CryptoDredge 0.27.0 — CUDA Miner

[KartingProfi]

0.9.6 for Windows CUDA 10.0 - where or when? Thanks.

[1713953758]

1713953758

[1713953758]

1713953758

[dragonmike]

EVGA 1080 SC at +100/+800 with ETHlargement pill: 913 h/s. Impressive. At 85% PL. The entire rig (5 cards) draws only 550W from the wall. That's insane.

Gonna have to see if I can push the core higher and gain some more hashrate. But as it stands, it's more efficient than Polaris cards at CN heavy. Actually very close to Vega even.

[Saphire]

I want to share my hashrates using CryptoDredge 0.9.6 miner for X22i algo

GTX 1070 (+50 Core, -300 Mem)
- MSI GTX 1070 Gaming OC - 8.2 MH/s
- ASUS GTX 1070 Stirx OC - 7.7MH/s
- Gigabyte GTX 1070 OC - 7.6MH/s

GTX 1070 Ti (+100 Core, -300 Mem)
- Gigabyte GTX 1070 Ti OC - 8.7 MH/s

GTX 1080 Ti (+100 Core, -300 Mem)
- MSI GTX 1080 Ti Trio - 14 MH/s
- MSI GTX 1080 Ti Gaming - 12.5 MH/s
- ASUS GTX 1080 Ti Strix - 13 MH/s
- ASUS GTX 1080 Ti Turbo - 11.5 MH/s

[Ragnarok01]

Hi dev,
how to make the configuration file (pools) work with the .bat file?
If in .bat file pool is disconnected, then program accept config pools?

[laik2]

Hello CryptoDredge developers,
I guess your miner inherits ccminer slow api response, could you take a look and possibly extend API to not report power/temp/fan via some param of your choice, on 6+ GPUs requesting `threads` takes at least 10+ seconds.

Thanks.

[peteycamey]

why there is no x22i version (0.9.6) link on the main page?

[Q_R_V]

There is. Just choose "Mirror on Github".

[IvanZib]

wich CUDA I need with win_64 GTX1060 6gb 9.2 or 10.0 ?

wich CUDA I need with win_64 GTX1080 9.2 or 10.0 ?

[Juniormkl]

CryptoDredge 0.9.7 Added on https://minermonitoring.com

Join our Discord: https://discordapp.com/invite/6HjxYQw

[xlockdownx]

Can Anyone help me ?

i Have this Problem with algo Phi2   

When i had drivers 399.24  with 6 1080ti's i was hitting 68 MH/s

Now That i have download  latest Driver 416.34  i'm at 57 MH/s

That is a Big lose for me.
i dont want to swith back to 399.24 because of Cuda 10 ( Would have to make other Changes )
PLS Help .

Thx

[kamak]

Up-Time Miner Stats Readout

Could you please implement an Up-Time Indicator within the console readout at the GPU stats interval points. This helps with quick calculations of shares per time, and the ability to see if the miner has restarted due to errors.....etc

[IvanZib]

Can Anyone help me ?

i Have this Problem with algo Phi2   

When i had drivers 399.24  with 6 1080ti's i was hitting 68 MH/s

Now That i have download  latest Driver 416.34  i'm at 57 MH/s

That is a Big lose for me.
i dont want to swith back to 399.24 because of Cuda 10 ( Would have to make other Changes )
PLS Help .

Thx

have the same problem with 5x1080

temporarily switched to old drivers (399)

[bitverzo]

Hi

Here is the new NVIDIA coin calculator. Chek this url: https://bitverzo.com/mining/nvidia

[niask]

[17:34:31] INFO  - 568/2 Rejected : diff=6.92048, reason: Job not found.
Is this can be a sign of an excessive overclock?

[Cryptotourist]

Hello,

I'm always reluctant when it comes to updating NVDIA drivers - or any drivers for that matter.
If everything is working as expected, I don't, until it comes down to the last thing to do.

I have a rig with 1080 ti's, CUDA 9.1, and NVIDIA driver 390.65, running on windows 10 64bit, using CryptoDredge 0.9.6 <3
I understand that these algos are not memory sensitive and bla bla, but...

I get a super stable performance of 14.5 MH/s on each ti (MSI GTX Trio) on average, and 63-64 KH/W, mining x22i.

Core: +100
Mem: +750
Plimit: 80%

Also running an instance of x16r with t-rex 0.7.2 on the same rig, and both algos are loving it

[17:34:31] INFO  - 568/2 Rejected : diff=6.92048, reason: Job not found.
Is this can be a sign of an excessive overclock?

No, less than 1% rejected shares in general is considered to be normal. That said, I just checked and I have none! 
Maybe tweaking your rig gave you the bad shares - or not.

[Techgtv]

I'm having trouble getting the miner stable on CN-Heavy.

I have my 12x 1050 TI rigs running stable on +175+750 (clock/mem) 72% PL ( same as 1050TI with 6 pin power at 60% PL ), but i cannot for the life of me figure out why some of my 1070 TI Rigs, 12X 1070 TI, are just not hashing at all, it seems none of them want to hash higher then 8KH/s as a total rig.

I can leave the OC completely off, or my original OC that i had, makes no difference

Windows 10 PRO, latest Nvidia Cuda 10 drivers, 82 GB swap file

Any insight?

[Cryptotourist]

Just a guess at this point, try CUDA 9.1 or 9.2 

[Ragnarok01]

Dear dev,
I really look forward to the function of the failover pool in pools.txt file!

[dragonmike]

Hey Cryptodredge, AMD devs have upped their game. Now Polaris cards hash CN_haven just as fast as a GTX 1080 with the pill... Is there more performance to be extracted for nVidia on heavy-type algos? Or is it just a question of boosting power limit and crank up clocks?

[vrm42]

Hi, how do you get API to work? I tried web socket mode and I got "connection reset" in the browser. I try tcp and browser is unable to connect.
I tried both 127.0.0.1:4068 and 0.0.0.0:4068, all the same. I also tried locally. I use cuda 10 drivers and tried 9.7 version of cryptodredge and the new 10 version. Thanks!