CCminer(SP-MOD) Modded GPU kernels.

[sp_]

@github. Some work left.  (will not work without fixing the bugs)

The padding bytes are incorrect.
I think the temp buffer needs to be moved from the stack out to globalmem etc..

Perhaps we should add cpu mining first to verfiy the methods.

[1714694663]

1714694663

[1714694663]

1714694663

[1714694663]

1714694663

[1714694663]

1714694663

[Slava_K]

@github. Some work left.  (will not work without fixing the bugs)

The padding bytes are incorrect.
I think the temp buffer needs to be moved from the stack out to globalmem etc..

Perhaps we should add cpu mining first to verfiy the methods.

Doesnot submit hashes (or even found any). GPU load very small...
GREAT WORK SP_!!!

[Epsylon3]

dunno what to answer... but the coin dev will not be happy about that... and that draft on your main branch

[sp_]

axioncoin
@github. Some work left.  (will not work without fixing the bugs)

It's not finished. Don't use it. I am sharing so somebody else can fix it..

Submitted a speedup in quark and x11.. Precalculated 1/32 of blake from cpu->const mem

quark is now hashing at 16.4 MHASH on the gigabyte windforce 970 factory clocks.

[dga]

axioncoin
@github. Some work left.  (will not work without fixing the bugs)

It's not finished. Don't use it. I am sharing so somebody else can fix it..

Submitted a speedup in quark and x11.. Precalculated 1/32 of blake from cpu->const mem

quark is now hashing at 16.4 MHASH on the gigabyte windforce 970 factory clocks.

Typo in Makefile.am --

Code:

@@ -54,7 +55,7 @@ ccminer_SOURCES       = elist.h miner.h compat.h \
                          x11/cuda_x11_luffa512_Cubehash.cu \
                          x13/x13.cu x13/cuda_x13_hamsi512.cu x13/cuda_x13_fugue
                          x15/x14.cu x15/x15.cu x15/cuda_x14_shabal512.cu x15/cu
-                         x15/whirlpool.cu x15\cuda_axiom.cu x15\axiom.cu \
+                         x15/whirlpool.cu x15/cuda_axiom.cu x15/axiom.cu \

(backslash instead of forward slash)

Also, for a bit, I thought that Wolf was bragging about the R9 Furry.  I'm glad I misread.

[Slava_K]

The R9 Fury (NOT Fury X, less CUs), does 20.5MH/s on Quark.

But hash per watt?

[sp_]

The gtx 970 has a tdp of 150W but the windforce has a 4pin and a 3pin power cable. It uses around 180W to produce 16.4 MHASH. (0,091 MHASH per watt)
the R9 Fury has a tdp of 275W.

[sp_]

Doesnot submit hashes (or even found any). GPU load very small...
GREAT WORK SP_!!!

if you know how to code, you can change a few lines and make a bitcoin in a couple of days with few rigs.
With some more workl the gpu miner can be 20 times faster than the cpu miner for an average cpu. That's why somebody is selling private kernals for 3 BTC++
Donate some Beers, and I will finish the kernal.

[sp_]

The gtx 970 has a tdp of 150W but the windforce has a 4pin and a 3pin power cable. It uses around 180W to produce 16.4 MHASH. (0,091 MHASH per watt)
the R9 Fury has a tdp of 275W.

Measurements from the wall? Because mine's in a machine with overvolted cards, and an FX-8370...

yes. I use 1x risers and have 1x enabled in the bios for all cards. standard voltage and just factory overclock. Other 970 cards use less power and produce less hash, but 2x 970 seems to cost the same, and produce more hash per watt. But nice job on the kernals.

[hashbrown9000]

damn. flaxscript really dumped hard today

[Grim]

The gtx 970 has a tdp of 150W but the windforce has a 4pin and a 3pin power cable. It uses around 180W to produce 16.4 MHASH. (0,091 MHASH per watt)
the R9 Fury has a tdp of 275W.

Measurements from the wall? Because mine's in a machine with overvolted cards, and an FX-8370...

yes. I use 1x risers and have 1x enabled in the bios for all cards. standard voltage and just factory overclock. Other 970 cards use less power and produce less hash, but 2x 970 seems to cost the same, and produce more hash per watt. But nice job on the kernals.

Ah, okay. I need to drop volts on the other cards and see how it does, then.

@Wolf0 you need to find yourself a memory bandwidth limited algo. That is the only place a fury can shine at.
A gcn 1.2 core is only slightly more efficient than a gcn 1.0 core so no wonder your hash per watt numbers on compute intensive algos is as bad as on any 280x.

Hint! Axiom Hint!

[hashbrown9000]

if you know how to code, you can change a few lines and make a bitcoin in a couple of days with few rigs.

HA! Here I am with a couple rigs, barely scraping by at about 1/10 that rate.  wish i knew what i was looking at when i open up some of this code.

[Slava_K]

if you know how to code, you can change a few lines and make a bitcoin in a couple of days with few rigs.
With some more workl the gpu miner can be 20 times faster than the cpu miner for an average cpu. That's why somebody is selling private kernals for 3 BTC++
Donate some Beers, and I will finish the kernal.

i am not a programmer... but some btc for beer? tomorrow in evening i send to you 0.045 btc (3 days of my mining)

[myagui]

@sp_ :

I'd advise looking at PoW reward schedules on a coin before committing time to something that is just about done and over with.
I respect your work and whatever you decide to publish, but in this particular case, I think people are just setting themselves up for disappointment. The PoW blocks were 50 coins per block a day ago, now sit at 25 coins per block, and it will be 10 coins per block in little over a day. I don't think anyone will find a Pot of Gold at the end of this Rainbow, but anyhow, I wish everyone the best of luck.  

If your comment about the BTC 3++ private kernels concerned me, as I was quite public about selling: I did close deals on a private version, with a total of 4 people at 0.5 BTC each (roughly matching my initial quote to the developers to open source it). That's it, and I closed sales at that point. I'm rather frustrated that something that could have sustained decent returns for a few lucky people, will quickly turn into 0 returns for just about everyone. The usual, to be perfectly honest.  

[djm34]

The gtx 970 has a tdp of 150W but the windforce has a 4pin and a 3pin power cable. It uses around 180W to produce 16.4 MHASH. (0,091 MHASH per watt)
the R9 Fury has a tdp of 275W.

Measurements from the wall? Because mine's in a machine with overvolted cards, and an FX-8370...

yes. I use 1x risers and have 1x enabled in the bios for all cards. standard voltage and just factory overclock. Other 970 cards use less power and produce less hash, but 2x 970 seems to cost the same, and produce more hash per watt. But nice job on the kernals.

Ah, okay. I need to drop volts on the other cards and see how it does, then.

@Wolf0 you need to find yourself a memory bandwidth limited algo. That is the only place a fury can shine at.
A gcn 1.2 core is only slightly more efficient than a gcn 1.0 core so no wonder your hash per watt numbers on compute intensive algos is as bad as on any 280x.

Hint! Axiom Hint!

Those hash per watt numbers are still getting better - I may be limited by the hardware, but it's capable of more.

still don't get it why every one telling it is limited, mem bandwitdth is supposed to be much higher, than any other card around, as lots of core as well, it should do great work on neoscrypt, yescrypt (lol... nah...), lyra.. unless the limitation is in opencl and dealing with new type of memory (the reason why I might wait for pascal + hbm2.0 + cuda N.x)

[scryptr]

BEER DONATION--

I flipped you another nickle today.  I hope that it confirms faster than the last donation, confirmation on that one took seven days.  Please finish the Axiom kernel!

My cumulative total went over 0.1 BTC today, if it wasn't already.  I've never solomined before, I'd like to try with spreadcoin.  About how many spreadcoin per GTX 970 per day should a miner expect?

Thanks again for your work!       --scryptr

[chrysophylax]

BEER DONATION--

I flipped you another nickle today.  I hope that it confirms faster than the last donation, confirmation on that one took seven days.  Please finish the Axiom kernel!

My cumulative total went over 0.1 BTC today, if it wasn't already.  I've never solomined before, I'd like to try with spreadcoin.  About how many spreadcoin per GTX 970 per day should a miner expect?

Thanks again for your work!       --scryptr

this obviously depends on the hashrate scryptr ...

and obviously how much processing and optimization you can cram into a single card with a single miner ...

#crysx

[pokeytex]

Not sure I am understanding correctly - there is a new algo out called axiom that is CPU only but a GPU miner has been created already?  Where I can I buy the miner and how much does it cost?  Please send me a PM if you are selling it.

Thanks - pokeytex

[chrysophylax]

Not sure I am understanding correctly - there is a new algo out called axiom that is CPU only but a GPU miner has been created already?  Where I can I buy the miner and how much does it cost?  Please send me a PM if you are selling it.

Thanks - pokeytex

to my understanding - the first half of your statement is correct ...

though the miner is no longer up for sale - and sp has already stated earlier that for a few donations from us ( the community ) he will be happy to implement and work on the algo in this fork of ccminer ...

myagui has already stated also - that due to the lower PoW rewards of axiom - it is wise to look at the time and effort spent on such things and whether it would be viable for that time and effort ...

i hope that helps pokeytex ... and if im incorrect in anything here - im sure someone will correct me ...

#crysx

[scryptr]

Not sure I am understanding correctly - there is a new algo out called axiom that is CPU only but a GPU miner has been created already?  Where I can I buy the miner and how much does it cost?  Please send me a PM if you are selling it.

Thanks - pokeytex

AXIOM MINER--

If you want to mine Axiom algo, donate to SP_ .  He may finish the algo in CCminer.  If your donations total more than 0.1 BTC, he'll also send you his Spreadcoin special miner.  Mostly, he'll keep working on CCminer and improving it.

My 750ti cards are mining Quark at 6.5-6.6 Mh/s each now.  My 960 card gets 320 kh/s on Neoscrypt, and 10.6 Mh/s on Quark.       --scryptr