further improved phatk_dia kernel for Phoenix + SDK 2.6 - 2012-01-13

[saykor]

Currently running 11.7 drivers, 2.1 SDK and poclbm.
Did some testing.
With 11.7, SDK 2.5 this kernel runs better than the poclbm phatk kernel.
With 11.7, SDK 2.1 this kernel runs a bit worse than the poclbm phatk kernel.

And with 11.7, SDK 2.1 I get the highest Mhash/s. So I'll be still using the poclbm phatk kernel.
Pretty much same goes for 11.6, which seems to perform slightly slower than 11.7, but otherwise all same.

I'm not sure how or why it seemed as if your kernel was boosting my Mhash, when I first tried it, but with my current setup, unfortunately poclbm phatk wins.

hi,
where you find SDK 2.5? i see only 2.4

[Diapolo]

New version online!

DL here: http://www.mediafire.com/?xlkcc08jvp5a43v

Dia

[Diapolo]

Currently running 11.7 drivers, 2.1 SDK and poclbm.
Did some testing.
With 11.7, SDK 2.5 this kernel runs better than the poclbm phatk kernel.
With 11.7, SDK 2.1 this kernel runs a bit worse than the poclbm phatk kernel.

And with 11.7, SDK 2.1 I get the highest Mhash/s. So I'll be still using the poclbm phatk kernel.
Pretty much same goes for 11.6, which seems to perform slightly slower than 11.7, but otherwise all same.

I'm not sure how or why it seemed as if your kernel was boosting my Mhash, when I first tried it, but with my current setup, unfortunately poclbm phatk wins.

hi,
where you find SDK 2.5? i see only 2.4

Use Cat 11.7 preview!

Dia

[1MLyg5WVFSMifFjkrZiyGW2nw]

Seems good, but it brings no gain for me ... weird. Will look into this and perhaps re-use your idea, if I may .
Can only think of a good compiler optimization...

Thanks,
Dia

Your modifications don't improve my speed either  

I'm using poclbm on Linux with the SDK version 2.4, card is 4670
currently 33.8 MHash/s with my change, hope I can get it up a bit more

i not have this line. what version you use? can you upload it?

http://uploading.com/files/69c85b51/phatk.cl/
what miner are you using?

[saykor]

http://uploading.com/files/69c85b51/phatk.cl/
what miner are you using?

i use Phoenix 1.5
i am with windows7 with 3 cards 5830
and windows server 2008 with 1 card 5830

i not have a file phatk.cl

[1MLyg5WVFSMifFjkrZiyGW2nw]

http://uploading.com/files/69c85b51/phatk.cl/
what miner are you using?

i use Phoenix 1.5
i am with windows7 with 3 cards 5830
and windows server 2008 with 1 card 5830

i not have a file phatk.cl

The file is named kernel.cl in Phoenix, it's in the "kernels\poclbm" folder.

EDIT:
this version of the kernel doesn't have the #define sharound
if you want to try this, rename your kernel.cl to something else,
then rename the downloaded file to kernel.cl and restart phoenix

[saykor]

yes i do exactly what you told in edit. but it not run on windows os
Phoenix just crash after connect to the server

[saykor]

with last Diapolo's patch here the result for my cards

win7:
1th card +1
2th card +2
3th card +5

win2008
1th card +1

[1MLyg5WVFSMifFjkrZiyGW2nw]

yes i do exactly what you told in edit. but it not run on windows os
Phoenix just crash after connect to the server

with last Diapolo's patch here the result for my cards

win7:
1th card +1
2th card +2
3th card +5

win2008
1th card +1

Strange, both work for me but with Diapolo's version I get 6 MH/s less, even though it also removes the redundant t1(n).

[mazmorbid]

Thanks for the code change, adds around 2% to my 5870's 

[Bert]

New version online!

DL here: http://www.mediafire.com/?xlkcc08jvp5a43v

Dia

Moved my 5870 cards, which were already using the previous version, from 456Mh/s to 456.75Mh/s, sent an additional donation, thanks.

[deepceleron]

Use Cat 11.7 preview!

