Gateless Gate Sharp 1.3.8: 30Mh/s (Ethash) on RX 480!

[zawawa]

6x1070

Code:

Total 1585.0 sol/s [dev0 265.9, dev1 268.7, dev2 270.8, dev3 263.0, dev4 260.2, dev5 267.8] 104 shares
Total 1585.4 sol/s [dev0 265.6, dev1 268.9, dev2 270.9, dev3 264.2, dev4 260.4, dev5 268.6] 104 shares
Total 1585.2 sol/s [dev0 264.9, dev1 269.8, dev2 270.2, dev3 263.2, dev4 260.0, dev5 267.9] 105 shares
Total 1585.1 sol/s [dev0 264.5, dev1 270.4, dev2 269.6, dev3 263.6, dev4 260.2, dev5 267.4] 105 shares
Total 1584.6 sol/s [dev0 265.5, dev1 270.4, dev2 268.8, dev3 261.8, dev4 259.0, dev5 267.9] 105 shares
Total 1584.9 sol/s [dev0 264.3, dev1 270.7, dev2 269.1, dev3 262.1, dev4 259.8, dev5 268.2] 106 shares
Total 1585.1 sol/s [dev0 262.9, dev1 271.3, dev2 268.6, dev3 262.4, dev4 261.0, dev5 269.3] 107 shares
Total 1584.8 sol/s [dev0 262.9, dev1 270.8, dev2 267.2, dev3 261.5, dev4 260.6, dev5 269.2] 107 shares

Note: the cpu fix for linux seems to be broken, I tested with dirty fix(LD_PRELOAD libtime.so), with cpu fix it will be faster 1-2%.

Those are very nice numbers to see... Thanks for your report.
I don't think the CPU fix for Linux was even applied to my fork to begin with.
I will fix this problem soon.

[1714057514]

1714057514

[1714057514]

1714057514

[1714057514]

1714057514

[1714057514]

1714057514

[zawawa]

I just updated the repository. 170 sol/s on RX 480 and 162 sol/s on GTX 1060 3GB. kernel_sol gets split up and now 200% faster with a parallelized duplicate search. NR_ROWS_LOG=12 is finally usable on RX 480. More speedups are coming soon.

[krnlx]

I just updated the repository. 170 sol/s on RX 480 and 162 sol/s on GTX 1060 3GB. kernel_sol gets split up and now 200% faster with a parallelized duplicate search. NR_ROWS_LOG=12 is finally usable on RX 480. More speedups are coming soon.

Code:

main.c:1434:12: error: too few arguments to function ‘solve_equihash’
   total += solve_equihash(dev_id, ctx, queue, k_init_ht, k_rounds, k_sols, buf_ht,

Code:

main.c:1489:3: error: too many arguments to function ‘mining_mode’
   mining_mode(*dev_id, program, ctx, queue, k_init_ht, k_rounds, k_potential_sols, k_sols, buf_ht,

[ioglnx]

You are on a good way to on with Claymore on AMD.
For nvidia it's still a need a double of current sol/s :-D

[Vetal_inside]

Try 15.12 drivers.

 The "Relive" junk does NOT work well for mining in particular, and is a major step backwards for pre-RX series cards in general while being a SERIOUSLY bad case of massive bloatware.
 It, like many of the Crimson 16.x series drivers, also refuses to work with modded BIOS on many cards.

 The Crimson stuff past 15.12 offered NO improvement from mining on pre-RX series cards.

Yes, I also thought that this was driver problem: I uninstalled 16.** with DDU, and install 15.12 back. But it have the same situation: GGv0.0.1 shows hashrate, GGv0.0.2 not.

[zawawa]

I just updated the repository. 170 sol/s on RX 480 and 162 sol/s on GTX 1060 3GB. kernel_sol gets split up and now 200% faster with a parallelized duplicate search. NR_ROWS_LOG=12 is finally usable on RX 480. More speedups are coming soon.

Code:

main.c:1434:12: error: too few arguments to function ‘solve_equihash’
   total += solve_equihash(dev_id, ctx, queue, k_init_ht, k_rounds, k_sols, buf_ht,

Code:

main.c:1489:3: error: too many arguments to function ‘mining_mode’
   mining_mode(*dev_id, program, ctx, queue, k_init_ht, k_rounds, k_potential_sols, k_sols, buf_ht,

Oops... Let me fix that right now.

[zawawa]

I just updated the repository. 170 sol/s on RX 480 and 162 sol/s on GTX 1060 3GB. kernel_sol gets split up and now 200% faster with a parallelized duplicate search. NR_ROWS_LOG=12 is finally usable on RX 480. More speedups are coming soon.

Code:

main.c:1434:12: error: too few arguments to function ‘solve_equihash’
   total += solve_equihash(dev_id, ctx, queue, k_init_ht, k_rounds, k_sols, buf_ht,

Code:

main.c:1489:3: error: too many arguments to function ‘mining_mode’
   mining_mode(*dev_id, program, ctx, queue, k_init_ht, k_rounds, k_potential_sols, k_sols, buf_ht,

I just pushed a fix for Linux. Please let me know if that works.

[zawawa]

Try 15.12 drivers.

 The "Relive" junk does NOT work well for mining in particular, and is a major step backwards for pre-RX series cards in general while being a SERIOUSLY bad case of massive bloatware.
 It, like many of the Crimson 16.x series drivers, also refuses to work with modded BIOS on many cards.

 The Crimson stuff past 15.12 offered NO improvement from mining on pre-RX series cards.

Yes, I also thought that this was driver problem: I uninstalled 16.** with DDU, and install 15.12 back. But it have the same situation: GGv0.0.1 shows hashrate, GGv0.0.2 not.

You could change the following line in param.h:

Code:

#define OPENCL_BUILD_OPTIONS_AMD    "-I.. -I. -O1 -DAMD"

to:

Code:

#define OPENCL_BUILD_OPTIONS_AMD    "-I.. -I. -O1 -DAMD -legacy"

and see how that goes with the new and old drivers. You need to rebuild the miner, though.

[zawawa]

You are on a good way to on with Claymore on AMD.
For nvidia it's still a need a double of current sol/s :-D

Eqminer is doing 220 sol/s on GTX 1060 3GB, so I just need another 36% boost, no?
I was hoping I wouldn't have to get my hands dirty with SASS...

[Vetal_inside]

You could change the following line in param.h:

Code:

#define OPENCL_BUILD_OPTIONS_AMD    "-I.. -I. -O1 -DAMD"

to:

Code:

#define OPENCL_BUILD_OPTIONS_AMD    "-I.. -I. -O1 -DAMD -legacy"

and see how that goes with the new and old drivers. You need to rebuild the miner, though.

I'm using precompiled binaries for win

Also, I also tested GG v0.0.2(precompiled for win) with gtx 1070 (~2050/4400/75%), 376.33, Win8.1x64 - up to 230 sols. For comparison - latest nicehash eqm miner gives up to ~400 sols.

[krnlx]

Code:

Total 1844.5 sol/s [dev0 295.5, dev1 308.1, dev2 314.9, dev3 314.3, dev4 314.0, dev5 315.9] 40 shares
Total 1844.3 sol/s [dev0 294.9, dev1 308.8, dev2 315.4, dev3 314.5, dev4 313.7, dev5 314.9] 41 shares
Total 1846.2 sol/s [dev0 294.3, dev1 310.0, dev2 315.7, dev3 314.3, dev4 314.3, dev5 316.3] 41 shares
Total 1845.8 sol/s [dev0 293.7, dev1 308.9, dev2 315.0, dev3 314.6, dev4 313.2, dev5 317.0] 41 shares
Total 1846.1 sol/s [dev0 294.3, dev1 309.5, dev2 314.6, dev3 314.4, dev4 312.9, dev5 315.8] 42 shares
Total 1845.1 sol/s [dev0 292.7, dev1 310.1, dev2 314.5, dev3 315.8, dev4 310.9, dev5 314.7] 43 shares
Total 1844.1 sol/s [dev0 292.9, dev1 309.4, dev2 314.0, dev3 315.8, dev4 310.2, dev5 313.4] 43 shares

6x1070 overclocked + working cpu fix


patched main.c with working cpu fix on linux

http://pastebin.com/USaqBN2e

[nerdralph]

I can send you Claymore's kernel .il files.  email me if you want them.

Thanks for the offer, but I am pretty sure I can figure things out on my own at this point.
Besides, I greatly admire Claymore's coding skills.
Peeking into his code against his will would be disrespectful, and I don't want to do that.
Now I'm 100% confident that I can catch up with  Claymore's, Optiminer, and Eqminer.
It's just a matter of time and effort.

You have much to learn if you think the way Claymore behaves is deserving of respect.  After discovering Claymore copied some of Marc's code, Marc stayed calm and rational, while Claymore acted like a woman at the wrong time of the month.  If you look at everything JW (Genoil) has written, he has shown himself to be competent, calm, rational, and even humble (something I know I lack).  Despite some complaints about Wolf's behavior, most complaints about him are subjective and emotional.  When evaluated objectively, I don't think you'll find any dishonest or irrational behavior.  He seems wise for his age, since many people never learn to stop worrying about what other people think about your behavior, and just do what you enjoy as long as it is not objectively going to cause harm to others.
You're young though, so you may gain wisdom with age...

p.s. I'm not going to claim I learned every bit of this through wisdom over the years.  As an Aspie I was lucky to be born immune to some common social manipulation techniques like shaming and rejection.  That left me to use logical/stoic analysis to evaluate much of human behavior.

[zawawa]

Code:

Total 1844.5 sol/s [dev0 295.5, dev1 308.1, dev2 314.9, dev3 314.3, dev4 314.0, dev5 315.9] 40 shares
Total 1844.3 sol/s [dev0 294.9, dev1 308.8, dev2 315.4, dev3 314.5, dev4 313.7, dev5 314.9] 41 shares
Total 1846.2 sol/s [dev0 294.3, dev1 310.0, dev2 315.7, dev3 314.3, dev4 314.3, dev5 316.3] 41 shares
Total 1845.8 sol/s [dev0 293.7, dev1 308.9, dev2 315.0, dev3 314.6, dev4 313.2, dev5 317.0] 41 shares
Total 1846.1 sol/s [dev0 294.3, dev1 309.5, dev2 314.6, dev3 314.4, dev4 312.9, dev5 315.8] 42 shares
Total 1845.1 sol/s [dev0 292.7, dev1 310.1, dev2 314.5, dev3 315.8, dev4 310.9, dev5 314.7] 43 shares
Total 1844.1 sol/s [dev0 292.9, dev1 309.4, dev2 314.0, dev3 315.8, dev4 310.2, dev5 313.4] 43 shares

6x1070 overclocked + working cpu fix


patched main.c with working cpu fix on linux

http://pastebin.com/USaqBN2e

I just pushed your update. Thanks a bunch!

[Stealth8368]

I can send you Claymore's kernel .il files.  email me if you want them.

Thanks for the offer, but I am pretty sure I can figure things out on my own at this point.
Besides, I greatly admire Claymore's coding skills.
Peeking into his code against his will would be disrespectful, and I don't want to do that.
Now I'm 100% confident that I can catch up with  Claymore's, Optiminer, and Eqminer.
It's just a matter of time and effort.

You have much to learn if you think the way Claymore behaves is deserving of respect.  After discovering Claymore copied some of Marc's code, Marc stayed calm and rational, while Claymore acted like a woman at the wrong time of the month.  If you look at everything JW (Genoil) has written, he has shown himself to be competent, calm, rational, and even humble (something I know I lack).  Despite some complaints about Wolf's behavior, most complaints about him are subjective and emotional.  When evaluated objectively, I don't think you'll find any dishonest or irrational behavior.  He seems wise for his age, since many people never learn to stop worrying about what other people think about your behavior, and just do what you enjoy as long as it is not objectively going to cause harm to others.
You're young though, so you may gain wisdom with age...

p.s. I'm not going to claim I learned every bit of this through wisdom over the years.  As an Aspie I was lucky to be born immune to some common social manipulation techniques like shaming and rejection.  That left me to use logical/stoic analysis to evaluate much of human behavior.

I have to say I agree with you here Claymore was happy to use Marc's code and charge for his miner offered a couple of zcash and let's move on, seems Claymore has a following here that think he's some kind of God, I think he's starting to think that way himself but I just see a greedy developer who gives very little back to the community,in this world if someone offers you a leg up you take it and be thankful I can't see a problem your doing same as claymore did using Marc's code but helping the community we are all hoping you will come through for us and we can have a miner that's as fast or faster, Claymore getting his knickers in a twist yet he ignored Marc's license and used his code maybe he thinks it doesn't apply to him, have a look at the kernel and make gateless gate the fastest miner out there!

[xeridea]

I can send you Claymore's kernel .il files.  email me if you want them.

Thanks for the offer, but I am pretty sure I can figure things out on my own at this point.
Besides, I greatly admire Claymore's coding skills.
Peeking into his code against his will would be disrespectful, and I don't want to do that.
Now I'm 100% confident that I can catch up with  Claymore's, Optiminer, and Eqminer.
It's just a matter of time and effort.

You have much to learn if you think the way Claymore behaves is deserving of respect.  After discovering Claymore copied some of Marc's code, Marc stayed calm and rational, while Claymore acted like a woman at the wrong time of the month.  If you look at everything JW (Genoil) has written, he has shown himself to be competent, calm, rational, and even humble (something I know I lack).  Despite some complaints about Wolf's behavior, most complaints about him are subjective and emotional.  When evaluated objectively, I don't think you'll find any dishonest or irrational behavior.  He seems wise for his age, since many people never learn to stop worrying about what other people think about your behavior, and just do what you enjoy as long as it is not objectively going to cause harm to others.
You're young though, so you may gain wisdom with age...

p.s. I'm not going to claim I learned every bit of this through wisdom over the years.  As an Aspie I was lucky to be born immune to some common social manipulation techniques like shaming and rejection.  That left me to use logical/stoic analysis to evaluate much of human behavior.

I have to say I agree with you here Claymore was happy to use Marc's code and charge for his miner offered a couple of zcash and let's move on, seems Claymore has a following here that think he's some kind of God, I think he's starting to think that way himself but I just see a greedy developer who gives very little back to the community,in this world if someone offers you a leg up you take it and be thankful I can't see a problem your doing same as claymore did using Marc's code but helping the community we are all hoping you will come through for us and we can have a miner that's as fast or faster, Claymore getting his knickers in a twist yet he ignored Marc's license and used his code maybe he thinks it doesn't apply to him, have a look at the kernel and make gateless gate the fastest miner out there!

Given little to community?  He had a top quality miner when all the other miners were pretty much garbage. He charges a fee, what is wrong with that?  So any developer who charges a small fee is a greedy bastard?  Programmers need to make money like everyone else.  I am a developer, I am 100% for open source, have led and contributed to some, but I have nothing wrong with compensating others for their efforts.  Most my work has been at a company, because I have need to support family. Many FOSS software has many paid devs btw. Copied code was a mistake, he meant to write his own, but was rushed.  It wasn't any code related to performance, like others claim. He settled things with Marc ages ago, some will just be butthurt forever.  If Marc is satisfied, then all you others shouldn't get into it.  He overlooked irritation for all the haters, and released Linux version, for all those who prefer it, now in return we should just disrespect him?  Every Zec miner that is even remotely profitable has a devfee, Claymore is most stable, and fastest on most cards (by volume of those mining....), is tied for lowest fee, and most features.  How does this make him greedy? How is disrespecting their licence any better?

You guys are trying to get others to steal someone else work, doesn't that make you hypocrites?  Some others claim they can code better than CM, and post lots of theoretical stuff, but no code.  Then, they insist others steal CM code, which they call subpar (because of course they claim they themselves can produce awesome code, but never do of course).  If these people can really code as awesome as they say, they should do it instead of wasting time bashing.

I applaud zawawa for staying away from all this nonsense, and writing code instead of wasting time on drama. He is the real FOSS contributor.

I will post no more here about this, don't want to pollute this thread, which should be about miner development, not drama.

[zawawa]

Alright, I'm finally done implementing the "variable" NR_ROWS_LOG.
Now you can customize all the relevant parameters specifically for each round.
I am almost hitting the 180 sol/s mark with RX 480, and GTX 1060 is reaching 170 sol/s.
My wife is now constantly complaining about my forgetfulness and absent-mindedness, but that's OK...

[xeridea]

You are on a good way to on with Claymore on AMD.
For nvidia it's still a need a double of current sol/s :-D

Eqminer is doing 220 sol/s on GTX 1060 3GB, so I just need another 36% boost, no?
I was hoping I wouldn't have to get my hands dirty with SASS...

EQM is doing ~180 sol on a 1060 3GB stock (1822/1901), I could get ~206 with my non extreme OC (~1936/2000). I think that was not latest version though (switched to EWBF) though, so perhaps ~200 stock, 220 with some OC.  EWBF about 204.  Haven't tried new EWBF miner, many say it is unstable. So stock you are close in speed, at least on 1060 3GB.  People flaunt NH miner for speed, but EWBF is similar, or better profit due to high NH fee.

Sapphire 480 8GB gets ~240 with CM 9.2 (stock 1306/2000, 1750 strap), so a bit farther a way on that, but making good progress.

[Genoil]

My wife is now constantly complaining about my forgetfulness and absent-mindedness, but that's OK...

Reality-check: NOT OK

[Eliovp]

My wife is now constantly complaining about my forgetfulness and absent-mindedness, but that's OK...

Reality-check: NOT OK

I honestly do not want to give those a dollar who are lost in the "crypto-world", or rather lost their wive/girlfriend because of this..

 

[ioglnx]

Zawawa find a proper balance with real life and this hobby.
So take some days off from coding and give your wife a sorry and packed in that you can explain her why you are doing this 

Women often forget that men easily get lost into their hobby and about each detail like a child playing with imagination