Author

Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 582. (Read 2347641 times)

legendary
Activity: 1510
Merit: 1003
gtx1070 319mhs lbry
https://goo.gl/photos/HwCU4255tVModkpz8

Damn slow on my poor gtx750.

legendary
Activity: 1512
Merit: 1000
quarkchain.io
Which intensity is best for Maxwells with mod#3 ?
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
I'm only interested in same-power comparisons.
Can someone please test sp-mod and tpruvot on low TDP (2/3 or 3/4 max), to have maximum efficiency and no throttle?

My testcard has a tdp of 150W. the screenshot was taken with the 372.70 driver and windows 8.1 (-500 on the memory clock)

Sp-mod LBRY #3 on a gtx 1070 with -i 29

Messured with a meter to 147Watt in the wall. (inno3d gtx 1070) default 150w tdp



member
Activity: 82
Merit: 10
302 mh on the 980ti wasnt stable for more than 4 hours. I think the core was near 1600. Still able to get 297 stable I think though. Might be able to increase intensity from 26 to 27 and get 300 stable.
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
#3


How fast is the opensource with the same settings on the same cards?

Can you hash @ 191 MHASH on the gtx 970 with the opensource?
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
_sp_ please provide a screenshot with miner and gpu-z/overclocking utility to support you're claims,...because we have no way of validating you're results prior too purchase

Scroll back a few threads and you see screenshots reported by many users.

Windows 10 has issues with high intensities..

Using -i 29 on the gtx 1070 and the 980ti only works in windows 7 and 8.1
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer

hashing well on w 8.1

How fast is the opensource with the same settings on the 980ti?

Can you hash @ 300 MHASH on the gtx 980ti with the opensource?
member
Activity: 130
Merit: 10
I get 20% increase on this card with the latest drivers and release #3. Tested with 6 cards in the rig. 24 hours of mining.

http://www.gainward.com.tw/main/vgapro.php?id=954&lang=en

_sp_ please provide a screenshot with miner and gpu-z/overclocking utility to support you're claims,...because we have no way of validating you're results prior too purchase
legendary
Activity: 2716
Merit: 1094
Black Belt Developer
I'm only interested in same-power comparisons.
Can someone please test sp-mod and tpruvot on low TDP (2/3 or 3/4 max), to have maximum efficiency and no throttle?
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
i never seen a 970 so low Wink stop saying crap
legendary
Activity: 1134
Merit: 1001
SP_

I think that if you take away 0.10 btc tax, you make it free, and use tax 2% for each miner .

It is much better, and you make more profit, many people in more send to you 2% profit .

sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
I get 20% increase on this card with the latest drivers and release #3. Tested with 6 cards in the rig. 24 hours of mining.
http://www.gainward.com.tw/main/vgapro.php?id=954&lang=en
strange
I try latest 372.70 drivers and give only 1-2% increase: on last 1.8.1 tpruvot miner 149 mh, on your #3 release 151-153 mh.
And 368.69 drivers give 161 mh (~8% more).
Which intensity you set for 970 and 1070 ?

Default intensity

The gainward trottle and only give 120MHASH with the opensource. and around 145 with release #3. (default intensity is -i 23)
This card has a reference cooler and a low clock (1178 MHz (boost) / 1051 MHz (base))
full member
Activity: 304
Merit: 100
I get 20% increase on this card with the latest drivers and release #3. Tested with 6 cards in the rig. 24 hours of mining.

http://www.gainward.com.tw/main/vgapro.php?id=954&lang=en

strange
I try latest 372.70 drivers and give only 1-2% increase: on last 1.8.1 tpruvot miner 149 mh, on your #3 release 151-153 mh.
And 368.69 drivers give 161 mh (~8% more).

Which intensity you set for 970 and 1070 ?
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
I get 20% increase on this card with the latest drivers and release #3. Tested with 6 cards in the rig. 24 hours of mining.

http://www.gainward.com.tw/main/vgapro.php?id=954&lang=en
full member
Activity: 304
Merit: 100


If you have a gtx 970, 980 or 980ti card (compute 5.2 maxwell) the increase is 20-25% on the standard clocks.(release #3).
I need more time to increase the pascal speed. Next up is a speedup for the gtx 1060. I have a testcard comming soon.

which drivers give 20-25% increase ?
I receive only 10% on release #3 for 970 MSI & ASUS
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
I only have a 1070 card in my testrig. With the opensource i can reach 275-280mhash with oc. My private kernel oc around 100mhz higher than the opensource so you get an additional 5% from the oc. 305-310mhash stable.
Receive a 5% increase and give 3.3% + 0.1 BTC - it's not fair.

If you have a gtx 970, 980 or 980ti card (compute 5.2 maxwell) the increase is 20-25% on the standard clocks.(release #3).
I need more time to increase the pascal speed. Next up is a speedup for the gtx 1060. I have a test card coming soon.

You can get 10% more if you know how to configure your rig properly. I have written many posts about this but I can write it again..

-Reduce the memclock (with nvinspector you can reduce -1000mhz, with a bios mod you can reduce to 850mhz or lower)You can use the nvidia-smi tool to get the supported clocks mem/core and use the lowest value for the memory
-Overclock the core to 2050-2100 or more. When you reduce the speed of the memory you free power that can be used to overclock. Since my kernel doesn't use memory. 0% load on the memory controller makes +100mhz stable oc on the core possible.
newbie
Activity: 25
Merit: 0
I only have a 1070 card in my testrig. With the opensource i can reach 275-280mhash with oc. My private kernel oc around 100mhz higher than the opensource so you get an additional 5% from the oc. 305-310mhash stable.

Receive a 5% increase and give 3.3% + 0.1 BTC - it's not fair.

You see, removing the initial payment of 0.1 BTC, you will attract more people who will bring you income. For those who have 1-2 cards in 1070 are not ready to buy your miner. Payback tends to infinity.

You will not hear the angry responses of those who gave money, and who were not satisfied with the speed of your miners. Using the "as is" - you will not reproach.

Claymore has made the right decision, and it is a success. You will stagnate.

Do 2% with no down payment, and earn more.

You are a good programmer, but a bad marketing.


PS. sorry for my English.
legendary
Activity: 3248
Merit: 1070

Hey look you regurgitated the part about intensity I posted a couple days ago... Good job bro and you have no idea what intensity actually does. Higher intensity does not mean you'll have better hashrate. Higher intensity can actually produce worse hashrate and use more power.

It's not a magic value too, where you increase intesnity by 5 and it's magically that much better. Your lack of understanding of... well... anything is astounding and you try to use it as 'proof' something is working wrong. Reminds me of the guy complaining that Wolf0 broke his rigs because he purchased a miner that utilized his GPUs better and produced more heat while drawing more wattage... then pretended like his cards would always use the same wattage.

If you don't know what you're talking about, don't act like you do.


You can't do math. What Yiimp lists as actual is what it paid for actual MH/BTC in the last day.

this guy is funny, keep posting retarded sentence, like regurgitating, no what i mean about intensity is that he changed how intensity work, by assigning it other value, but as usual you didn't understood shit, and i was hoping for a code rewrite instead....

and yes you can do the math about any coin, only because you don't know how to do it, it does not mean that it can not be done, please stay away with your ignorance is infecting me
sp_
legendary
Activity: 2954
Merit: 1087
Team Black developer
I've tested a lot of Pascals today with SP_mod#3.
The approach is only 5-7% , according to several different models I have:
1080 FE - 6 cards rig from 330 to 365MH
Palit 1080 GameRock ( Without Premium) - 6 cards rig - from 365 to 384-7MH
Palit 1080 SJS  - 6 cards rig - from 370 to 398 MH
Palit 1070 SJS - 6 cards rig - from 288  to 301 MH

So guess I'll be waiting for this 12% gain ...

I only have a 1070 card in my testrig. With the opensource i can reach 275-280mhash with oc. My private kernel oc around 100mhz higher than the opensource so you get an additional 5% from the oc. 305-310mhash stable. I don't have a 1080  card to test
legendary
Activity: 1764
Merit: 1024

There is no 'maximum' we should all be doing the same speed on the same kernels at the same clocks as long as we aren't limited by TDP. This is why your logic is bat shit insane, you don't even have any idea how mining actually works. There is no way for someone to do faster with the same hardware unless their clocks/tdp/drivers/OS are different and in weird cases memory types (such as with Ethereum). Most cases Windows 7/8 are all the same, occasionally W10, but usually it's just slower.

Which is why on the last page rednow corrected himself and fell inline with speeds everyone else is getting. The only other thing that could possible effect it is intensity, but default intensity is almost always spot on.

This completely puts aside that you're comparing apples to oranges. rednoW isn't magically going to get the same speed as another guy on the other page running at 290Mhs, nor is it magically going to be 5% higher unless everything I posted above is the same, but it isn't... Mentioning hashrate without clocks or other information is dumb... even more so when people try to 'pad' their scores.


i know that you actually shot crap on purpose just to talk with someone, but you are again wrong on everything, if someone can do 290 with heavy oc, this must be considered as example, and the % or boost should be added on top of that

not on the lowest possible, that would be dumb like your post, so please go ahead and stop with this bullshit propaganda that you are doing


Oh yeah? Because he owns a copy of the program? You're comparing apples to oranges dude. You can't do that in comparisons. You were comparing someone with a heavy OC to someone without one... Specifically the guy you mentioned didn't even post his OC, just the 'results' he's getting.

what i mean is, that he is not redoing the code, but just tweaking intensity, for paying fee plus donation i was expecting a code redo to boost pascal


Hey look you regurgitated the part about intensity I posted a couple days ago... Good job bro and you have no idea what intensity actually does. Higher intensity does not mean you'll have better hashrate. Higher intensity can actually produce worse hashrate and use more power.

It's not a magic value too, where you increase intesnity by 5 and it's magically that much better. Your lack of understanding of... well... anything is astounding and you try to use it as 'proof' something is working wrong. Reminds me of the guy complaining that Wolf0 broke his rigs because he purchased a miner that utilized his GPUs better and produced more heat while drawing more wattage... then pretended like his cards would always use the same wattage.

If you don't know what you're talking about, don't act like you do.

Look. Each gpu is different. I need to write seperate kernals for each of the series. This is just release #3 and compute 5.0 is using the 5.2 kernel.

Yes we all know that, but then don't advertise as 5-7% on compute 5.0 cards if it's simply not true.
Increase it to levels of 20-30% then people will be interested. FYI I have one 1060 incoming, so I'm interested into that card as well.

Having 0,1 btc subscription and a 3,3% fee it's too much, have one or another but whatever you do, you have to increase the performance of you're miners.

Nothing is false advertising... He advertised 5-7%, that's what people are getting... You want better performance, that does not make what people are getting wrong.

all pool payments for lbry is completely fucked, based on a simple math should be way higher, block reward is at 350 now, the diff mean shit basically

You can't do math. What Yiimp lists as actual is what it paid for actual MH/BTC in the last day.
Jump to: