Anybody else having problems with their GPUs?
My R9 Fury X started behaving really strangely. It idles at 65C instead of 27, and as soon as I start mining/playing a game it starts overheating, and after a time the computer just stops.
My rigs are totally fine, but the thing why I think it is related to mining is because I was running 6.0 on my computer when it restarted for the first time (this morning) and since then I can not make it work again.
I tried with a different windows, and I got the same behaviour. Any ideas? (the GPU is still under warranty so I am not worried)
It sounds like the fan is dead. |
|
|
|
i'm waiting the new FIJI , lol VEGA gpus next year, when ZEC pumping
AMD will need to pump ZEC heavily to sell a lot of those expensive Vegas )) Hopefully the 1TB/s HBM2 can be properly taken advantage of. |
|
|
|
Right now, I'm focusing on making cards hash more on memory-using algos across the board with VBIOS mods.
If you have any timings for Hynix BFR ICs that are similar in performance to The Stilt's Hynix AFR and Elpida BBBG timings I'd love to test them out  I've not yet experimented with values enough to find out EXACTLY how they interact for best performance. Oh well it was worth a shot. I guess it is a very time-consuming thing to figure out. Stilt said himself it took 50 hours to do the timing set for one card. |
|
|
|
Right now, I'm focusing on making cards hash more on memory-using algos across the board with VBIOS mods.
If you have any timings for Hynix BFR ICs that are similar in performance to The Stilt's Hynix AFR and Elpida BBBG timings I'd love to test them out |
|
|
|
|
Just when I was about to switch to ETC....
|
|
|
|
|
Is Claymore's miner any faster if not dual mining?
|
|
|
|
6хR9 380 periodically resets GPU0 speed, and the miner errors appear non-existent cards, with which it can be connected?
It's a bug, will be fixed in next update. Divide card number by 2. How soon will be next update ? Do you want more speed? Yep and it looks like we are going to need a lot of it to kill this botnet or private miner or whatever. https://www2.coinmine.pl/zec/index.php?page=statistics&action=poolFor those of you that want to hold right now, thanks!  |
|
|
|
|
No, no....they just can't SEE it. And also don't know what time it is.
Come to think of it...you should also add a flashing countdown timer at the bottom of the screen that blinks on and off in bright red!
And every 15 seconds it adds a little message of encouragement.
00:75: Doing good.
00:60: You got this.
00:45: Keep it up!
00:30: Almost there.....
00:15: Al...most...there....
00:00: You did it!
|
|
|
|
Measure time between "DevFee: start mining" and "DevFee: stop mining and disconnect" strings. It's always 90sec.
I do this every time the purple text appears since 2 days. Maybe I interpret something wrong. You are the "boss" (and a very good one). Ok I'll make these lines purple too so you can see it. Did anyone else LOL at this? Better add a time stamp to that line too just to be sure! And use bold letters...and CAPS! And maybe inside some '=' signs. And make it really bright purple, too! Like this: ================= 11/16/16-19:35:00 DEV FEE: START MINING =================
...
========== 11/16/16-19:36:30 DEV FEE: STOP MINING AND DISCONNECT ========== |
|
|
|
Like Quint said, nVidia is way better at F@H than AMD. A GTX 1070 can get at least 600K PPD while folding, which equates to about 25 CURE and 2250 FLDC per day...which in turn equates to about 0.0038BTC per day for a single 1070 at current prices. Looking at whattomine.com I am seeing that for 150 Sol/s (about what a 480 or 290 can do) you net the equivalent of 0.003 - 0.0032 BTC per day. So nVidia has things it is better at and can be more profitable. You just need to mine (or fold for) a coin that is more suited to the hardware. It's not the devs' fault that most algos are better suited to AMD.
i'm curious what altcoin you are folding exactly, the old gridcoin, or foldingcoin? i don't remember such profit for folding since a very long time, also folding is very cpu dependent, you have an hevay load there CURE and FLDC merge folding. FLDC also pays out MAGICFLDC as well as PEPECASH, but those are not worth much. Folding is not CPU intensive at all (unless you are CPU folding) With a decent CPU though you can get another 100k+ PPD if you want to fold on it as well. |
|
|
|
So according to ethman i see 12000 h/s on my rigs but on coinmine.pl zclassic pool i only see 10k never even seen the hashrate get close to what my rigs actually say
i did not have that issue on pool.mn , could it be a latency issue or is coinmine stealing hashes ?
