Pages:
Author

Topic: DiabloMiner GPU Miner - page 9. (Read 866206 times)

legendary
Activity: 1162
Merit: 1000
DiabloMiner author
October 03, 2012, 06:50:39 PM
A user at my pool retrying the same wrong password 10 times per second with DiabloMiner. Is that how it's supposed to run?

Might be an idea to give up and shut down if the password is wrong, or at least use a much longer delay. It may be that 10 times per second is because of running multiple instances, but still if it is once per second it is pretty extreme.


Most pools do not follow the HTTP spec, so there is no way for DM to detect the difference between a malfunctioning pool and a malfunctioning user.

Plus, no one should EVER run multiple instances of DM.

If the user has added multiple pools, DM will just go to the next pool after a few failed attempts.
legendary
Activity: 2730
Merit: 1034
Needs more jiggawatts
October 03, 2012, 06:23:47 PM
A user at my pool retrying the same wrong password 10 times per second with DiabloMiner. Is that how it's supposed to run?

Might be an idea to give up and shut down if the password is wrong, or at least use a much longer delay. It may be that 10 times per second is because of running multiple instances, but still if it is once per second it is pretty extreme.
full member
Activity: 155
Merit: 100
September 27, 2012, 11:49:49 PM
enable unofficial overclocking = 2, and delete any files in the Profile folder.

Already on mode 2, deleted profiles, still nothing on the voltage slider.
legendary
Activity: 952
Merit: 1000
September 27, 2012, 11:43:00 PM
It was the "Force Constant Voltage" setting. Before I checked it, changing the slider did nothing.

I tried that, and nothing changed. The "Core Voltage" slider is still unavailable.
The "Power Limit (%)" slider is active, and currently on +20%.
Any ideas how to undervolt this sucker? I'd love to reduce the power footprint a bit.
At full burn on the miner and gaming on the GTX480, this rig pulls over 900W at the wall.
I'd sure love to trim that a bit. Smiley

enable unofficial overclocking = 2, and delete any files in the Profile folder.
full member
Activity: 155
Merit: 100
September 27, 2012, 11:38:35 PM
It was the "Force Constant Voltage" setting. Before I checked it, changing the slider did nothing.

I tried that, and nothing changed. The "Core Voltage" slider is still unavailable.
The "Power Limit (%)" slider is active, and currently on +20%.
Any ideas how to undervolt this sucker? I'd love to reduce the power footprint a bit.
At full burn on the miner and gaming on the GTX480, this rig pulls over 900W at the wall.
I'd sure love to trim that a bit. Smiley
hero member
Activity: 497
Merit: 500
September 27, 2012, 11:41:27 AM
Those temps are fine.
full member
Activity: 155
Merit: 100
September 27, 2012, 08:16:10 AM
VDDC (VRM) temps are located on the sensors tab in GPU-z, usually toward the bottom.

I monitored the VRM temps with GPU-Z overnight. At full burn, VRM1 is at 70c and VRM2 is at 58c.
The GPU temp is currently at 61c. I think it was as high as 71c last night while I was playing a game on the other card.
Those temps all sound fine to me, what do you guys think?
full member
Activity: 155
Merit: 100
September 27, 2012, 08:06:18 AM
It was the "Force Constant Voltage" setting. Before I checked it, changing the slider did nothing.


Cool, I'll give that a try, thanks!
legendary
Activity: 952
Merit: 1000
September 27, 2012, 12:48:10 AM
That's my thread, and in my last post, I mention that I finally got undervolting in MSI AB to work, and now my fan is much lower, with much lower VRM temps.
How did you get the voltage working in AB? I haven't been able to get that figured out yet.
It was the "Force Constant Voltage" setting. Before I checked it, changing the slider did nothing.
full member
Activity: 155
Merit: 100
September 26, 2012, 11:55:20 PM

That's my thread, and in my last post, I mention that I finally got undervolting in MSI AB to work, and now my fan is much lower, with much lower VRM temps.

How did you get the voltage working in AB? I haven't been able to get that figured out yet.
legendary
Activity: 952
Merit: 1000
September 26, 2012, 10:31:16 PM
Jesus. Those VRMs AMD uses are usually safe up to 120c for 24/7 usage (the same way GPUs are safe up to 85c for 24/7 usage), hes really trashing them. Also 90% fan? More like dead fan.
That's my thread, and in my last post, I mention that I finally got undervolting in MSI AB to work, and now my fan is much lower, with much lower VRM temps.
legendary
Activity: 1162
Merit: 1000
DiabloMiner author
September 26, 2012, 10:21:31 PM

Jesus. Those VRMs AMD uses are usually safe up to 120c for 24/7 usage (the same way GPUs are safe up to 85c for 24/7 usage), hes really trashing them. Also 90% fan? More like dead fan.
legendary
Activity: 952
Merit: 1000
September 26, 2012, 12:29:41 PM
Wow, I wonder if your VRMs are overheating. My 7970 sometimes overheats to the point that the machine hangs on boot at 1125 on really hot days, so maybe thats what happened with you.

I guess it's possible, but if it was going to overheat, wouldn't it have done it during the summer?
My office stays a bit warm (78F+/-) but is pretty constant throughout the year.
AB and GPU-Z report full burn temps of around 74c, but maybe they aren't monitoring the VRMs?

VDDC (VRM) temps are located on the sensors tab in GPU-z, usually toward the bottom.

Also:
https://bitcointalksearch.org/topic/m.1138807
full member
Activity: 155
Merit: 100
September 26, 2012, 12:13:05 PM
GPU-z should show VRM temps. In afterburner press the arrow to the right of volts should pop up the adjustment bar.

Ok, I'll check out both of those options when I get home, thanks!
hero member
Activity: 497
Merit: 500
September 26, 2012, 11:59:25 AM
GPU-z should show VRM temps. In afterburner press the arrow to the right of volts should pop up the adjustment bar.
full member
Activity: 155
Merit: 100
September 26, 2012, 11:38:22 AM
I think consumer/prosumer/enthusiast watercooling is pointless. That said, there is absolutely no reason you shouldn't have a full block on there that covers everything: gpu, memory, and VRMs. If there are any chips exposed at all and they dont have normal air cooled heatsinks with a fan blowing on them, you have a problem.

Not pointless at all if you work in your office 8-10 hrs per day and don't want a turbine screaming in your ear all day! Smiley
My rig is heavily overclocked, but is quieter than my ceiling fan on low, so I can work or play in comfort.
The water block I have is full coverage, so I doubt that's the problem.
Is there any way to measure the temps of the VRMs?
legendary
Activity: 1162
Merit: 1000
DiabloMiner author
September 26, 2012, 11:30:45 AM
Also on the 7970's in AB you can control your mem volts. I have mine all the way down and my mem stays relatively cool. Just a thought.

I don't see anything about mem volts in AB. In fact, AB has never really worked as advertised for me.
With unofficial overclocking in mode 2, I can get the core clock to where I want it, and the mem clock down to about 650 before it gets unstable, but I never have been able to get anything related to voltage working.

@DiabloD3: The water block I have seems to be working fine. I wonder if the lack of fans on the VRMs is affecting temps? Shouldn't the water block cool the VRMs also?

I think consumer/prosumer/enthusiast watercooling is pointless. That said, there is absolutely no reason you shouldn't have a full block on there that covers everything: gpu, memory, and VRMs. If there are any chips exposed at all and they dont have normal air cooled heatsinks with a fan blowing on them, you have a problem.
full member
Activity: 155
Merit: 100
September 26, 2012, 11:27:17 AM
Also on the 7970's in AB you can control your mem volts. I have mine all the way down and my mem stays relatively cool. Just a thought.

I don't see anything about mem volts in AB. In fact, AB has never really worked as advertised for me.
With unofficial overclocking in mode 2, I can get the core clock to where I want it, and the mem clock down to about 650 before it gets unstable, but I never have been able to get anything related to voltage working.

@DiabloD3: The water block I have seems to be working fine. I wonder if the lack of fans on the VRMs is affecting temps? Shouldn't the water block cool the VRMs also?
legendary
Activity: 1162
Merit: 1000
DiabloMiner author
September 26, 2012, 11:23:50 AM
Wow, I wonder if your VRMs are overheating. My 7970 sometimes overheats to the point that the machine hangs on boot at 1125 on really hot days, so maybe thats what happened with you.

I guess it's possible, but if it was going to overheat, wouldn't it have done it during the summer?
My office stays a bit warm (78F+/-) but is pretty constant throughout the year.
AB and GPU-Z report full burn temps of around 74c, but maybe they aren't monitoring the VRMs?

Unless your card has a really retarded heatsink setup, 74c is fine, I wasn't having issues until I was >80c for 8 hours at a whack.
hero member
Activity: 497
Merit: 500
September 26, 2012, 11:23:44 AM
Also on the 7970's in AB you can control your mem volts. I have mine all the way down and my mem stays relatively cool. Just a thought.
Pages:
Jump to: