Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 1181. (Read 2347601 times)
First feedback of GTX960 on Qubit - 7.6- 7.7 MH/s with Palit GTX960 Super Jetsteam...
if you are familiar with reprogramming bios - then you will be able to recompile the kernels of the operating systems alongside the a hacked / reprogrammed bios ...
but for now - if you are using standard motherboards that are the 'norm' these days - then the limitation usually always ends up being 8 gpus ...
but for now - if you are using standard motherboards that are the 'norm' these days - then the limitation usually always ends up being 8 gpus ...
I am not a cuda developer, I an expert in assembly languages and reverse engineering for different platforms. Used to remove copy protection for games and commercial software 15-20 years ago. If I want my motherboard's to use more than 8 gpu's I think I can fix it. But it might take some time.
as it is his code and he can do what he wants with it ...
Obviously. I'm just contemplating donating myself but I believe that if everybody have an advantage, than nobody really have an advantage.
Correct - and it will inevitably leak; likely quickly.
@ chrysophylax:
Don't do the math about advantages. Think of a donation as a way to say thank you that really says so.
Otherwise your "thank you" is worth as much as "excuse me" from somebody passing by at the grocery store...
At least these are my thoughts on the matter. Just like I have my phone jailbroken and install cracked apps,
but when I find one that makes my life easier (saves some of my time, or in any way benefits me) I don't
hesitate to buy it through appstore. You can call it Karma if you wish... but that's how I roll.
I know a lot of people are not like that but I am very fond of it and also I am happy
as it is his code and he can do what he wants with it ...
Obviously. I'm just contemplating donating myself but I believe that if everybody have an advantage, than nobody really have an advantage.
Correct - and it will inevitably leak; likely quickly.
He could distribute slightly different (marked) versions for different donators but that's another story.
Anyway, I'm in. I'm curious to see the performance on the 750 Ti's.
Hardly matters - all that helps with is damage control.
and obviously this cant' be done in the GPL framework either) 
as it is his code and he can do what he wants with it ...
Obviously. I'm just contemplating donating myself but I believe that if everybody have an advantage, than nobody really have an advantage.
Correct - and it will inevitably leak; likely quickly.
He could distribute slightly different (marked) versions for different donators but that's another story.
Anyway, I'm in. I'm curious to see the performance on the 750 Ti's.
Hardly matters - all that helps with is damage control.
true - but its a brainstorm session in place mate
...im thinking maybe a pool that is in trust for the devs - though im falling asleep now and im not quite there yet with what to do with a trust account ...
thinktank tomorrow for me - for both the farm redesign and the donation methodology ...
will get there ...
nite all ...
#crysx
as it is his code and he can do what he wants with it ...
Obviously. I'm just contemplating donating myself but I believe that if everybody have an advantage, than nobody really have an advantage.
Correct - and it will inevitably leak; likely quickly.
He could distribute slightly different (marked) versions for different donators but that's another story.
Anyway, I'm in. I'm curious to see the performance on the 750 Ti's.
Just a few donators. I will keep improving the hashrate for every build, to make sure that my donators get an advantage. in 1 month from sunday I will start publish some of the improvements opensource. Perhaps 10% first, and then 10% each month after I cannot optimize any more. My goal is to publish everything in the end, but it will take time. And kernals needs to be tested.
a stepped approach may be a good way to do this sp ...
i know it will keep me interested - especially when the people that donate have a slight ( or possibly more ) advantage for a short while before everyone else ...
either way - i donate NOT for the advantage ... i do so because YOU as a dev produce better software and optimizations that i USE ...
so you and wolf and djm and others deserve to be paid and have an open wallet for donations ...
#crysx
as it is his code and he can do what he wants with it ...
Obviously. I'm just contemplating donating myself but I believe that if everybody have an advantage, than nobody really have an advantage.
Correct - and it will inevitably leak; likely quickly.
not all software / kernels / optimizations leak ... but i guess judging by past history ( yes - im indicating to your history on the matter wolf ) then i see where your statement is highly justified ...
there needs to be a better way to minimize the carnage of highly optimized software in order to satisfy the general public AND donors alike without sacrificing the earnings ( if you call a few beers 'earnings' ) of the devs ...
there has to be a better way than what we have now ... im thinking ...
bathrobe ... there is ALWAYS an advantage helping the devs create a better miner - always ...
its just the degree of advantage that varies ...
:|
#crysx
Just a few donators. I will keep improving the hashrate for every build, to make sure that my donators get an advantage. in 1 month from sunday I will start publish some of the improvements opensource. Perhaps 10% first, and then 10% each month after I cannot optimize any more. My goal is to publish everything in the end, but it will take time. And kernals needs to be tested.
as it is his code and he can do what he wants with it ...
Obviously. I'm just contemplating donating myself but I believe that if everybody have an advantage, then nobody really have an advantage.
I have solved a bug wich causes the cpu to use alot of cpu. Now x11 is hashing with 0-1% cpu usage. Fixed in the unpublished spreadcoin fork for now, but will submitt to github later. Less CPU-> less energy and more stable hashrates.
A noticable difference when using the computer while mining.
Hi sp_.A noticable difference when using the computer while mining.
We need the spreadx11 opt. miner asap, dif. is skyrocketing...
So is the price. 600% in one month. This coin is gold
I will test on my 970/980 rig on friday/saturday and release the first beta to the donators on sunday.
Can I ask how long do you intend to keep the miner exclusive to donators?
I think enough time to do not hurt the SPR coin as he stated before.
'enough time' doesnt really specify how long ...
i think 20-30days would be long enough - but ultimately that is up to sp ...
its his code ...
#crysx
IMHO it wouldn´t be fair to release to public since only few donated to support him.
well im one of those who donated - and i do think its fair ...
BUT - as i said - differing opinions mean nothing when it comes to sp's decision ...
as it is his code and he can do what he wants with it ...
though i do see where you are coming from ...
#crysx
I have solved a bug wich causes the cpu to use alot of cpu. Now x11 is hashing with 0-1% cpu usage. Fixed in the unpublished spreadcoin fork for now, but will submitt to github later. Less CPU-> less energy and more stable hashrates.
A noticable difference when using the computer while mining.
Hi sp_.A noticable difference when using the computer while mining.
We need the spreadx11 opt. miner asap, dif. is skyrocketing...
So is the price. 600% in one month. This coin is gold
I will test on my 970/980 rig on friday/saturday and release the first beta to the donators on sunday.
Can I ask how long do you intend to keep the miner exclusive to donators?
I think enough time to do not hurt the SPR coin as he stated before.
'enough time' doesnt really specify how long ...
i think 20-30days would be long enough - but ultimately that is up to sp ...
its his code ...
#crysx
IMHO it wouldn´t be fair to release to public since only few donated to support him.
I have solved a bug wich causes the cpu to use alot of cpu. Now x11 is hashing with 0-1% cpu usage. Fixed in the unpublished spreadcoin fork for now, but will submitt to github later. Less CPU-> less energy and more stable hashrates.
A noticable difference when using the computer while mining.
Hi sp_.A noticable difference when using the computer while mining.
We need the spreadx11 opt. miner asap, dif. is skyrocketing...
So is the price. 600% in one month. This coin is gold
I will test on my 970/980 rig on friday/saturday and release the first beta to the donators on sunday.
Can I ask how long do you intend to keep the miner exclusive to donators?
I think enough time to do not hurt the SPR coin as he stated before.
'enough time' doesnt really specify how long ...
i think 20-30days would be long enough - but ultimately that is up to sp ...
its his code ...
#crysx
I have solved a bug wich causes the cpu to use alot of cpu. Now x11 is hashing with 0-1% cpu usage. Fixed in the unpublished spreadcoin fork for now, but will submitt to github later. Less CPU-> less energy and more stable hashrates.
A noticable difference when using the computer while mining.
Hi sp_.A noticable difference when using the computer while mining.
We need the spreadx11 opt. miner asap, dif. is skyrocketing...
So is the price. 600% in one month. This coin is gold
I will test on my 970/980 rig on friday/saturday and release the first beta to the donators on sunday.
Can I ask how long do you intend to keep the miner exclusive to donators?
I think enough time to do not hurt the SPR coin as he stated before.
I have solved a bug wich causes the cpu to use alot of cpu. Now x11 is hashing with 0-1% cpu usage. Fixed in the unpublished spreadcoin fork for now, but will submitt to github later. Less CPU-> less energy and more stable hashrates.
A noticable difference when using the computer while mining.
Hi sp_.A noticable difference when using the computer while mining.
We need the spreadx11 opt. miner asap, dif. is skyrocketing...
So is the price. 600% in one month. This coin is gold
I will test on my 970/980 rig on friday/saturday and release the first beta to the donators on sunday.
Can I ask how long do you intend to keep the miner exclusive to donators?
Quark:
v28= ~11,160 kh/s
v30= ~11,441 kh/s
v31b= ~11,650 kh/s (and decrease in CPU usage)
x11:
v28= ~5,730 kh/s
v30= ~5,850 kh/s
lyra2:
v28= ~1,350 kh/s
v30= ~1,350 kh/s
(2 750Ti's, no overclock)
v28= ~11,160 kh/s
v30= ~11,441 kh/s
v31b= ~11,650 kh/s (and decrease in CPU usage)
x11:
v28= ~5,730 kh/s
v30= ~5,850 kh/s
lyra2:
v28= ~1,350 kh/s
v30= ~1,350 kh/s
(2 750Ti's, no overclock)
nice...
I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
You probobly need a modded driver. I usually have 2-3 amd cards, and fill the rest of the slots with nvidia cards. Good for the powersupply as well.I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
This looks interesting. 16 GPU's on one motherboard using 4 PCI-e slots.
http://amfeltec.com/products/gpu-oriented-cluster/
But Might have to modify the driver...
EDIT: just checked the price.. £140 for pciE splitter 1->4 slots. Bether to buy another motherboard. Not worth it.
sp - i checked these out a while ago - even contacted the people ...
this will work ONLY if the bios of the motherboard allows it to ...
most standard bios only allows ( its own memory limitation ) 8 gpu's per motherboard ...
this limitation is only compensated when you have motherboards with enhanced bios - like some of the supermicro motherboards - that are designed to have more than 8 gpus via a bios augmentation ...
if you are familiar with reprogramming bios - then you will be able to recompile the kernels of the operating systems alongside the a hacked / reprogrammed bios ...
but for now - if you are using standard motherboards that are the 'norm' these days - then the limitation usually always ends up being 8 gpus ...
you can definitely add more ( i have had 12 gpus on the one system in linux ) and the os ( in my case ) could see ALL the gpus - but could only use 8 at a time due to this limitation ...
if there was no limitation - i would be running them right this minute
btw - the supermicro dual xeon boards that can handle 12gpus are all around $600-700AUD ... not worth the effort for mining unless you have the cash for it as a project ...
#crysx
nice...
I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
You probobly need a modded driver. I usually have 2-3 amd cards, and fill the rest of the slots with nvidia cards. Good for the powersupply as well.I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
This looks interesting. 16 GPU's on one motherboard using 4 PCI-e slots.
http://amfeltec.com/products/gpu-oriented-cluster/
But Might have to modify the driver...
EDIT: just checked the price.. £140 for pciE splitter 1->4 slots. Bether to buy another motherboard. Not worth it.
nice...
I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
You probobly need a modded driver. I usually have 2-3 amd cards, and fill the rest of the slots with nvidia cards. Good for the powersupply as well.
ya, only thing that amd is left good at ...
nice...
I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
I can't get more than 6 working in win 8.1 (the same mobo had 7 amd cards running)
You probobly need a modded driver. I usually have 2-3 amd cards, and fill the rest of the slots with nvidia cards. Good for the powersupply as well.
Jump to: