|
Genoil just dropped v 0.5
|
|
|
|
 Anyone know why the gpu voltages are not in actual mV? Is it safe to just enter a value in mV? This is Mis rx 480 4gb btw, thanks! You can enter mV values but they will be ignored. GPU core voltage is calculated at runtime in relative steps to its base voltage. What you will need is an offset either hardcoded into the bios using a hex editor ( complicated ) or a tool like Strixx or Watttool ( simple ) to load the core voltage offset at system startup. |
|
|
|
Anyone know why the gpu voltages are not in actual mV? Is it safe to just enter a value in mV? This is Mis rx 480 4gb btw, thanks! So you have card bios with timings same from 800 all the way up AND IT IS WORKING? Yes, you can put voltages... You know that's not possible. Its probably just the first half of the Straps that are the same/similar. |
|
|
|
I'm going to keep it brief this time  : https://github.com/Genoil/ZECMinerBinaries only, sorry. So until the big boys are ready with their 100Sol/s cannons, you can mine some ZEC with this piece of shit. Highly unstable but pretty fast thanks to @mrb's SilentArmy kernels. yay Don't undersell yourself, your miners are great man. We may not say it often but your hard work is seriously appreciated. Many thanks for all you do. Cheers. |
|
|
|
This OpenCL miner based on modified John Tromp's and "silent army" kernel Github repository: https://github.com/eXtremal-ik7/xpmclient/tree/version/zcashCurrent releases: 0.1.1: https://drive.google.com/file/d/0ByzvxWEyrTINamFFLTFhODVYT1k/view?usp=sharing http://coinsforall.io/distr/zcashclient-0.1.1.zip0.2.0: https://drive.google.com/file/d/0ByzvxWEyrTINeEFyTWVuelZkU2c/view?usp=sharing http://coinsforall.io/distr/zcashclient-0.2.0.zip0.2.1: https://drive.google.com/file/d/0ByzvxWEyrTINRkgzZG5vdU9ZQkU/view?usp=sharing (silentarmy: automatically starts 2 instances, you can change instances number in config) http://coinsforall.io/distr/zcashclient-0.2.1.zip0.2.2: https://drive.google.com/file/d/0ByzvxWEyrTINQlM2bm13ZnF6Sms/view?usp=sharing (silentarmy: NVidia support) http://coinsforall.io/distr/zcashclient-0.2.2.zipOS Supported: Windows, Linux (binaries for Ubuntu 16.04) Drivers: Catalyst 16.xx or amdgpu-pro 16.30 recommended GPUs supported: AMD & NVidia. Optimal choice AMD Polaris GPU RX470 & RX480 with custom ROM https://bitcointalk.org/index.php?topic=1584617.0 or NVidia Pascal GTX10xx Radeon RX470: 25-40
Radeon RX480: 28-4x (45 sol/s only from silentarmy kernel with 2 running instances)
Other GPUs
GeForce GTX1070: 30 sol/s with silentarmy, 25sols/s with old kernel
Pool mining: Miner uses custom binary protocol first time implemented in xpmclient (Primecoin GPU miner). Working pool instance: http://coinsforall.io/#!/coin/ZCASHTo start mining: - Put your T-address or Z-addess to config.txt (need ZCash client installed, online wallet or exchange with ZCash support). Pool support both ZCash address types for payments. server = "coinsforall.io";
port = "6668";
# Your ZCASH payout address (z-addr)
address = "zthTb2ToG5kuAc11Ni6NxoYjoxQdEr2XvST7XEdLvmHG9k7kotEcwYCaAn1SH4zRdPoXn9kNeFm8CQo k9SNtNgn8d9WcdTK";
- Configure your GPUs (edit corefreq, memfreq, powertune and fanspeed parameters) - Run zcashgpuclient in Linux or zcashgpuclient.exe in Windows - Put you address to 'Your Address' field on web site for monitoring workers and payments Pool system: PPLNS (Pay Per Last N Shares); accounting procedure uses all your shares received during last hour. Known problems with ZCash:1. Corruption wallet bug: https://github.com/zcash/zcash/issues/1694 Author says, that he send T -> Z transaction, but it not neccessary, wallet can be corrupted at any moment. Pool wallet goes down an 01.00UTC 29Oct, about 2 hours pool not working. And 7-8 hours needed by repair wallet and send mined ZEC back to pool. 2. Slow and sequential transactions. Z->Z taken an ultimate long time (up to 30 min for creation and up to 5 hours to delivery) and CPU/RAM resources. Z -> T quite faster, it takes 3min for creation and same interval for delivery. If I found way for make transactions more quickly, I implement it immediately, but now I see only one method - return back minimal payment to 0.05 - 0.1 ZEC. All your payments queued, size of this queue now is 1500 and grows after decrease minimal payment size and wallet corruption incident. You need a long time for wait payments. 