T-Rex 0.26.8 ETHW, ETC (+dual mining) GPU miner

[chocoborider]

I am a little new, my 3090 asus strix is only doing about 92-95 MH/s, I am using t-rex-0.19.5-win-cuda11.1.zip on windows 10, is that the correct one?

Yes, CUDA 11.1 build is the only one that works with 3090. I'm assuming you're talking about ethash? You may want to try overclocking memory to achieve better results.

Gotcha thanks. I have it at 2000mhz GPU boost. Watching my clocks, it really only sits at 800-1700mhz when mining. My memory is locked at 19004 though while mining, it never fluctuates.

[DGBfanster]

Hi which build to use for 1080ti cards? CUDA 11, 10 or 9?

[trexminer]

Yeah, it turned that the problem for me was running windows 7. Updating to windows 10 fixed the problem. Now sometimes when driver fail, overclock settings from msi afterburner get reset, miner restart and start mining with default clocks.
Is there a way to make miner exit or set watchdog like in claymore to watch hashrate and if it drops below certain level - exit? CMD script will do restart thing afterwards.

There's no builtin hashrate monitoring at the moment. The miner can detect various errors and reboot the PC if needed, but low hashrate is not an error obviously. We'll see what we can do about that.

[trexminer]

Hi which build to use for 1080ti cards? CUDA 11, 10 or 9?

CUDA 10 should be fine

[bg17aw]

I have 1080ti and 1060 cards (mixed rig) on SimpleMining

I am getting this error:

Running miner t-rex-v0.19.5-cuda11.1
Options: -a ethash -o stratum+tcp://eth-eu1.nanopool.org:9999 -u wallet.name -p c=XVG -i 20
20201227 14:47:01 ERROR: Can't start miner, video driver is too old. Please update the video driver or use T-Rex compiled with CUDA v10.10 or older

How to update driver in SimpleMining? Thanks

Also, is -i 20 ok? Not sure what is best setting

Is -p c=XVG ok? Not sure what is best setting

If I switch to cuda10, I am still getting an error:

Running miner t-rex-v0.19.5-cuda10.0
Options: -a ethash -o stratum+tcp://eth-eu1.nanopool.org:9999 -u wallet.rigname -p c=XVG -i auto
20201227 14:58:36 ERROR: Can't start miner, stod
20201227 14:58:36 T-Rex finished.
Miner ended or crashed. Restarting miner in 30 seconds...

[alucard20724]

I have 1080ti and 1060 cards (mixed rig) on SimpleMining

I am getting this error:

Running miner t-rex-v0.19.5-cuda11.1
Options: -a ethash -o stratum+tcp://eth-eu1.nanopool.org:9999 -u wallet.name -p c=XVG -i 20
20201227 14:47:01 ERROR: Can't start miner, video driver is too old. Please update the video driver or use T-Rex compiled with CUDA v10.10 or older

How to update driver in SimpleMining? Thanks

you need to ask that in the simplemining thread.  this question is geared to the OS, not the miner.

https://bitcointalk.org/index.php?topic=1541084.0

[trexminer]

I have 1080ti and 1060 cards (mixed rig) on SimpleMining

I am getting this error:

Running miner t-rex-v0.19.5-cuda11.1
Options: -a ethash -o stratum+tcp://eth-eu1.nanopool.org:9999 -u wallet.name -p c=XVG -i 20
20201227 14:47:01 ERROR: Can't start miner, video driver is too old. Please update the video driver or use T-Rex compiled with CUDA v10.10 or older

How to update driver in SimpleMining? Thanks

Also, is -i 20 ok? Not sure what is best setting

Is -p c=XVG ok? Not sure what is best setting

If I switch to cuda10, I am still getting an error:

Running miner t-rex-v0.19.5-cuda10.0
Options: -a ethash -o stratum+tcp://eth-eu1.nanopool.org:9999 -u wallet.rigname -p c=XVG -i auto
20201227 14:58:36 ERROR: Can't start miner, stod
20201227 14:58:36 T-Rex finished.
Miner ended or crashed. Restarting miner in 30 seconds...

