Actually 295X2 is an excellent solution for winter, just run the furmark and voila - a 650W heater is ready for service!Nothing will ever be as power hungry and hot as GTX 480. My friend heated his room during winter with it.
AMD: Volcanic Islands R1100/1200 (8***/9*** series) Speculation/ Rumour Thread
You would have to remove the water cooler first...Actually 295X2 is an excellent solution for winter, just run the furmark and voila - a 650W heater is ready for service!
Won't be needed.
All those Joules of heat will stay in your room regardless.
R9 290 really is in whole another room-heating league compared to any other GPU I ever owned. During GPU mining I was amazed just how FAST it heats my open window room.
AMD's power management keeps it at 300W max at default I gather. Of course it does have an adjustable power limit.... If power wasn't a problem they'd just let it run so-called "boost clock" full time.
itsmydamnation
Veteran
its got a target temperature/fan speed/clock speed thing going on, under most game conditions you are pulling around 280 watts. The fan increases when target temperature is exceeded, if temperature cant be maintained clocks are dropped.
The problem with stock 290/x is its coolers inability to dissipate that much heat without a stupidly loud fan. if you look at nonref 290x's with the same temp/fan/clock targets you will see increased performance with increased power consumption with much lower sound.
edit: CCC allows for more specific throttling/OC conditions
Built-in OC and power limit adjustment in CCC was in fact the reason I switched back to the red team lately-ish (that and the fact R9 GPU was the only high-end GPU on the market that wasn't 1.5 year+ old already at that time.)
Of course, the card I picked was already factory overclocked and it can't be clocked any higher without some serious artefacting in 3DMark fire run test (arguably artefacts even at "stock" clocks in some conditions, like if you pre-heat it before running a test or run more than one loop of testing), so I didn't get as much mileage out of this feature as I'd liked to. However, I can at least DOWNclock the card and bind the clocks to different hotkeys much more easily now than with an NV board, I don't need any external software to do that.
Of course, the card I picked was already factory overclocked and it can't be clocked any higher without some serious artefacting in 3DMark fire run test (arguably artefacts even at "stock" clocks in some conditions, like if you pre-heat it before running a test or run more than one loop of testing), so I didn't get as much mileage out of this feature as I'd liked to. However, I can at least DOWNclock the card and bind the clocks to different hotkeys much more easily now than with an NV board, I don't need any external software to do that.
Yeah I know. I saw some tests with the power limit jacked up and in some cases they saw about 335W power consumption.
Source engine games for example don't have functional vsynch/triple buffer implementation (super mega mouse lag when enabled; makes game near unplayable.) In that case it's nice to have a handbrake to pull to limit power consumption, otherwise the game renders upwards of 300fps for no reason near all the time.
Also: summer heatwave + no aircon. Or, distributed computing. Might not want to have fans roaring and card belching heat all night long...!
Me too, usually.
CaptainGinger
Newcomer
Most (all?) Source engine games have a cvar for that, fps_max 60 or whatever.
I don't know why but valve defaults it to 300.
I don't know why but valve defaults it to 300.
Well i will join CaptainGinger on this, for old source games: limit maxfps ( seriously ouside benchmark you dont need to run at 230fps min. What i like the most is the games who cut off the v-sync for menu... make you feel your run the 3Dmark 2003 benchmarks where you see 3000+fps..
Last edited by a moderator:
I think remember it is in Nvidia inspector. For AMD user, RadeonPro. ( but for source games, there's allways a the config of the game you can edit )
The MSI On-Screen Display Server can easily limit fps in any game I've tried (lots) to whatever you want. It's a hard lock (CPU will not issue commands to the GPU to render >(the_FPSlimit_you_set)). You don't have to have an MSI GPU to use it or anything.
Protip: if you limit any Bethesda Elder Scrolls or Fallout games to 60fps (on a 60Hz monitor, if the fps drops below that it's still lag time) and enable vsync you get the vsync without the usual ~1/3sec mouse lag you get by enabling vsync in those games. That trick works in most games that introduce apocalyptic mouse lag when vsync is enabled.
Protip: if you limit any Bethesda Elder Scrolls or Fallout games to 60fps (on a 60Hz monitor, if the fps drops below that it's still lag time) and enable vsync you get the vsync without the usual ~1/3sec mouse lag you get by enabling vsync in those games. That trick works in most games that introduce apocalyptic mouse lag when vsync is enabled.
Just for update the thread, AMDFirePro S9150 Server, 5.03Tflops SP / 2.57Tflops DP ...
AMD today announced the AMD FirePro S9150 server card -- the most powerful server Graphics Processing Unit (GPU) ever built for High Performance Computing. Based on the AMD Graphics Core Next (GCN) architecture, the first AMD architecture designed specifically with compute workloads in mind, the AMD FirePro S9150 server card is the first server card to support enhanced double precision and break the 2.0 TFLOPS double precision barrier. With 16 GB of GDDR5 memory -- 33 percent more than the competition -- and maximum power consumption of 235 watts, AMD FirePro S9150 server GPUs provide supercomputers with massive compute performance while maximizing available power budgets.
"Today's supercomputers feature an increasing mix of GPUs, CPUs and co-processors to achieve great performance, and many of them are being implemented in an environmentally responsible manner to help reduce power and water consumption," said David Cummings, senior director and general manager, professional graphics, AMD. "Designed for large scale multi-GPU support and unmatched compute performance, AMD FirePro S9150 ushers in a new era of supercomputing. Its memory configuration, compute capabilities and performance per watt are unmatched in its class, and can help take supercomputers to the next level of performance and energy efficiency."
The AMD FirePro S9150 server card is supported by a powerful software ecosystem, enabling developers to better harness its compute performance. In addition to AMD support for OpenCL, including support for OpenCL 2.0 planned later this year, AMD today announced a collaboration with Pathscale to support OpenMP 4.0 for AMD FirePro S9150 and selected other AMD FirePro graphics cards.
More infos . here ..
http://www.guru3d.com/news-story/amd-firepro-s9150-server-card.html
AMD today announced the AMD FirePro S9150 server card -- the most powerful server Graphics Processing Unit (GPU) ever built for High Performance Computing. Based on the AMD Graphics Core Next (GCN) architecture, the first AMD architecture designed specifically with compute workloads in mind, the AMD FirePro S9150 server card is the first server card to support enhanced double precision and break the 2.0 TFLOPS double precision barrier. With 16 GB of GDDR5 memory -- 33 percent more than the competition -- and maximum power consumption of 235 watts, AMD FirePro S9150 server GPUs provide supercomputers with massive compute performance while maximizing available power budgets.
"Today's supercomputers feature an increasing mix of GPUs, CPUs and co-processors to achieve great performance, and many of them are being implemented in an environmentally responsible manner to help reduce power and water consumption," said David Cummings, senior director and general manager, professional graphics, AMD. "Designed for large scale multi-GPU support and unmatched compute performance, AMD FirePro S9150 ushers in a new era of supercomputing. Its memory configuration, compute capabilities and performance per watt are unmatched in its class, and can help take supercomputers to the next level of performance and energy efficiency."
The AMD FirePro S9150 server card is supported by a powerful software ecosystem, enabling developers to better harness its compute performance. In addition to AMD support for OpenCL, including support for OpenCL 2.0 planned later this year, AMD today announced a collaboration with Pathscale to support OpenMP 4.0 for AMD FirePro S9150 and selected other AMD FirePro graphics cards.
More infos . here ..
http://www.guru3d.com/news-story/amd-firepro-s9150-server-card.html
D
Deleted member 13524
Guest
So Hawaii originally has a 2:1 DP ratio, but the consumer versions were cut down to 8:1?
Also, passive coolers on 200W cards?!
Also, passive coolers on 200W cards?!
It's not meant to remain passive, the chassis fans are supposed to be positioned directly in front of the card, blowing through to the exhaust on the back. It's not like the people buying these things don't know what they're doing wrt cooling, so it works better in saving room and removing redundant fans if they allow them to take care of the fans themselves.
Yes. This isn't exactly news, the FirePro had exactly the same 2:1 DP ratio already. FWIW this card is pretty much exactly the same as a FirePro W9100, just minimally lower clock, somewhat different cooling (not passive, provided by server case) cutting the display outputs in the process (which should help with cooling). No idea which one is more expensive though
.In addition they are installed on server chassis with thermal air controlled systems ( refreshed with cold water ).. basically its like if you put your motherboard + gpu in your kitchen fridge .. When you buy them on this form.. offtly they are used on blades system ( 2 to 7U )
Yes. This isn't exactly news, the FirePro had exactly the same 2:1 DP ratio already. FWIW this card is pretty much exactly the same as a FirePro W9100, just minimally lower clock, somewhat different cooling (not passive, provided by server case) cutting the display outputs in the process (which should help with cooling). No idea which one is more expensive though
Usually, compute-oriented cards (S models) are cheaper.