 Show Posts
|
|
Pages: [1]
|
|
If apt-get isnt available, youre using the wrong OS.
|
|
|
|
|
Unless you are pushing insane overclocks, the GPU will become unprofitable before it becomes unstable (or dies).
|
|
|
|
|
That's really low for a 1050ti (I can get 13-14 MH/s with minimal tweaking, and 11-12 with no tweaking). Which miner software? You shouldnt need to touch GPU BIOS or worry about motherboards.
|
|
|
|
|
Very nice work with CUDA daggerhashimoto -- getting about 14.2 MH/s @ 52W on a 1050Ti with minimal card tweaking, which is about 5% faster with 15% less energy than Claymore, which was the previous best for my card.
I'll see how stable and profitable it is over the next day or so, then try some dual mining.
|
|
|
|
|
Mine is without a 6-pin, and generally hovers around 1.8GHz core clock during mining. It does have Samsung memory, which I have had 0 issue longterm running at +667 (for ethash). Its a EVGA SSC, so probably has better binned chips than some other cards.
Again, I haven't tried to tune equihash much, because ethash has been more profitable for a while.
|
|
|
|
|
170-175 sol/s with EWBF on a 1050Ti with a mild overclock. Could probably push it to 175-180, but not sure how stable it would be.
|
|
|
|
|
If you can really buy those AMD cards for those prices today, you should do it -- and forget mining, sell them on eBay. They are impossible to find right now (in the US), and going for 50-100% over retail cost on eBay.
|
|
|
|
|
Is anyone having the pool actually consistently pick the most profitable coins to mine? I let NemosMiner 1.7.6 run for an hour or so, and it never picked the most profitable algo, despite switching every 5 mins or so. Some of the algos picked were 3-5x less profitable, so something is wrong with the algo-switching logic (I also don't believe profitability can possibly switch that quickly.)
I dont think this is a NemosMiner problem, but curious if anyone else sees it.
Back to single algo mining for me.
|
|
|
|
|
1050 Ti:
~13MH/s ETH with +500 mem. ~87H/s ZEC stock, ~89H/s with +250 mem.
|
|
|
|
Higher clocks at equal voltage == better energy efficiency.
Well, no. Voltage is only one part of power usage, higher clocked cards use more current than lower clocked cards at the same voltage. That's why overclocked cards get hotter even when not overvolted. |
|
|
|
Getting numerous errors along these lines: 2012-01-26 20:45:16.563426 > Unhandled Error
2012-01-26 20:45:16.563545 > Traceback (most recent call last):
2012-01-26 20:45:16.563710 > File "/usr/lib/python2.7/site-packages/twisted/python/log.py", line 69, in callWithContext
2012-01-26 20:45:16.563815 > return context.call({ILogContext: newCtx}, func, *args, **kw)
2012-01-26 20:45:16.563914 > File "/usr/lib/python2.7/site-packages/twisted/python/context.py", line 118, in callWithContext
2012-01-26 20:45:16.564013 > return self.currentContext().callWithContext(ctx, func, *args, **kw)
2012-01-26 20:45:16.564120 > File "/usr/lib/python2.7/site-packages/twisted/python/context.py", line 81, in callWithContext
2012-01-26 20:45:16.564219 > return func(*args,**kw)
2012-01-26 20:45:16.564316 > File "/usr/lib/python2.7/site-packages/twisted/internet/posixbase.py", line 581, in _doReadOrWrite
2012-01-26 20:45:16.564457 > why = selectable.doRead()
2012-01-26 20:45:16.564554 > --- <exception caught here> ---
2012-01-26 20:45:16.564651 > File "/usr/lib/python2.7/site-packages/twisted/internet/udp.py", line 131, in doRead
2012-01-26 20:45:16.564749 > self.protocol.datagramReceived(data, addr)
2012-01-26 20:45:16.564845 > File "/home/adam/p2pool/nattraverso/pynupnp/upnp.py", line 462, in datagramReceived
2012-01-26 20:45:16.564943 > result = client.getPage(url=loc)
2012-01-26 20:45:16.565038 > File "/usr/lib/python2.7/site-packages/twisted/web/client.py", line 605, in getPage
2012-01-26 20:45:16.565135 > *args, **kwargs).deferred
2012-01-26 20:45:16.565231 > File "/usr/lib/python2.7/site-packages/twisted/web/client.py", line 580, in _makeGetterFactory
2012-01-26 20:45:16.565328 > scheme, host, port, path = _parse(url)
2012-01-26 20:45:16.565424 > File "/usr/lib/python2.7/site-packages/twisted/web/client.py", line 552, in _parse
2012-01-26 20:45:16.565521 > host, port = host.split(':')
2012-01-26 20:45:16.565616 > exceptions.ValueError: too many values to unpack
Cant tell whether this is a problem with twisted or p2pool, though it doesnt actually appear to affect function of p2pool. Just a tad annoying to see it frequently scrolling by. For what its worth I'm running arch linux with twisted 11.1.0. Tried both p2pool 0.8.1 and latest git, happens on both. |
|
|
|
The latest CGMiner Windows executable fails to start mining giving these errors Error: clSetKernelArg of all params failed.