Oh so that power bill you posted was for 1 month!? Shit I thought it was for a whole year lol! So are you paying residential rates? Seems like you should get a better rate for that kind of operation. |
|
|
|
Like Quint said, nVidia is way better at F@H than AMD. A GTX 1070 can get at least 600K PPD while folding, which equates to about 25 CURE and 2250 FLDC per day...which in turn equates to about 0.0038BTC per day for a single 1070 at current prices. Looking at whattomine.com I am seeing that for 150 Sol/s (about what a 480 or 290 can do) you net the equivalent of 0.003 - 0.0032 BTC per day. So nVidia has things it is better at and can be more profitable. You just need to mine (or fold for) a coin that is more suited to the hardware. It's not the devs' fault that most algos are better suited to AMD.
True about needing to do proper workload for profit, but by your calculations, 1070 is 22% more profit per day, which is good until you factor in you can get a 480 for $200 on a good sale, and it is hard to find a 1070 much below $400. 480 per day has been between ~0.0030 BTC and ~0.0042, so it is best for mining right now. Also true about up-front costs. However, ppl complaining that zec miners need to be more optimized for nVidia are barking up the wrong tree. Even if a miner was optimized for nVidia, how much better do you think it's going to be? It's still not going to outperform a 480 in a memory-hard algorithm. Best to find a different coin to mine. |
|
|
|
|
Like Quint said, nVidia is way better at F@H than AMD. A GTX 1070 can get at least 600K PPD while folding, which equates to about 25 CURE and 2250 FLDC per day...which in turn equates to about 0.0038BTC per day for a single 1070 at current prices. Looking at whattomine.com I am seeing that for 150 Sol/s (about what a 480 or 290 can do) you net the equivalent of 0.003 - 0.0032 BTC per day. So nVidia has things it is better at and can be more profitable. You just need to mine (or fold for) a coin that is more suited to the hardware. It's not the devs' fault that most algos are better suited to AMD.
|
|
|
|
|
Ummmm....I'm not sure what it is you're trying to calculate but you might want to check your work again.
|
|
|
|
Factors of 1000 aside...it still seems like 13.5¢/kWh to me.
the coinwarz calculator is what i used to figure out the equivalent price per day with what I am paying for the 14000kwh , it comes to .20 c/kwh Ok but you posted your power bill which was $1985. If you paid 20¢ per kWh on 14MWh, then your bill should have been $2800. $1985 / 14650kWh = $0.135/kWh |
|
|
|
Honestly I am not an expert on all of this but I do know a bit so I just wanted to give you a bit better understanding and more concrete numbers to base your wiring solution off of. Looking at the wiki for Molex though, I am seeing 11A per pin. Since it is @ 12V, this translates to 132W of power. If you look at it this way, one molex chain should be good to provide 66W of power to two cards. Although a card can draw up to 75W through the PCI-E slot, keep in mind that it is sharing some of the load with the mobo/riser.If you really want to get better/more specific advice I'd suggest the forums at http://www.jonnyguru.com/. These guys know their shit. Err I just realized you were talking about risers with USB cables but I was thinking about ribbon cables when talking about load sharing...so disregard that part. The USB cables are only used for signal pins and don't actually provide any power, so in that case all of the power would need to go through the molex. Still though, 1 molex chain per 2 risers is fine. |
|
|
|
|
By 6-pin and 8-pin I mean the end that connects to devices...not the PSU end. So just talking about PCI-E connectors here not Molex. With the risers, there is still a path for power through the motherboard and riser. The molex just adds another path for power in parallel, so not all of the power coming into the card from the bottom is going through the molex cable. Even if it is carrying a little more current than it's supposed to be according to specifications...specs have safety margins built-in already so it's not an absolute maximum. In reality two risers powered by one molex chain isn't going to melt anything. If you're worried about that you could always limit yourself to one molex per riser.
|
|
|
|
|
Show Posts
