CPU PCI-E lines are irrelevant. All slots besides the main x16 go through the chipset. |
|
|
|
The miner cannot run with 16 or more GPUs, P106 to be precise: [11:46:05]Phoenix Miner 5.1c Linux/gcc - Release build
--------------------------------------------
CUDA version: 10.0, CUDA runtime: 8.0
/hive/miners/phoenixminer/h-run.sh: line 23: 22678 Segmentation fault (core dumped) ./PhoenixMiner
phoenixminer exited, waiting to cooldown a bit
5.0.21-200905-hiveos |
|
|
|
|
3
|
Alternate cryptocurrencies / Mining (Altcoins) / Re: NBMiner v31.1, GPU Miner for ETH, RVN, GRIN, AE, BTM, SERO, HNS, BFC, SIPC, TRB
|
on: September 13, 2020, 08:53:45 AM
|
The miner cannot run with 16 or more GPUs, P106 to be precise: [11:46:05] INFO - ------------------- System -------------------
[11:46:05] INFO - OS: Ubuntu 18.04.5 LTS, 5.0.21-200905-hiveos
[11:46:05] INFO - CPU: Intel(R) Celeron(R) CPU G3930 @ 2.90GHz
[11:46:05] INFO - RAM: 6829 MB / 7860 MB
[11:46:06] INFO - CU_DRV: 10.2, 440.100
[11:46:06] INFO - ------------------- Config -------------------
[11:46:06] INFO - ALGO: ethash
[11:46:06] INFO - URL:
[11:46:06] INFO - USER:
[11:46:06] INFO - URL1:
[11:46:06] INFO - USER1:
[11:46:06] INFO - TEMP: 85C
[11:46:07] ERROR - !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
[11:46:07] ERROR - Mining program unexpected exit.
[11:46:07] ERROR - Code: 11, Reason: Process crashed
[11:46:07] ERROR - Restart miner after 10 secs ...
[11:46:07] ERROR - !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! |
|
|
|
The PCI-E extenders can run at PCIE 1.0 X1 with no hashrate drop. What makes you think that you will benefit from PCIE 4.0?  |
|
|
|
12 1060 micron (no strap) @ 635mV (little bit more efficient then pl 70)
How are you able to set low voltage like that? In my afterburner in curve editor voltage starts from 800mv, so there is probably a different way  NVIDIA Inspector command line lets you go as low as 650mV. There are guides around the internet, look it up. |
|
|
|
This is usually caused by nvidia driver bug, try using low intensity dag loading or reduce your memclk( usually the pill crashes it, not the miner ).
Switching briefly from P2 to P0 is not a driver bug, that's how the P-states work on nvidia. There's a workaround on Windows, but Linux has no such settings. I'll try to check the GPU clocks when a dag is being switched to see whether my suspicions are correct. All ethash miners experience the same issue, except Claymore. Edit - looks like my suspicions are correct: #Date Time gpu pwr temp sm mem enc dec mclk pclk
#YYYYMMDD HH:MM:SS Idx W C % % % % MHz MHz
20190519 14:27:01 0 159 67 100 100 0 0 5613 1531
20190519 14:27:02 0 159 67 100 100 0 0 5613 1531
20190519 14:27:03 0 159 67 100 100 0 0 5613 1531
20190519 14:27:04 0 159 67 100 100 0 0 5613 1531
20190519 14:27:05 0 159 67 100 100 0 0 5613 1531
20190519 14:27:06 0 159 67 100 100 0 0 5613 1531
20190519 14:27:07 0 160 67 100 100 0 0 5613 1531
20190519 14:27:08 0 63 66 0 0 0 0 5613 1544
20190519 14:27:09 0 69 65 3 0 0 0 6106 1556
20190519 14:27:10 0 < this is where it crashed This is while the DAG switches to another one. Note the mclk column. |
|
|
|
|
Can't really search the thread, so I'll report a bug here. On NVIDIA rigs, the main display card (GPU0) crashes during DAG switch - this usually happens on 1080 and 1080 Ti. The main reason seems to be that the cards briefly go from loaded P2 to unloaded P0 during the switch, which brings them to unstable memory clocks.
|
|
|
|
RX sires cards do not look too bad for progpow https://progpow.pro/. Looks like a 580 can get 80% of the performance of a 1070 for less then 1/2 the price. RX 570 = 11.23 1070 = 13.3 That's at default clock with default power draw. The card will draw roughly 200-230W while mining, so it is pointless. Undervolting it down to 1150-1200 core should achieve nearly half the power draw. |
|
|
|
|
I'm using this motherboard with two 1-to-4 PCI-E multipliers put in the x1 slots for 12 cards in total. Works flawlessly with no additional bios tweaks besides enabling mining mode. You must convert your windows install to GPT in order to avoid error code 43, though. Linux should work out of the box with 6+ cards.
|
|
|
|
|
No way, ETC already stopped with rigs having a 3GB main card. The temporary workaround is to put a 4GB+ card into the first x16 slot, but it will break again in roughly 2 weeks.
ETH mining will be dead as well on 3GB cards by the end of the year, so I'd recommend you to find an alternative.
|
|
|
|
The performance of the RX590 is increased by 12-15% (Core - 1469-1545 Mhz), so its hashrate during the mining of the Ether will give 30-32 mh, which allows to compete with GTX1070.
Ethereum hashrate has almost nothing to do with core speed, as long as it is high enough (around 1150-1200MHz mark). Increasing it further has no effect on the hash. |
|
|
|
|
I wouldn't be surprised if RX 590 can handle 200MHz higher core at the same voltage even while undervolted. This means that 1400MHz @0.9V or 1200MHz @0.8V is possible, which brings ~15% higher hash OR ~10% lower power draw on core-bound algorithms, including cryptonight. Ethash hashrate will most likely be unaffected, so this card wouldn't be viable for ethereum.
|
|
|
|
|
Please learn the difference between break even and return of investment before making educational charts. Assuming that the price and difficulty stays the same (it won't), the device will pay itself in 43.5 months ("1305 days ROI" in your chart) and after that it will start mining for profit.
|
|
|
|
|
Any chance to add support to other CN variants? 8GB cards are great with 4MB scratchpad algorithms.
|
|
|
|
-knows what he's doing -mines ethereum lol |
|
|
|
How the hell do you dual mine two core-bound algorithms?  |
|
|
|
What hashrate increase are we talking about?
As stated in message subject, 50-90% increase. Talk to me in numbers, percentage means nothing. |
|
|
|
|
What hashrate increase are we talking about?
|
|
|
|
I'm expecting at least 45 MH/s on Ethash with RX 680 (as I've a feeling that AMD just might have optimized these for mining).
Why would you expect that? GDDR5 is good up to 9000-9500MHz and the change to GDDR6 would require a brand new memory controller and a large rewrite of the chip. Ethash is NOT core starved on the current cards - it's running as well as possible with the 256-bit bus without getting limited by the core. This is literally the limit of GDDR5. The only thing we can safely expect is either a lower voltage for the same frequency, or a higher frequency for a given voltage. Which turns into better hash/watt ratio - small gains on memory-based algorithms like CryptoNight variants and Ethash, larger gains on core-based like X16R, lyra2z/lyra2v2, equihash. |
|
|
|
|
Show Posts
