- Restored and added performance on gtx 1060 / 970
- gtx 1070 31MHASH++
- Fixed exit up on crash
- Included 64 bit build
My testrig is up 6% from signatum sp-mod #4.
Sending out now.
Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 256. (Read 2347601 times)
I am working on signatum #5 with some fixes and improvements.
Thanks SP_
I am working on signatum #5 with some fixes and improvements.
hi SP_ "gpux: unspecified launch failure " about this error, is it got to do with to low tdp ?
running on 1080ti
tdp - 70%
core + 90
and mem -400
i'll try with more settings. to see.
will try core + 100 to see if it helps stabilize a bit
running on 1080ti
tdp - 70%
core + 90
and mem -400
i'll try with more settings. to see.
will try core + 100 to see if it helps stabilize a bit
I saw an overall 5% jump on my 1080s and 1080 TIs.
The biggest jump in on my non FE 1080s. all are solid 37 now.
Thanks SP
The biggest jump in on my non FE 1080s. all are solid 37 now.
Thanks SP
Still running with 60% tdp?
wasn't this new mod supposed to be good for low tdp? just tried , it's indeed faster on lower tdp i get 0.6MH more hash on each card
On my 1080Ti rig, mod4 does give a little boost and it's happy.
On my 1060 rig, I have to reduce the intensity to 20 or it would crash after 4-5mins with something about skein512 cpu not validate blablabla I don't remember. I would just leave it 20 intensity. Not a big deal on this.
On my 1070 rigs, mod4 does give a little boost but is unstable. It would just stop mining after like 2 mins, like stuck at receiving job but doesn't crash and api will start reporting crazy hashrate like "1234GHs".
If I use Mod3, they are all stable and fine.
All cards run at 100%TDP, also tested different core and mem clock, stock/oc/underclock, it would still stop mining.
All using latest driver (the one mentioned in the email)
Can you fix it?
Also, I don't have to pay electricity so i don't really care about the power consumption, but if you would make a power saving version, I would be happy to hear less from the fans
On my 1060 rig, I have to reduce the intensity to 20 or it would crash after 4-5mins with something about skein512 cpu not validate blablabla I don't remember. I would just leave it 20 intensity. Not a big deal on this.
On my 1070 rigs, mod4 does give a little boost but is unstable. It would just stop mining after like 2 mins, like stuck at receiving job but doesn't crash and api will start reporting crazy hashrate like "1234GHs".
If I use Mod3, they are all stable and fine.
All cards run at 100%TDP, also tested different core and mem clock, stock/oc/underclock, it would still stop mining.
All using latest driver (the one mentioned in the email)
Can you fix it?
Also, I don't have to pay electricity so i don't really care about the power consumption, but if you would make a power saving version, I would be happy to hear less from the fans
I saw an overall 5% jump on my 1080s and 1080 TIs.
The biggest jump in on my non FE 1080s. all are solid 37 now.
Thanks SP
The biggest jump in on my non FE 1080s. all are solid 37 now.
Thanks SP
Still running with 60% tdp?
OMG, please, do something with tdp, The card will die with this mod @100% power. They working under 75C+, this is not what they want.
75C is not a problem. 85C is a problem The cards will trottle and hash is lost.
Hmm seems it doesn't work as smooth as sp#3.
Same settings but this time I'm getting "gpux: unspecified launch failure " after a short period of time.
Hope it will work more stable after I reduce the overclocking settings... but strange at all thought the settings couldn't stay the same.
@SP
Could you force a restart if an error occured like "gpux: unspecified launch failure "?
Right now it's an endless loop...
same hereSame settings but this time I'm getting "gpux: unspecified launch failure " after a short period of time.
Hope it will work more stable after I reduce the overclocking settings... but strange at all thought the settings couldn't stay the same.
@SP
Could you force a restart if an error occured like "gpux: unspecified launch failure "?
Right now it's an endless loop...
how much was your memory ? mine was set to -450
gtx 1080ti aorus extreme
memory clock: - 500
core clock: + 120
TDP: 65%
all the same gpus' and it was working with sp#3.
guess I need to switch back to sp#3 need the restart function if an issue occured.
In sp#4 it's just an endless loop if an issue occured.
