T-Rex 0.26.8 ETHW, ETC (+dual mining) GPU miner - page 32.

Mapuo

hero member

Activity: 910

Merit: 1002

Quote from: NikiKa on March 30, 2021, 12:23:48 PM

Hi everybody ,
what mean "!" in DAG generating string ?

20210330 00:28:48 GPU #1: DAG generated [crc: bdc4b7d7(!), time: 5825 ms], memory left: 4.95 GB
20210330 00:28:48 GPU #2: DAG generated [crc: 36ddfb5a(!), time: 5833 ms], memory left: 4.95 GB
20210330 00:28:48 GPU #3: DAG generated [crc: a56cf2cd, time: 5823 ms], memory left: 4.95 GB
20210330 00:28:48 GPU #5: DAG generated [crc: 5483d6c7(!), time: 5836 ms], memory left: 4.95 GB

From the dev

Quote

xray — 03/13/2021
DAG checksum is not valid for GPU0. This is a sign that you can get invalid shares with this GPU0. The amount of bad shares is variative depending on % of bad data in a DAG. If you get too many invalid shares then you need to sort it out. Usually can be fixed by lowering OC. @Mapuo

In your case "GPUO" are GPUs 1, 2 and 5.

rednoW

legendary

Activity: 1510

Merit: 1003

Quote from: 1337leet on April 01, 2021, 07:52:29 AM

Hey guys,

I run 2 weeks old drivers and trex.

Will the resizable bar driver + newest Trex make my 3080 and 3090 faster?

nope

1337leet

hero member

Activity: 1308

Merit: 508

Hey guys,

I run 2 weeks old drivers and trex.

Will the resizable bar driver + newest Trex make my 3080 and 3090 faster?

highlander1992

newbie

Activity: 8

Merit: 0

Quote from: dedizonesv2 on March 31, 2021, 12:29:30 AM

Quote from: leojoyms on March 29, 2021, 08:28:30 PM

Anyone have stale shares increased in eth with this latest version??

Yes: news version stale bad share, a rollback to "t-rex-ver. 0.19.12-cuda111" no stale

We are still waiting for me to dev, participate in the topics a little ... and explain to us !!!

Quote from: highlander1992 on March 30, 2021, 12:47:05 PM

any fix for memory leak rtx 2060 gddr6? I applied "--no-nvml" command and nvlm patch and still getting memory leak.

there are also problems see here https://github.com/trexminer/T-Rex/issues

Post issue Wink

I manage to solve this out.It was some kind of virus @ mscorsvw.exe hidden in net run time network with windows files.Some bastard tried to rob me

CoolTomik

newbie

Activity: 5

Merit: 0

Is there way how to get OC info about mining GPU cards like core Voltage/Mem frq/core frq over json api?

chemz

newbie

Activity: 3

Merit: 0

Quote from: JayDDee on March 29, 2021, 01:20:38 PM

They aren't related. Validate shares tests the validity with the CPU before submitting it. Stale shares
are valid, just too late to be accepted.

Hi! Does that mean that if we're not using this argument in the .bat file we're going to have potentially less valid shares? How does it differ from the Rejected statistics? If I get an Invalid share in the miner, but not a Rejected one, am I not having 1 valid share less either way?

Thanks!

dedizonesv2

jr. member

Activity: 152

Merit: 3

Quote from: leojoyms on March 29, 2021, 08:28:30 PM

Anyone have stale shares increased in eth with this latest version??

Yes: news version stale bad share, a rollback to "t-rex-ver. 0.19.12-cuda111" no stale

We are still waiting for me to dev, participate in the topics a little ... and explain to us !!!

Quote from: highlander1992 on March 30, 2021, 12:47:05 PM

any fix for memory leak rtx 2060 gddr6? I applied "--no-nvml" command and nvlm patch and still getting memory leak.

there are also problems see here https://github.com/trexminer/T-Rex/issues

Post issue Wink

highlander1992

newbie

Activity: 8

Merit: 0

any fix for memory leak rtx 2060 gddr6? I applied "--no-nvml" command and nvlm patch and still getting memory leak.

NikiKa

newbie

Activity: 7

Merit: 0

Hi everybody ,
what mean "!" in DAG generating string ?

20210330 00:28:48 GPU #1: DAG generated [crc: bdc4b7d7(!), time: 5825 ms], memory left: 4.95 GB
20210330 00:28:48 GPU #2: DAG generated [crc: 36ddfb5a(!), time: 5833 ms], memory left: 4.95 GB
20210330 00:28:48 GPU #3: DAG generated [crc: a56cf2cd, time: 5823 ms], memory left: 4.95 GB
20210330 00:28:48 GPU #5: DAG generated [crc: 5483d6c7(!), time: 5836 ms], memory left: 4.95 GB

leojoyms

newbie

Activity: 54

Merit: 0

Anyone have stale shares increased in eth with this latest version??

dedizonesv2

jr. member

Activity: 152

Merit: 3

Quote from: JayDDee on March 29, 2021, 11:34:31 AM

The miner doesn't request jobs, the stratum server sends them.

Latency is the network delay between sending a share and receiving the acknowledgement.

Validate shares means the CPU will verify the GPU hash is correct by performing a CPU hash on the same data
before submitting it as a share.

High latency has nothing to do with the miner or validate shares. It's a networking issue.

Tks Bro

I have a network problem? therefore on a RIG but not the other two, sometimes it is the others but this one. that's why I ask

No more worries since this morning, I just rebooted the TREX miners 3/4 times

JayDDee

full member

Activity: 1424

Merit: 225

Quote from: hb81 on March 29, 2021, 12:50:22 PM

What does this mean?

Quote

--send-stales Don't drop stale shares

Also curious of this.

Quote

--validate-shares Validate shares before sending to a pool. Also enables share diff info

They aren't related. Validate shares tests the validity with the CPU before submitting it. Stale shares
are valid, just too late to be accepted.

Higher latency means higher probability of stale shares being submitted. The miner can detect if a job
has gone stale while it was hashing a valid share. Since the miner spent the effort to find a valid share
it can choose whether to submit it anyway and risk it being rejected, or discard it.

hb81

newbie

Activity: 3

Merit: 0

Hmm.. now that i looked in the pool and there was now stale share. Right about same time what miner reported OK share with longer time.
Any ideas or is this just situation that ia cant do nothing? Grin

hb81

newbie

Activity: 3

Merit: 0

What does this mean?

Quote

--send-stales Don't drop stale shares

Also curious of this.

Quote

--validate-shares Validate shares before sending to a pool. Also enables share diff info

What is the actual point of those?
Could that Validate shares cause more stales?
I usually have 1-2% of stales. Now there is not any stales about 9 hours. I think its my internet connection.
Can't have wires so have to play with 4G mobile network. 5G spreading all around. (Don't worry.. here in Finland we have quite good mobile connections)
4G modem and CAT6 to rig.
Ping to pool is avg 31ms. So not bad. Just tested.
Sometimes OK share tells quite long times. Now that i look miner there is one share with 317ms. Right after that there is one with 54ms.

JayDDee

full member

Activity: 1424

Merit: 225

The miner doesn't request jobs, the stratum server sends them.

Latency is the network delay between sending a share and receiving the acknowledgement.

Validate shares means the CPU will verify the GPU hash is correct by performing a CPU hash on the same data
before submitting it as a share.

High latency has nothing to do with the miner or validate shares. It's a networking issue.

dedizonesv2

jr. member

Activity: 152

Merit: 3

Quote from: blueknight32 on March 22, 2021, 06:52:49 AM

Okay So I'm seriously pissed about this issue. I have my usual T-rex miner config setup (Same one I have been using since before my network change) but wen the screen shuts off it seems the miner connection fails which is ridiculous. The only thing changed was I went from the crappy cox all in one gateway modem/router to a Ubiquiti Dream machine pro. My computer is connected via hardwire cat6 and get 7ms pings, 850 mb/s down and no network problems other wise BUT it seems like maybe the dream machine is blocking something at the network level? I have added the entire trex folder to the windows 10 exclusion list and have actually just fully disabled the win10 antivirus for now. The computer is not set to sleep only the screen is. All low power settings are disabled for things such as the the Ethernet NIC and video card. Once the screen shuts off after 10 minutes about 3 minutes later the miner stops connecting with the errors listed in the image attached to link. I am using t-rex-0.19.11-win-cuda11.1 for my MSI RTX 3090 with current

code : t-rex.exe -a ethash -o stratum+tcp://us2.ethermine.org:4444 -u 0xmineraddress -p x -w rig0 pause

Yes, check pcie / monitor powersafe in windows

Disable shutoff computer and juste poweroff screen with button on/off or connected electrical outlet.

For me i have open topic: https://bitcointalksearch.org/topic/latency-gpu-5327048

When I look at "screen -r miner" I sometimes randomly get a #GPU which displays 1875 Ms ... GPU # 0 | 5.04 G
Then everything becomes correct again, could someone tell me what is really this latency time?

Is this the time it took him to request a JOB, calculate it and then send it to the pool? who could be involved in the GPU? CPU? a network disturbance ?

And for this I don´t understand the last added command line "--validate-shares

how to add in json config ?

"validate-shares":

OR

"validate-shares": 1

It would be good if the software creator came more often and answered questions, we participate in the dev fee but also more explanatory documentation would be welcome because currently it is the community that provides the support.

aavvaallooss

newbie

Activity: 5

Merit: 0

Quote from: willi9974 on March 29, 2021, 03:58:04 AM

came invalid shares more from a to low "Power Limit" or more from a to high "memory frequence"?

[FAIL] 989/1016 - Provided PoW solution is invalid!, 55ms ... GPU #1

In my case always because too much memory overclocking

willi9974

legendary

Activity: 3500

Merit: 2792

Enjoy 500% bonus + 70 FS

came invalid shares more from a to low "Power Limit" or more from a to high "memory frequence"?

[FAIL] 989/1016 - Provided PoW solution is invalid!, 55ms ... GPU #1

dedizonesv2

jr. member

Activity: 152

Merit: 3

Quote from: aavvaallooss on March 28, 2021, 09:45:03 AM

I don´t understand the last added command line "--validate-shares Validate shares before sending to a pool. Also enables share diff info".
What is for? It´s good or bad for avoid stale or invalid shares?

Me to i don't understant this parametre

how to add in json config ?

"validate-shares":

OR

"validate-shares": 1

thank you for helping me for the others it was easy but I don't know what this parameter is for but also how it is activated in hiveOS the configuration file is in json!

miner29

full member

Activity: 1275

Merit: 141

They are definitely not the exact same...see one report Asus RTX 3070 and next reports RTX 3070.....

Topic: T-Rex 0.26.8 ETHW, ETC (+dual mining) GPU miner - page 32. (Read 174982 times)