CUDA 10.0 build should work on your system. The error "Can't start miner, stod" is about "-i auto" argument passed in, it can't be "auto", so either set it to an integer, so better remove it altogether.

[bg17aw]

thanks, removing "i -auto" and using cuda10 solved the issues.

[th00ber]

Hi which build to use for 1080ti cards? CUDA 11, 10 or 9?

The 3090's Have thermal throtlle.
I had the exact same issue, but I solved it by adding a DIY backplate cooler (2 fans on the backplate).
So now it do not thermal throttle and I can get 110 MH/s with powerlimit to 82% and core +0 / mem +500

[dirton]

The miner can detect various errors and reboot the PC if needed, but low hashrate is not an error obviously.

Well if it could shutdown right after such error it would be great, now it just restarts with low hashrate after every single error like this:

Code:

ERROR: Can't find nonce with device [ID=0, GPU #0], cuda exception in [StreamContext<struct search_results,struct Ethash::KernelLaunchTag>::synchronize, 51], hardware stack error

[dlh1963ua]

tell me, how can i control a decrease of rig total hashrate?
maybe there is some parameter below which the rig will restart? such as in Claymore "-minspeed"

[socks435]

Had you been running Claymore with straps? If so, try "--mt 5" with t-rex, it will increase the hashrate.

Nice it improve my hashrate a bit compared to phoenix but the problem is the shares drop.
Any recommendation to increase shares or config for 1080ti? Tried to set it to intensity 20 and played around 19 to 22 but 20 is more stable good hashrate the only problem is the shares are drop.

[miner29]

Had you been running Claymore with straps? If so, try "--mt 5" with t-rex, it will increase the hashrate.

Nice it improve my hashrate a bit compared to phoenix but the problem is the shares drop.
Any recommendation to increase shares or config for 1080ti? Tried to set it to intensity 20 and played around 19 to 22 but 20 is more stable good hashrate the only problem is the shares are drop.

Shares are what you get paid for....unless solo mining...then its only blocks.

[BogdanCo]

Hello, having an issue with T-rex mining ETH on a rig with 6x 1080ti:

Code:

20210101 02:04:08 GPU #1: Gigabyte GTX 1080 Ti - 53.71 MH/s, [T:54C, P:82W, F:68%, E:271kH/W], 1770/1783 R:0.73%
20210101 02:04:08 GPU #2: Gigabyte GTX 1080 Ti - 53.52 MH/s, [T:53C, P:83W, F:71%, E:270kH/W], 1790/1799 R:0.5%
20210101 02:04:08 GPU #3: Gigabyte GTX 1080 Ti - 44.16 MH/s, [T:53C, P:90W, F:54%, E:222kH/W], 1764/1773 R:0.51%
20210101 02:04:08 GPU #4: Gigabyte GTX 1080 Ti - 53.72 MH/s, [T:53C, P:78W, F:61%, E:273kH/W], 1738/1744 R:0.34%
20210101 02:04:08 GPU #5: Gigabyte GTX 1080 Ti - 53.70 MH/s, [T:53C, P:77W, F:66%, E:273kH/W], 1724/1731 R:0.4%
20210101 02:04:08 Hashrate: 312.54 MH/s, Shares/min: 2.949 (Avr. 3.434) R: 0.529%, Avr.P: 1187W, Avr.E: 263kH/W
20210101 02:04:08 Uptime: 2 days 3 hours 9 mins 39 secs | Algo: ethash | T-Rex v0.19.5
20210101 02:04:08 WD: 2 days 3 hours 9 mins 41 secs, shares 10540/10596
20210101 02:04:08
20210101 02:04:08 GPU #0: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #1: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #5: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #3: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #4: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #2: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:36 WARN: WATCHDOG: T-Rex has a problem with GPU, terminating...

From time to time it hangs and then nothing happen... it stays like this for hours until I have to manually restart the rig. Isn't there a watchdog who supposed to restart it automatically when an error occur? Thanks!

[socks435]

Hello, having an issue with T-rex mining ETH on a rig with 6x 1080ti:

Code:

20210101 02:04:08 GPU #1: Gigabyte GTX 1080 Ti - 53.71 MH/s, [T:54C, P:82W, F:68%, E:271kH/W], 1770/1783 R:0.73%
20210101 02:04:08 GPU #2: Gigabyte GTX 1080 Ti - 53.52 MH/s, [T:53C, P:83W, F:71%, E:270kH/W], 1790/1799 R:0.5%
20210101 02:04:08 GPU #3: Gigabyte GTX 1080 Ti - 44.16 MH/s, [T:53C, P:90W, F:54%, E:222kH/W], 1764/1773 R:0.51%
20210101 02:04:08 GPU #4: Gigabyte GTX 1080 Ti - 53.72 MH/s, [T:53C, P:78W, F:61%, E:273kH/W], 1738/1744 R:0.34%
20210101 02:04:08 GPU #5: Gigabyte GTX 1080 Ti - 53.70 MH/s, [T:53C, P:77W, F:66%, E:273kH/W], 1724/1731 R:0.4%
20210101 02:04:08 Hashrate: 312.54 MH/s, Shares/min: 2.949 (Avr. 3.434) R: 0.529%, Avr.P: 1187W, Avr.E: 263kH/W
20210101 02:04:08 Uptime: 2 days 3 hours 9 mins 39 secs | Algo: ethash | T-Rex v0.19.5
20210101 02:04:08 WD: 2 days 3 hours 9 mins 41 secs, shares 10540/10596
20210101 02:04:08
20210101 02:04:08 GPU #0: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #1: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #5: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #3: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #4: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #2: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:36 WARN: WATCHDOG: T-Rex has a problem with GPU, terminating...

From time to time it hangs and then nothing happen... it stays like this for hours until I have to manually restart the rig. Isn't there a watchdog who supposed to restart it automatically when an error occur? Thanks!

What version did you use cuda 10 or 11?
Switch to cuda 10 to fix this issue.

Hey... can you share your config/cc it seems that you get almost 90watts with 53MH/s what is your current driver for gtx 1080ti?

I only get around 41mh/s with 150watts.

[BogdanCo]

Hello, having an issue with T-rex mining ETH on a rig with 6x 1080ti:

Code:

20210101 02:04:08 GPU #1: Gigabyte GTX 1080 Ti - 53.71 MH/s, [T:54C, P:82W, F:68%, E:271kH/W], 1770/1783 R:0.73%
20210101 02:04:08 GPU #2: Gigabyte GTX 1080 Ti - 53.52 MH/s, [T:53C, P:83W, F:71%, E:270kH/W], 1790/1799 R:0.5%
20210101 02:04:08 GPU #3: Gigabyte GTX 1080 Ti - 44.16 MH/s, [T:53C, P:90W, F:54%, E:222kH/W], 1764/1773 R:0.51%
20210101 02:04:08 GPU #4: Gigabyte GTX 1080 Ti - 53.72 MH/s, [T:53C, P:78W, F:61%, E:273kH/W], 1738/1744 R:0.34%
20210101 02:04:08 GPU #5: Gigabyte GTX 1080 Ti - 53.70 MH/s, [T:53C, P:77W, F:66%, E:273kH/W], 1724/1731 R:0.4%
20210101 02:04:08 Hashrate: 312.54 MH/s, Shares/min: 2.949 (Avr. 3.434) R: 0.529%, Avr.P: 1187W, Avr.E: 263kH/W
20210101 02:04:08 Uptime: 2 days 3 hours 9 mins 39 secs | Algo: ethash | T-Rex v0.19.5
20210101 02:04:08 WD: 2 days 3 hours 9 mins 41 secs, shares 10540/10596
20210101 02:04:08
20210101 02:04:08 GPU #0: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #1: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #5: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #3: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #4: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:08 GPU #2: generating DAG 4.01 GB for epoch 385 ...
20210101 02:04:36 WARN: WATCHDOG: T-Rex has a problem with GPU, terminating...

From time to time it hangs and then nothing happen... it stays like this for hours until I have to manually restart the rig. Isn't there a watchdog who supposed to restart it automatically when an error occur? Thanks!

What version did you use cuda 10 or 11?
Switch to cuda 10 to fix this issue.

Hey... can you share your config/cc it seems that you get almost 90watts with 53MH/s what is your current driver for gtx 1080ti?

I only get around 41mh/s with 150watts.

Cuda 10 I'm using, as far as I know cuda 11 is only for newer NVidia cards.... and it's not working for 1080ti. However I have 10 rigs with 6x 1080ti, only one of them is making this kind of issues.

Config: -a ethash -o stratum2+tcp://daggerhashimoto.eu.nicehash.com:3353 -u [wallet]  -w $rigName --kernel 0 --mt 6

OC:  core +100 / +125 depends of the card;
        memory +600
        PL 210W
and the OhGodAnETHlargementPill activated.

It's not 90 Watts, this is the SW readings.... I get around 1400w at wall for the whole rig (6x1080ti) so I would say 220W for one 1080ti.

[dlh1963ua]

From time to time it hangs and then nothing happen... it stays like this for hours until I have to manually restart the rig. Isn't there a watchdog who supposed to restart it automatically when an error occur? Thanks!

I have the same problem...
That's why I asked above for the mechanism of restart when the total hash rate decreases.
I understand that this is a watchdog task, but even using "--watchdog-exit-mode" does not help ((

[chup]

Cuda 10 I'm using, as far as I know cuda 11 is only for newer NVidia cards.... and it's not working for 1080ti. However I have 10 rigs with 6x 1080ti, only one of them is making this kind of issues.

Config: -a ethash -o stratum2+tcp://daggerhashimoto.eu.nicehash.com:3353 -u [wallet]  -w $rigName --kernel 0 --mt 6

OC:  core +100 / +125 depends of the card;
        memory +600
        PL 210W
and the OhGodAnETHlargementPill activated.

It's not 90 Watts, this is the SW readings.... I get around 1400w at wall for the whole rig (6x1080ti) so I would say 220W for one 1080ti.

What driver version? I don't think You need OhGod and --mt option , they basically do the same.
I believe that using Nicehash is more stressful for Your rigs because of more often DAG change and creation. Try pointing problematic rig to Eth pool to see how it behaves.

[BogdanCo]

Cuda 10 I'm using, as far as I know cuda 11 is only for newer NVidia cards.... and it's not working for 1080ti. However I have 10 rigs with 6x 1080ti, only one of them is making this kind of issues.

Config: -a ethash -o stratum2+tcp://daggerhashimoto.eu.nicehash.com:3353 -u [wallet]  -w $rigName --kernel 0 --mt 6

OC:  core +100 / +125 depends of the card;
        memory +600
        PL 210W
and the OhGodAnETHlargementPill activated.

It's not 90 Watts, this is the SW readings.... I get around 1400w at wall for the whole rig (6x1080ti) so I would say 220W for one 1080ti.

What driver version? I don't think You need OhGod and --mt option , they basically do the same.
I believe that using Nicehash is more stressful for Your rigs because of more often DAG change and creation. Try pointing problematic rig to Eth pool to see how it behaves.

Driver version 430.40 on linux. You're right that mining to an ETH pool is more stable and as far as I remember never has this problem mining on ETH pools... however, my question here is why the rig is not restarting since the Watchdog was triggered?
If it's restarting, then i have no problem, loosing only some seconds to mine, but if it hangs up then it burns electricity for 3-4-5 hours maybe....

[socks435]

Driver version 430.40 on linux. You're right that mining to an ETH pool is more stable and as far as I remember never has this problem mining on ETH pools... however, my question here is why the rig is not restarting since the Watchdog was triggered?
If it's restarting, then i have no problem, loosing only some seconds to mine, but if it hangs up then it burns electricity for 3-4-5 hours maybe....

That's pretty old driver I recently updated my driver from 450 to 455 Nvidia version.
Are you paying cheap Electricity? I can't able to reach the 52Mh/s but I got around 49 with your setup but it's not profitable I'm wasted too much power compared to my setup with 44mh/s t-trex just got some idea from your setting and it works with PL: +160w

I more than happy if I can get around 48mh/s with PL:+160