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

[joseph32]

Hey T-Rex,

would be awesome if you could add in a future version CuckooCycle and CuckaToo31/32. You are still the fastest miner in most algos and I think you can be also much faster than Gminer here too.

[Lucky - Luciano]

I tried the T-Rex miner today and it was kind of stupidly set for situations when the internet connection disappears. Namely, the miner switches himself off after a failed internet connection ?! I often lose my internet connection due to a power outage. Today I was unpleasantly surprised to see that T-Rex is not working and Team Red miner is working. How do I solve this problem?

[Clocker1000]

Since I switched my 1080Ti from Phoenix to T-Rex, I'm noticing that after about a day of mining, when I come back to the PC the miner is still happily hashing away but I cannot access the internet with my browser.  I have to reboot the PC and then I can access websites again.   Anyone else have a similar issue?   I'm mining ethash and using the pill as well.

[morticio777]

Problem when mining octopus for the last 2 day, before your miner are the best one with realy nice speed... Plz fix ASAP

i try 2 different pool, same problem


https://i.imgur.com/DPfIxKj.png

[trexminer]

Problem when mining octopus for the last 2 day, before your miner are the best one with realy nice speed... Plz fix ASAP

i try 2 different pool, same problem

Thanks for reporting. As a temporary workaround you can try starting the miner with your GPU(s) on stock settings and then apply overclock after DAG has been generated. We're working on making DAG generation more stable.

[ildotoress]

Hi I am new to T-Rex and tried just to run it to see how it goes but it can't connect to any pool with the default setup:

Code:

d:\Miner>t-rex.exe -a ethash -o stratum+tcp://eth-eu1.nanopool.org:9999 -u 0x1f75eccd8fbddf057495b96669ac15f8e296c2cd.rig0/some@email.org -p x
20210107 23:48:52 T-Rex NVIDIA GPU miner v0.19.5  -  [CUDA v11.10 | Windows]
20210107 23:48:52 r.5f0b2f67355c
20210107 23:48:52
20210107 23:48:52 NVIDIA Driver v461.09
20210107 23:48:52 CUDA devices available: 1
20210107 23:48:52
20210107 23:48:52 WARN: DevFee 1% (ethash)
20210107 23:48:52
20210107 23:48:52 URL : stratum+tcp://eth-eu1.nanopool.org:9999
20210107 23:48:52 USER: 0x1f75eccd8fbddf057495b96669ac15f8e296c2cd.rig0/some@email.org
20210107 23:48:52 PASS: x
20210107 23:48:52
20210107 23:48:52 Starting on: eth-eu1.nanopool.org:9999
20210107 23:48:52 ERROR: No connection, retry after 10 seconds
20210107 23:48:52 ApiServer: HTTP server started on 0.0.0.0:4067
20210107 23:48:52 -------------------------------------------------------
20210107 23:48:52 For control navigate to: http://192.168.0.119:4067/trex
20210107 23:48:52 -------------------------------------------------------
20210107 23:48:52 ApiServer: Telnet server started on 0.0.0.0:4068
20210107 23:48:52 WARN: GPU #0(002b00): MSI GeForce RTX 3080, intensity set to 22
20210107 23:49:02 ERROR: No connection, retry after 10 seconds
20210107 23:49:12 ERROR: No connection, retry after 10 seconds

I have allowed the trex.exe in Bitdefender firewall but it will not connect to any pool...

I have the latest driver 461.09.
Miner is latest version from first page on this thread with CUDA 11.1.
I can telnet in the pool URL and port but the miner never manages it... any help would be appreciated.

[kardonxtx]

Sorry for the newb question but I haven't mined since x11\13. What are you guys doing to get such high hash rate at low wattage out of your 1080tis?

right now at 2000 core  5200 mem and I'm only seeing 32MH/s at 200watts.

I'm just running the vanilla config t-rex -a ethash -o stratum+tcp://us1.ethermine.org:4444 -u [wallet] -w rig0

[kardonxtx]

adding --mt 6 seems to have given me a decent boost up to ~46 once I realized my batch file wasn't working because running as admin changes the starting directory lol.

-- kernel flag seems to be ignored, in windows at least. I tried mirroring the --kernel 0 setting but the client still indicates it's using 4. (or maybe it's working and it auto benchmarks that quickly and is choosing 4?)

And another newb question.... How is the wattage being determined? I tried lowering my overclocks and my P actually increased. Then I moved them back and they stayed pretty much the same. Is this just a rubbish stat and I should bust out the killowatt?

[trexminer]

-- kernel flag seems to be ignored, in windows at least. I tried mirroring the --kernel 0 setting but the client still indicates it's using 4. (or maybe it's working and it auto benchmarks that quickly and is choosing 4?)

How is the wattage being determined? I tried lowering my overclocks and my P actually increased. Then I moved them back and they stayed pretty much the same. Is this just a rubbish stat and I should bust out the killowatt?

--kernel is not ignored (for ethash and octopus), 0 is the default value which is "benchmark all of them and choose the fastest".  You can set it manually with "--kernel 1" (2, 3, ..)

Wattage readout is what nvidia driver gives us. It isn't exactly what you'll see with a kilowatt but if the miner says the power consumption increased it is very likely the case, and vice versa.

[trexminer]

Problem when mining octopus for the last 2 day, before your miner are the best one with realy nice speed... Plz fix ASAP

i try 2 different pool, same problem

Please try 0.19.7 and let me know if it fixes the issue https://github.com/trexminer/T-Rex/releases/tag/0.19.7

[trexminer]

T-Rex 0.19.7

* ("ethash", "octopus", "kawpow") Improve stability for 30xx series GPUs
* ("ethash") Verify overclock stability after DAG rebuild ("Instability detected" message is printed in case there are issues)
* Add "--gpu-report-interval-s" parameter to control hashrate summary report frequency based on the number of share submissions
* (API) Add new "pause" command for "control" handler to pause and resume mining
* (Watchdog) Display a list of GPUs caused miner restart with "GPU is idle" error

Bug fixes:
* ("x21s") "incorrect size of extranonce2" error
* (UI) Incorrect GPU index in share submission report when "--pci-indexing" mode is enabled
* (API) Miner fails to switch algorithms on the fly for some ""<algo_from>" -> "<algo_to>"" combinations

[kardonxtx]

--kernel is not ignored (for ethash and octopus), 0 is the default value which is "benchmark all of them and choose the fastest".  You can set it manually with "--kernel 1" (2, 3, ..)

Wattage readout is what nvidia driver gives us. It isn't exactly what you'll see with a kilowatt but if the miner says the power consumption increased it is very likely the case, and vice versa.

Thank you.

[ubik_paris]

Hello,

I have a strange issue on a 3090.
 
The MH/s is slowing down very quickly, it start a 106 MH/S and go down litlle by little to nearly 50MH/s and then stabilize to that value. The GPU load is 100% but what is strange is the Core Mhz that is very low.
I have another computer with a 3090 and it is always nearly 110Mh/s. it run both on windows 10 with the lastest Nvidia Studio driver.
Is someone know what is the problem and is there a workaroung?

Thanks

[ZeeeN]

Hello,

I have a strange issue on a 3090.
 
The MH/s is slowing down very quickly, it start a 106 MH/S and go down litlle by little to nearly 50MH/s and then stabilize to that value. The GPU load is 100% but what is strange is the Core Mhz that is very low.
I have another computer with a 3090 and it is always nearly 110Mh/s. it run both on windows 10 with the lastest Nvidia Studio driver.
Is someone know what is the problem and is there a workaroung?

Thanks

did you try game ready driver?

[maXonja]

Is there any command to reboot rig after every DAG epoch change or if instability is detected?
Can you program that when miner detects instability/incorrect shares it activates reboot.bat in his folder?
This will solve problems with nVidia 3X00 cards and memory overclock issues at DAG epoch change
Because we have no use of this ""Instability detected" message is printed in case there are issues"

[ubik_paris]

Hello,

I have a strange issue on a 3090.
 
The MH/s is slowing down very quickly, it start a 106 MH/S and go down litlle by little to nearly 50MH/s and then stabilize to that value. The GPU load is 100% but what is strange is the Core Mhz that is very low.
I have another computer with a 3090 and it is always nearly 110Mh/s. it run both on windows 10 with the lastest Nvidia Studio driver.
Is someone know what is the problem and is there a workaroung?

Thanks

did you try game ready driver?

no I didnt tried the game ready driver. I also use these machines to work so studio driver are better for me. It would be surprising if it was the reason   

[ubik_paris]

Hello

Windows defender is detecting a Beareuws.A!ml virus on 0.19.7 release.
What about it? false positive?

[joseph32]

Hello

Windows defender is detecting a Beareuws.A!ml virus on 0.19.7 release.
What about it? false positive?

Do you really think he puts a virus inside?

[trexminer]

Is there any command to reboot rig after every DAG epoch change or if instability is detected?
Can you program that when miner detects instability/incorrect shares it activates reboot.bat in his folder?
This will solve problems with nVidia 3X00 cards and memory overclock issues at DAG epoch change
Because we have no use of this ""Instability detected" message is printed in case there are issues"

There is no command yet because we hope we solved the instability problem with 30xx cards in 0.19.7 for good. Please give it a try and let us know if you still experience issues.

[trexminer]

Hello

Windows defender is detecting a Beareuws.A!ml virus on 0.19.7 release.
What about it? false positive?

Yes, it is a false positive. We charge 1% fee, so spreading viruses and having bad reputation because of it is not what we're after. See https://github.com/trexminer/T-Rex#antivirus-alerts for more details.