Topic: Claymore's ZCash/BTG AMD GPU Miner v12.6 (Windows/Linux) - page 278. (Read 3839163 times)

What straps?

815 on a 480 8G, clocks 1190/2100 and an undervolt of -62

Obviously not stock 

810 h/s on stock clock and memory R9 390, undervolted -150mv

XMR is the best mining option right now. 686 h/s on my Rx 480  1235/2145 

What version of driver do you use? Also do you get any error like "Code 43 ? ( unlikely on two cards configuration)

Anyone know how to solve this?

DevFee: ZEC: Stratum - connecting to 'zec.coinmine.pl' <177.54.158.139> port 7017 (SSL/TLS)
12:26:58:702   6ec   ZEC: Stratum - Cannot connect to zec.coinmine.pl:7017
12:26:58:705   6ec   DevFee: ZEC: Stratum - Failed to connect, retry in 20 sec...


This for mining
This for mining

win10
RAM 8gb
SSD 120
2x r7 370

yes. I cleaned up the drivers and reinstalled from scratch

Of course, that's always an option. It's just that I installed every second or third iterations since 16.7.x for these RX4xx cards and none of them seemed to make any meaningful difference for mining (neither performance, nor stability), except this one. That's why I thought it's worth reporting. And it's always possible that things like this are not driver issues but a software bug exposed by some otherwise neutral changes in the driver, or they are driver bugs but trivial to work around in software...

the rx480s have a design flaw where the power peaks for a split second above the pci-e specs and becomes very dangerous at 220 watts when OC to the max. I'm not an electric engineer but from what I've read on the net these spikes will eventually cause the motherboard to fail. ZCASH mining takes these spikes to the max. The same spikes are minimal on XMR. It is going to be interesting to read about mobo failures in the near future.

What is your hardware configuration? driver version?

has anyone ever faced an issue "can't detect AMD cards?

Try to go to a previous driver version (16.10.1) certainly works
Then you need to disallow Windows 10 from updating your drivers automatically because after you revert to 16.10.1 Windows will try to update them

There is something strange going on with the 16.12.2 AMD driver (I skipped 16.12.1) and these mining softwares (Claymore ZEC and ETH):
- after a fresh OS boot I rarely (if ever) get all cards (usually 6 per machine) up to full speed, some get stuck at their lowest GPU clock speeds
- restarting the same miner with the same settings usually reproduces the same pattern (the same GPUs remain stuck as before)
- starting a different miner (ETH after stopping ZEC or vice versa) usually draws a different pattern of stuck GPUs (there is overlap but it's not the same)
- starting two processes of the same miner in parallel with 3-3 different cards instead of all 6 usually gets all cards up to full speed
- starting a single process after getting all cards "unstuck" once with parallel processes usually results in all cards running at full speed (even after switching miners)
There is no detectable/reported driver crash (it's no like those card instantly crash and get reset to "safe" speeds, they never get up to full speed).
Sometimes even smaller machines have similar problems, like 1 card getting stuck from 3 (but it's more often for 6 cards and the 3-3 parallel trick is usually enough for 6 cards)

I like your analysis

But if I were you and had 7 gpu's without being able to make them work good together I would probably do 2 things:

1. Sell only the 7th Gpu and make a very efficient 6-GPU rig which is quite easy skipping a lot of costs with the double (mobo, psu, risers, case, RAM, CPU)
2. Buy 5 more cards and make 2 very efficient 6-GPU Rigs

     I'm a bit late to the party, but I'm getting 250-255 h/s on Sapphire Nitro+ rx 480 OC 8gb cards totaling 1750 - 1785 h/s (stock bios). According to GPU-Z, each card draws an average of 115w (core), but shows intermittent spikes no greater than the maximum specified power consumption for the card (225w). Running Windows 10 on both rigs with crimson V. 16.11.5 (this is the most recent update, that I've found, that will recognize a bios mod).


GPU Settings:

Core clock = 1400
Mem clock = 2040
Voltage = +6 mV
Fan = 100%
Pow lim = +50%


.bash Settings
                                                                         _____
setx GPU_FORCE_64BIT_PTR 0                                     |
setx GPU_MAX_HEAP_SIZE 100                                    |-----------> The hashrate was more consistent once I added these variables. You don't have to use
setx GPU_USE_SYNC_OBJECTS 1                                  |                  them for 4gb cards or above.
SET GPU_SINGLE_ALLOC_PERCENT = 100            _____|
Stratum-Port: 3336
US-Server: zec-us1.dwarfpool.com
-i 8
-allpools 1


First Rig:

Motherboard: MSI z97 Gaming 5
CPU: Pentium Haswell 3258
PSU: EVGA Supernova P2 1200w Platinum (ran all 7 cards on this rig for a while, but could only run at -i 2 and no overclock)
SSD: 120gb (partitioned for Ubuntu (30gb) and Windows 10 (90gb))
Memory: 8gb DDR3 1600 (2 x 4gb (2x > 1x))
GPU: 4x Sapphire Nitro+ rx 480 OC 8gb (stock bios).
Risers: The green ones with the black usb cable??
Display: HDMI Headless Display Adapter (Only one is needed for the FIRST PCI-e slot recognized by the motherboard).
*Power button needed
**WIFI
***Remote access only


Second Rig:

Motherboard: Gigabyte ga-z97x Gaming 7
CPU: Pentium Haswell 3258
PSU: Rosewill Photon 1050w Gold Rated + SeaSonic SSR-650RM 650W Gold Rated (connected with dual PSU adapter)
SSD: 120gb (partitioned for Ubuntu (30gb) and Windows 10 (90gb))
Memory: 8gb DDR3 1600 (2 x 4gb (2x > 1x))
GPU: 3x Sapphire Nitro+ rx 480 OC 8gb (stock bios).
Risers: The black ones with the blue usb cable??   <------Better quality than the green ones. I recommend these.
Display: HDMI Headless Display Adapter (Only one is needed for the FIRST PCI-e slot recognized by the motherboard).
*Power and reset buttons are included ON this motherboard! Amazing.
**WIFI
***Remote access only


     I could do 8 cards on the MSI Gaming 5 (7x PCI-e + 1x M.2 ---> PCI-e adapter), but the rig lost some stability once I went over 6x GPUs, even with enough power. The Gigabyte ga-z97x Gaming 7 can house 7 cards (6x PCI-e + 1x M.2 ---> PCI-e adapter), but as I said before, stability seemed to become more of an issue with more than 6x GPUs. Each rig has been running for about a week without any resets or crashes. Just in case the miner or machine does crash, I have made some simple .bash scripts to restore all of my settings and programs upon reset. I have also used the Windows Event Log and Manager to reset in the event of a freeze crash. So all in all, I plan to have 6 cards per rig and I'll probably have to add another PSU to the first rig when I reach that point.


*DAS IMPORTANT*

     When it comes to power supplies, it is almost ALWAYS cheaper to buy 2x units and an adapter, than it is to shell out the money for even a 1200w PSU. My EVGA Supernova P2 1200w Platinum was about $220 (on sale), but I could have bought 2x  700w-850w PSUs + a convergence adapter for 2/3 the price. Also, Never purchase a PSU with a rating less than gold; it will save you both time and money in the long run.

*This is a great PSU reference guide. https://i.imgur.com/tgrbCnr.jpg

     On a final note, I am aware that there are mods to allow for more than 4 GPUs on Windows 7 and more than 5 GPUs on Windows 8, but save yourself from a giant, time consuming pain in the d!

Cheers,
SweaterJacket

I'm running a Rosewill Photon 1050 with 5 cards currently! Been going about a week with absolutely no stability issues, etc.

1333mhz core, 2100mhz memory clock with -30mV under keeps them cool and has generally run well for me. I'm using 16.11.5.

So in short... Yes, you could very easily do like 6 cards with your 1300w power supply.

Can you miner be faster than the  private miners?

Win 7 64 bit + Crismos 16.10.1 or 16.10.2 ( don't forget about 6gpu_mod) it works for sure!

how many sols on 270 AMD ?