-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
summer is coming
- Thread starter wagex
- Start date
- Joined
- May 30, 2004
- Location
- Folding@Home
Doh... Phone typing of text walls sucks. I accidentally hit back.. I will try to remember what I typed...
I think we have confusion because we are debating assumptions and opinions, not single stated points.
First, You were stating that he was going to have throttling issues because of the non climate controlled area, though you did state that this was an "if" depending on headroom. My thoughts are that he will be okay due to his increased airflow and larger volume of air (whole warehouse/ outside air) as well as the open air rack system, but I mostly agree with you because there could be some cards that bump up through their headroom. Though see 2nd point.
second, you disagreed with my statement that, "the difference (in temperature) is not as substantial as some may think." This is not necessarily something that can be debated well, because it is a generality of an opinion. My main point was, that in my experience, most mining cards are not effected by temperature and throttling as much as some people think. I have run many cards hotter and for longer time periods at those temperatures than most would believe. See my previous post #66 in this thread- I have a couple cards that I have run at 95c for months. I think that the main problem with this is the fan life. They do not last long when run at 100%, but that is not really an issue to myself or moho, because, as he stated, you can simply tie a better fan on after the original dies.
Last,(I think) Is the statement that temperature of the gpu goes up 1 degree for every one degree of ambient temperature. I disagree with this as a statement of fact. It is perhaps a good rule of thumb, or a general rule, but there are temperature deltas that will effect this as temperatures between the air and heatsinks get closer or further apart, or as the efficiencies of different heatsinks interact at more extreme temps. I don't have any specific examples or scientific proof for this now, and I could be mistaken. Mostly, I am sure there are some differences, but it may be more towards the more extreme temperatures, so it may not be relevant for this discussion. Thus, my disagreement, but only as a matter of fact.
Edit: I also comprehend the fact that 95f is 95f. I understand that the temperature is the same, but our perception of the temperature can be different, depending on wind speed, due to evaporation of perspiration. I also realize that there is no evaporation involved with computer cooling, except in watercooling setups that are built to take advantage of that.
I think we have confusion because we are debating assumptions and opinions, not single stated points.
First, You were stating that he was going to have throttling issues because of the non climate controlled area, though you did state that this was an "if" depending on headroom. My thoughts are that he will be okay due to his increased airflow and larger volume of air (whole warehouse/ outside air) as well as the open air rack system, but I mostly agree with you because there could be some cards that bump up through their headroom. Though see 2nd point.
second, you disagreed with my statement that, "the difference (in temperature) is not as substantial as some may think." This is not necessarily something that can be debated well, because it is a generality of an opinion. My main point was, that in my experience, most mining cards are not effected by temperature and throttling as much as some people think. I have run many cards hotter and for longer time periods at those temperatures than most would believe. See my previous post #66 in this thread- I have a couple cards that I have run at 95c for months. I think that the main problem with this is the fan life. They do not last long when run at 100%, but that is not really an issue to myself or moho, because, as he stated, you can simply tie a better fan on after the original dies.
Last,(I think) Is the statement that temperature of the gpu goes up 1 degree for every one degree of ambient temperature. I disagree with this as a statement of fact. It is perhaps a good rule of thumb, or a general rule, but there are temperature deltas that will effect this as temperatures between the air and heatsinks get closer or further apart, or as the efficiencies of different heatsinks interact at more extreme temps. I don't have any specific examples or scientific proof for this now, and I could be mistaken. Mostly, I am sure there are some differences, but it may be more towards the more extreme temperatures, so it may not be relevant for this discussion. Thus, my disagreement, but only as a matter of fact.
Edit: I also comprehend the fact that 95f is 95f. I understand that the temperature is the same, but our perception of the temperature can be different, depending on wind speed, due to evaporation of perspiration. I also realize that there is no evaporation involved with computer cooling, except in watercooling setups that are built to take advantage of that.
Last edited:
Just a bump for earthdog. It seems that we have for the most part resolved our different opinions and were debating for no reason other than to come to the conclusion that we agree. And perhaps that no one can clearly state every opinion, every time they say something, especially in a forum setting where you are limited in time and in space permitted to reply, and be 100% correct in your thoughts, facts and your opinions.
- Joined
- Feb 21, 2008
- Location
- Baltimore, MD
I've had the same systems running since the coldest part of winter here. Well below freezing.
I've had the door to the room cracked about an inch to let the cold air in. At it's coldest, it was about -10C to -15C outside. And that cracked door allowed the room to get quite chilly. The hottest card ran about 72C. Now, with the door cracked the same amount, and it's now about +20C outside, and considerably warmer in the room, and the same hottest card is only reaching about 77C.
So I'll estimate that the ambient temp in the room has gone up about 10-15C, but GPU temps have only gone up about 5C
I've had the door to the room cracked about an inch to let the cold air in. At it's coldest, it was about -10C to -15C outside. And that cracked door allowed the room to get quite chilly. The hottest card ran about 72C. Now, with the door cracked the same amount, and it's now about +20C outside, and considerably warmer in the room, and the same hottest card is only reaching about 77C.
So I'll estimate that the ambient temp in the room has gone up about 10-15C, but GPU temps have only gone up about 5C
Yes, if one does not normally run fans at 100%, then there is always room for higher ambients. Which is likely most of the explanation for gsr's temps.
(Edited for clarity)
Last edited:
There is simply too many unknowns to me to continue to preach it like the Gospel. Perhaps I took too many liberties in the reference I linked in expanding it to GPUs... not sure.
And while I appreciate the input above (gsr.. two posts came in before I posted), its kind of anecdotal to me as the intake temps of the cards are unknown. So while the room may be - this or that, the intake area can be significantly different...or not. Who knows. But I would imagine there should have been more difference there for whatever reason. I know when benching from ambient to to freezing (~32F) I see load temps drop close to that difference... the delta doesn't really change much in that respect. So your results are curious to me in that light.
What I know though is the reference I linked and my personal experiences in reviewing and benchmarking for literally hundreds of hours and seeing 1:1 differences at a non extreme level (for example, 70F to 80F) when I bothered to check (trust me, I wasn't looking often, LOL!) took me down this path.
Anyway, I appreciate the lively discussion and hashing things out like this...and though we ended up with nothing concrete, I believe this is what forums are all about.
And while I appreciate the input above (gsr.. two posts came in before I posted), its kind of anecdotal to me as the intake temps of the cards are unknown. So while the room may be - this or that, the intake area can be significantly different...or not. Who knows. But I would imagine there should have been more difference there for whatever reason. I know when benching from ambient to to freezing (~32F) I see load temps drop close to that difference... the delta doesn't really change much in that respect. So your results are curious to me in that light.
What I know though is the reference I linked and my personal experiences in reviewing and benchmarking for literally hundreds of hours and seeing 1:1 differences at a non extreme level (for example, 70F to 80F) when I bothered to check (trust me, I wasn't looking often, LOL!) took me down this path.
Anyway, I appreciate the lively discussion and hashing things out like this...and though we ended up with nothing concrete, I believe this is what forums are all about.
- Joined
- May 30, 2004
Sounds like you have the headroom to play then... which is a great thing!
Yeah, it may peak there, but your averages are 10C below that at PEAK sumer with 8 months of the year being, according to you, 'sweater weather'. Usually that type of stance are from people where it is acutally 90F for a couple months out of the year, not days... so I found your comment quite funny!
The averages that TWC reports are a joke, because they aren't even downtown Springfield..... Looks like they get the data from Chester or roughly where Weathersfield is..... And have the nerve to call it Springfield.....
Those are a good at least 5 F cooler. And if that's even near the airport, that's miles away from my place..... LOL.....
Interestingly, the blocks and fields around me match the averages for Lebanon, NH better. (My area is near the NH border)
Especially where the town hall is and east from there. Especially, because I'm in a rain shadow, so T-storms often don't come for a while when west of me are getting a lot of activity. Likely the Green mountains to the west blocking the rain.
I do interestingly get the phenomenon that some seem to report, where T-storms seem to divide and go around downtown.
A T-storm YouTuber in Rhode Island seems to get more T-storms head on..... I guess he don't have to water the hibiscus plant a ton like I do.....
I can see where you think that I just dreaming, but the typical weather has been convenient for me, especially when I want to do outdoor activities without mosquitoes biting me.....
Last edited:
- Joined
- Feb 21, 2008
- Location
- Baltimore, MD
no i dont use any kind of features in cgwatcher, nor do i use the autofan feature in cgminer.
i set the fan speeds to static levels. something like ~75% at all times (different cards have different static levels based on what i felt was appropriate).
but i remember one day, when the wind blew the door wide open, and i came down to see the cards running at something silly like 30C WHILE MINING
i set the fan speeds to static levels. something like ~75% at all times (different cards have different static levels based on what i felt was appropriate).
but i remember one day, when the wind blew the door wide open, and i came down to see the cards running at something silly like 30C WHILE MINING
My 2 mining only rigs are going out in the morning. I got all of my power run and put my old router to work as a wireless bridge, so I don't have to run an ethernet cable out. I still have to do the ducting to bring in outside air, but I have it figured out for the most part.
It will be nice to have my space in my benching area and have enough outlets to plug in my single stage and a computer to bench, at the same time. I have spent too much time in the 4ghz spi thread lately, I am itching to get some benching in.
It will be nice to have my space in my benching area and have enough outlets to plug in my single stage and a computer to bench, at the same time. I have spent too much time in the 4ghz spi thread lately, I am itching to get some benching in.
I got both dedicated mining rigs moved out. I had a little issue with my connection on one rig, but it was a single checkmark in the router settings that I had missed.
Temps are better than they were when inside, but airflow is improved. They rigs were in cramped corners of my basement before. With 70f outside temps, GPU temps on the two worst cards (xfx), are at 76-78c. The box fan is on the lowest speed. I will keep updating as outside temps warm up.
It is nice to have a cooler, quieter basement, since my bedroom and benching room are both down there.
Temps are better than they were when inside, but airflow is improved. They rigs were in cramped corners of my basement before. With 70f outside temps, GPU temps on the two worst cards (xfx), are at 76-78c. The box fan is on the lowest speed. I will keep updating as outside temps warm up.
It is nice to have a cooler, quieter basement, since my bedroom and benching room are both down there.
I have a small, detached garage that I moved them to. It should be secure, as well as keep the weather out. I put a place for cool air to get pulled in and be ducted toward the rigs, then the air should be exhausted out through the eaves. I put a grill at the intake, but I still need to put a screen to keep bugs out. I thinking about putting a light timer on the fan, so it doesn't run at night when it is cooler. Perhaps some sort of a thermostat would work better. I will have to see what I can find that will work.
- Joined
- Jan 28, 2011
- Location
- CA
I've heard that the latency induced by wireless links can cause problems for mining. Something to keep in mind in case of future problems.
- Joined
- Sep 7, 2013
I can confirm that is of no importance.
I run on satellite internet with over 800 ms pings
- Joined
- Sep 7, 2013
we have cooled off and been having rain for the last few days. I had to turn the furnace on for a little while last night...
summer is coming though
Curecoin.
GPU's are below 60C all the time + uber low power draw