Dia

The ATI 110619a-121104E file doesn't include a 11.7 video driver for OSs other than Windows 7. It does have a new AMD APP SDK, version 2.5.684.24. The package also installs drivers for ATI TV Wonder 600 USB and Hydravision (WTF?), so if you want to try out the new SDK, be sure to do a custom install and uncheck the other stuff.

That being said, the new SDK makes no difference for me. I've ran 10.11 and 11.6 drivers on 2.4 and 2.5 SDK with identical benchmarks on all in a very repeatable setup. However your patch does make a difference! The improvement on my overclocked 5830 (1070/392, WinXPsp3 stripped, Sempron 2.7ghz):

Before: 340.21 Mhash/s
After: 341.59 Mhash/s
Improvement: 0.41%

Update: Total accept/reject rate for three miners running this kernel: 11644/253 = 97.9% efficiency, no change there.

[saykor]

Use Cat 11.7 preview!

Dia

The ATI 110619a-121104E file doesn't include a "11.7" video driver. It does have a new AMD APP SDK, version 2.5.684.24. The package also installs drivers for ATI TV Wonder 600 USB and Hydravision (WTF?), so if you want to try out the new SDK, be sure to do a custom install and uncheck the other stuff.

That being said, the new SDK makes no difference. I've ran 10.11 and 11.6 drivers on 2.4 and 2.5 SDK with identical benchmarks on all in a very repeatable setup. However your patch does make a difference! The improvement on my overclocked 5830 (1070/392, WinXPsp3 stripped, Sempron 2.7ghz):

Before: 340.21 Mhash/s
After: 341.59 Mhash/s
Improvement: 0.41%

I haven't run it long enough to see if it increases rejects or stales, but I will report back. This is looking good (but it will take a week of mining with the patch to even make up for my 30 minutes downtime...)

What is your setting? How you make 340 with 5830? i am on 307max with same cards.
You know from where to download the "11.7" video driver?

Thanks for your help

[h2]

thanks a lot! brought me from 421 to 423 (x2) for my 5970 (already had the 3% Ma patch applied)

everybody getting "FATAL kernel error: Failed to load OpenCL kernel!" run phoenix once with sudo prefix. afterwards you can launch it again without sudo.

greets
H²

[figvam]

340 Mh/s -> 344 Mh/sec in peak on a 5850.

[Soak]

Try to uninstall SDK 2.1 and use that Cat 11.7 preview here: http://developer.amd.com/Downloads/110619a-121104E.zip

390 with new Catalyst and old kernel.
388 with new Catalyst and new kernel.

ATI Radeon 6970 on Windows 7.

[strictlyfocused]

Went 232 to 233 on an MSI 5770   

Sending a donation your way!

[sturle]

Hello,

you might want to change this

Code:

#define sharound(n) { t1 = t1(n); Vals[(131 - n) % 8] += t1(n); Vals[(135 - n) % 8] = t1(n) + t2(n); }

to

Code:

#define sharound(n) { t1 = t1(n); Vals[(131 - n) % 8] += t1; Vals[(135 - n) % 8] = t1 + t2(n); }

This got me a 25% performance increase!

Seems good, but it brings no gain for me ... weird. Will look into this and perhaps re-use your idea, if I may .
Can only think of a good compiler optimization...

It is not obvious that this will gain anything.  On the contrary, unless the compiler optimize it away, it makes the second and third instruction dependant on the first.  This is bad on a GPU which issues 4 or 5 instructions in parallel every clock cycle.

[Apopfis]

http://i55.tinypic.com/2q0kq69.png

http://i55.tinypic.com/2q0kq69.png

Testing my new Sapphire 5850
I need new PSU and watercooling for the card to make it stable. Currently running cards (overnight) steady at 412 Mh/s @ 1030 MHz core and other one ~360 Mh/s @ 950 core because only AGGRESSION=8. Psu is Corsair 450W so cannot OC & OV both cards  to the max. For the first card this mod gave increase from 401-> 412 while card is running @ 1030 core.

[rcocchiararo]

downloaded newest file.

307 > 310
278 > 280

cant't complain, it was free