"gpux: unspecified launch failure "
"gpux: unspecified launch failure "
"gpux: unspecified launch failure "
What is your aorus hashrate? I can't oc over +60 core but in TDP 100% I have 48-49 Mh. #SP3
nexus 7 sp mod - I have set intensity for all cards to 2 , now it's not crashing because cpu usage and the hashrate was lowered just a bit, but 12 hours and no block with a bunch of 1060, do you see the actual hashrate when you use low intensity? thanks
OMG, please, do something with tdp, The card will die with this mod @100% power. They working under 75C+, this is not what they want.
Skunk sp-mod #4
New World record hashrate on the gtx 1070
New World record hashrate on the gtx 1070
i did not receive anything on my mail, but what tdp you are using for that speed?
is there a built in 2% dev fee + miner purchase
or just 1 of the above ?
or just 1 of the above ?
The opensource still does 19MHASH on the gtx 1070. Sp-mod #4 is 60% faster than the opensource. And around 10% faster than the palgin binary (Released 3 days after sp-mod #3)
My work was stolen and spread. What do you expect me to do??
My work was stolen and spread. What do you expect me to do??
I don't know, honor your promises?
That's just me though...
#4 has actually caused a decrease in my output. I went from 315MH/s to 307 with #4, same settings. I thought #4 was supposed to work better at lower TDP? Which you said repeatedly.... Oh well...
Glad I held onto my rewards until the price re-peaked at .00001500, at least it sort of helped me get closer to being in the black.
Also... why is this thing using 100% of my CPU? around 75% for ccminer and 25% for System?
EDIT: Even using recommended settings 100% power, -500Mem +100 (anything faster crashes) I still get better numbers with spmod3, 344MH/s with spmod3, 333 MH/s with spmod4. I have 6x1080Ti and 1x1070, 1080Ti's are getting roughly 53Mh/z and 1070 28MH/s.
Signatum #4 has now been sendt to the donators.
check your spam inboxes.
Unfortunately no speed increase at 1060 cards. Pity check your spam inboxes.
In 24hours I will release a new mod with 0.1BTC buy in and 2% fee. The algo is very profitable with my modded kernel.
Just to vouch for SP. I purchased the sigt mod at the very beginning. With 21 - 1080 Ti, I made $7000 mining 152,000 sigt.
His miner was 30% faster at 70% TDP, so his mod contributed $2100 in earnings.
0.1 BTC, $268 might be a bit steep for some people, especially if it's a new coin. We wont know how profitable it will be until it hits the exchange.
Sigt definitely paid off though. Even if I cost 0.1 BTC, I personally will still be purchasing it.
You made $2100 with 21 cards. Some of my buyers have more than 1000 cards.
Also why is mod4 compiled with CUDA 7.5, isnt CUDA 8 better?
I saw an overall 5% jump on my 1080s and 1080 TIs.
The biggest jump in on my non FE 1080s. all are solid 37 now.
Thanks SP
The biggest jump in on my non FE 1080s. all are solid 37 now.
Thanks SP
lol what? World record? You're kidding right?
https://i.gyazo.com/dfaebe590821806de4869310efe544b4.png
Nice try though
edit: And this is on SP mod 3... Was getting the same on the CWI miner. Send me SP mod 4 so I can see what kind of boosts it really gives... For all we know, you just OC'd the shit out off your card.
Hmm seems it doesn't work as smooth as sp#3.
Same settings but this time I'm getting "gpux: unspecified launch failure " after a short period of time.
Hope it will work more stable after I reduce the overclocking settings... but strange at all thought the settings couldn't stay the same.
@SP
Could you force a restart if an error occured like "gpux: unspecified launch failure "?
Right now it's an endless loop...
same hereSame settings but this time I'm getting "gpux: unspecified launch failure " after a short period of time.
Hope it will work more stable after I reduce the overclocking settings... but strange at all thought the settings couldn't stay the same.
@SP
Could you force a restart if an error occured like "gpux: unspecified launch failure "?
Right now it's an endless loop...
how much was your memory ? mine was set to -450
gtx 1080ti aorus extreme
memory clock: - 500
core clock: + 120
TDP: 65%
all the same gpus' and it was working with sp#3.
guess I need to switch back to sp#3 need the restart function if an issue occured.
In sp#4 it's just an endless loop if an issue occured.
"gpux: unspecified launch failure "
"gpux: unspecified launch failure "
"gpux: unspecified launch failure "
Same error here. We need quick fix so it can restart if get this error.
Also why is mod4 compiled with CUDA 7.5, isnt CUDA 8 better?
Jump to: