CCminer(SP-MOD) Modded GPU kernels.

[antantti]

@antantti
Can you validate my kernel's hashrate on your machine too?
Some basic math makes me expect that you should see (1466/1453.3)*3.303 ~= 3.33Gh/s  _{(That's not the case for memory bounded algos.)}

March version 3250 on 970, 4020 on 980, 1466 -i 30
Latest 3280 on 970, 4070 on 980, 1466 -i 30

My 970´s start hitting power limiter, that's why they don't go past 3300... So your version is a bit faster than sp´s on my machine.

[1715025932]

1715025932

[1715025932]

1715025932

[1715025932]

1715025932

[1715025932]

1715025932

[ltc_bilic]

@sp and @other private buyers for decred

what's the hashrate of sp_mod_decred #8 on 750ti?

[sp_]

Nice responses here today. My decred is the fastest. I have sendt the sourcecode out to trusted miners. The code is different than yours and faster. I don't have a broken hash. in #5 and up.

[sp_]

@sp and @other private buyers for decred
what's the hashrate of sp_mod_decred #8 on 750ti?

With OC 680MHASH.

[joblo]

This is getting out of hand. But let's remember how this all started, when SP went closed after
importing lots of open source improvements that were contributed in good faith by various developpers.

He has added some of his own improvements, but not enough, IMO, to justify what he is now charging
for his miners. Whether he is misrepresenting the work of others as his own, playing tricks to increase
performance, or it's just a little marketing hype is a matter between SP and the individuals involved.

I must be careful in my criticism as my project has primarilly been a copy/paste integration effort.
I'm not a young guy trying to learn, but an old fart trying to learn some new things.

Notwithstanding that cpuminer-opt is not a marketable product I would not consider charging users
for it giventhe level of original content. Furthermore I recognized that any original content I did add
would always remain open and free. I would also resent anyone forking my product and closing it

Furthermore his distribution of binary only packages derived from open source may be in violation
of the GPL.

I, for one, do not like SP's current strategy and have said so in the past, and choose not to support this
activity.

@SP My apologies for referring to you in the third person in your own thread, however my post was directed
openly to all developpers and users.

[Nikolaj]

In a unregolated market like this, these behaviours are simply called "greed". The amount of accusations are so heavy that replying with just 2 repeated sentencies (fastest.., fastest..) it's a passive admission of guilt. Too many in  this market donate for nonsense, especially when a donation it's not legally prosecutable. If it's prooven that some of the sp miners were more fast due a more aggressive intensity, being published like this it's simply called theft. I was interested in some of "his"works but the amount of greed, in his pm's, forced me to ignore his "kind and gelid" requests of a forced donation.

3 days later the same speed was released open source.. need we to add anything else? Don't donate without a reason guys, your money leads to other's guilt and greed (G&G 'tm :-) )

As the vastness of space, the emptyness of human nature.

[sp_]

I work hard to give you the fastest kernals. 0.1 btc is nothing. Decred is a simple algo with few lines of code. Many developers are working in private to reach the optimal speed. Because faster kernals is giving a bigger profit.

[sp_]

I have decided to release the sourcecode for decred private #4.

https://github.com/sp-hash/ccminer/commit/3a726d90efd528ed386407bbd5b223583d80b378

You can test it on linux.

Use whole number  intensities  to get the correct hashrates o  the pool 28,29,30,31 (ie -i 29.6 will give higher rates in the miner, but not on the pool. Known overflow bug corrected in #5)

Download the latest 1.7.6 and copy and rename the decred_ok.cu to decred.cu into the algo256 folder. Then rebuild with cuda 7.5


Cheers

[sp_]

That's what happens when you copy-paste code you dont understand for over a year from algo to algo Mr. "3lit3".

I released decred #4 opensource. My work is original,real and fast.

Yet again, by the time you fixed your binaries and brought to your buyers a fixed product, this code was already uploaded in my github fixed, since Feb of this year.
And it's been already a week since the pallas comit on the open-sourced repo.

I didn't know you had A repo. I don't give a fuck. The Pallas commit is not giving me any improvement on the 750ti on windows. (less than 1%)
Asus strix 1280Mhz cuda 7.5 build windows 32bit:

1.7.5: 477MHASH
1.7.5 pallas: 478MHASH
Decred sp-mod #4(opensource): 550MHASH(15% faster than the pallas version)
Decred sp-mod #8: 590MHASH (23% faster than the pallas version

The reason it was not working in CUDA7.5 came a month later when you reported bigger hashrate in CUDA7.5 and I was trusting your results.

My results are real, but when people use crazy intensities like -i 29.6 my code had a bug. It was corrected quickly in release #5

You made me look into it and i found the flaw which was preventing CUDA7.5 to translate the source the same way CUDA6.5 did.
I remember that chrysophylax said both private decred miners were reporting almost the same results, with yours being slightly faster.
But mine was already 15% faster in CUDA6.5 version rather than CUDA7.5 which was reporting less than 1% boost.

My code is optimized for the 750ti, the 970 is not important since you don't use the 970 to mine decred, you mine etherum or something more profitable.

That said: From SP_VAPORMOD#1 to SP_VAPORMOD#3 you were selling vaporware.
And you just admitted that from SP_BROKENMOD#4 until, SP_MOD#8 you were selling broken code?  

I am not selling anything. People donate and they get access to private bins. My private kernals are always the fastest.


Stop trolling my thread newbie!..

[sp_]

If you wonder why quark is +62% and x11 is only up 18%, it's because x11 need some more work and donations...

I just did a comparison between my buyable private miner (0.1BTC) and the latest 1.7.5 release.
I have written 0-10% faster, but I should have written 17-74% faster

Sp-Mod #5 default intensities hashrates in MHASH/s. (980ti 1290 coreclock/1650 memclock)

nist5: 43.80 (+17%)
quark: 27.85 (+61,91%)
lyra2v2 16.10 (+16.6%)
x11:   12.85 (+18%)
x13:   10.27 (+19.97)
x15:    8.83 (+18.52 )
neoscrypt: 0.99 (+73.68%)

1.7.5: (32bit windows build cuda 7.5) default intensities in MHASH/s (980ti 1290 coreclock/1650 memclock)

nist5: 37.45
quark: 17.20
lyra2v2: 13.81
x11: 10.88
x13: 8.56
x15: 7.45
neoscrypt: 0.57

1.7.5: (32bit windows build cuda 6.5) (980ti 1290 coreclock/1650 memclock)

nist5: 38.93
quark: 17.5
lyra2v2: 13.18
x11: 10.95
x13: 8.88
x15: 7.7
neoscrypt: 0.71

If you call my hard work scam, then fuck off. All ccminer kernals have been optimized. It's alot more work than just changing the intensity parameter..

[alexis78]

I came here because you reported 30%, 15%, 8% increase in the open sourced vcash kernel.
Since nobody validated this increase and people were reporting that your hashrate is rather similar or slightly lower
i admit i got out of temper (my bad).
Especially when your argument about the boost was that you hardcoded a bigger intensity.

As i said above, I dont care where people invest their btcs.
I was here to find out if you were speaking the truth about the increase.

[djm34]

That's what happens when you copy-paste code you dont understand for over a year from algo to algo Mr. "3lit3".

I released decred #4 opensource. My work is original,real and fast.

Yet again, by the time you fixed your binaries and brought to your buyers a fixed product, this code was already uploaded in my github fixed, since Feb of this year.
And it's been already a week since the pallas comit on the open-sourced repo.

I didn't know you had A repo. I don't give a fuck. The Pallas commit is not giving me any improvement on the 750ti on windows. (less than 1%)
Asus strix 1280Mhz cuda 7.5 build windows 32bit:

1.7.5: 477MHASH
1.7.5 pallas: 478MHASH
Decred sp-mod #4(opensource): 550MHASH(15% faster than the pallas version)
Decred sp-mod #8: 590MHASH (23% faster than the pallas version

The reason it was not working in CUDA7.5 came a month later when you reported bigger hashrate in CUDA7.5 and I was trusting your results.

My results are real, but when people use crazy intensities like -i 29.6 my code had a bug. It was corrected quickly in release #5

You made me look into it and i found the flaw which was preventing CUDA7.5 to translate the source the same way CUDA6.5 did.
I remember that chrysophylax said both private decred miners were reporting almost the same results, with yours being slightly faster.
But mine was already 15% faster in CUDA6.5 version rather than CUDA7.5 which was reporting less than 1% boost.

My code is optimized for the 750ti, the 970 is not important since you don't use the 970 to mine decred, you mine etherum or something more profitable.

That said: From SP_VAPORMOD#1 to SP_VAPORMOD#3 you were selling vaporware.
And you just admitted that from SP_BROKENMOD#4 until, SP_MOD#8 you were selling broken code?  

I am not selling anything. People donate and they get access to private bins. My private kernals are always the fastest.


Stop trolling my thread newbie!..

he isn't a newbie, he wrote the fast whirlpoolx algo you did paste in your release.

I think we are several to call your copy and paste a scam... selling to someone a program where the only thing you change is the intensity, is a scam. There is no hard work. You are not fooling any one at that level.

[sp_]

Pools can be picky sometimes, quick tests showed me that xvc gives reliable results on some major pools.

xvc@1466 -i 30 #5
970 ~3170
980 ~3920

xvc@1466 -i 30 #6
970 ~3260
980 ~4030

I think release #3 did 3300@1466. So revert to #3 on the 970.

[airwalker2662]

I came here because you reported 30%, 15%, 8% increase in the open sourced vcash kernel.
Since nobody validated this increase and people were reporting that your hashrate is rather similar or slightly lower
i admit i got out of temper (my bad).
Especially when your argument about the boost was that you hardcoded a bigger intensity.

As i said above, I dont care where people invest their btcs.
I was here to find out if you were speaking the truth about the increase.

I am curious of seeing what the default intensity is per miner. Is their a command argument that will show intensity in ccminer? I know debug does not.

[sp_]

I came here because you reported 30%, 15%, 8% increase in the open sourced vcash kernel.
Since nobody validated this increase and people were reporting that your hashrate is rather similar or slightly lower

The different NVIDIA graphic cards behave different. If you test your kernal on gtx 950, gtx 960 and the gtx960M. I did the first vanilla mod on the gtx 960m. and my results are real and can be verified. Why don't you start with decred #4 opensource. What are you getting on Linux?

[sp_]

I am curious of seeing what the default intensity is per miner. Is their a command argument that will show intensity in ccminer? I know debug does not.

There is no command, but for some algos you can check the gpu memory being used with gpu-z, and then try to match. Or just look in my opens sourcecode.

[sp_]

he isn't a newbie, he wrote the fast whirlpoolx algo you did paste in your release.
I think we are several to call your copy and paste a scam... selling to someone a program where the only thing you change is the intensity, is a scam. There is no hard work. You are not fooling any one at that level.

I base my work on opensource just like the rest of you. I made whirlpoolx 10-15% faster after Alexis's first commit.. Some of the speedup was increasing intensities and a rewrite to more than one nonce per thread. All my work is documented on github and the credits are preserved. (first commit 8 march 2015)
the whirlpoolx kernal have 18 commits:

https://github.com/sp-hash/ccminer/commits/windows/x15/whirlpoolx.cu

[airwalker2662]

I am curious of seeing what the default intensity is per miner. Is their a command argument that will show intensity in ccminer? I know debug does not.

There is no command, but for some algos you can check the gpu memory being used with gpu-z, and then try to match. Or just look in my opens sourcecode.

Wouldn't this be a simple addition to include with debug? To just pass the number in the algo used for the card being used? I think it would be very useful..

[sp_]

In my fork I have made a new parameter called -X (xintensity) just like in sgminer. Most of the kernals have -X as default and not -i

[sp_]

I came here because you reported 30%,

Because your kernal used a default intensity of 24 on windows. That's pretty stupid isn't it. You loose 30%, it doesn't matter how good the code is, if you launch your kernal with retarded default parameters.. In your code, you case on windows, and reduce the speed with 1 line of code..
Cryptominers only care about two things. Speed and Stability.