2009-01-02, 19:28
There's ONE very vocal guy on the AVS forums claiming issues with the cooling on this board. NVIDIA had issues with their GPUs awhile back pertaining to their surface mount product selection, you probably heard about it. This guy seems to think the problem is still there and that heat makes it an issue on these boards - it's complicated.
Me personally? I am not sweating it. NVIDIA\ASUS are smarter than me, the product is warranted, heat would be a valid issue THEY would be responsible for, and while we will probably see higher temps with hardware acceleration I do not think it will be too bad. Fixing it with a small fan would probably do it.
One thing that is valid though - manufacturers build these boards sort of expecting an Intel style heatsink. That is a heatsink blowing DOWN with the air then passing over whatever radiators these guys put next to the socket. Heatsinks like I have in my desktop are huge and stand up blowing air a totally different way, little radiators around the socket would get little air. I doubt many of US are doing that in an HTPC that we want to be silent but it's something to consider - especially for guys trying to passively cool a CPU!
I'll digress a little more... My first water cooled CPU was done way back before hardware could be purchased and anyone knew jack about what worked, I built it all and I used Peltiers too. My first setup had no radiator, just a BIG contractor sized bucket. In a few days the water was literally steaming hot. So I put a radiator (trans cooler from a car) on it and stuck it outside in the January cold. A few days later the water was AGAIN steaming hot! Snow melting on it worked to extend things but was a PITA. In desperation I put a small fan on it, temps dropped like a ROCK!
Lesson learned - even a small amount of airflow removes the boundary layer of hot air and dramatically improves cooling. Afterward I learned about fun with condensation on CPU pins and then later ICE buildup. (lol) Learned about algae growth in cooling lines screwing up cooling too!
So, airflow is important even if it's only a little bit, go water if you want ultimate cooling. I think we'll be fine with this board. <shrug>
Me personally? I am not sweating it. NVIDIA\ASUS are smarter than me, the product is warranted, heat would be a valid issue THEY would be responsible for, and while we will probably see higher temps with hardware acceleration I do not think it will be too bad. Fixing it with a small fan would probably do it.
One thing that is valid though - manufacturers build these boards sort of expecting an Intel style heatsink. That is a heatsink blowing DOWN with the air then passing over whatever radiators these guys put next to the socket. Heatsinks like I have in my desktop are huge and stand up blowing air a totally different way, little radiators around the socket would get little air. I doubt many of US are doing that in an HTPC that we want to be silent but it's something to consider - especially for guys trying to passively cool a CPU!
I'll digress a little more... My first water cooled CPU was done way back before hardware could be purchased and anyone knew jack about what worked, I built it all and I used Peltiers too. My first setup had no radiator, just a BIG contractor sized bucket. In a few days the water was literally steaming hot. So I put a radiator (trans cooler from a car) on it and stuck it outside in the January cold. A few days later the water was AGAIN steaming hot! Snow melting on it worked to extend things but was a PITA. In desperation I put a small fan on it, temps dropped like a ROCK!
Lesson learned - even a small amount of airflow removes the boundary layer of hot air and dramatically improves cooling. Afterward I learned about fun with condensation on CPU pins and then later ICE buildup. (lol) Learned about algae growth in cooling lines screwing up cooling too!
So, airflow is important even if it's only a little bit, go water if you want ultimate cooling. I think we'll be fine with this board. <shrug>