3. Another zcashd wallet crashes. Need a time for remove blockchain and download it back. Very often zcashd stucks on block 145 (why 145? I don't know). Pool issues:1. "Fake payments" Some clients see payment, but really transaction not sent to network. All this transaсtions will be replayed. TODO:
1. Payments without Z-address support. No payments to clients from Z-Address. No payments to client's Z-Addresses. All payment problems caused by new 'z' API which can't be used for pool. For anonymous transactions - OK, but not for pool. (TODAY)
Done! payouts is fixed.. finally. Need select a procedure for move already mined coins from your Z to T address
2. Rollback 'Fake' payments and return funds to balance (TODAY)
Started. All 'Fake' payments will be deleted, coins return to balance
Final pure stable version will be until 1 Nov. You can continue mining here, or switch pool/miner and return later after launch non-Z version. Or leave this pool, if you think that it's scam.
Windows defender keeps deleting the latest version, marking it with "Virus Detected" even before I can open it. Am I the only one having this issue? |
|
|
|
|
So I dove into Hex editing for the BIOS and its not as complicated as its made out to be. For the layman, I still feel what watttool offers is pleanty. With HEX what you can additionally do is create intermediate memory timing straps and can include an offset in the bios to reduce core volts consumption.
With all the tweking I could do, the BEST I can get my RX 470's is to hash at 25.6 Mh/s while consuming 90W from the wall (GPUZ is like 54W), with a 5 card system at 510W with a lot of bios editing in hex (not polaris). This was a part of my pet 250Mh/s with 1KW at the wall single frame rig/project that I finally achieved yesterday. Its got two 5 card systems in one compact open frame. I will be setting up my 4th such rig to touch 1Gh/s soon.
I have a power ceiling at my place that limits how many rigs I can have running, so most of my research and efforts have been focused into efficiency and not peak performance. Also, the cards run cool, have 0 memory errors, are significantly undervolted and most importantly the risers run cool so i'm insuring a good long run for my hardware.
I think we all know by now that these cards can be pushed pretty hard. But unless you have a proper cooling setup with ducts and air conditioners, and unless you are sure that you are going to dump these cards when ETH goes POS, I would not recommend pushing them to the edge. Find a suitable balance and stick to it, even for perf enthusiasts 27 Mh/s is what you should be happy at with a RX 470 4GB running 24/7. I've made my peace with less.
|
|
|
|
|
So I dove into Hex editing for the BIOS and its not as complicated as its made out to be. For the layman, I still feel what watttool offers is pleanty. With HEX what you can additionally do is create intermediate memory timing straps and can include an offset in the bios to reduce core volts consumption.
With all the tweking I could do, the BEST I can get my RX 470's is to hash at 25.6 Mh/s while consuming 90W from the wall (GPUZ is like 54W), with a 5 card system at 510W with a lot of bios editing in hex (not polaris). This was a part of my pet 250Mh/s with 1KW at the wall single frame rig/project that I finally achieved yesterday. Its got two 5 card systems in one compact open frame. I will be setting up my 4th such rig to touch 1Gh/s soon.
I have a power ceiling at my place that limits how many rigs I can have running, so most of my research and efforts have been focused into efficiency and not peak performance. Also, the cards run cool, have 0 memory errors, are significantly undervolted and most importantly the risers run cool so i'm insuring a good long run for my hardware.
I think we all know by now that these cards can be pushed pretty hard. But unless you have a proper cooling setup with ducts and air conditioners, and unless you are sure that you are going to dump these cards when ETH goes POS, I would not recommend pushing them to the edge. Find a suitable balance and stick to it, even for perf enthusiasts 27 Mh/s is what you should be happy at with a RX 470 4GB running 24/7. I've made my peace with less.
|
|
|
|
Only ETH mode gives 32.8 Mh/s for overclocked R9 390X.
ETH + DCR mode gives 30.1 Mh/s ETH + 440 Mh/s DCR for the same graphic card.
As I see in your readme file, I will get higher hashrate if I use dual mode. But it s not. Is it possible ?
Reduce your -dcri to 1 for claymore's miner ( for 440MH in DCR it must be around 30 currently) and you will see your cards max ETH hash rate which should be around 34 Mh/s. Then increase dcri steadily till you start losing ETH hash rate, find the optimum balance point where you get almost your max ETH hash but also significant DCR. This point should be between -dcri 16 to 24. |
|
|
|
Eth mining uses the ethash algorithm, which does a keccak hash (essentially sha3-512), 64 x 128-byte random DAG reads, then another keccak hash. The keccak hashing is GPU intensive, while the DAG reads are memory intensive. All the public miners use a similar opencl implementation, and so have similar GPU/MEM requirements. For AMD cards with a 256-bit memory interface (R9 380, Rx 470/480...), that ratio works out to 0.56. So with a 1500Mhz memory clock, the optimal core clock is 840Mhz. Increasing the core clock beyond 840 will not increase hashrate, and ends up using more power. Similarly, a Rx 470 with a 1750Mhz memory clock should have a core clock of 980Mhz.
One additional condition is that the GPU needs enough compute units so that it can saturate the memory bandwidth. With the publicly available miners, that minimum is around 22-24 compute units, so cards like the R7 370 do not max out their memory bandwidth.
This ratio is not a fundamental limit of the ethash algorithm, so a kernel with a highly optimized keccak implementation (such as Wolf's private kernel) likely has a lower ratio and a lower minimum number of compute units required for optimal performance. This means you won't see any miner get more than 28Mh from a Rx 470 at 1750Mhz, someone could release a miner that gets its maximum hashrate with a core clock of much lower than 980Mhz, and therefore reduces power consumption.
Can confirm, a lot of trial and error over weeks got me to 1050 core/1870 mem as the most efficient for my RX 470's and it fits the 0.56 ratio. |
|
|
|
To start tweaking the card with watttool i just put the card in system and then run the program in windows, nothing else?
Or is there some other steps like, i need to flash vbios first?
I would appreciate a little step-by-step tutorial just to be sure Wink
I just haven't done this before, only tweaked in msi afterburner which doesnt let you tweak that much. voltage control locked etc.
For wattool there is no problem. You dont need to flash anything. But to change the memory straps(and get higher hashrate) you need to save the cards bios with atiflash, open the rom file with polaris bios editor. And change the straps. Save the bios file(save as new file, keep the original you saved at first) and then flash with atiflash, reboot. Which card you have exactly? 600W Bronze running 4 cards? I can only run 3 cards. The system will boot on the 4th but when heavy hashing then the PC will hang. 750W Gold with 5 cards? Interesting...  Probably better configuration for the cards. And maybe slight difference in PSU. And also maybe not dual mining but eth only. Yes, ETH only. I gave up on dual mining with SIA dropping in value a lot over the past month, and my risers heating up. Each card is at 1050/1870 - 870/870 with 1500 strap, and the 4 card system draws ~475W from the wall so the 600W Bronze works just fine. |
|
|
|
ETH is dead end, say Hallo! to ZCash - together we conquer the Moon
Yes, yes, yes, move all your ETH machines to ZCash ASAP. Please. Cant wait for the difficulty drop on ETH. |
|
|
|
Here are some info correlate PSU and GPU. Based on rx470
550w Bronze - 2 GPU
650w Bronze - 3 GPU
650w Gold - 4 GPU
850w Gold - 5 GPU
1000w Gold - 6 GPU
anyone else have other setup able to achieve more? these are my current setup.
750W Gold running 5 cards 600W Bronze running 4 cards |
|
|
|
Hello everyone, please spare some time to give some advice What is the best OS to control ETH mining farm with 20+ hosts? There are windows 7s installed on most of the PCs and it's hard to remotely control them, most problems come in when one of the host freezes. I use claymore's miner to mine ETH only. Claymore's monitoring software I installed on my home server which is connected to farm with VPN and I can monitor them when I'm at home, I also use python scripts that sends me e-mail and sms when something is wrong on site. At first everything seems to be fine, but I know I need a lot of work to do there. First I'm interested what would be the best OS to install on hosts and why? Also if there are any solutions to remotely reboot freezed host just like I can power them up with Wake On Lan packets (settings from bios are changed, the host PCs power up themselves whenever electricity goes down). I'm new to ETH mining so if these questions are obvious sorry for that, I need to somehow maintain that farm. there are ways to power off and power on the pc's some pdus' allow for turn on and turn off. I have some 8 plug 30 amp pdus that can be turned on and off via an ethernet connection. server technology https://www.servertech.com/ would allow for power fully off up to to 8 pc's in one pdu with windows 7 set to reboot for a black out they may have exactly what you want. this would be a best solution I guess, Thank you for your advice. I could not find them locally so I think I have to buy them online. Hello everyone, please spare some time to give some advice What is the best OS to control ETH mining farm with 20+ hosts? There are windows 7s installed on most of the PCs and it's hard to remotely control them, most problems come in when one of the host freezes. I use claymore's miner to mine ETH only. Claymore's monitoring software I installed on my home server which is connected to farm with VPN and I can monitor them when I'm at home, I also use python scripts that sends me e-mail and sms when something is wrong on site. At first everything seems to be fine, but I know I need a lot of work to do there. First I'm interested what would be the best OS to install on hosts and why? Also if there are any solutions to remotely reboot freezed host just like I can power them up with Wake On Lan packets (settings from bios are changed, the host PCs power up themselves whenever electricity goes down). I'm new to ETH mining so if these questions are obvious sorry for that, I need to somehow maintain that farm. AMD or nVidia ? AMD rx480 I got them used from eBay and saved a lot of money http://www.ebay.com/itm/Server-Technologies-Sentry-Switched-Cabinet-Distribution-Unit-PDU-CX-8H2C211-/162214816828?this looks like the right one. it has an ethernet link It means I can remotely reboot any host right? And what about the OS? I tried etherOS, but to be honest it's too complicated to me, so I decided either win7 or win10 which one is better Windows 10 all the way if you are using RX 4XX series cards. Bios mods, settings tweaks are all waaay easier in windows. |
|
|
|
Rubbing salt on wound??  Hehe I never miss a chance So finally my setup is better than someone's else one . Currently, I'm at 28,6 MH/s for 95W. Did you only played with the mV or did you changed something else ? I'll try to reduce again my consumption by removing a few mV. Umm, im not sure, the 95W that you see in GPUZ or some other SW monitor is just the PCIe consumption. Add another 45W atleast that the card would consume through the 6 pin power plug. So the 67W listed above is more like 67+40=112W. Yours would be around 135W. What h331m4n quoted looks like the actual power draw at the wall at 115W, which includes GPU 6 pin power draw, and PSU inefficiencies. No worries though, some simple settings tweaks using even the bundled Wattman would get you under 120W per card. |
|
|
|
setx GPU_MAX_ALLOC_PERCENT 95
setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_SINGLE_ALLOC_PERCENT 100
ethminer -G -S eth.pool.minergate.com:45791 -O you_email --farm-recheck 200 -G --cl-local-work 256 --cl-global-work 8192
Thanks, will try it out Would be real interested in the results. |
|
|
|
Thank you very much mate ! You changed my life ! For now, I'm at 30 MH/s at 110W. Do you have an Elpida memory or an Hynix one ?
If I remember correctly, you swore by the GTX 1070's a little while ago, does that make you a convert? Do you BELIEVE now? |
|
|
|
@ markAZ. How much mh you got ?
I am now at 25 cards around 550mh.
I do no bios flashing and run pretty much stock 20/21 for rx 470 and 25 for rx 480
Hey Philip, I went ahead and got me a Biostar Z170 after reading your posts about it, and the board works perfectly with 4 cards so I owe you. Here is my two cents then : With 25 cards ( RX 470/480's ) some simple bios mods and NO overclocking ( underclocking and undervolting infact ) they can easily get ~25.5Mh each, totaling to ~640Mh. There is next to no reason to not do the bios mods, since there is zero chance of bricking (if you mod your own bios instead of flashing someone else's) and undervolting/underclocking insures your power draw and heat generation stays low. I currently have 28 RX 470 cards at 730Mh/- running in 6 systems with a combined power draw of ~3.1KW and all these are in my small house and the heat and noise are low enough to live with. I'm sure you can easily add more to your garage and hit that 1Gh mark I wish I had 10KW capacity id go crazy and spend all my money on this stuff |
|
|
|
Anyone got any ideas why windows locks up when I start claymore. It gets to loading dag then locks up.
All cards are over locked. Most of the time it's the card with active video for teamviewer
Cards were at 1130/auto core 1945/955 mem
Moved them too 1130/auto core 2000/1000 mem
To much of an overclock?
Didn't move power.. maybe Up the watts?
I have noticed that on some of my machines the card that is the primary display adapter, that you connect to a monitor etc, needs to be set to a slightly lower overclock/higher voltage than the rest for better stability. This card has to run additional system threads and can cause freezes/lockups. Your clocks are pretty high already, I would recommend not getting greedy and sticking to 1130/1945. Raising them further will hardly net you any noticeable gains while pushing up power usage as well as causing instability. Happy Mining. |
|
|
|
Exact model of cards? Watt-meter?
Using an APC ups for power measurements, no wattmeter. GPUZ readings at ~70W per card, APC reading for whole system ~620W. Exact model : Sapphire RX 470 4GB OC (Ref design) It seems pretty down low. How much do you undervolt?
I have some rx 470 4Gb and my power usage is almost 1.5 yours
Bios Mod with 1500 mem strap all the way up Core : 1050/870 Mem : 1870/870 Dcri -18 for Siacoin dual mining (22 is the most efficent profit wise but I wanted lower heat for my sata-molex risers). ~26.8 ETH / 240 SIA Running for almost a month, hiccup free. |
|
|
|
And I'm certain it would be possible with a titanium efficient PSU.
No - it would not be I can definitely see power draw from the wall at around 900w, and less.
You will not get 900W from wall hashing with 6 RX 480 with rate of 180MHs Well im running several 5 X RX470 rigs at ~620W each, hashing at 133Mh/s+ and while I have no 6 slot motherboards to test, adding one card to this setup would push the Hash to 170Mh/s and the power to 750W. So I see no reason why 6 Rx480's cant run under 850W or even, possibly 800W. Single or dual mining? Dual, with each card at ~26.8/240 pulling ~100W from the wall (~70W in GPUZ). |
|
|
|
|
Show Posts
