If I block the internet from bitcoin-qt, changing the settings works, but there's still ~5secs of unresponsiveness. So then I checked out what it wrote in debug.log after it got up again:

2014-02-09 19:21:57 No valid UPnP IGDs found
2014-02-09 19:21:57 upnp thread exit
2014-02-09 19:21:57 upnp thread start

So I disabled upnp and it now works like normal.


Fixed I guess...

Was about to try bitcoin payment on humble bundle. Didn't know what the "miner fee" part was all about so I was unsure if there would be any fee on my transaction. Went to options to see that my fee was at 0.000000000, pushed reset options. Went in options again because I didn't want to see the ui in my mother language. "Need to restart client until this option is in effect"... ok. I shut it down but didn't relaunch, instead I decided to update it to latest 0.8.6 (I had 0.8.5). After updating, I went into options again as it was in crappy language. Now it will always go into unresponsive-mode after pressing "ok"(even if I didn't change anything, apply greyed) or "apply", cancel works correctly. Giving it tens of minutes didn't work either, I press the "x" and windows wants to shut it down forcefully.

Stuff that doesn't work:
- Reinstalling latest version
- Uninstaller(or windows uninstall, same thing I guess)
- Deleting roaming\bitcoin (I have backup)
- Deleting everything "bitcoin"-related with regedit
- Restarting PC
- Trying installing with different user

Stuff that does work:
- Installing version 0.8.1 or older (0.8.2 or later doesn't seem to work)

I did notice that from 0.8.1 to 0.8.2 the installer-icon has changed : ) If that has to do with anything. So atm if I want to change any settings in options, I need to install 0.8.1 to change them, then install latest bitcoin-qt back. Otherwise it seems to work fine, it's finding peers and collecting data. But I don't even want to use this software for finance stuff as it's not working fully.

I have changed settings many times after 0.8.2 in the past, but for some reason after updating to 0.8.6 it doesn't work anymore. Using Windows 8.0.

How to fix?

If you are going to use them for gaming, having multiple GPU and just 1GB memory makes no sense.

stock: 225W per card + 95W for cpu = 770W, now add HDD+fans+mobo+overclocks+usb+misc, it gets pretty close if even higher than 850W. Even if it's left below 850W, the PSU will degrade over time and you might see results of it in 1-2years.

I would go for single 7970(their price should drop as 680 is ~equal/better with ~same/lower price), which is about same gaming performance than 2x 6950. Also it gets rid of nasty crossfire/sli problems. Or wait for 680 gtx optimized bitcoin mining results : )

As the mobo only has pci-e x1 and not x4, for my knowledge it will require some extra modding to give the card enough power. Or just a proper x1->4x/x8/16 adapter.

What people are saying in the reviews is that you can't go over specific powerdraw which is about ~+30%(up to 250W). So the card can't go past 1200Mhz-1300Mhz depending on load. GPU boost seems to look tricky shuffle between temperature, voltage and coreclock. GPU boost impossible to disable?... hope not. Memory clock isn't changed.

Hopefully MSI/ASUS will release their custom cooler cards with 6+8 pin : )

Can people post room temperatures also with gpu and %fan. As room temp has pretty huge impact on %fan.

For me 275 gtx mining, from previous 20110428 to current 20110627 -> -12% performance.

132409 block was very quick and I noticed weird things.

01:47:24 -> 01:48:02.

my miners did following(my time is +0-5sec):

1:
bitcoins.lc:8080 22/06/2011 01:47:05, 496c61e4, accepted
bitcoins.lc:8080 22/06/2011 01:47:58, long poll: new block 000005cd29c1e557
bitcoins.lc:8080 22/06/2011 01:47:59, Using new LP URL /LP
bitcoins.lc:8080 22/06/2011 01:47:59, LP connected to bitcoins.lc:8080
bitcoins.lc:8080 22/06/2011 01:48:05, 6d389633, accepted
bitcoins.lc:8080 22/06/2011 01:48:18, 42704dbd, invalid or stale
bitcoins.lc:8080 22/06/2011 01:48:18, long poll: new block 0000013b88e0686d

