Bitcoin Forum
April 25, 2014, 07:38:49 AM *
News: Due to the OpenSSL heartbleed bug, changing your forum password is recommended.
 
   Home   Help Search Donate Login Register  
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15  All
  Print  
Author Topic: further improved phatk_dia kernel for Phoenix + SDK 2.6 - 2012-01-13  (Read 29428 times)
conspirosphere.tk
Hero Member
*****
Offline Offline

Activity: 1022


Regime change begins at home


View Profile WWW

Ignore
August 20, 2011, 01:54:19 PM
 #241

Strange: now I am getting the max speed with the original phatk2 included in Phoenix 1.6.2 with my 5750. Here the results of some tests:

phoenix-1.6.2 original
phatk VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=168 Mhs

phoenix-1.6.2 original
phatk2 VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=170 Mhs

phoenix-1.6.2 original
-k phatk2 VECTORS2 FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=169 Mhs

2011-08-11 kernel 
phatk VECTORS2 FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
= 165 Mhs

2011-08-11 kernel
phatk2 VECTORS2 FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
= 153 Mhs

phatk 2011-07-17 kernel
phatk VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=167 Mhs

poclbm VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
= 157 Mhs


Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1398411529
Hero Member
*
Offline Offline

Posts: 1398411529

View Profile Personal Message (Offline)

Ignore
1398411529
Reply with quote  #2

1398411529
Report to moderator
1398411529
Hero Member
*
Offline Offline

Posts: 1398411529

View Profile Personal Message (Offline)

Ignore
1398411529
Reply with quote  #2

1398411529
Report to moderator
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
August 20, 2011, 02:40:33 PM
 #242

Strange: now I am getting the max speed with the original phatk2 included in Phoenix 1.6.2 with my 5750. Here the results of some tests:

phoenix-1.6.2 original
phatk VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=168 Mhs

phoenix-1.6.2 original
phatk2 VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=170 Mhs

phoenix-1.6.2 original
-k phatk2 VECTORS2 FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=169 Mhs

2011-08-11 kernel 
phatk VECTORS2 FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
= 165 Mhs

2011-08-11 kernel
phatk2 VECTORS2 FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
= 153 Mhs

phatk 2011-07-17 kernel
phatk VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
=167 Mhs

poclbm VECTORS FASTLOOP=false AGGRESSION=7 WORKSIZE=256 BFI_INT
= 157 Mhs

Seems mixed up a bit Smiley.

phatk = VECTORS2 (with my version)
phatk2 = VECTORS

Dia

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
hash
Newbie
*
Offline Offline

Activity: 11



View Profile

Ignore
August 22, 2011, 04:20:48 AM
 #243

This worked very good. I've updated my miners.
mute20
Sr. Member
****
Offline Offline

Activity: 265


21


View Profile

Ignore
August 22, 2011, 09:53:17 PM
 #244

Anymore updates planned for the future Grin.

Also I saw a 3% boost someone else was touting can I use that in conjunction with this?
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
August 23, 2011, 05:45:52 AM
 #245

Anymore updates planned for the future Grin.

Also I saw a 3% boost someone else was touting can I use that in conjunction with this?

Yes, there are updates planned, but it's really hard to squeeze more performance out of this. I'm currently testing a few different paths to achieve better performance!

Where have you seen a 3% tweak? Can you post a link here, so that I can check if this can be included (if it's not in already).

Thanks,
Dia

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
mute20
Sr. Member
****
Offline Offline

Activity: 265


21


View Profile

Ignore
August 25, 2011, 05:37:57 PM
 #246

Can't seem to find it ,but I assume it had similar fixes to yours.
supersquish
Newbie
*
Offline Offline

Activity: 17


View Profile

Ignore
August 26, 2011, 11:31:59 AM
 #247

Thanks for this great kernel mod, it did result in a solid 7~ mhash increase on my 5570, up from 63~. Will donate some if I manage to get amazingly lucky solo mining (I figure it's more fun to try for freak luck, It's a better habbit than lotto tickets right?)
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
August 27, 2011, 12:11:09 PM
 #248

Download version 2011-08-27: http://www.mediafire.com/?697r8t2pdk419ji

This version is a bit faster on 58XX cards, reports indicate it can be faster on 69XX cards, too ... I guess this is because of the optimized writing to the output buffer.
You can leave out the BFI_INT switch, but remember to supply the VECTORS2 switch Smiley! This version takes care of wrong WORKSIZE arguments, too ... if you forget that switch, if it has an too big value or if it's not a power of 2, the maximum supported WORKSIZE for each device is used.

Thanks,
Dia

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
swapper
Newbie
*
Offline Offline

Activity: 17


View Profile

Ignore
August 27, 2011, 12:58:57 PM
 #249

Thanks, we will try.
pekv2
Hero Member
*****
Offline Offline

Activity: 742



View Profile

Ignore
August 27, 2011, 01:21:21 PM
 #250

What Mhash/s rates the 5830 users receiving on version 2011-08-27?

Stay Safe. - First Amendment to the United States Constitution. - BTC 1NsTtfLyfcDjvJQEr6gEAgCiMBWi11jdd7 LTC LQUH8kSZ31qML4K8rCUMByqMSgb68dTmga
Crs
Member
**
Offline Offline

Activity: 103



View Profile

Ignore
August 28, 2011, 11:42:22 PM
 #251

here's my magic formula:

-k phatk2 AGGRESSION=13 FASTLOOP VECTORS2 WORKSIZE=128 PLATFORM=1
Ati 5850
before:348 Mh/s
after:359 Mh/s

If my post has been useful, you might want to donate: 1V7zr2jmQQXS5gv2kKNpUkgG6qmDffTZp
pk
Newbie
*
Offline Offline

Activity: 12


View Profile

Ignore
August 29, 2011, 12:44:27 AM
 #252

here's my magic formula:

-k phatk2 AGGRESSION=13 FASTLOOP VECTORS2 WORKSIZE=128 PLATFORM=1
Ati 5850
before:348 Mh/s
after:359 Mh/s

Saw an increase of 20MH/s, thanks.
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
December 27, 2011, 11:40:41 AM
 #253

Download version 2011-12-21: http://www.mediafire.com/?r3n2m5s2y2b32d9

Should restore some of the speed loss for 58XX owners, who switched to SDK / runtime 2.6 and is the best for 69XX owners, too.

Edit: Guys, try a setting of 64 for the WORKSIZE, it showed good results for me, but still depends on the card!

Dia

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
jhajduk
Newbie
*
Offline Offline

Activity: 26


View Profile

Ignore
January 02, 2012, 05:26:49 PM
 #254

Hi, I noticed a potential improvement

you can replace
Code:
W(121);
sharoundW(121);
W(122);
sharoundW(122);
W(123);
sharoundW(123);

// Round 124
Vals[7] += Vals[3] + P4(124) + P3(124) + P1(124) + P2(124) + s1(124) + ch(124) + H[7];
with
Code:
W(121);
Vals[2] += t1W(121);
W(122);
Vals[1] += t1W(122);
W(123);
                Vals[0] += t1W(123);

// Round 124
Vals[7] += Vals[3] + P4(124) + P3(124) + P1(124) + P2(124) + s1(124) + ch(124) + H[7];

Because you don't need Vals[4],Vals[5], and Vals[6] to compute the final Vals[7]
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
January 02, 2012, 06:42:48 PM
 #255

Hi, I noticed a potential improvement

you can replace
Code:
W(121);
sharoundW(121);
W(122);
sharoundW(122);
W(123);
sharoundW(123);

// Round 124
Vals[7] += Vals[3] + P4(124) + P3(124) + P1(124) + P2(124) + s1(124) + ch(124) + H[7];
with
Code:
W(121);
Vals[2] += t1W(121);
W(122);
Vals[1] += t1W(122);
W(123);
                Vals[0] += t1W(123);

// Round 124
Vals[7] += Vals[3] + P4(124) + P3(124) + P1(124) + P2(124) + s1(124) + ch(124) + H[7];

Because you don't need Vals[4],Vals[5], and Vals[6] to compute the final Vals[7]

I'll have to look into this, during the first test all performance relevant numbers were identical to my latest kernel. But perhaps reordering of the operations will help.
Thanks for your input Smiley!

Dia

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
jhajduk
Newbie
*
Offline Offline

Activity: 26


View Profile

Ignore
January 05, 2012, 05:15:07 AM
 #256

I also noticed that if the pools used the 7th 32 bit component of the hash ("g") rather than the first 32 bit ("a") for computing shares you could stop after the 61st round of computation rather than the 63rd.  That would be maybe a 2% efficiency for pools that implemented it.
gijs007
Newbie
*
Offline Offline

Activity: 17


View Profile

Ignore
January 08, 2012, 10:26:05 AM
 #257

Hi, I'm new to bitcoin mining and I'm not sure if I'm posting this in the correct forum (I can't post in the phoenix thread), I've used guiminer for a few months now and decided to give phoenix a try.
And its great, phoenix is slightly faster and has a stable hashrate.
However I noticed that phatk2 is much slower(50-60Mhashes/s) then pathk, and I believe this optimized kernel is supposed to replace the original pathk2 kernel, because when I overwrite the original pathk kernel with this one my hash rate is almost the same as with pathk2.

Am I using incorrect settings? I have an unlocked 6950 card with 910 mhz core and 1440 mhz memory.

