Comment by lijok 2 months ago what do you mean they can’t get rid of heat? radiators exist 12 comments lijok Reply ceejayoz 2 months ago https://en.wikipedia.org/wiki/External_Active_Thermal_Contro...All that gets you 70kW of cooling. Radiating to vacuum isn't very efficient. lijok 2 months ago And that’s sufficient for roughly 100 unoptimized DC grade H200s.Not efficient, and it doesn’t have to be, because the cooling system has 0 opex cost. And capex clearly can be made to wor ceejayoz 2 months ago OK, so you've got a football-sized solar and radiator array to support 100 H200s.Why are we not building it on land again in some abandoned mall's parking lot? 1 reply → dzhiurgis 2 months ago Cooling space station is very different from cooling off chips. One requires extensive piping, other - a simple radiator. ceejayoz 2 months ago Both require the same thing - moving heat - and you’ll find plenty of piping in a datacenter for this reason. scottyah 2 months ago Dang, sucks we can never improve any technology. Let's just call it quits, guys. cyclopeanutopia 2 months ago Maybe one day we can, but it's definitely not in a category "there is no doubt". 1 reply → ceejayoz 2 months ago Technology is wonderful.Physics still gets a say. joe_mamba 2 months ago Space radiators are not very efficient due to lack of airflow in space. lijok 2 months ago Efficiency in the cooling loop is of no consequence as it has 0 opex cost in space. Do the capex numbers make sense?
ceejayoz 2 months ago https://en.wikipedia.org/wiki/External_Active_Thermal_Contro...All that gets you 70kW of cooling. Radiating to vacuum isn't very efficient. lijok 2 months ago And that’s sufficient for roughly 100 unoptimized DC grade H200s.Not efficient, and it doesn’t have to be, because the cooling system has 0 opex cost. And capex clearly can be made to wor ceejayoz 2 months ago OK, so you've got a football-sized solar and radiator array to support 100 H200s.Why are we not building it on land again in some abandoned mall's parking lot? 1 reply → dzhiurgis 2 months ago Cooling space station is very different from cooling off chips. One requires extensive piping, other - a simple radiator. ceejayoz 2 months ago Both require the same thing - moving heat - and you’ll find plenty of piping in a datacenter for this reason. scottyah 2 months ago Dang, sucks we can never improve any technology. Let's just call it quits, guys. cyclopeanutopia 2 months ago Maybe one day we can, but it's definitely not in a category "there is no doubt". 1 reply → ceejayoz 2 months ago Technology is wonderful.Physics still gets a say.
lijok 2 months ago And that’s sufficient for roughly 100 unoptimized DC grade H200s.Not efficient, and it doesn’t have to be, because the cooling system has 0 opex cost. And capex clearly can be made to wor ceejayoz 2 months ago OK, so you've got a football-sized solar and radiator array to support 100 H200s.Why are we not building it on land again in some abandoned mall's parking lot? 1 reply →
ceejayoz 2 months ago OK, so you've got a football-sized solar and radiator array to support 100 H200s.Why are we not building it on land again in some abandoned mall's parking lot? 1 reply →
dzhiurgis 2 months ago Cooling space station is very different from cooling off chips. One requires extensive piping, other - a simple radiator. ceejayoz 2 months ago Both require the same thing - moving heat - and you’ll find plenty of piping in a datacenter for this reason.
ceejayoz 2 months ago Both require the same thing - moving heat - and you’ll find plenty of piping in a datacenter for this reason.
scottyah 2 months ago Dang, sucks we can never improve any technology. Let's just call it quits, guys. cyclopeanutopia 2 months ago Maybe one day we can, but it's definitely not in a category "there is no doubt". 1 reply → ceejayoz 2 months ago Technology is wonderful.Physics still gets a say.
cyclopeanutopia 2 months ago Maybe one day we can, but it's definitely not in a category "there is no doubt". 1 reply →
joe_mamba 2 months ago Space radiators are not very efficient due to lack of airflow in space. lijok 2 months ago Efficiency in the cooling loop is of no consequence as it has 0 opex cost in space. Do the capex numbers make sense?
lijok 2 months ago Efficiency in the cooling loop is of no consequence as it has 0 opex cost in space. Do the capex numbers make sense?
https://en.wikipedia.org/wiki/External_Active_Thermal_Contro...
All that gets you 70kW of cooling. Radiating to vacuum isn't very efficient.
And that’s sufficient for roughly 100 unoptimized DC grade H200s.
Not efficient, and it doesn’t have to be, because the cooling system has 0 opex cost. And capex clearly can be made to wor
OK, so you've got a football-sized solar and radiator array to support 100 H200s.
Why are we not building it on land again in some abandoned mall's parking lot?
1 reply →
Cooling space station is very different from cooling off chips. One requires extensive piping, other - a simple radiator.
Both require the same thing - moving heat - and you’ll find plenty of piping in a datacenter for this reason.
Dang, sucks we can never improve any technology. Let's just call it quits, guys.
Maybe one day we can, but it's definitely not in a category "there is no doubt".
1 reply →
Technology is wonderful.
Physics still gets a say.
Space radiators are not very efficient due to lack of airflow in space.
Efficiency in the cooling loop is of no consequence as it has 0 opex cost in space. Do the capex numbers make sense?