[~1000 GH/sec] BTC Guild - 0% Fee Pool, LP, SSL, Full Precision, and More

[Clipse]

mmm starting to think long polling isnt working all that well(or at all?) on btcguild.

I am consistently >4% stales with btcguild and my deepbit is consistently <2%

[compro01]

miner idle message don't seem to be working.  my rig just went down for an hour due to a power outage and no email came. 

next to the miner idle setting is a warning saying that my donation for the previous round was under 2%, but my donation is set for 2.5% and the unconfirmed reward is 0, with everything in confirmed, so that is working.

[nomnomnom]

...

Hey, somebody made poclbm with the phatk kernel, maybe this is something for you xD
http://forum.bitcoin.org/index.php?topic=19169.0

[Veldy]

mmm starting to think long polling isnt working all that well(or at all?) on btcguild.

I am consistently >4% stales with btcguild and my deepbit is consistently <2%

Using Phoenix?  It isn't stable with long polling.    Yesterday, three blocks were solved in rapid succession leading to three long polling pushes in very close proximity to each other; for me, two in a row with no submitted shares between the two.  Went for another share or two if I remember correctly and then I noticed the red color of the worker on the site and went to take a look.  Found it sitting there, looking as if it was running [showed full hash rate, history, accepted/rejected shares], but GPU was at 0% utilization and I noticed that phoenix couldn't even be stopped via the normal Ctrl-C ... bye.  Killing the shell window stopped the process easily enough and I restarted.  Problem is that Phoenix is always giving me trouble [active development die on phoenix or what?].

[Clipse]

Yeh using phoenix, I have no issue with phoenix on deepbit ?

[Druas]

For the record, since Phoenix is never fixed, I hate Phoenix ... love hate ... I like the output a little better, but it is the most unreliable miner ... I wish they would fix it, but no news is bad news.
Problem is that Phoenix is always giving me trouble [active development die on phoenix or what?].

I am sure you have tried it, but Diablo Miner is my favorite. Very active developer and with the vectors and -f1 I get the same hash rate as phoenix.

[grod]

Still showing as no credit for any work starting with block 989.  I guess you're welcome to the 15 or so 5830-hours as an unexpected donation -- hope it's the pool op who got the bonus instead of a random person. It's been a great pool until then.

Also confirming I'm getting way fewer connection/stale issues at deepbit with the phoenix miner.  Very short and very occasional miner idles, and actual credit for work done. 

[mike85123]

Still showing as no credit for any work starting with block 989.  I guess you're welcome to the 15 or so 5830-hours as an unexpected donation -- hope it's the pool op who got the bonus instead of a random person. It's been a great pool until then.

I got credit for 989.

[Veldy]

Yeh using phoenix, I have no issue with phoenix on deepbit ?

I have an issue with phoenix on both when something changes with the pool [or several rapid long polling push go through].  POCLBM ALWAYS recovers for me.

EDIT:  I just noticed that they finally came out with phoenix 1.50 (never saw a 1.49).  Anyway, I tentatively updated the software [shared among machines making for easy upgrade ].  It is way to early to tell, but so far not a single stale; having said that, no stales in poclbm either, so your mileage may vary.  If it proves to be stable (no real statistical difference in stales compared to poclbm and no lockups or faults on long polling or idles) then I will move my other workers to Phoenix and away from poclbm.  Why?  Phoenix seems to maintain hashrate a little better than poclbm when the machine is in use (doesn't seem to matter what -f value I pass, but if I go much below 30 on my workstation that I use for work every day [9-13 hours straight] and play, the lower -f value causes a bit of desktop lag and hash rate still drops when moving windows around for instance [I have Aero turned on for this machine].  For whatever reason, Phoenix, doesn't give me that problem while maintaining a good balance (virtually no noticeable desktop lag, unless some site pops a YouTube video which results in first video and then a hard lock ... damn ATI/AMD drivers (Catalyst 11.6).  They have never been known to have all that stable of drivers for their latest series of hardware.

Here is what I use with Phoenix on my work station (heavy use).  Dual monitors with high resolution on both [not that cheap 1920x1080 crap they are selling now days, 2048x1152 and 1680x1050 respectively ... second is older and my side monitor, but both are 16x10 ratio which is much nicer than 16:9 if you do any development or documentation work IMHO].

MSI Radeon 6970 (essentially reference board) with core clock pushed just a little to 920MHz running at about 76-78C (have had to increase the fan ... I think it is time to blow dust and dander out ).

VECTORS BFI_INT FASTLOOP WORKSIZE=128 AGGRESSION=9 -k poclbm
VECTORS BFI_INT FASTLOOP WORKSIZE=128 AGGRESSION=9 -k poclbm

[Bitonetta]

NICE!
Account workers are now alphabetized.   

[eleuthria]

NICE!
Account workers are now alphabetized.   

3 updates so far this morning:
1) Account workers alphabetized.
2) Getwork spammer detection script optimized (very quickly fixed NL1's issues).
3) Failed login timer - If your IP address has been failing logins, it will begin to enforce a timeout window before accepting new login credentials.  This timer grows exponentially after the third failed login.  It is entirely IP based, not account based, so a brute force attempt cannot circumvent the timer by successfully logging into a known account.

[Veldy]

NICE!
Account workers are now alphabetized.   

3 updates so far this morning:
1) Account workers alphabetized.
2) Getwork spammer detection script optimized (very quickly fixed NL1's issues).
3) Failed login timer - If your IP address has been failing logins, it will begin to enforce a timeout window before accepting new login credentials.  This timer grows exponentially after the third failed login.  It is entirely IP based, not account based, so a brute force attempt cannot circumvent the timer by successfully logging into a known account.

1) is quite nice.  2) and 3) ... EXCELLENT!  That is better than anything any other pool does that I am aware of.  Later tonight, when I get the stats I am gathering, I will move back to your pool, payoff my ~-0.25BTC debt and then earn some more.  I am very happy indeed to see such an innovative approach to solving a common attack vector.  Further, it may keep some miners [i.e. Phoenix 1.48 .. jury is still out on 1.50 for me] from crashing outright in many cases.

As always, you are the best!

Thank you.

[hugolp]

1) Account workers alphabetized.

Why? I and Im sure a lot others had the workers in order and now they are in disorder.

I know you are just trying to improve the site, but maybe making it optional would be better?

[Carnth]

1) Account workers alphabetized.

Why? I and Im sure a lot others had the workers in order and now they are in disorder.

Some people have their workers setup like this:

Computer_A-GPU_A
Computer_A-GPU_B
Computer_A-GPU_C
Computer_B-GPU_A
Computer_B-GPU_B
Computer_C-GPU_A
Computer_C-GPU_B


So, having everything alphabetized makes everything nice-n-tidy™

[mike85123]

3 updates so far this morning:
1) Account workers alphabetized.

Just did some renaming and they didn't alphabetize themselves... was your alphabetizing a one time deal or should renames and new workers put themselves in the correct order?

[eleuthria]

3 updates so far this morning:
1) Account workers alphabetized.

Just did some renaming and they didn't alphabetize themselves... was your alphabetizing a one time deal or should renames and new workers put themselves in the correct order?
[/quote]

It pulls them on an alphabetical sort on the My Account page.  Adding the sort to My Workers page now.

[mike85123]

It pulls them on an alphabetical sort on the My Account page.  Adding the sort to My Workers page now.

I figured out what was going on. I had 1 machine go down the other day so I hid the workers. Just got it fixed and they were still hidden but submitting shares. The workers were out of alpha order and sitting at the bottom of the list, prob because they were hidden but submitting. Once I unhid them they popped up into correct order.

[mike85123]

Anyone see how hard we have been killing it the last 45 mins??

7 BLOCKS in 45 mins

back (133108) to back (133109) to back (133110) to back(133111) to back (133112) and then a couple blocks later another back (133115) to back (133116).

[zerokwel]

Anyone see how hard we have been killing it the last 45 mins??

7 BLOCKS in 45 mins

back (133108) to back (133109) to back (133110) to back(133111) to back (133112) and then a couple blocks later another back (133115) to back (133116).

WOOT come on team BTCG... keep pushing them mumber's daddy needs another video card.

kiss em rabbits feet. etc

[MSJHWT]

Anyone see how hard we have been killing it the last 45 mins??

7 BLOCKS in 45 mins

back (133108) to back (133109) to back (133110) to back(133111) to back (133112) and then a couple blocks later another back (133115) to back (133116).

Crazy good luck, "Luck in past 24 hours       975621 shares   (+41.4%) " lol