Pages:
Author

Topic: T-Rex 0.26.8 ETHW, ETC (+dual mining) GPU miner - page 64. (Read 174982 times)

member
Activity: 283
Merit: 63
Did an update to latest nVidia drivers and now t-Rex is not working.  Here's the error message it gives - "Can't start miner, unable to get number of CUDA devices, is nvidia driver installed?"

My install went fine and it was work fine before the install.  I suspect this latest version (419.67) for Windows 10 is not compatible with this latest version on t_Rex.

I have 3 Windows 10 rigs, and I have only updated this one.  The other 2 are working (well creeping along really slow due to such high difficulty levels).

I can install the earlier version of nVidia drivers, but thought you might want to take a look at whats causing this.

-Rodger

419.67 should be working fine. Someone from Discord had the same issue and then said it was fixed by "properly" installing the new drivers. Not sure what that means but I suspect the driver update screwed something up in the system so they had to clean everything up and reinstall the drivers.
sr. member
Activity: 658
Merit: 250
I'm developing a custom stratum server that seems to be working fine now (miners submit valid work), but t-rex is outputting these messages constantly:

WARN: New iteration find shares for device [ID=0, GPU #0]: job_id=2e61

What could be the cause? I don't get that connecting to normal stratum servers.

EDIT: Sorry, I realized there were extra warnings too with protocol debugging enabled. That's happening on other servers, too. Seems like it warns about receiving the same notify twice, is that somehow bad?
newbie
Activity: 7
Merit: 0
Did an update to latest nVidia drivers and now t-Rex is not working.  Here's the error message it gives - "Can't start miner, unable to get number of CUDA devices, is nvidia driver installed?"

My install went fine and it was work fine before the install.  I suspect this latest version (419.67) for Windows 10 is not compatible with this latest version on t_Rex.

I have 3 Windows 10 rigs, and I have only updated this one.  The other 2 are working (well creeping along really slow due to such high difficulty levels).

I can install the earlier version of nVidia drivers, but thought you might want to take a look at whats causing this.

-Rodger
newbie
Activity: 87
Merit: 0
How to solve this?
OS: SMOS
algo: X16R

20360207 10:04:51 x16r block 628215, diff 133433.339
20360207 10:05:06 x16r block 628216, diff 133658.400
20360207 10:08:06 Wait on data is timed out, 180 seconds have passed, 0 bytes in socket remained
20360207 10:08:06 WARN: Connection with pool timed out. Trying to reconnect...
20360207 10:08:06 Authorizing...
20360207 10:08:07 Authorized successfully.
20360207 10:08:29 x16r block 628217, diff 133093.467
20360207 10:09:02 x16r block 628218, diff 133366.705
20360207 10:09:51 x16r block 628219, diff 132969.159
20360207 10:11:53 x16r block 628220, diff 131603.105


Happens often on random rigs, rig stuck at this and not hashing until I reload miner.

Thanks!
member
Activity: 438
Merit: 27
There are different 1070 and 1080, you need to tell us what powertarget you have.
It might be an option to use Simplemining. It will start crashed GPUs by itself.
You should go for around 120 Watt.

Example:
My 1080 (Gigabyte G1) are running on 120 Watt and each gets around 21 to 22 Mhs.

110.62 MH/s   42℃
30%   1544 1594 1569 1569 1607 (120)
4519 4519 4519 4519 4519

newbie
Activity: 1
Merit: 0
How about setting up intensity and cpu priority do you think there is hashrate changes or improvement?

Just looking for good settings not in settings on MSIafterburner do you recommend what settings should I touch and change in order to improve the hashrate?

I use -i 21 for 1070 and -i 22 for 1080\1080Ti... there are improvement in h\r if 23 is used, but power draw becomes spiky. For 1070 there is no improvement if 22+ is used. For overclocking\undervolting cards I use custom curve in AB, I tune it so 1070 draw 120wt 1080 145wt and 1080ti 185wt. With fixed clock\voltage you don't need to adjust PL.
Hello.
Can you send me screenshots with overclocking\undervolting 1070 and 1080 cards in AB? and curve AB? I know how to properly overclock cards for eth, but I do not know how to deal with x16r. Thank you very much!
legendary
Activity: 3164
Merit: 1003
Any X16R results for 2060 and 1660TI?

On BITCORE with bitcore algo or some call it timetravel10 algo.  Smiley

EVGA XC Ultra 1660 Ti    50 pw  75+ core  -502 mem

20190315 21:21:01 GPU #0: EVGA GTX 1660 Ti    - 30.85 MH/s, [T:61C, P: 79W, F:25%, E:390kH/W]

newbie
Activity: 13
Merit: 0
Any X16R results for 2060 and 1660TI?
member
Activity: 283
Merit: 63
Can we maintain the hash order? I mean that I want to main the hash order removing red and only yellow and green must be shown in hash order?
Can we adjust that on the T-rex miner is it possible?

Because the hashrate is not stable sometimes the hashrate shows 100% and sometimes half, I just want to make the miner stable.

Unfortunately, it's not possible. Hash order is determined by the blockchain, and T-Rex only displays it.
legendary
Activity: 1638
Merit: 1046
How about setting up intensity and cpu priority do you think there is hashrate changes or improvement?

Just looking for good settings not in settings on MSIafterburner do you recommend what settings should I touch and change in order to improve the hashrate?

I use -i 21 for 1070 and -i 22 for 1080\1080Ti... there are improvement in h\r if 23 is used, but power draw becomes spiky. For 1070 there is no improvement if 22+ is used. For overclocking\undervolting cards I use custom curve in AB, I tune it so 1070 draw 120wt 1080 145wt and 1080ti 185wt. With fixed clock\voltage you don't need to adjust PL.
Can we maintain the hash order? I mean that I want to main the hash order removing red and only yellow and green must be shown in hash order?
Can we adjust that on the T-rex miner is it possible?

Because the hashrate is not stable sometimes the hashrate shows 100% and sometimes half, I just want to make the miner stable.
jr. member
Activity: 312
Merit: 2
How about setting up intensity and cpu priority do you think there is hashrate changes or improvement?

Just looking for good settings not in settings on MSIafterburner do you recommend what settings should I touch and change in order to improve the hashrate?

I use -i 21 for 1070 and -i 22 for 1080\1080Ti... there are improvement in h\r if 23 is used, but power draw becomes spiky. For 1070 there is no improvement if 22+ is used. For overclocking\undervolting cards I use custom curve in AB, I tune it so 1070 draw 120wt 1080 145wt and 1080ti 185wt. With fixed clock\voltage you don't need to adjust PL.
legendary
Activity: 1638
Merit: 1046
If u will see on mined coins, t-rex will be much faster
t-rex is faster, power consumption is the same
How about setting up intensity and cpu priority do you think there is hashrate changes or improvement?

Just looking for good settings not in settings on MSIafterburner do you recommend what settings should I touch and change in order to improve the hashrate?
full member
Activity: 251
Merit: 100
If u will see on mined coins, t-rex will be much faster
jr. member
Activity: 312
Merit: 2
t-rex is faster, power consumption is the same
legendary
Activity: 2128
Merit: 1657
hi does anyone have experience with both T-Rex 0.9.2 and Z.enemy 1.28, when mining X16r, which in your opinion is better overall

in terms of hashrate, stability and power consumption ?.. Thanks
newbie
Activity: 3
Merit: 0
#!/bin/bash

sudo nvidia-smi -pm 1
sudo nvidia-smi -i 0 -pl 78
sudo nvidia-smi -i 1 -pl 78
sudo nvidia-smi -i 2 -pl 78
sudo nvidia-smi -i 3 -pl 78
sudo nvidia-smi -i 4 -pl 115
sudo nvidia-smi -i 5 -pl 78
sudo nvidia-smi -i 6 -pl 78
sudo nvidia-smi -i 7 -pl 78
sudo nvidia-smi -i 8 -pl 115
sudo nvidia-smi -i 9 -pl 115
sudo nvidia-smi -i 10 -pl 115
sudo nvidia-smi -i 11 -pl 78

nvidia-smi

sudo nvidia-smi -i 0 -pl 78 - 1060 6 Gb
sudo nvidia-smi -i 4 -pl 115 - 1070 8 Gb
newbie
Activity: 9
Merit: 0
Power: 80%
Core: +95
Memory: -300

When someone says Power 80% is that the same thing as:
sudo nvidia-smi -i 1 -pl 137
This factors in factory setting for power is 151 for a 1070.
newbie
Activity: 3
Merit: 0
newbie
Activity: 4
Merit: 0
Hello!

I am using 2 sets - P106-100 and 2 sets - 1050ti in a single rig. GPU Driver Version: 388.31

Setting on P106-100:

Power: 80%
Core: +95
Memory: -300

Setting on 1050ti
Power: 80%
Core: +95
Memory: -250

Miner version 0.9.2 CUDA 0.9.2 NVDIA GPU Miner

Problem:
Miner will run in awhile and it will hang afterwards. The miner is not stable. Is there a problem with mix GPU's in a rig?

Thank you.

I think your power limit may be too low on those 1050s. It looks like you'd be around 60W per card which is a bit too low IMO. Id try upping the 1050tis to 65-70W per card.
newbie
Activity: 18
Merit: 0
Dear dev,
I really look forward to the function of the failover pool in pools.txt file!
Miner already supports this fucnction

Anyone tried the new 1660 ti with T-Rex on x16rt by any chance?
Could be worth a cheeky investment... Been a while since I built a new mining rig... Now might not be the worst time.

That 1660 ti looks like it could be very, very efficient...
Expected hashrate from 1660 ti will be same like gtx 1070, but some optimisations may be required
Pages:
Jump to: