I have few rigs with mixed Radeons (380/290x/580/fury) and those rigs are rock solid stable. The only issues i had was with PSU or burned cable.
All rigs are on Linux from USB flash drive. Uptime over month is nothing strange.
No expierience with nvidia mining rigs though.
Make notice that you can modify Radeons bios, some people doesn't understand the settings and flash best bios from internet. Not every card will work on those stable. Sometimes memory are to high or voltate to low.
I don't say ATI is better or Nvidia is better. I just think that if someone have issues with something then he should focus on resolving problem, not complaining.
Topic: To all you AMD mining rig owners..I solute you! - page 2. (Read 726 times)
I have zero problems with my two rigs running Saphire Nitro + RX580, Linux is the key 
I have zero issus with my Nvidia cards but my amd rigs are picky with drivers and overclocking but I have them all on the beta blockchain drives and don't push the clocks that hard and they've been running for 2 months without a crash it's pushing they hard is when shit happens and those compute drivers all my rigs hate them
The problem is because of those piece of crap Vega cards. They dont play well with anything else and they mine everything but XMR like shit. You should sell those as quick as possible and buy some better cards.
there is nothing to do with vega, there is to do with amd drivers.
The problem is because of those piece of crap Vega cards. They dont play well with anything else and they mine everything but XMR like shit. You should sell those as quick as possible and buy some better cards.
Which windows do you guys use? Windows 10 is a crap, better use windows 7
MY friend disabled 70+ different things on Windows 10 just to avoid updates. Crazy.
MY friend disabled 70+ different things on Windows 10 just to avoid updates. Crazy.
still hanging around with Windows 7, downside is I couldn't get the latest driver and squeeze out more juice out of it.
but still happy for stability.
the problem is amd drivers, i posted this a while ago. You have to flash every single card to work properly, amd did this to its drivers. It was 10000 easier before the dag fix, now is frustrating.
likely too much underclocked or overclocked, so it just freezes without telling you the problem where mining eth is okay and have no problems, every algo is different, so for this not to happen, you will need to change the core and memory clocks, try very low and then increase bits by bits until the system froze again, that means you found the card and will have to lower its core or memory clock. This is not rocket science, mining with amd is very hard, especially now that amd is trying to block any underclock or overclock with their drivers. Nowadays you have to flash your bios to make it work right on that algorithm, before used to have watttool which could underclock or overclock to everything you wanted, now you cant and people cant do anything because the dag fix is after amd fucked all the drivers up.
I am using my Radeon RX570 GPU rig to mine Monero with Claymore's CryptoNote v11.3. All starts well and the rig works fine for hours then for no obvious reasons it breaks. All I can see is this screen:
GPU #5 Ellesmere (AMD Radeon (TM) RX 470 Graphics), 385 MB available, 32 compute units (pci bus 6:0:0)
GPU #5 recognized as Radeon RX 470/570
POOL version
GPU #0 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
GPU #1 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
GPU #2 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
GPU #3 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
Does anyone know what could be the problem here?
GPU #5 Ellesmere (AMD Radeon (TM) RX 470 Graphics), 385 MB available, 32 compute units (pci bus 6:0:0)
GPU #5 recognized as Radeon RX 470/570
POOL version
GPU #0 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
GPU #1 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
GPU #2 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
GPU #3 algorithm ASM, -h 896, -dmem 1 (Memory used:3672MB)
Does anyone know what could be the problem here?
likely too much underclocked or overclocked, so it just freezes without telling you the problem where mining eth is okay and have no problems, every algo is different, so for this not to happen, you will need to change the core and memory clocks, try very low and then increase bits by bits until the system froze again, that means you found the card and will have to lower its core or memory clock. This is not rocket science, mining with amd is very hard, especially now that amd is trying to block any underclock or overclock with their drivers. Nowadays you have to flash your bios to make it work right on that algorithm, before used to have watttool which could underclock or overclock to everything you wanted, now you cant and people cant do anything because the dag fix is after amd fucked all the drivers up.
My AMD GPU's go down every 7 or 8 days when mining Ethereum on Ethermine and probably 3-4 days when mining Pirl on Minerpool.net
I am runing the problematic rig out of two for two days now without any issues. And I bet it could run for a long time if I don't tuch it. But some days it decides to crash every few hours.
Probably issue with mixed cards, but can't help it.
Probably issue with mixed cards, but can't help it.
Which windows do you guys use? Windows 10 is a crap, better use windows 7
MY friend disabled 70+ different things on Windows 10 just to avoid updates. Crazy.
MY friend disabled 70+ different things on Windows 10 just to avoid updates. Crazy.
Running 4000+ Sapphire rx470 mining edition without any issues.
Then from the past some batches of rx480/rx570/rx580's. All doing good!
Just remember to use linux in stead of windows. HiveOs.farm is a really good OS.
Any questions, just ask!
(I really hate NVIDEA, so i got your issue turned around)
Then from the past some batches of rx480/rx570/rx580's. All doing good!
Just remember to use linux in stead of windows. HiveOs.farm is a really good OS.
Any questions, just ask!
(I really hate NVIDEA, so i got your issue turned around)
I bought those 560's for fun because they were cheap and I figured I can get some ETH out of it. To be honest with you..that one click bio wasn't all that bad and when I missed with 5 560's and 2 580's...not much problems and it was pretty stable once I had it running...
you know what they say...don't mess some thing that ain't broken. Well, I did and it's still turned off after 5 hours of messing with last night (maybe I'll try again tonight before I throw in the towel). My 3 Vega 64's were less problematic when I had them on the test machine alone....figured I'll save some electricity by moving down to my little mining farm.
Power is definitely not the issue...I'm running 1800w 240v psu. Maybe I should increase my memory to 16gb instead of 8? I'm just spit balling here trying to get some ideas from you guys
you know what they say...don't mess some thing that ain't broken. Well, I did and it's still turned off after 5 hours of messing with last night (maybe I'll try again tonight before I throw in the towel). My 3 Vega 64's were less problematic when I had them on the test machine alone....figured I'll save some electricity by moving down to my little mining farm.
Power is definitely not the issue...I'm running 1800w 240v psu. Maybe I should increase my memory to 16gb instead of 8? I'm just spit balling here trying to get some ideas from you guys
Almost all of my rig fully AMD, but I have one rig with two cards only (1060).
My AMD card was:
R9: 290, 290x, 295X2, 390, 390X.
RX: 470, 480, 580.
All of my AMD rigs was mixed (R9+RX), I never have one rig with the same brand, series, type. Giving me a lot of issue at first, but now I can handle em.
Hear hear! AMD cards really suck once you are configuring them. I know because i have 470s 570s and VEGAs! Compatibility is shit. I have to DDU everytime i install drivers. All the reboots are because of AMD!
But in the end, when you get them calm and stable, DAMN. They are sweet babies that earn more than most NVIDIA counterparts.
But in the end, when you get them calm and stable, DAMN. They are sweet babies that earn more than most NVIDIA counterparts.
Couldn't agree more
I think you are doing something wrong in configuring your cards. Most of the mining computers are running with them and only a few people mine with Nvidia.
My colleagues have 20 mining computers, each with six amd graphic cards and one b250 motherboard with 11 cards. All the computers with 6 cards are running stable from more than one month. The only one crashing was the one with 11 but a linux distro fixed it too, so overall Amd cards are better for mining compared to nvidia.
My colleagues have 20 mining computers, each with six amd graphic cards and one b250 motherboard with 11 cards. All the computers with 6 cards are running stable from more than one month. The only one crashing was the one with 11 but a linux distro fixed it too, so overall Amd cards are better for mining compared to nvidia.
Hear hear! AMD cards really suck once you are configuring them. I know because i have 470s 570s and VEGAs! Compatibility is shit. I have to DDU everytime i install drivers. All the reboots are because of AMD!
But in the end, when you get them calm and stable, DAMN. They are sweet babies that earn more than most NVIDIA counterparts.
But in the end, when you get them calm and stable, DAMN. They are sweet babies that earn more than most NVIDIA counterparts.
Yeah, when they're bad they're very bad, but it'll settle down. You'll gradually smooth out the clocks for each card so they're stable and it'll be fine. Unless you have a cheap psu or some other glitch ....
I have a few RX 580 rigs, some times they give me trouble, but they are usually alright.
Yea I know. I have two rigs.
First is 3x RX 470 + 1x R9 380
Second is 4x RX 580 + 1x R9 290x + 1x R9 380.
The first one is rock stable, the second one gives me problems. But due to the prices of the cards that was the best I could put together. All cards are used and working fine. When I stabilize the second rig it works days without problems, but some days it just crashes every hour or so. Oh well buy cheap, play cheap
Waiting for card prices to go down so I can buy a proper rig. NVIDIA most likely as its more versatile atm.
First is 3x RX 470 + 1x R9 380
Second is 4x RX 580 + 1x R9 290x + 1x R9 380.
The first one is rock stable, the second one gives me problems. But due to the prices of the cards that was the best I could put together. All cards are used and working fine. When I stabilize the second rig it works days without problems, but some days it just crashes every hour or so. Oh well buy cheap, play cheap
Waiting for card prices to go down so I can buy a proper rig. NVIDIA most likely as its more versatile atm.
So I had a mixed rig (5 560's, 2 580's and 4 1080's and 1 1060) rig which it didn't give me much problems (with the exception of me only running Phoenix miner / that Zechash one)...I also have 3 Vega 64's which I was running it on it's own because I didn't have enough power to combine them (no big deal)....today, I decided to remove my Nvidia cards and put all AMD's together...this is when hell break loose!!!! At first, window keeps crashing on me so I was able to get it stablzed @ 8 gpu (instead of 10). No big deal, but then trying to get it to run sgminer and the damn thing just won't stop crashing (kept saying my GPU is sick)....after 5 hours or messing with it..I had enough!
I don't know how you guys do it...but you guys deserve a holiday just for putting up with this crap....all my nvidia cards....ZERO problem. AMD...nothing but problem!
anyway, just needed to vent a little..
I don't know how you guys do it...but you guys deserve a holiday just for putting up with this crap....all my nvidia cards....ZERO problem. AMD...nothing but problem!
anyway, just needed to vent a little..
Jump to: