December 7th, 2013 
CryptoBadger Now that the excellent Radeon 7950 video cards have essentially become unobtainable, I’m getting a lot of questions from people about how to get the most out of the new Radeon R9 280X in terms of mining. I managed to get my hands on a few Sapphire 280X cards this past week, and have done a little bit of experimenting.
I’ll get around to updating my guide for the 280X at some point in the next couple weeks, but if you’re looking for a good starting point as far as cgminer and undervolting settings go, read on.
Undervolting and Power Consumption
First, the 280X is essentially a re-branded 7970, which means they can be undervolted using VBE7 and the method I previously outlined for the 7950. My Sapphire cards run completely stable at 1137 mV, so if you’re following my undervolting guide, simply use 1137 in place of the 1081 that I recommended for the 7950. I’ll test lower voltages at some point soon, but the 7970/280X doesn’t have a stellar reputation as far as undervolting goes, so I wouldn’t expect too much more.
Running at 1137 mV, you can expect a 3 GPU rig to use between 850-900 watts at the wall, assuming you’re using the rest of the hardware that I recommend in my 7950 guide. You will, of course, need to upgrade the power supply to something a bit larger to accommodate the 280X’s larger power draw—I recommend a 1000+ watt gold or platinum rated PSU. The best choice for those of you without access to very cheap electricity is probably something like this platinum-rated Seasonic PSU. This gold-rated Corsair is a bit less expensive and also a good choice. If you’re looking to run four GPUs, this eVGA PSU has enough PCIe connectors without dealing with splitters.
Cgminer Settings and Performance
These are early test results, but they should get most of you up and running with acceptable performance. My tests were done on a platform with all of the hardware outlined in my guide, the only exception being that the PSU was substituted for a larger 1000 watt unit, and of course the 7950 GPUs were replaced with 280X cards.
For my tests, I was running Xubuntu 12.04 with the latest beta AMD Catalyst drivers installed, AMD SDK 2.7, and cgminer 3.7.2. I’m also using a 280X-optimized scrypt kernal in cgminer. I’ll be looking at Xubuntu 13.x and other driver versions (and Windows, too!) in the coming weeks.
The best results I saw were with the following cgminer settings:
-I 13 -g 2 -w 256 --lookup-gap 2 --thread-concurrency 8192 --gpu-engine 1000 --gpu-memclock 1500 --temp-target 70 --auto-fan
This gets me roughly 710 Kh/sec per GPU, completely stable. Oddly enough, setting the core clock speed any higher than 1000 mhz results in lower cgminer performance.
I’ve read about some people getting speeds up to 750 Kh/sec with other brands and/or other configurations, but in my limited testing that was the best that I was able to achieve without compromising stability.
If you have another brand of 280X, you may want to try 1080 mhz or 1060 mhz for the core clock speed (--gpu-engine 1080), as many people report the best success in that range.
One interesting note is that the 280X seems to be pretty sensitive to heat, and will self-throttle its performance down whenever GPU temperatures get over 70C or so (at least, my Sapphires do). Setting cgminer’s auto-fan temp at 70C prevents this, although I run my rigs in a fairly cold room—some of you may need to point a box fan at your rigs to get the best performance out of your 280Xs.
I’ll be trying to get my hands on other brands of 280X cards, especially the ones by Gigabyte and MSI (although many are reporting good results with the Asus and PowerColor models, too)—although I’m sure most of you have noticed that they’re fairly difficult to come by right now.
If you’re currently running a 280X-based rig, please feel free to leave a comment below with your settings and results!
Posted in 
Cryptobadger!
I have three sapphire r9 280x dual-x gpu’s and I have spent over a week trying to find perfect settings for the card, no matter what I have tried the computer would get one of the mistakes: 511 degrees on one of the cards and then freeze, or a black screen and then freeze. But I did get a high hash rates up to 740-750
Yesterday I have found one of the forums on the internet and one of the guys there told me his settings, everything like yours but the –gpu-memclock is 1250!
I have tried this settings. My miner is working non stop, no mistakes, no nothing! The hash rate is around 700.
My question is, how come making the gpu-memclock settings at 1250 solved my problem? Does anyone else on here use the same settings?
3 of my cards are sapphire r9 280x dual xs and if you look at my settings on the last page they run just fine with them.
Did you fix this? Having the same issue when running 4 GPUs.
Hi Cryptobadger!
I noticed you posted a dead link to the optimized scrypt kernel bins for the r9 280x. I can’t manage to find it anywhere on the net, so I was wondering if you could share?
Robert
You rock! I have been struggling to get the right settings for the Gigabyte 280x’s and stumbled upon your site, using -I 13 -g 2 -w 256 –lookup-gap 2 –thread-concurrency 8192 –gpu-engine 1080 –gpu-memclock 1500 –temp-target 70 –auto-fan I get 753-760. Weird because the stock clock is 1100 changing it to 1080 actually yields better hash, THANX!!!
I’m running 3x R9 280x. It used to be 3x ASUS but one broke after 2 weeks so now its 2x ASUS and 1x MSI. I’m having around 740-750 on two of them, but one card is always about 80 lower than the rest at 670. It’s not the MSI card and it was the same when I had 3x ASUS, which I find a bit strange, anyone else seen this behavior?
Would someone also be so kind to explain the whole optimized scrypt kernel thing for a relative new miner?
Is the card thats slower just happen to be the one use also use to power your screen? That normal.. You have to tweak differently than the others. If u asking 100% out of your cards and yet also want to have a responsive screen then the khs will drop for that card.. Doing to much on the screen and u might have more driver faliures. Even if u go headless and vpn into your rig.. The gpu thats supposed to be the one providing the visuals will tend to run differently than the other gpus that only exist to be dedicated processing workhorses. Hope this helps
vpn.. Remote.. You know what I mean (use another comp to monitor your rig type deal).
How to set the fan to auto adjust my temp? ex I want it to run at 54% at 94 degree, 53% at 93 degree, 52% at 92 degree?
Did you use vbe7? I think you can add a fan profile there.
no, I use cgminer 3.5.0 and .bat file
For those with 280x toxics try profion’s 800khs setup but run at 1175mhz clock/1870 or 1875mhz mem – getting me solid 825khs per gpu 71c (well ventilated open rig).
Yeah, with this settings i realy get close to 825KHash
BUT:
– system crashes after few minutes
– can’t get that Thread-concurrency 11136 running get a error from cgminer(running version 3.7.2)
add sdk is 2.9
gpu driver 13.25.18
catalyst 13.11
2 Cards 1020 Watt PSU
try TC 11200. But after playing around more, though it does give me these ridiculous numebers and can maintain system stability.. it’s not good. causing regects when blocks accepted. OC gaming sites overclock the mem mhz to a max in the high 1600s.. realise that gmaing & mining are different animals but there only so much a gpu can take. I’Ve moved back down to 1100 clock, 1500mem giving me 765khs stable, low temps, no regected blocks. Might play around with the sky high speeds in the future but for 24/7 mining it seems pointless. Would be interested to get Profion’s opinion on this. Is he actually getting accepted block from these highly tweaked 800khs gpus of not.
Hello everyone!
This post is about ELECTRICAL OUTLETS – any feedback is greatly appreciated.
Scenario: I have just finished building 3 machines running 3 GPUs each – all of them being the MSI R9 280Xs. All machines are stable (for the most part) and producing between 2000-2100 hash power. The machines are setup in my home office.
Situation: I tried running all machines of one outlet and kept blowing the circuit breaker. Then I ran one machine of an outlet in the office and ran the other two off an outlet in a room next to the office. This was running fine but the room got incredibly hot and then all the power went off and all the smoke detectors in the house went off after that. This was slightly terrifying to my wife and dog.
What can I do? It seems that all machines are down until I figure a safe way to get them running 24/7. I thought about having an electrician come out and run dedicated circuits to each machine. This obviously and additional cost. There has to be safe more effective way to run multiple home run machines. Any comments/suggestions are greatly appreciated. Safety is obviously important.
Please help. Thank you all!
the short of it is, dont skimp. its your house.
2 hours of electrician is waaay cheaper than a week of hotel. :p
@flagit – LOL, I could not agree with you more! Does this seem to be a common occurrence? Pulling 9.1 amps at the outlet per machine seems pretty intense. Are you running multiple machines? Any lessons learned there?
From other posts and considerations it seems that the safest route is to have a circuit breaker dedicated per mining rig.
Moving forward, I am certainly going with hiring an electrician to have all four rigs running in the same area. Just seems more convenient and safe – piece of mind is worth a few hundred bucks. Also, want to explore adding a few more machines as well.
Happy Mining!
yea, materials are set in stone.
the labor is what you really pay for, and even @60/hr, 3 hours = less than $200.
think of it like, insurance. for you, your family, and equipment.
Like the other have stated, get dedicated breakers for each of your rigs. Typical North American are 15amps.. One machine will run you about 7-10 depending on your setup (4 280x gpu running at 1400w sucks back 12amps..) so the breakers will constantly pop if anymore loads are added. Also by seperating your rigs from other things in your office you avoid problems of voltage variance or harmonics caused by say turning on a light switch that happens to run on the same breaker.
+1
if its winter where your at (like it is here), try opening a window. (after you fix the electrical problem.
and MAKE SURE the power cords running to the PSU are HEAVY gauge (thick wires) and not some cheapo thin cords. Voltage drop on thin wire draws more current.
Hi,
I got a R9 280x tri-x (11221-08-40G), this card needs 320 W for mining. So, is it possible to flash the bios on this model to undervolt it ? I’m already looking for the switch who’s located on the 7950, but I don’t find it on my R9 280x.
Thanks for your help…
Just a few general questions…
1) What exactly does –lookup-gap 2 do?
2) Why stop at 3 video cards per mobo? There are guides out there for building 6 GPU rigs, surely that’s gonna be the most efficient setups?
Hi. I recommend you read the readme.txt that comes with cgminer it is very good and explains a lot. Version 3.7.2 has a GPU_ readme also which helped me a great deal.
Btw cryptobadger what currently is the fastest mining card to buy? AMD r290? And best £ per hash?
Cheers
I have a Gigabyte R9 280X (Rev 2)
I am getting 770KH/S using the settings
cgminer –scrypt -o stratum+tcp://blahblah:0000 -u username.pass -p luvmining777 -g 2 -w 256 -I 13 –shaders 2048
You will also need (OC Guru II) Open and set the GPU Clock to 1075 & Memory Clock to 5980 (Anything higher or lower then these settings will cause speed drops)
I am running 100% stable with 78-79C temps. Fan RPM is 3619
Hope this helps.
Undervolted my Gigabyte 280X’s to 1094, stable so far for few hours and temps are sub 70, average rate 720kh/s. Once I get my pci-e risers I’ll probably try to get more kh’s with expense of temperature.
Hi,
there’s a big black button on the card, is this the bios switch button ? If not, what is the use of it ?
@Cryptobadger and everyone owning an 280x Vapor-x,
It seems that the issue with the vapor-x is the latest BIOS version. Apparently since 015.041.xxxxxxx the Vapor-X has an built-in temperature limit.
Also, flashing your BIOS to the previous version seems a no go..
Source: https://litecointalk.org/index.php?topic=10313.0
Hello everyone
What is a better choice
MSI R9 280x or Asus R9 280x?
according to
http://www.tomshardware.com/reviews/radeon-r9-280x-third-party-round-up,3655-5.html
The Asus makes much lower noise has more power phases & has slightly worse temperatures
But i see MSI recommended many more times and hardly any recommendation for Asus in mining builds,also on non mining builds people say Asus is more reliable than MSI(less failures,better vrm)
Am I missing something that gives MSI an advantage over Asus for mining?
I have 2x ASUS TOP II running now, they are great. They undervolted to 1.118 V and hit up to 765 KHash/sec @ 1080/1500. Temps stay down in LOW 70’s, much better than my XFX and Gigbyte (both HIGH 70’s at same settings).
When I bought, they were only ~$20 more, so if you can swing that price differential still, I say go with ASUS.
Cryptobadger… First thank you for your posts, they’re great and very very helpful and well explained.
Secondly… does anybody know where to find the optimized bins to test? Apparently the guy who made it got a bit mad cause people were asking him to make bins and has deleted the post, and with the post all the links to the optimized bins… a real pity!
I’m not good enough to compile my own bins or modify the source code, therefore wanted to try the pre-made bins in a trial and error fashion… but now I don’t even have that possibility 🙁
thanks!
I yhink the sempron recommendation is not the best idea.. Perhaps on a linux box but certainly not for windows. Tested it over last fee days as was building new machinr and decided to try some semprons… Being single core chips they struggle to manage an open wallet, trixx, chminer and whatever else is running in the background. Removing everything and running the rig as a slave node to a head node with the wallet greatly increase hash rate to something close to my usual rigs with multicore fx chips but still not the same. As soon as i remote into the sempron nodes, the hash speed drops. Change the chip to an fx (95wtdp) and output jumps and becomes much more stable. Could be the mobos i use or the fact that windows 8.1 is more demanding on the cpu by nature and so a multicore processor can better manage things. Just tought it be good for other the heard about it if you’ve tried everything and can get to the realm of 740+ kh/s which pretty much all 280x should be able to manage.
That being said have several sempron chips, just bought, box is opened but only a couple have seen actual use for a day or so. If anybody is interested let me know figure at current valuations 1LTC (+shipping/taxes etc) would be fair. While I’m at it, 16x-16x 8inch risers & 1x-16x 8inch risers sealed and brand new are also avail (think 10-20 of then I’ll have to check..)
Well windows is just bloated like hell with that monolitic kernel that does EVERYTHING and can basically not be “downfeatured”.
There’s a reason basically all servers all over the world don’t run with that shit.
On my rig a sempron can easily handle apt-get doing stuff + mining and still doesn’t drop even a tiny bit.
Of course that might be because the USB stick slows every operation down like shit, but well…at least it’s working in my favor 😉
Lol I hear ya. A properlly setup linux box can’t be beat, just dont have the patience or time for it these days. The extra cost that comes with windows is worth it, for me at least.
Good point about the usb stick, hadnt thought about that, like u said, linux can be pared down into a lean mean machine should one so choose to invest the time to do so (and learn how to). Maybe in the future it’s something I’ll look into.
Gah.. Cgminer.. Lol chminer thx for that iphone. Also reference for those whom “can’t” get to 740kh/s and above.. Not those who can..
Hi there,
I just builded up a rig using all of the great tutorials provided here.
My rig is now running on two Sapphire R9 280x, under Xubuntu 12.10. I use cgminer 3.7.2. Both GPU are up and showing in cgminer.
http://img585.imageshack.us/img585/6872/7h9x.jpg
As you can seen everything could be ok, except for one thing: for some strange reason, the second GPU seems to be idling (clockspeed stuck at 300 Mhz). Regarding clockspeed, GPU 0 seems to be just fine: it is now at 1000 Mhz and adapts each time I change the settings. At the moment, there is no monitor attached to any of the GPU (connecting to the rig through SSH), and no dummy plugs either (since there is no VGA output on the cards).
What’s ever stranger, is that both GPU seems to be delivering the same (very poor) hashrate: around 200 – 300 Khz each. How come, since one is running at full speed and the other is idle ? I’ve already tried several settings, changing TC, clockspeed and memspeed, fan speed and such, but I haven’t seen any significative difference so far.
Does anybody have an idea how to solve this, as it’s really driving me nuts ! ^^
Thanks a lot in advance for your help.
Yours,
SYN
SYN, using this for my two Sapphire vapor-X cards and get around 730 each :
./cgminer –scrypt -I 13 –gpu-powertune -20 -g 2 –gpu-engine 1050 –gpu-memclock 1500 –temp-target 70 –auto-fan
What config do you use?
with those settings i am getting ~560khash on my vapor-x using windows and cgminer 3.7.2 :S
I think something is wrong with them :S
I would not use Windows really, it seems to be a bit of an overhead compared to using linux when it comes to litecion-mining
do you have any tutorial on how to instal linux and run litecoin?
Well, I wouldn’t presume I would do a better job at it then Cryptobadger, his guide is what I used, apart from how I did not use the stock ATI-drivers.
Could be a bit of a bother to actually install linux if you don’t have a cdrom or a second usb-stick though. I happened to have two USB-sticks, where I put the install image on one of them and installed XUbuntu on the second.
Windows is fine (just cost a bit more, 1 software licence, 1 ssd drive, better cpu than a sempron).. provided you have a multcore cpu.. Semprons dont handle all the multitasking very well. With linux they ok since the background services etc can be paired down. I use fx6300 for all my rigs running windows and its fine.. since the tdp of the vishera is 95W its only adding around 50W to the total wattage of the system.. So if you have a sempron chip.. I’ve found that the only way to get it to the realm of 700kh/s+ is to remove all pointless services and run it as a headless rig. Only downside is if the remote into the rig.. the hashing rate will drop.
Ivan.. If you’ve never used linux distros before.. just dont do it lol. It’s not a simple as it sounds and its not like windows at all. Unless you a prepeared to do alot of learning on linux and how it works.. you may be biting off more than you can chew esp if all you want out of it is to run a mining rig. To each their own of course.
@Erebus:
Windows is fine for a mining rig, but then you start to list all the stuff that sucks ? lol !
I can start a 1680×1050 xrdp-session on my rig while it is doing a shitload of apt-updates + mining – and the KH/s DON’T drop.
All on a sempron for 17€.
So either windows IS shit on a low-cost-low-power-consumption rig or I am lucky as hell. Ah well, I guess I’m just lucky.
guys at least add either “shaders” or “thread-concurrency” (either the default 8192 or 11200.. I use the later) to your file.
otherwise.. so the idling card.. do you have a dummy plug on it? Not use if ubuntu requires one or not.
Oh, and be sure to update your drivers. Stock ATI Catalyst drivers that comes with XUbuntu are useless, you need to grab the latest ones from ATI (if you are feeling brave, get the newest beta-drivers, i personally don’t use them).
Got my two cards up to around 750 using the below a few minutes ago, don’t know if it’s stable long-term though.
./cgminer –scrypt –lookup-gap 2 –thread-concurrency 8192 -g 2 -I 13 -w 256 –gpu-memclock 1500 –gpu-engine 1080 –gpu-powertune 0 –expiry 1 –scan-time 1 –queue 0 –auto-fan –temp-target 79 –temp-cutoff 85
hi, i have one rig with 2 280x vapor x, and i am having trouble with it. The card on which is monitor connectend only gives me 620mhash od 1015/1500, what ever i try it loweres the hash down.
Does anybody have any sugestions?
What are your computer specs, OS and contents of your conifg or bat file for cgminer pls. That will help greatly. Also litecoin hardware comp list is always a good place to checkout for some config/bat file programming inspiration.
Hi ivansfm,
I have the same problem with my sapphire Vapor-X OC-Boost – Engine Clock is 1000 MHz, Memory 1500, Intensity 20, look-up gab 2 – anything else drives the has-rate down.
I use the setup of cryptobadgers guide under linux except that I have 2x Sapphire HD 7970 and 1x Sapphire R9 280
(latest version of Obuntu, cgminer 3.7.2)
I’m running the HD7970 with 1100 MHz, Engine Clock and 1500 Memory Clock giving me stabel 650 – 670 KH/s.
The R9 280 is running stable at 620 KH/s.
The System with only 2x HD7970 uses 550 Watt at the power plug. With the additional R9 280x it needs 850 Watt, giving the HD7970 a better Hash to power ratio.
The installation of the 79xx optimized cgminer-bin-file with concurrency of 24000 didn’t show any significant improvement (however I’m not quite sure, if i installed it correcty)
I used the VEB-GPU-Bios changer, running the cards stable with 1054 mV voltage, reducing now my power consumption to 700 Watt (150 Watt difference)
Without the change in Voltage the temperature of the 7970 cards was at 79 – 84 °C, now they are running at 69 – 73°C
One annoying problem I have, is that the dummy plug for the GPUs is not working. I have to connect a monitor, otherwise cgminer does not detect any gpus.
Best regards
velniukas
“Engine Clock is 1000 MHz, Memory 1500, Intensity 20, look-up gab 2 – anything else drives the has-rate down”
.. intensity of 20 is good for 7950 radeons… not 7970 or 280x so lower intensity to 13. Look at typical setting for your cards on the litecoin hardware comparison chart. Also its lookup gap not gab. use 2 threads not 1, use thread concurrency of 8192 or 11200 not 24000.. also clock is too slow. But then again this may not apply since you guys use linux. Start using the linux boards and other forums to obtain info and for better “optimized” kernels/drivers etc. If you can’t customized these yourself, then you are at the mercy of the linux community for solutions, and pray that those whom have the technical knowledge to create these files are willing to freely share.
Hi Erebus,
thanks for your comment – in the hardware comparison chart people report better results up to 750 kH/s for intensities of 13 (for the same 7970 and R9 280x cards). For myself, reducing the Intensity to 19,18, … 13 just decreases has-rate proportionally. Might this be influenced by my Moterhboard memory (4 GB)? With increasing clock speed, I get a thermal problem, that’s why, I stick to 1100 MHz for the 7970.
Thread_concurrency of 8192 or 24000 does not influence my has-rates.
Best regards
Where are u putting the thread concurrency line? In a conf or bat file? You technically dont have to specify one, you can use “shaders” instead and let cgminer determine the largest size for TC. You can then check the bin file in the cgminer folder and see which TC number was attributed to it. Before every tun you should delete the bin file. The 280x should all run 720khs and above. May come down to specific hardware combos, hard to say, with windows there are less “software hurdles” to jump over.. Just plug in lastest drivers and press go.. Linux is a different animal all together.. This is why I gave up linux and when full on windows. To each their own, i just got tired of tweaking kernels or asking the linux community for help. Hardware recquirements are lower for linux so 4gb should be sufficient as is stated in Cryptobadger’s guide. For windows 8.1 though, multi-core processor would be a must over the sempron recomendation (unless its for a headless node thats not remoted in much). Lastly, windows, linux etc.. Make sure to turn off anything u don’t need to be running in the background. If this is a dedicated computer for mining.. Then randol programs in the background should be removed (dropbox, antivirus blablabla, make it lean and mean).
add
“delete *.bin”
to your .bat file.
I find that the option “-g 2” makes all the difference on my two Sapphire 280x Vapor-X OC.
The problem with overclocking I think has to do with the specific bios that your cards might have.
I run my two cards stable since yesterday with the following:
./cgminer –scrypt –lookup-gap 2 –thread-concurrency 8192 -g 2 -I 13 -w 256 –gpu-memclock 1500 –gpu-engine 1080 –gpu-powertune 0 –expiry 1 –scan-time 1 –queue 0 –auto-fan –temp-target 79 –temp-cutoff 85
Giving me about 749 a piece @ around 77-78 degrees and 50-60% of full RPM on the fans…
try TC of 11200.. gives me stable gpus running 755 (engine @ 1080) for HIS 280x and 765 (engine @ 1100) for Toxics. Otherwise our setup seem pretty indentical. Would be curious to see what it does for your gpus. As they are Sapphires I’d assume that they’d perform close to their Toxic counterparts.. albeit prob a little louder/hotter.
Oh well, tc of 11200 did not change much, maybe a bit more close to 750 than before, but nothing else changed really.
What other parameters are you using on your cards?
There badically identical to yours. Perhaps somoly comes down to hardware and software combinations. I run the lastest drivers on windows 8.1 machines. Though to be honest the amd drivers have been finicky, sometimes thy’ll run for days, sometimrs they keep crashing for a few tries. A little frustrating.
Hey Erebus, sorry to butt in on this thread but I wanted to ask for your contact e-mail if you still have those risers and/or CPUs available..
Hey Catharicist, yes still have some (erebusmine at gmail).
Live with it mate! Im only getting 520 KH/s on me HIS 7970 ‘s on linux.
Weird thing is when I plugged it into my main pc running Wind 7 it had a lot better hash rate.
Ive now bought Asus r9 280x Direct CU TOP and getting 730 KH/s from that using windows.
Its noisy though! Its noisier than my 3 HIS cards running on my mining rig they are sooo quiet just a shame they’re hash rate is low?
So just an update, I have cut my 5×280 rig down to 4. I had trouble keeping all 5 up and hashing happily. Driver crashes, CGMiner marking a GPU as dead, etc; It is MUCH more stable at 4. Even with the LEPA PSU I think I just wasn’t pushing enough watts to keep them all fed and happy.
What wattage did your Lepa manage?
5×280 means you need at least 1650W to keep your system happy…
I bought the Gigabyte R9 280X (Rev 2) as a 2nd gpu. It went to 90 C in 2 minutes. Took it back and got a replacement Gigabyte R9 280X (Rev 2). Same problem. Don’t have any problems with my Sapphire Vapour x . I am not able to get another Vapour x or Toxic. Does the Asus , XFX or MSI 280x have as good a stability ( re: temp) as the two Sapphire Models? Maybe the Assus DC2T?
its got a lot to do with your ambient air temperature, and the auto fan settings you are using. if you are in a case, you might need to aim a fan at it, or add an additional case fan.
I had the exact same problem with the gigabyte r9 280x’s. I was able to fix it.
1. You must under volt these cards to 1080-1137 (the cards are voltage locked so I used a third party software to flash them down)
2. Set your fan speed to 70% when mining (when the fans got faster than that i got higher temps)
3. Until you get risers (mine are shipping from China) you’ll need a good fan to remove the heat from the back end of the cards.
4. Run both cards at 1020 Mhz and you’ll get 715 Kh/s.
On which voltage are you runing your Sapphire TOXIC GPU’s?
i can’t go lower than 1,200V (seems locked)
one of my both cards seem to be faulty
first card is running smooth with 1080/1500 at 45% Fan speed 75°C
other cards runs on 1000/1500 Fan Speed 55-60% at 75°C
both are set to 1,235V and Powertune 20
it seems that the second card sets a higher voltage(i think powertune allows the Card to set higher voltage) so the heat generation is bigger. Can’t read out the exact voltage cause i’m using BAMT.
I have Asus r9 280x direct cu top running at 730 and 73c and my His 7970 are about the same. Mind you it’s pretty cold here in Wales and I have a huge office fan cooling the buggers.
Better to wear out a couple of desk fans than GPUs!
2x MSI R9 280X and 715 kh/s per each, two weeks stable. temp. around 70-77 when the monitor is disconnect. and the best, i have only approx 590-610 watt for all my rig. without changing voltage (coz i can’t manage this by afterburner i don’t know why). for core and fan speed i used afterburner 1050/1500 and this is my script TC 8192 Shaders 2048 g2 I13 w256
my rig conf.
Corsair rm850 watt, 2x2GB ddr3 hyperx, CPU athlon ii 250u,hdd 160gb 2,5″
What are U think about power consumption?
and asrock extreme4 970 :))
How can i change the voltage?
And i have question it is possible to add one more msi r9 280x into this rig?