89.
It's rediculos
When apply this little 'tweak'
my mh/sec rate increses by seeming 2mhash/s per 6870 core But, the value at what im hashing at jumps all the fuck over the place. like 20-150-270-279-410-4-89.
It's rediculos
Yup, T'was compiled wrong.
Everyone update your GUIminer client
This beautiful tweak fix has been intergrated into the newest update, No weird fiddling required
And i can confirm that i've gained 10mh/s/core now 6870BE
Seeing all these success stories makes me wonder if the file i downloaded was actually assembled correctly.
When apply this little 'tweak'
my mh/sec rate increses by seeming 2mhash/s per 6870 core But, the value at what im hashing at jumps all the fuck over the place. like 20-150-270-279-410-4-89.
It's rediculos
When apply this little 'tweak'
my mh/sec rate increses by seeming 2mhash/s per 6870 core But, the value at what im hashing at jumps all the fuck over the place. like 20-150-270-279-410-4-89.
It's rediculos
Anyone tried this using guiminer? I can see the file but not sure if I just edit the portion:
// Ma can also be implemented in terms of BFI_INT...
#define Ma(x, y, z) amd_bytealign((y), (x | z), (z & x))
// Ma can also be implemented in terms of BFI_INT...
#define Ma(x, y, z) amd_bytealign((y), (x | z), (z & x))
This patch works fine with guiminer as well. And yes, just edit the portion you posted.
thank you very much , this has increased my rate
Jen
Jen
Krypta:
I've seen that before as well, I didn't change it, but it sure looks like it shouldn't even compile!
I haven't tried the alterations yet, but will do so when I come home!
UPDATE: It works great, increase on both 6870's from 287 to 296 (> 3%!)
I've seen that before as well, I didn't change it, but it sure looks like it shouldn't even compile!
I haven't tried the alterations yet, but will do so when I come home!
UPDATE: It works great, increase on both 6870's from 287 to 296 (> 3%!)
Before: 353 MHps
After: 364 MHps
Difference: 11 MHps (3.0%)
Thanks!
After: 364 MHps
Difference: 11 MHps (3.0%)
Thanks!
Hy all, I'm new here and pretty new with this bitcoin thing.
I tried to modify phatk kernel as explained and I was able to speed up my mining as all of you.
Really don't know when and how this amd_bytealign function is called, anyway I think it should be better to write it as follows
#define Ma(x, y, z) amd_bytealign(((z) ^ (x)), (y), (x))
That said, I took a look at the rest of the phatk kernel and found what seems to be a sintax error. In my file it's on line 74:
#define r1(x) ((rot(x],15)^rot(x,13)^((x)>>10U)))
Mmmm... that poor and lonely square bracket... rewrite it as follows
#define r1(x) ((rot(x,15)^rot(x,13)^((x)>>10U)))
These modifications did not speed up my mining power but it seems that my rejecteds almost halved.
If this helps you as well, donations are obviously welcome
1CbRmFAYrGYhzQibmE5A1CWMp7srFT4upy
Krypta
I tried to modify phatk kernel as explained and I was able to speed up my mining as all of you.
Really don't know when and how this amd_bytealign function is called, anyway I think it should be better to write it as follows
#define Ma(x, y, z) amd_bytealign(((z) ^ (x)), (y), (x))
That said, I took a look at the rest of the phatk kernel and found what seems to be a sintax error. In my file it's on line 74:
#define r1(x) ((rot(x],15)^rot(x,13)^((x)>>10U)))
Mmmm... that poor and lonely square bracket... rewrite it as follows
#define r1(x) ((rot(x,15)^rot(x,13)^((x)>>10U)))
These modifications did not speed up my mining power but it seems that my rejecteds almost halved.
If this helps you as well, donations are obviously welcome
1CbRmFAYrGYhzQibmE5A1CWMp7srFT4upy
Krypta
Anyone tried this using guiminer? I can see the file but not sure if I just edit the portion:
// Ma can also be implemented in terms of BFI_INT...
#define Ma(x, y, z) amd_bytealign((y), (x | z), (z & x))
// Ma can also be implemented in terms of BFI_INT...
#define Ma(x, y, z) amd_bytealign((y), (x | z), (z & x))
400 -> 412 on my 5870. great!
6970 @ 975/400 from ~420 Mhash/S to ~432 Mhash/S
Getting some good results, but one out of 7 workers always locks up after some minutes. GPU usage goes down to zero, shown MHash rate stays locked at last value, cursor is still flashing. Occurs only when the cards are overclocked, but the same and even higher OC values work flawless without this fix.
I did this with phatk in phoenix-1.50 and I got a 2.4% increase on both my 5870 and my 5770. Thanks!
5870 (930/350), 410->421, 2.6%, thx 
I tried using Phoenix, but my GPU isn't recognized. Neither with phatk.
I'm running Arch Linux with an Ati Radeon HD4650. DiabloMiner detects my GPU, and I installed AMDStream, so that isn't the problem.
I'm running Arch Linux with an Ati Radeon HD4650. DiabloMiner detects my GPU, and I installed AMDStream, so that isn't the problem.
Dude, awesome so this accounts for 3% out of your 6.4% improvement, where are the other 3.4% coming from? I don't have 250 btc, in fact I only have ~2 that have been donated to me so far. (Please note - if you decide to share this, post it,don't pm me )
so this accounts for 3% out of your 6.4% improvement, where are the other 3.4% coming from? I don't have 250 btc, in fact I only have ~2 that have been donated to me so far. (Please note - if you decide to share this, post it,don't pm me )I think I deserve the right to keep this remaining ~3% a secret of mine. I have contributed a lot to this community. AFAIK I was the...
* first to show how to down-plug x16 cards in x1 slots by cutting PCIe slots and/or shorting pins A1-B17 http://blog.zorinaq.com/?e=42
* first to use BFI_INT, and document how to patch a kernel with it: http://blog.zorinaq.com/?e=43
* first to document the PCIe 12V rail redirection hack http://blog.zorinaq.com/?e=44
* first to present a cheap solution to cool 4 double width cards http://blog.zorinaq.com/?e=47
Now all the miners are using BFI_INT; shops are selling pre-modified PCIe extenders; etc. Enjoy :-)
Sure, of course you do, dude! It was just a joke
sorry! By the way, I absolutely love your blog - I've read it a lot, just didn't know it was yours. is it possible this doesnt work on nVidia cards? both with OpenCL miner & CUDA miner i'm still getting the exact same hashrate, even after restarting Guiminer.
or am i missing something? (using windows 7 btw)
or am i missing something? (using windows 7 btw)
Sorry, this only works for ATI cards that support BFI_INT. In general, nVidia cards are much worse for mining, so I'm not sure if there's effort at all going into making nVidia's stuff mine faster...
Yeah i know Nvidia cards aren't too good for mining, but i was hoping for a slight increase... oh well
Thanks for pointing it out tho, much appreciated
Dude, awesome so this accounts for 3% out of your 6.4% improvement, where are the other 3.4% coming from? I don't have 250 btc, in fact I only have ~2 that have been donated to me so far. (Please note - if you decide to share this, post it,don't pm me )
so this accounts for 3% out of your 6.4% improvement, where are the other 3.4% coming from? I don't have 250 btc, in fact I only have ~2 that have been donated to me so far. (Please note - if you decide to share this, post it,don't pm me )I think I deserve the right to keep this remaining ~3% a secret of mine. I have contributed a lot to this community. AFAIK I was the...
* first to show how to down-plug x16 cards in x1 slots by cutting PCIe slots and/or shorting pins A1-B17 http://blog.zorinaq.com/?e=42
* first to use BFI_INT, and document how to patch a kernel with it: http://blog.zorinaq.com/?e=43
* first to document the PCIe 12V rail redirection hack http://blog.zorinaq.com/?e=44
* first to present a cheap solution to cool 4 double width cards http://blog.zorinaq.com/?e=47
Now all the miners are using BFI_INT; shops are selling pre-modified PCIe extenders; etc. Enjoy :-)
Yay! got the expression:
Y&(Z^X)|X&(~Z^X)
Y&(Z^X)|X&(~Z^X)
Jump to: