No you're not... If i calculated right that is 15 cent /kwh. Thats really high for american standards, but cheap compared to germany. We currently pay 22 (euro)cent per kwh ~ 25 $cent. I'm running two rigs with 1300W. Thats almost 7€ a day just for power.

Anyone knows why I'm getting messed up artifacts in the afterburner windows after setting mem clocks to anything below 400? The artifacts come with load, as soon as I start the miners they show up, nothing when the cards are idle.

BTW my second, exactly the same machine doesn't have this problem.

bring the memory clocks down to 300-350Mhz. Saves a lot of power and the card stays cooler. You should also apply the phoenix tweaks to speeze out more MH/s or try playing around with agression and worksize. My cards are giving me 285MH/s @ 900Mhz core.

sapphire's 5850 fans seem to fail much more likely. I'm running my fans all at 95%, getting a little worried though because of all those reports here. I have to maintain those speeds because the cards are so stuffed together, getting almost 80°C on the hottest card. I bought some risers now, hope that will help. Four cards on one board just cannot safely be kept running without AC...

So the summary tab is updating reeeealy slow (at least for me). It refreshing like every minute or so. "Connecting..." also shows up from time to time even though the worker is mining fine/shares going up. Any way to speed up the update rate?

unless you're using only one GPU, I don't see how this could be true.

Are those some old recycled IDE cables???

Halo zusammen,

kann mir jemand einen o.g. Riser empfehlen? Hab hier im Forum schon von durchgeschmorten Risern gehört und will daher nicht irgendeinen kaufen. Auf Ebay gibts welche aber die verkauft der Vorbesitzer grade deshalb weil sie auf seinem Board nicht funktionieren, und das selbe habe ich auch  (MSI 790FX)

Ich brauch die Dinger nur um mit der Belüftung besser an die Karten zu kommen, PCIex16 Slots hab ich genug, brauche also keine Adapterkarten oder sonstiges.

meiner Meinung nach definitiv ein Hitze-Problem. Meine Karten verrecken mir schon bei 83°C, und das ohne Gehäuse. Mit Gehäuse ist das die reinste Sauna. Dabei solltest du nicht nur an die GPU-Temperatur denken sondern auch an andere Bauteile die von der Abwärme benachbarten Grafikkarten erhitzt werden. Die vertragen zwar auch einiges aber auf keinen Fall über Monate eine sowieso schon überhitzte Grafikkarte nebendran. Da werden als erstes die Spannungswandler den Geist aufgeben.

Lass die Gehäuse weg, besorg dir paar ordentliche Lüfter (laut!) und leg die direkt auf die Grafikkarten. Dann noch ein paar Ventilatoren damit die Lüfter nicht ständig die warme Luft, die sie an den Seiten ausblasen, oben wieder ansaugen.

Anyone tried older drivers (2010) and stream 2.1?

Any news on the unneccesary CPU usage while mining?

The summary tab is showing me "connecting..." for every worker, but in fact is already mining. The accepted shares go up and GPU usage is at 99% as well. If I stop the worker and restart it, the mh/s rate is show for about two seconds, then switches again to "Connecting...". Any way to fix this?

Another question, does GUIMiner have a fail detection to restart the miner automatically if it locks up (or can the backup option somehow be abused for this)?

Just FYI, it's the driver that is crashing and recovering. I still don't know what is causing this though...

Wasn't there some kind of Catalyst / SDK combination that worked well without CPU usage or was I getting something wrong? I heard of 10.12 + 2.1 somewhere, anyone using this?

GPU usage is not as maxed out as it could be compared to just running phoenix in console. Console only gives me always 99% load. With 3 workers I get 98-99% for each, 4 workers even worse, very unstable between 95-99%. Could it be cpu usage? AOCLBF uses up to 21% with 4 workers every two seconds, with 3 workers only up to 10%.

EDIT cpu usage of AOCLBF is even higher when using the temperature control in the overlay, goes up to 40% with 4 cards and reducing MHs rate from 294 to 291 My guess is that this is caused by the constant overlay update that is gathering data every second times 4. Can i somehow adjust the updating speed? Or how about a console view only.

Hi all,

my two mining rigs were running flawless for two weeks but lately it occurs more and more frequent that two workers for always the same GPUs just stop hashing. The console kind of freezes with the last Mh/s rate it was working with, but the cursor at the end of the line is still flashing.
This is how it looks like:


What makes me wonder is why the GPU usage of the other cards goes down as well when the third card fails. Afterburner updating interval is set to 10 seconds and I can see this usage drop on the other cards in the graph every time, so it must take at least some seconds.

I'm using win7/x86, ATI 11.6 + APP 2.4, phoenix miner with phatk kernel, settings: -k phatk DEVICE=2 VECTORS BFI_INT WORKSIZE=256 AGGRESSION=12 FASTLOOP
I also altered the phatk kernel with the formula given in this forum to improve the hash rate. I had some issues with this on our first machine, one miner chrashed the same way every two minutes with this change, but it also happened once just before I made the change so the issue is not cause by this but it makes it more likely to occur.

Anyone knows what is causing this behaviour?

312Mh/s @ 975Mhz core??? Thats impressive.

It's enough, at least for this kind of OC - I'm running the same system. But it should be a gold certified PSU.

Try OC'ing the outside card just a little more, don't touch the other ones. Bring up the fans to 90-100% and try getting the gaps between each card a little bigger to give them some more space to breathe, ideally putting an extra fan ontop.
83°C is close to the  limit, I never go above 80, my hottest card is currently running at 76°C with 960Mhz on each and the outside card at 1000Mhz.

I bought two rigs just two weeks ago and they are running great. This is my setup:

4x Sapphire 5830
MSI 790FX-GD70 (or MSI890FXA-GD70 if not availiable)
a GOOD PSU, 80+ Gold of Platinum certified (Chieftec, OCZ, ...) with at least 800W, 900+ for OC
AMD Sempron 140
2GB DDR3 cheapest availiable
Two cheap HDDs or USB drives
a good fan (12cm or more) to put on the cards

The PSU should be you most expensive part, since it has to supply a big amount of power over a long period. The 80+ certification tells the efficience the PSU converts drained power into BTC You want to lose as little power as possible into warm air - depending on your location power costs can be a major issue (like in germany). Look which PSUs are availiable at the shop of your coice (There shouldn't be many that match all criteria) and check some reviews first!

Case-less is the way to go for cooling matters, just put a fan on the block of cards to blow some fresh air into the gaps. Ideally a big fan faced to the board to spread the warm air. Get a solid, non-burnable surface (I use a big tile) and put all that stuff on top. I used long screws and nuts on the screwholes of the mainboard to get some space between board and tile.

You won't need any riser cards or PCI-E adapters, the MSI boards are just plug and play. All four cards nicely lined up. You should (gently!) try pushing them outwards to get a bigger gap size between the cards and fix those position with the power connector cables or any other construction of your choice. With that done, OC to 950Mhz should be no problem with the hottest card at 74°C on warm days The outside cards can be clocked even higher.

I don't think that is what the OP consideres "cheapest".

I'm working with two MSI 790FX-GD70 - good price for four PCI-E slots. If you still can get it somewhere, go for it, but MSI replaced it with the 890FXA-GD70 which is still very good but costs a bit bit more due to new features which you won't need for mining.