[~1000 GH/sec] BTC Guild - 0% Fee Pool, LP, SSL, Full Precision, and More - page 119.

Veldy

member

Activity: 98

Merit: 10

Quote from: Kehaar on June 05, 2011, 08:07:08 AM

Quote from: Allenssmart on June 05, 2011, 08:00:58 AM

I've been mining for 13 hours straight with 100% usage on both gpus and temps at 65C and 67C. Is it safe to run this 24/7?

That temperature is fine for chips. Worry when it gets to 80C+ (even then it'll probably be fine). Some cards idle at like 60+

Hardware will stop itself from overheating - either by drawing less power and slowing down / increasing fan speed - or turning off the computer alltogether.

Fatigue happens if the the hardware expands and contracts a lot (increase/decrease in temperature).

I imagine going between 40 - 70C wouldn't be much of a problem tbh, though I'm not an expert.

Temperature cycling is NOT fine for anything. It causes "fatigue" wear. Any 3rd year mechanical, civil or aerospace engineer or materials science major will know this without thinking 1 second about it. Since there are MANY MANY parts in a board and only one needs to fail to kill it, fatigue is a nightmare for them. Mining is not the norm for a gaming card ... gaming is. Gaming runs at whatever is required while you are playing and then goes back to typical idle when finished. This is cycling over long periods of time and much few times during the life of the card. With mining, it is happening as often as once per minute in this pool.

So, sorry, but you are absolutely incorrect about the damage temperature cycling can do, even if below 70C the entire time.

Veldy

member

Activity: 98

Merit: 10

Quote from: Allenssmart on June 05, 2011, 08:00:58 AM

I've been mining for 13 hours straight with 100% usage on both gpus and temps at 65C and 67C. Is it safe to run this 24/7?

It seems Phoenix will continue to process stale work if new work is not available and old work still is. My stale rate went up massively overnight compared to when I was mining on deepbit. One miner was nearly 9%. One of my machines runs poclbm and the many of the idles were too short to cause the CPU usage and temperature to drop, but not all ... in fact, there were enough that there was always one showing on my graph in MSI Afterburner. I had to ditch the pool until this is correct as my hardware is too important to fatigue it unnecessarily.

niooron

full member

Activity: 193

Merit: 100

I'm also having problems with phoenix sending stale shares after LP pushes new work. What's up with that?

Allenssmart

newbie

Activity: 23

Merit: 0

http://img718.imageshack.us/img718/835/mineridle.png

Allenssmart

newbie

Activity: 23

Merit: 0

Quote from: eleuthria on June 04, 2011, 11:50:50 PM

Both the frequency and severity of the idles had decreased significantly over the last hour, but they are still poking around once in a while. At 650 GH/sec+, it's becoming quite obvious that a single server is not only a liability, but a handicap.

I'm now working on setting up another server, and tying the two together so that stats, login credentials, and block allocations are properly split between them. This one will probably still be in the US, but I'll be looking for one on the east coast to compliment our current Los Angeles server. Once setup, I will include instructions on the website for pointing your miners towards the closer server if you are inclined to do so.

The issue is getting worse. In the past 5 minutes, around 45% of my hashes are getting miner idle. I've already forwarded port 8332 so I don't think that's the problem.

The Koolio

hero member

Activity: 560

Merit: 500

Minds are like parachutes they work best when open

Quote from: Allenssmart on June 05, 2011, 08:00:58 AM

I've been mining for 13 hours straight with 100% usage on both gpus and temps at 65C and 67C. Is it safe to run this 24/7?

should be fine, Ive been mining for over a week 100% on my 5870. These cards are tested for way longer and for much higher fan speeds, i.e. 150-200%

Kehaar

newbie

Activity: 12

Merit: 0

Quote from: Allenssmart on June 05, 2011, 08:00:58 AM

I've been mining for 13 hours straight with 100% usage on both gpus and temps at 65C and 67C. Is it safe to run this 24/7?

That temperature is fine for chips. Worry when it gets to 80C+ (even then it'll probably be fine). Some cards idle at like 60+

Hardware will stop itself from overheating - either by drawing less power and slowing down / increasing fan speed - or turning off the computer alltogether.

Fatigue happens if the the hardware expands and contracts a lot (increase/decrease in temperature).

I imagine going between 40 - 70C wouldn't be much of a problem tbh, though I'm not an expert.

Allenssmart

newbie

Activity: 23

Merit: 0

I've been mining for 13 hours straight with 100% usage on both gpus and temps at 65C and 67C. Is it safe to run this 24/7?

MrGaSp

sr. member

Activity: 418

Merit: 253

So instead of people not being able to mine fast enough, the issue is the pool not keeping up. XD

Really hope the second server works for you

jasonk

full member

Activity: 168

Merit: 100

Quote from: eleuthria on June 04, 2011, 11:50:50 PM

Both the frequency and severity of the idles had decreased significantly over the last hour, but they are still poking around once in a while. At 650 GH/sec+, it's becoming quite obvious that a single server is not only a liability, but a handicap.

I'm now working on setting up another server, and tying the two together so that stats, login credentials, and block allocations are properly split between them. This one will probably still be in the US, but I'll be looking for one on the east coast to compliment our current Los Angeles server. Once setup, I will include instructions on the website for pointing your miners towards the closer server if you are inclined to do so.

Thanks I'd like to come back to your pool, but I can't because of so many idle miners, my profits will be less.

Veldy

member

Activity: 98

Merit: 10

Quote from: russelljohnson on June 05, 2011, 01:09:12 AM

Quote from: IlbiStarz on June 05, 2011, 12:09:29 AM

And who the heck is user 2631??

Seriously... ~6.5BTC every block...that's like over 50Gh/s...

Somebody NOT using GPUs to mine.

AMD?

russelljohnson

member

Activity: 84

Merit: 10

Quote from: IlbiStarz on June 05, 2011, 12:09:29 AM

And who the heck is user 2631??

Seriously... ~6.5BTC every block...that's like over 50Gh/s...

Somebody NOT using GPUs to mine.

marcus_of_augustus

legendary

Activity: 3920

Merit: 2349

Eadem mutata resurgo

Quote from: IlbiStarz on June 05, 2011, 12:09:29 AM

And who the heck is user 2631??

Seriously... ~6.5BTC every block...that's like over 50Gh/s...

ArtForz?

IlbiStarz

full member

Activity: 336

Merit: 100

And who the heck is user 2631??

Seriously... ~6.5BTC every block...that's like over 50Gh/s...

IlbiStarz

full member

Activity: 336

Merit: 100

Yayy

gogogo Eleuthria

eleuthria

legendary

Activity: 1750

Merit: 1007

Both the frequency and severity of the idles had decreased significantly over the last hour, but they are still poking around once in a while. At 650 GH/sec+, it's becoming quite obvious that a single server is not only a liability, but a handicap.

I'm now working on setting up another server, and tying the two together so that stats, login credentials, and block allocations are properly split between them. This one will probably still be in the US, but I'll be looking for one on the east coast to compliment our current Los Angeles server. Once setup, I will include instructions on the website for pointing your miners towards the closer server if you are inclined to do so.

Veldy

member

Activity: 98

Merit: 10

Quote from: The Koolio on June 04, 2011, 09:55:11 PM

Quote from: russelljohnson on June 04, 2011, 09:48:55 PM

Eleuthria, are you still working on the connection issues?

Still picking up idle messages from poclbm

I have a few too, but this time its once every 10 mins... Im fine with that

I'm not, That allows your GPU to go from 100% utilized to 0-1% utilized for a small period of time, however, that small period of time will cause significant temperature drops and then it will heat back up again when the connection is restored and work streams in. The effect is fatigue wear and tear on your hardware that you do not want to be putting it through if you want it to last. Constant load is FAR better than cycling.

The Koolio

hero member

Activity: 560

Merit: 500

Minds are like parachutes they work best when open

Quote from: russelljohnson on June 04, 2011, 09:48:55 PM

Eleuthria, are you still working on the connection issues?

Still picking up idle messages from poclbm

I have a few too, but this time its once every 10 mins... Im fine with that

russelljohnson

member

Activity: 84

Merit: 10

Eleuthria, are you still working on the connection issues?

Still picking up idle messages from poclbm

eleuthria

legendary

Activity: 1750

Merit: 1007

It may be too early to call the issue resolved...but it may be. ArtForz noticed a limit on the connections backlog for being accepted in pushpool set to 100. The server sometimes get hammered FAR faster than it can process 100 connections, and those connections were simply timing out (thus causing idles). This explains why the idles were often centered around LP attempts.

This has been corrected, and after a brief (5 minutes) of the server locking up, it seems to be back on track.

Topic: [~1000 GH/sec] BTC Guild - 0% Fee Pool, LP, SSL, Full Precision, and More - page 119. (Read 379078 times)