Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 904. (Read 2347659 times)

Thanks !
The 960 or more 750Ti cards seem to be most bang for the buck then.

Prices here where i live:

EVGA 750Ti FTW: 180usd
Asus Strix 950: 219usd
Asus Strix 960: 279usd

My EVGA ssc 950 is only doing 8,5 MHASH on the standard clocks in quark.
The 960 strix is doing 10,7 MHASH (standard clocks)

I am thinking about adding five gpu:s shortly.
Not sure what is the most bang for the buck gpu right now, i have been a bit inactive in this thread for a while.

I am looking at:

EVGA 750Ti FTW (Quark 7.3Mhs stable, running ten of these already)
Asus Strix 950
Asus Strix 960

Anyone got some fresh hashrates on the 950 and 960 cards with the latest ccminer mod ?

That's why I sometimes release private miners as well.
The latest sp-modded pentablake miner was 100% faster and sold to 5 people that gave me 1,5BTC in a few hours.
Then I give something back to the community and optimize the other blake's in x11,x13,x15,x14,x17,quark and lyra2v2 and opensource.  
Just a few KHASH, but over time they matter.

A few KHASH in the most profitable algo, is worth more than MHASH'es in a less profitable algo.

Thanks for your support.

I have 3 commits in the lyra2v2 after release 70. Small steps that sums up to more free hashpower..

Thanks for your support.


But if you have 1000ti's you will earn 20k USD a month and then you don't need to work any more.
But GPU miners have been burned before in the bitcoin/litecoin rally.
The risk is extreme. Big initial costs, big electricity bills. High risk.. With the current bitcoinprices it's not worth it..
When NVIDIA release their 16NM chips next year you have 1000 useless GPU's

It would work if there weren't mining farms that can easily pay 5 BTC for a miner ;-)
When there is money... game over!

My GTX 960 get a speedup in lyra2v2 around 100 kh\s

i send you some vtc bears

best regards

Can confirm speed up in lyra2rev2 for 750 Ti.

Nice, and thank you.

this is for you:
3eab6f22bb361680190f12cd9f12bde1292752b29f4e801e554c501eaed99747

P.S. i don't want to continue that thread about opensource, but i decided to say my personal opinion.
Everybody of you guys had a good point: about private dev miners, about license, donations etc etc...
But i want to say that before that it was really interesting to participate in nvidia mining. just for fun.
Why only for fun? everything is simple. most of us - IT guys, everybody knows salary in IT. but if we will check number in mining, even if you have 100 750Ti cards you CAN'T earn more that 2k usd a month, but it still less than even average salary.
and everybody understand that 100 750Ti it's big... big farm and you need additional time to keep it up and running. So... after numbers, why not to do something just for fun, for good community, for cooperation.

Ok, this was just my 2 cents. don't get it close on your own.

Yes I submitted another speedup in lyra2v2 as well. Most of the first round and some of the second round of blake 256 precalculated(100 assembly instructions removed and replaced by 16 32bit constmem reads). The first round of Keccak256 unrolled, and 24 xor instructions removed. (x^0=x)

More speedups are comming

In addition to ScryptJaneLeo being hard to set up, it's not very profitable. It needs to double in value to be on par with quark.

BTW W10 machines get 2/3 the hashrate of W8.1 for ScryptJL

Speedup is not more then 30-40kH on quark. But lyra2rev2 is sooo good!

else scrypt:n and jane are not the same :p

Submitted another speedup in quark/x11. +50-100KHASH on the gtx 970.

Not sure about the other cards. Please test..

(More of blake512 - 80 was  precalculated.)

Here are my configs with cudaminer for

970: --algo=scrypt-jane:15 -L 8 -i 1 -l t64x4 -C 1 -b 65536

750ti: --algo=scrypt-jane:15 -H 2 -i 0 -m 1 -l t5x24 -L 8 -b 65536

i don't remember the hashrates but they were what most of other people were getting.

So I found best params.

GTX 980 Ti: -L 16 -l t22x23
GTX 950: -L 7 -l t6x32

What are best ones for 970 and 980? Don't have these cards...

Right on point bathrobehero. I neglected the SMM/SMX difference between the 980 and 970 upthread.

That explains it why GTX 980 Ti likes L=16

Last I checked scrypt-jane in SP's fork didn't work.

LEOcoin was stupidly profitable a few months ago but for a few weeks now it's not that profitable and now with nicehash adding it, it will probably keep diving.
The next n-factor increase will be in 38 hours though (n-factor 16) so don't spend too much time trying to find the best n-factor 15 config.

As a tip though, the fastest kernel launch configs are based on how many SM units a card have:
750 Ti's have 5 so the fastest launch config will look like t5x?, 970's have 13 (so t13x?), 980 have 16 and the 980 Ti have 22. With the correct lookup-gap of course.

I'm using an old version of sp_'s fork (somewhere around commit 800) and added a few basic command line arguments that are helping with scrypt-jane mining like being able to start autotuning from a set amount instead of having to wait from the beginning (so you can start autotuning from like t13x1), added another one that allows setting how much GPU memory ccminer can use (which is 88% by default on windows and 98% on linux), reintroduced the batchsize parameter since it was replaced with api-bind which slightly speeds the hashrate up. Just basic things saving some time and allowing to change stuff without having to recompile each time.
If someone need it just let me know.

For n-factor 15 I get:
750 Ti: 1.26khs (-L 5 -l t5x16)
970: 3.27khs (-L 10 -l t13x24)

N-factor 16:
750 Ti: 0.53khs (-L5 -l t5x15)
970: 1.15 (-L 14 -l t13x28)

These are for windows with oc, linux.

https://github.com/cbuchner1/CudaMiner/blob/master/README.txt#L131

Ideally, one should run with low'ish -L values, but there's different sweet spots for maximum performance vs best efficiency.
A higher -L gives you the most performance potential for a given compute capability, especially useful in high end cards. A lower -L gives you the most efficiency, and it is especially useful in lower end cards.