The settings I use are: -k phatk DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

or if I want to run on phatk2:  -k phatk2 DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
January 08, 2012, 11:49:18 AM
 #258

Hi, I'm new to bitcoin mining and I'm not sure if I'm posting this in the correct forum (I can't post in the phoenix thread), I've used guiminer for a few months now and decided to give phoenix a try.
And its great, phoenix is slightly faster and has a stable hashrate.
However I noticed that phatk2 is much slower(50-60Mhashes/s) then pathk, and I believe this optimized kernel is supposed to replace the original pathk2 kernel, because when I overwrite the original pathk kernel with this one my hash rate is almost the same as with pathk2.

Am I using incorrect settings? I have an unlocked 6950 card with 910 mhz core and 1440 mhz memory.

The settings I use are: -k phatk DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

or if I want to run on phatk2:  -k phatk2 DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

I will only comment on my kernel here, which has to be used with:
Code:
-k phatk DEVICE=0 VECTORS2 AGGRESSION=11 WORKSIZE=128 FASTLOOP=false

The normal Phoenix kernel, which is in the default Phoenix download package doesn't have VECTORS2 and needs BFI_INT switch supplied (I activate this by myself if cl_amd_media_ops extension is available).

Dia

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
gijs007
Newbie
*
Offline Offline

Activity: 17


View Profile

Ignore
January 08, 2012, 12:02:50 PM
 #259

Hi, I'm new to bitcoin mining and I'm not sure if I'm posting this in the correct forum (I can't post in the phoenix thread), I've used guiminer for a few months now and decided to give phoenix a try.
And its great, phoenix is slightly faster and has a stable hashrate.
However I noticed that phatk2 is much slower(50-60Mhashes/s) then pathk, and I believe this optimized kernel is supposed to replace the original pathk2 kernel, because when I overwrite the original pathk kernel with this one my hash rate is almost the same as with pathk2.

Am I using incorrect settings? I have an unlocked 6950 card with 910 mhz core and 1440 mhz memory.

The settings I use are: -k phatk DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

or if I want to run on phatk2:  -k phatk2 DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

I will only comment on my kernel here, which has to be used with:
Code:
-k phatk DEVICE=0 VECTORS2 AGGRESSION=11 WORKSIZE=128 FASTLOOP=false

The normal Phoenix kernel, which is in the default Phoenix download package doesn't have VECTORS2 and needs BFI_INT switch supplied (I activate this by myself if cl_amd_media_ops extension is available).

Dia
Thanks,I replaced the normal pathk kernel with yours, the performance is now the same in Mh/s although shares seem to come slightly quicker.
Hmm with a worksize of 64 I get slightly better performance(0.5 Mh/s more).
I'm now at ~408 MH/s drops to 407.91 sometimes, and with aggression on 12 I get 408+ Mh/s
Fine tuned aggression to 16 and I get 409+ Mh/s nearly 410.
Diapolo
Hero Member
*****
Offline Offline

Activity: 766


Bitcoin-Qt co-developer


View Profile WWW

Ignore
January 08, 2012, 01:12:23 PM
 #260

Hi, I'm new to bitcoin mining and I'm not sure if I'm posting this in the correct forum (I can't post in the phoenix thread), I've used guiminer for a few months now and decided to give phoenix a try.
And its great, phoenix is slightly faster and has a stable hashrate.
However I noticed that phatk2 is much slower(50-60Mhashes/s) then pathk, and I believe this optimized kernel is supposed to replace the original pathk2 kernel, because when I overwrite the original pathk kernel with this one my hash rate is almost the same as with pathk2.

Am I using incorrect settings? I have an unlocked 6950 card with 910 mhz core and 1440 mhz memory.

The settings I use are: -k phatk DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

or if I want to run on phatk2:  -k phatk2 DEVICE=0 VECTORS BFI_INT AGGRESSION=11 worksize=128  FASTLOOP=false

I will only comment on my kernel here, which has to be used with:
Code:
-k phatk DEVICE=0 VECTORS2 AGGRESSION=11 WORKSIZE=128 FASTLOOP=false

The normal Phoenix kernel, which is in the default Phoenix download package doesn't have VECTORS2 and needs BFI_INT switch supplied (I activate this by myself if cl_amd_media_ops extension is available).

Dia
Thanks,I replaced the normal pathk kernel with yours, the performance is now the same in Mh/s although shares seem to come slightly quicker.
Hmm with a worksize of 64 I get slightly better performance(0.5 Mh/s more).
I'm now at ~408 MH/s drops to 407.91 sometimes, and with aggression on 12 I get 408+ Mh/s
Fine tuned aggression to 16 and I get 409+ Mh/s nearly 410.

What's your setup? Driver, OS, card?

Like my work for Bitcoin-Qt?
1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x
bitcoin:1PwnvixzVAKnAqp8LCV8iuv7ohzX2pbn5x?label=Diapolo
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15  All
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!