2:
bitcoins.lc:8080 22/06/2011 01:46:57, 85778ec9, accepted
bitcoins.lc:8080 22/06/2011 01:48:00, ffe68765, invalid or stale
bitcoins.lc:8080 22/06/2011 01:48:00, long poll: new block 000005cd29c1e557
bitcoins.lc:8080 22/06/2011 01:48:03, 30cf1cb1, accepted
bitcoins.lc:8080 22/06/2011 01:48:06, aa5e0a73, accepted
bitcoins.lc:8080 22/06/2011 01:48:15, d61ffd3b, invalid or stale
bitcoins.lc:8080 22/06/2011 01:48:15, long poll: new block 0000013b88e0686d

Unlucky for me that both of them did nothing in the 47 minute(every 13sec avg for both), but weird is that I did work for this fast round, that got accepted. Does server just look how many accepted shares there was between 01:47:24 -> 01:48:02 instead of calcing the true accepted for that round?

Edit: I had no shares for this fast round. Rx: N/A +0.00000000

Seems like my slow nvidia 275 gtx has less problems, but my 7 times faster 6950s are constantly dropping to backup pool.

Getting rpc timeouts every now and then, forcing the miner to connect to backup pool for atleast 50 works.

Tnx :) Updated it to look better.

Just made that to show the relative nice stats on miners point of view. I guess these infos would be easy to add on the "round-information" page.

I found a rare thing to be "stuck" while mining. Somehow the pool send work for over 30minutes(manually shut it down at this point) or was sending all the time, but every completed work was "invalid/connection error" as the server was down or full.

So maybe implement to swap pool if there's suddently a lot of invalids.

ah.. must be side effect of this problem:

http://forum.bitcoin.org/index.php?topic=7760.msg194957#msg194957

guess we will see ~random earnings until this is fixed? :)

Weird earned statistics:

475   130056    2011-06-11 15:03:10    0:31:12 duration    47 my shares 0.00550424 earned. This would be 107Mh/s for me during this round.

That makes like 1.5 shares per minute, I had a look at poclbm and I wasn't able to find this slowdown within 2hours. I have 2x 6950 doing at ~670Mh/s total.

Previously I also got this kind of round:

464   129962    2011-06-11 05:38:38    0:40:54 duration   1117   my shares 0.04807224 earned. 1955Mh/s over 40mins...


So there's something weird going on..

I noticed 300-400 memory range gave huge amount of more heat than 760.

I was mining at US-Eligius and it went down. But the miner just spammed this:

[6/8/11 2:13:51 PM] Rejected block 1075 found on Cayman (#2)
[6/8/11 2:13:51 PM] ERROR: Connection failed: connect timed out

It kept working for 30 minutes, every found block was labeled as rejected and connection failed spam. I had a backup miner pointed at BTC-Guild, but it was still working with minimum of 2Mh/s, the miner pointed to Eligius was still working at full load(692Mh/s). I had to manually shut it down for the backup to kick in.

Is this a problem with DiabloMiner or with the pool giving 30+ minutes worth of work? :)

Eligius personal info support would be great too.

http://eligius.st/~luke-jr/

Personal infos found mainly in "balances.json", which is further in. EU and US are seperate.

Eligius uses the bitcoin address as username, so can use that to pull data.

Expensive one is better for longer time, cheaper one is more safer.

no case, best case

About the dropping GPU randomly... My 6950 seems to sometimes hop from 77-> 80 for one second and then go back to 77, increasing fan speed in the same time. Maybe hashkill for some reason detects this spiky(maybe more than 80c) temperature measure and shuts GPU down?

Problem went away after 2 first tries... weird. Anyway now it's telling that my hashrate is up to 684, which is kind of lower than previous versions 704(altough it never provided no where near 700).