Thread 0 failure, exiting
Error: clSetKernelArg of all params failed.
Thread 1 failure, exiting
et cetera et cetera... Same problem on self-compiled OSX (10.7) version. Linux 1.5.4 appears to be working well. |
|
|
|
|
"At no point has AMD specified when a GPU using GCN will appear, so it’s very much a guessing game". It is rumored and extremely likely that the 7xxx GPUs, which have reportedly taped out and are quite possibly in production, are NOT this "future generation" AMD GPU. They will be based on a die-shrunk and likely tweaked 5xxx/6xxx-like GPU, meaning they should have similar performance characteristics to older generations while likely improving some workloads due to evolutionary changes, and improving performance/watt because of the die shrink.
No one knows how they will perform in any given situation of course, but I would expect them to be comparable to previous generations.
|
|
|
|
|
+1 for a sticky. This is working great for me with ~same Mh (maybe 1-2% slower) as phoenix/phatk in Win7 64-bit, yet with CPU essentially at idle instead of 1 core maxed. System is noticeably quieter and I expect it uses less energy as well. Will be sending you some BTC if it proves to be stable with low reject rate.
Highly recommended for ATI users, especially in these hot months where every watt counts.
|
|
|
|
|
Great news on the new server, thanks for all your hard work eleuthria! I'll definitely be upping my donation percentage once the new servers are online and (hopefully) the idling and disconnect problems are resolved.
Are we moving to a uswest.btcguild.com/useast.btcguild.com/eu.btcguild.com type of situation, or will connecting to btcguild.com attempt to auto-negotiate the closest server?
|
|
|
|
Obligatory NVIDIA cards suck for mining comment  . You should really see if the amount of electricity you are using is going to get you enough coins to pay the power bill. Otherwise it would make more sense to just buy them. That being said, which miner are you using? What settings? |
|
|
|
So if I were to get some 5830s, what it the current recommended config?
Catalyst 10.11 with sdk 2.1?
Newest catalyst with sdk 2.4?
linux/windows?
And, under each of those settings, would you be using poclbm or phatk?
I only ask because as time progresses, I notice that what people do changes, and I'm wondering what has been established as "best".
Haven't done extensive testing but I'm currently getting about 300Mh/s +/- a few with: Win 7 x64 Catalyst 11.5 phoenix 1.48/phatk TriXX to adjust to 996/300 (MSI afterburner couldn't underclock my memory nearly that far) It's been stable for 24+ hours. Would attempt slightly higher clocks, but I'm not sure its worth it.... already near 25% overclock. Would love to see some linux results, but as near as I can tell, the clocks I'm getting in windows require a BIOS flash to attain in linux, which just isn't a risk I'm willing to take. |
|
|
|
|
Thanks PcChip, that worked like a charm! I should've mentioned I was using a sapphire card in my orig. post.
Sapphire users, check out TriXX, it seems to work better on our cards than the oft-recommended MSI Afterburner. No need to mess with ATI driver install, the default standard set is fine.
>300Mh/s, 5830 @ 996/300. Custom fan control keeps it around 65C, which I think is acceptable for long term use.
|
|
|
|
|
Trying to lower memory clocks on my 5830 below 650 is giving me instant blank screens with vertical blue lines, seemingly these speeds are not accepted by my hardware. I've tried 600, 500, 300, all with the same problem. Moving from 800/1000 (stock speed) to 950/650 has been rock stable but i'd like to drop it the memory clock farther if it will save me temp, energy and probably card life. Anyone else have this problem?
I saw the posts about completely removing drivers and reinstalling a limited set, but I haven't had trouble changing clocks until I try to apply mem clocks 600Mhz or less... so I don't think its a problem with the ATI drivers interfering.
|
|
|
|
|
