No support for Stratum or GBT, I believe.

It's been mentioned here that the 97% PPS bonus doesn't show up in stats, but you should be getting 97/95 of your rewards when you get to that level. I got paid exactly the amount that was shown in stats for last week, while I had more than the required number of shares for 97% PPS. The higher payout obviously can't be calculated using the 97/95 formula anymore, with higher PPS periods during the last few weeks. Anyway, I definitely should be getting more than what was shown in stats, since those don't show the bonus if you reach 1.125M shares for the week.

I agree with GenTarkin on this one. They clearly defined the rule, stating that loyalty points are earned until price hits $2.50 / 1GH/s day. That mention about "a couple of months" can't be anything but their estimate on how long BTC will stay above that price floor, when you also look at the actual rule. Which is now being changed?

Is the new client going to get something similar to cgminer's intensity setting? IMO this is a must-have feature for non-dedicated rigs, and it looks like you are also trying to reach the less professional miners. I can easily play Battlefield 3 at a stable 60fps, and have cgminer in the background using whatever GPU resources are left over.

Sent an email to support, because my payment is still missing. I do have over 1 BTC earned.

I'm a bit confused about the fee reduction bonus. The following is stated in OP:


Yet the week just changed in stats. I thought I still had a day to gather those 1.125M shares, which I could've just managed to achieve. Is the actual fee bonus calculation done differently? I've only been mining for five days now at CoinLab.

EDIT: Never mind, I missed the announcement on second post. It's weird to have conflicting information on those two first posts. Guess I don't get the fee reduction for my first week then, even though I averaged over 10GH/s.

Payments don't work for me either. Site says that payment is sent, but I can repeat that every 2 minutes and nothing actually happens.

My hashrate was reported correctly (6000-6100MH/s) during 100% PPS, but when it dropped back to 95% my hashrate is now reported as 5700-5900. I read about similar problems in this thread earlier. Nothing critical if the shares are still counted properly.

Isn't the client supposed to have an API? Because AMD drivers already require running X, the GUI requirement shouldn't be too much of a problem if the client can be accessed from the API.

Are you still using SDK 2.1? You said you tried several different SDKs, but uninstalling them is actually quite tricky. SDK 2.1 was optimal for the 5000 series, but with newer driver it causes exactly the problem you describe: card hashes, but never produces shares.

Start your client with -rescan option, that should scan the blockchain for missing transactions.

Do I still get the fee reduction bonus for my first week, if I joined in the middle of the week and have been hashing at a bit over 10GH/s? That won't reach 1.125M shares before monday.

I am also getting that same long poll error when connecting to hashpower.com. With 2.8.7 long polling works there.

I'm getting this with cgminer:

No suitable long-poll found for pool http://hashpower.com

Is long-polling really not supported, or is there something wrong? My rejects seem to be quite high.

EDIT: Never mind, this seems to be a problem with the new cgminer 2.9.1.

I have a bunch of receive addresses on my wallet, mostly for different mining pools. Sometimes when renaming a label, another label undergoes the same rename operation. I investigated this more closely and noticed that every time I rename the label of a certain address, it always causes the additional renaming of a certain other address. For example, every time I rename my Deepbit address, my Mt.Gox address is renamed to whatever I renamed the Deepbit address to. The other label isn't actually permanently renamed, because it reverts to it's original name after I restart the client. This is certainly a bug, and I could help debugging it if I had some instructions.

I'm getting noticeably higher rejects when mining through rollproxy. Long poll is enabled on both cgminer and rollproxy, but my rejects are still 1-2% compared to the 0.5% max when directly connecting. There's no additional network delay since rollproxy is running on localhost. Has anyone else noticed higher reject rates? What might be the cause?

Maybe you have ULPS enabled on your graphics card? That can cause the crashes you describe, because the GPU will be powered off when there's no load or display activity. There are plenty of posts about it here.

I'm not very experienced in debugging driver problems like this, but can you check your C:\Windows\System32\OpenCL.dll file stats? Mine shows "Date modified 23.10.2012" even though I just reinstalled 12.8 today. The file size is 53.5kB. There's also another file C:\Windows\SysWOW64\OpenCL.dll with the same modification date, but 49.0kB size.

Just trying to shoot randomly here, from what little experience I have reading about similar problems... It's a little odd that the file modification date is 23.10.2012, maybe it's left over from the 12.10 release? I don't know how a file from 12.8 could have that date.

Oh and also, my OpenCL version reported by cgminer is 1084.2. I found this with google:

http://hashcat.net/forum/thread-1692-post-9659.html

"If the kernel contains 1084.2_(VM) inside its filename you know your catalyst driver version wasnt downgraded correctly since this is the version of some version of catalyst 12.9 beta and higher. As other members said above, you need exactly catalyst 12.8 to run oclHashcat-* release versions."

Suggests that the OpenCL wasn't downgraded properly to the version that comes with 12.8.

After some digging around it seems that 10.0.938.2 is the OpenCL version that comes with 12.8, while 10.0.1084.2 comes with 12.11 beta. I tried deleting the OpenCL.dll files and reinstalling 12.8, but I just got the same version back. Different MD5 hash and modification date, but still 1084.2. Looks like it'll take a while to learn how to downgrade this.

OK, at least I got the OpenCL version downgraded finally. I had to manually delete these files after express uninstalling all AMD software:

C:\Windows\System32\OpenCL.dll
C:\Windows\System32\amdocl64.dll
C:\Windows\SysWOW64\OpenCL.dll
C:\Windows\SysWOW64\amdocl.dll

I don't have time to test the games currently but this looks promising. My hashrates are back to 12.8 speeds, I actually got 1% more with the newer OpenCL at same clocks. Not too bad if it stops games from crashing, of course.

This sounds exactly like what I'm experiencing. Haven't tried BF3 yet, but other games were crashing. I uninstalled everything and returned to 12.8, but playing games still result in display driver crashes when mining at the same time. Before trying 12.10 / 12.11 beta, I didn't have these problems. I quickly changed from 12.10 to 12.11 beta, because 12.10 was causing constant explorer.exe crashes, so I can't say which one caused these problems. I also had constant artifacting on desktop with the 12.11 beta, only when mining.

A few suggestions to help users track their rewards (any of these would be fine):

1) Add a column to the Blocks statistics page, showing the users reward for each block. I know this info is buried on the Transaction history page, but the Blocks page is more readable currently (see next alternative).

2) Give the Transaction history page similar filtering options that are available on Blocks page (BTC/NMC/ALL).

3) Show daily rewards (maybe averaged over 7 days) somewhere.