 Show Posts
|
|
Pages: « 1 2 3 [4]
|
Is xintensity 224 still recommended for nvidia cards mining on miningpoolhub? |
|
|
|
Try miningpoolhub.com I always get good returns there. 100 000 accepted shares 0 rejected. 1335/48/23 (96.53%)
Problems with some of your cards? To high clocks perhaps? Look for cpu validation errors in the logfile. Are you using xintensity 224 or 256 while mining from miningpoolhub? |
|
|
|
Thanks for that link. Switched to CrazyPool and changed xintensity to 256. Is that still high based on the Difficulty value of 2.00? |
|
|
|
Ethermine seems to be making -9.29% ethereum than the best paying pool. (last 7 days) (-7.02% the last 56 days) 07-october-2021:  Are you able to provide a link on how that data is generated? |
|
|
|
PING: 50ms DIFFICULTY: 1.00 EPOCH: 445
Too low difficulty for --xintensity 256. (default) try --xintensity 224 Just changed xintensity to 224. Should I wait a few hours for the Current/Average Hashrate to adjust properly on Ethermine? |
|
|
|
Tested it on Ethermine for 12 hours each with default intensity (256), rig of NV 30** series. TRex : 563 mh/s TBM : 555 mh/s (reported : 601 mh/s) That's really a huge gap between the reported hashrate and the pool average effective hashrate. Even though they pay for the stales (2%) and there's no invalid block. At this point, just provide the right xintensity and the chosen pool to actually see good results. Because I tested 4 different pools and many different xintensity values, and I still didn't find any way to improve my rentability. Actually quite the opposite. Been running TBM v1.11 with xintensity 256 on Ethermine and I am getting great results so far. 868m [2021-10-07 09:35:05.381] 2577.33 kH/W 390.50 MH/s 390.54 MH/s 4543/8/0 (99.82) Please keep up the great work!  Is 390 mh/s actually your average effective hashrate on ethermine? Just looking at it on your side doesn't mean it's "great results". Thanks for that insight. Based on your reply, it looks like it may not be "great results" based on the average hashrate reporting on Ethermine? Current: 351.8 MH/s Average: 362.7 MH/s Reported: 390.5 MH/s How close should they be? |
|
|
|
Been running TBM v1.11 with xintensity 256 on Ethermine and I am getting great results so far. 868m [2021-10-07 09:35:05.381] us1.ethermine.org (ethash) PING: 50ms DIFFICULTY: 1.00 EPOCH: 445 868m [2021-10-07 09:35:05.381] ID BOARD TYPE TEMP FAN CORE MEM WATT kW/h COST/h 868m [2021-10-07 09:35:05.381] GPU0 3060 Cuda 61/0 70 1320 8117 143 0.14 0.01USD 868m [2021-10-07 09:35:05.381] GPU1 3060 Cuda. 61/0 79 1633 8117 172 0.17 0.01USD 868m [2021-10-07 09:35:05.381] GPU2 3060 Cuda. 65/0 70 1354 8117 143 0.14 0.01USD 868m [2021-10-07 09:35:05.381] GPU3 3060 Cuda. 53/0 79 1651 8117 172 0.17 0.01USD 868m [2021-10-07 09:35:05.381] GPU4 3060 Cuda. 63/0 70 1420 8117 143 0.14 0.01USD 868m [2021-10-07 09:35:05.381] GPU5 3060 Cuda. 59/0 70 1358 8117 143 0.14 0.01USD 868m [2021-10-07 09:35:05.381] 916 0.92 0.07USD 868m [2021-10-07 09:35:05.381] 868m [2021-10-07 09:35:05.381] ID BOARD HASHRATE/W HASHRATE AVERAGE SHARES 868m [2021-10-07 09:35:05.381] GPU0 3060 455.28 kH/W 65.11 MH/s 65.10 MH/s 802/2/0 (99.75) 868m [2021-10-07 09:35:05.381] GPU1 3060 378.53 kH/W 65.11 MH/s 65.11 MH/s 738/1/0 (99.86) 868m [2021-10-07 09:35:05.381] GPU2 3060 454.90 kH/W 65.05 MH/s 65.05 MH/s 769/5/0 (99.35) 868m [2021-10-07 09:35:05.381] GPU3 3060 378.20 kH/W 65.05 MH/s 65.11 MH/s 737/0/0 (100.00) 868m [2021-10-07 09:35:05.381] GPU4 3060 455.06 kH/W 65.07 MH/s 65.06 MH/s 743/0/0 (100.00) 868m [2021-10-07 09:35:05.381] GPU5 3060 455.36 kH/W 65.12 MH/s 65.11 MH/s 754/0/0 (100.00) 868m [2021-10-07 09:35:05.381] 2577.33 kH/W 390.50 MH/s 390.54 MH/s 4543/8/0 (99.82) Please keep up the great work! |
|
|
|
How do you determine the amount of power that is sufficient for the xintensity value? If I change the xintensity to 256, should I increase my PL to 75%, or is that overkill?
You need to test it yourself. Tune the rig. Different cards need different tuning. With the right pool and more power --xintensity 512 can give a few more MHASH. If the Dag creation is unstable, you can lower the coreclock and memclock, increase intensity and power and still get a better hashrate. Just acquired my 6th 3060 TI NON-LHR and modified the --xintensity value to 256. https://imgur.com/a/EzKtBevLet's see how this runs for the next 24 hours. Great work by the way!!!!! |
|
|
|
Depends on the pool, but less than 5% stales doesn't seem to affect the profit on ethereum/ethereum classic mining.
For bitcoin and other coins stales are bad.
Increasing the xintensity will increase the hashrate, increase the amount of stale shares and also the power.
How do you determine the amount of power that is sufficient for the xintensity value? If I change the xintensity to 256, should I increase my PL to 75%, or is that overkill? |
|
|
|
My second test-rig is under 1.09 int 64, does not work good with newer versions:
1004m [2021-10-05 15:56:21.147] ID BOARD TYPE HASHRATE/W HASHRATE AVERAGE SHARES
1004m [2021-10-05 15:56:21.147] GPU1 3060Ti Cuda 467.57 kH/W 63.12 MH/s 63.20 MH/s 293/0/0 (100.00)
1004m [2021-10-05 15:56:21.147] 1441.14 kH/W 177.37 MH/s 177.42 MH/s 814/0/0 (100.00)
What OC settings are you using for the 3060 TI?
Depends on the pool. If the pool pay for stale shares the default 256 is fine. Flexpool, ethermine( ? ), nicehash and a couple of other doesn't pay for stale shares, so the miner need a lower --xintensity. Usually when a pool display the stale share amount, they eigther pay nothing or a reduced amount for stale shares. Sometimes the stale payout percentage is documented in the FAQ section of the pool. On flexpool if you see this:  It meens that do don't get payed for around 5% of the shares that you submit (stales), and you have to lower the intensity of the miner. Miningpoolhub.com, 2miners.com, nanopool.org, woolypool.com and many others pay the full amount for stale shares. The ethereum blockchain is paying for stale work. They are called uncle blocks, and this is different from other chains like Bitcoin, Doge, Dash etc. A mining pool can be configured to submit uncle blocks and not credit you for the work. Stealing 5% of your profit. Got it. I was always under the impression that it was best to ensure your miner submits a higher percentage of Accepted Shares vs Stale Shares? Is that not the case anymore? [moderator's note: consecutive posts merged] |
|
|
|
With the updated TBM v1.11 and new --xintensity recommended setting (--xintensity 224), I am getting 64+ MH per card (Total 322.14 MH) and with 0 stales/rejects so far (93+ mins running).  Looking good. The default --xintesity on nvidia is 256 in v1.11 but 224 produce less stale shares, so we might change the default in v1.12. If default is 256 for nvidia, should I change xintensity? |
|
|
|
GPU setup: 4x Nvidia 3060 TI FE 1x EVGA 3060 TI FTW3 ULTRA (NON-LHR) GPU Overclock Settings: Memory - +1050 (8,050 total) Core - -200 Power Limit - 71% Fan Speed - 70% With the updated TBM v1.11 and new --xintensity recommended setting (--xintensity 224), I am getting 64+ MH per card (Total 322.14 MH) and with 0 stales/rejects so far (93+ mins running). https://imgur.com/a/ekTsjndIs it safe to assume that it is working as intended? |
|
|
|
I've been running TBM for about 20 hours now with x-intensity setting 32 and so far it's been doing extremely well compared to T-Rex Miner. https://imgur.com/a/f4D0w7PGreat work so far! |
|
|
|
Hello.
Does anyone know if it is best to let T-Rex Miner auto define the Intensity and Kernel Values or should we set one ourselves?
 I set it myself. I noticed that 22 intensity vers. 20 -i had no difference as far as hashrate goes on my 1660ti's ect. And I go through each card to see what is the best hashrate for each Kernel. Thanks for the explanation! |
|
|
|
|
Hello.
Does anyone know if it is best to let T-Rex Miner auto define the Intensity and Kernel Values or should we set one ourselves?
|
|
|
|
Should I be able to see the CORE/MEM/WATT reporting by default?  you could try to open a cmd promt and run: nvidia-smi --id=0 --query-gpu=power.draw,clocks.sm,clocks.mem --format=csv nvidia-smi --id=1 --query-gpu=power.draw,clocks.sm,clocks.mem --format=csv nvidia-smi --id=2 --query-gpu=power.draw,clocks.sm,clocks.mem --format=csv nvidia-smi --id=3 --query-gpu=power.draw,clocks.sm,clocks.mem --format=csv nvidia-smi --id=4 --query-gpu=power.draw,clocks.sm,clocks.mem --format=csv Do you get any information out? Disregard the previous post. I updated to the latest Nvidia driver (471.12) and now it shows the Power Draw, Core Clock, and Memory Clock. |
|
|
|
Current Setup: 4x Nvidia RTX 3060 TI FE 1x EVGA RTX 3060 TI FTW3 OC Settings: MEM Clock: 1200,1200,1200,1200,1200 CORE Clock: -200,-200,-200,-200,-200 Power Limit: 70,70,70,70,70 Fan Speed: 65,65,65,65,65 Should I be able to see the CORE/MEM/WATT reporting by default?  |
|
|
|
|
