CCminer(SP-MOD) Modded GPU kernels.

[tbearhere]

ok, i've found how to reset to p2 state without rebooting your computer and without using inspector, just this nvidia-smi -rac

very useful, was a pain in the ass to reset each time i need to use other algo, which otherwise would result in a drivers crash with p0 state, like quark...probably due to the high overclock...

quark is the one with the most problems and then neoscrypt and lyra2v2 ... but quark is the worst.. even with no oc'ing.
There was a command in cuda mining  - C 0  - C 1  -C 2 but there not in ccminer and I don't know if that would help in the oc'ing or not.

yeah when i shift from ethereum to quark with the same oc(+150 core/mem) i always get a crash, qubit is the same, there is a bug on the miner maybe? i don't remember such thing with 750ti and i had 10 of those baby...well there was not ethereum back then...

Do you use different intensities for different algos? The default -i may be too high when combined with OC on some
algos.

Yes and it seems to be stabilizing... but I cant use my 2nd 970gtx. When I put that 2nd one in the system.. quark crashes. There is something else i'm looking into... why... cpu? bottle necking?  the build?  So I have a 970 setting doing nothing calling gigabyte and asrock maybe I can do something with the clocking. Im trying everything i can think of. On my 2 core but how many lanes does it have? ect  My other 2 core does the same on quark only.

There is a new driver out that seems to help... but.

[tbearhere]

NVIDIA Pascal Mythbusting

https://linustechtips.com/main/topic/462486-nvidia-pascal-mythbusting/

would still be nice to see one in action ...

even if it isnt '10x' the speed and all that ...

will get a chance soon i think ...

#crysx

This is all old... The unveil for Pascal was on Friday which someone else announced in the thread.

A 1070 is a bit faster then a Titan and a 1080 is a decent bit faster. Benchmarks aren't out yet.

Ok thx  Ben

[Amph]

ok, i've found how to reset to p2 state without rebooting your computer and without using inspector, just this nvidia-smi -rac

very useful, was a pain in the ass to reset each time i need to use other algo, which otherwise would result in a drivers crash with p0 state, like quark...probably due to the high overclock...

quark is the one with the most problems and then neoscrypt and lyra2v2 ... but quark is the worst.. even with no oc'ing.
There was a command in cuda mining  - C 0  - C 1  -C 2 but there not in ccminer and I don't know if that would help in the oc'ing or not.

yeah when i shift from ethereum to quark with the same oc(+150 core/mem) i always get a crash, qubit is the same, there is a bug on the miner maybe? i don't remember such thing with 750ti and i had 10 of those baby...well there was not ethereum back then...

Are you using a 2 core cpu?

yes g1840, is that a problem? and same thing as you, with one gpu only this does not happen, by the time i add the second it happen(even without riser, so it's not the riser the problem)

[tbearhere]

Maybe why I can't get 4 high end cards to work...not enough lanes on my cpu.

https://www.youtube.com/watch?v=rctaLgK5stA

This could be reason only for extremely high-bandwidth algo.
And high-bandwidth for gpu to RAM access

But I'm using a 2 core cpu with I don't know how many lanes that make a big difference I think on high end cards. I'm not sure.

[tbearhere]

Maybe why I can't get 4 high end cards to work...not enough lanes on my cpu.

https://www.youtube.com/watch?v=rctaLgK5stA

tanx tbear ...

im going to test this in the coming week with a more powerful cpu on the board ...

I don' think this applies to mining else we wouldn't be able to mine from x1 slots.

Mining  gaming  folding  doesn't it have to do as in the video with how many lanes a cpu has?

[tbearhere]

I'm looking now to see the cheapest 4 core cpu to get for the asrock.

You must be expecting a massive increase in hash to justify a new CPU? Isn't it more profitable to continue hashing on the existing equipment?

My plans where for this rig was 6x  980ti's  but I'm just trying to get 2x 980ti  and 2x 970 gtx running stable.

[tbearhere]

Also the amount of cache a cpu has could play a roll.
1150 CPUs only allow for 16 lanes of PCI Express lanes

[pallas]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

[tbearhere]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

What cards pallas?
And are you using ccminer latest build? On quark ?

[pallas]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

What cards pallas?

some 980ti and some 970

[tbearhere]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

What cards pallas?

some 980ti and some 970

And are you using sp ccminer latest build or yours?  On quark ?

[pallas]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

What cards pallas?

some 980ti and some 970

And are you using ccminer latest build or yours?  On quark ?

I've been using many different forks, sp and tpruvot based, depending on the algo.
Quark is sp's, used to crash a couple times a day once, but it has been pretty stable during the last months.
Of course they include some code of mine.
Maybe it has to do with the operating system (ubuntu in my case) or the fact it's a headless machine (not even X windows running).

[sp_]

Quark can mine for weeks in the latest build. (version 80.) The last bugs where fixed by t-nelson back in release 6x something.

[tbearhere]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

What cards pallas?

some 980ti and some 970

And are you using ccminer latest build or yours?  On quark ?

I've been using many different forks, sp and tpruvot based, depending on the algo.
Quark is sp's, used to crash a couple times a day once, but it has been pretty stable during the last months.
Of course they include some code of mine.
Maybe it has to do with the operating system (ubuntu in my case) or the fact it's a headless machine (not even X windows running).

Ok thx    I'm going try try to add the 970gtx in again and using -i 23 and the new drivers seems stable now.
It's been running for hrs and did a switch to quark algo.. no problem.
See what happens.  
ooo it is an asrock 81 btc  pro mb open air .... very cool here right now.
EDIT:when I add the 2nd 970 the stock clocks show 1418 the other 970 shows 1392.... i don't know why the difference and the same on the card i replaced 1418.

[AzzAz]

I'm running a 5 cards rig with a sempron (single core, very slow, very little cache).
All the algos work fine. I didn't try ether, but ccminer is just ok. Cpu load under 20%.

What cards pallas?

some 980ti and some 970

And are you using ccminer latest build or yours?  On quark ?

I've been using many different forks, sp and tpruvot based, depending on the algo.
Quark is sp's, used to crash a couple times a day once, but it has been pretty stable during the last months.
Of course they include some code of mine.
Maybe it has to do with the operating system (ubuntu in my case) or the fact it's a headless machine (not even X windows running).

Ok thx    I'm going try try to add the 970gtx in again and using -i 23 and the new drivers seems stable now.
It's been running for hrs and did a switch to quark algo.. no problem.
See what happens.  
ooo it is an asrock 81 btc  pro mb open air .... very cool here right now.
EDIT:when I add the 2nd 970 the stock clocks show 1418 the other 970 shows 1392.... i don't know why the difference and the same on the card i replaced 1418.

Do you have 2 version cards? Their bios is diferent? I have that same case with 970's and (!) with radeons 280x. MSI AB sets them diferently.

[sp_]

 I wonder if the new gtx 1070/1080 cards will beat the 980ti in the quark algo.

Here is sp-mod #6 private compute 5.2 unreleased:

[Amph]

they will for sure, their power target is much lower, one 1080 has a power target about the same as 970, and i doubt it will be lower, it's already comfirmed that they are faster

yes it was no mining or gaming, but there is no way that a new generation is lower than the previous one, never happened

[flipclip]

Another point is that different cards have different intensity limits so if you have different cards in the same rig
you will have to set it to a value that works on all cards which may not be optimum on some cards.

You can use comma separation to set different intensities per card. ie -i 27.2,25,24.9

[bensam1231]

SP are you actually working on a dual miner? ~_~

[socks435]

I wonder if the new gtx 1070/1080 cards will beat the 980ti in the quark algo.

Here is sp-mod #6 private compute 5.2 unreleased:

Looks like your speed is 30mh and how much you can mined altcoin with this speed.. are you mining ETH.?