- Status
Disc_Junkie
Call me D_J!
I know the definition of Tesla in Physics, it is SI unit of magnetic field intensity denoted by T.
dOm1naTOr
Wise Old Owl
Tesla is nvidia's new techniology which uses CUDA architecture, or GPU computing.
In a general purpose computer, there are limitations to CPu computing power due to limitation of performance per CPU and no. of CPU sockets in mobo. SO u can boost systems performance using high end GPUs.
These GPUS too can perform generam purpose computing and GPUs are gud at performing longer and complex instructions whereas CPU is very gud at computing simple, small and large number of instructions.
With this combination of multicore CPUs and multipipeline GPU[ around 240 for GTX280] we can almost have a supercomputer at home.
btw it has nothin to do with magnetic field or its intensities.
In a general purpose computer, there are limitations to CPu computing power due to limitation of performance per CPU and no. of CPU sockets in mobo. SO u can boost systems performance using high end GPUs.
These GPUS too can perform generam purpose computing and GPUs are gud at performing longer and complex instructions whereas CPU is very gud at computing simple, small and large number of instructions.
With this combination of multicore CPUs and multipipeline GPU[ around 240 for GTX280] we can almost have a supercomputer at home.
btw it has nothin to do with magnetic field or its intensities.
Tesla is the name of a brilliant scientist who did a lot of pioneering work on electricity and magnetism. The unit of magnetic field intensity was named in his honour, so was the Tesla coil.
Man-made things are often named after famous people, either to honour them or as catchy names for commercial products. E.g., Indira Gandhi Airport, Joe Walsh guitar, Nike Air Jordan shoes, etc.
Man-made things are often named after famous people, either to honour them or as catchy names for commercial products. E.g., Indira Gandhi Airport, Joe Walsh guitar, Nike Air Jordan shoes, etc.
dOm1naTOr
Wise Old Owl
haha
the thread starter wsa askin bout nvidia tesla.
nowadays GPUs are not limited to simple gfx functions and effects. with the introduction of physics processing in gpu, ccompanies realized that GPUs are very much capable and advanced than most CPUs in computing other complex calculations like that in nuclear research, space research and other high performance computing needs.
moreover GPUs are massively pipelined than CPUs. With unified shader pipeline, these pipelines became programmable to handle any kind of instruction> be in pixel shader procesing/vertex shader processing/geametry shader/physics/or general instructions.
The efficiency of the execution increases as the complexity of these instructiuon increases and the number of them decseases. But this stands true for only nvidia. Coz they uses longer pipeline architecture.
Whereas ATI uses more number of shorter pipelines and shader units in pair of 5. But only one of em is capable of computing complex instructiuons and others are just for computing shorter and simple instructions. So ATI cards might be better at executing general instructions coz of mu8ch more number of processing units and shorter but fast pipelines.
lets see how this story goes. I dun think ATi hasnt announced their entry into general purpose GPU computing like CUDA fron nvidia. All we can do is wait.
the thread starter wsa askin bout nvidia tesla.
nowadays GPUs are not limited to simple gfx functions and effects. with the introduction of physics processing in gpu, ccompanies realized that GPUs are very much capable and advanced than most CPUs in computing other complex calculations like that in nuclear research, space research and other high performance computing needs.
moreover GPUs are massively pipelined than CPUs. With unified shader pipeline, these pipelines became programmable to handle any kind of instruction> be in pixel shader procesing/vertex shader processing/geametry shader/physics/or general instructions.
The efficiency of the execution increases as the complexity of these instructiuon increases and the number of them decseases. But this stands true for only nvidia. Coz they uses longer pipeline architecture.
Whereas ATI uses more number of shorter pipelines and shader units in pair of 5. But only one of em is capable of computing complex instructiuons and others are just for computing shorter and simple instructions. So ATI cards might be better at executing general instructions coz of mu8ch more number of processing units and shorter but fast pipelines.
lets see how this story goes. I dun think ATi hasnt announced their entry into general purpose GPU computing like CUDA fron nvidia. All we can do is wait.
What's so funny? I explained the origin of the name. Maybe I should have concluded by saying that NVidia took a name that's well known in electronics and used it as a catchy code name. But that should have been obvious from the examples I gave.
Ahhh crap i completely forgot i had posted this thread,just got reminded when i got up few mins ago.
anyways i should have mentioned that i am not taking about magnetic flux intensity.
back to topic, i have read tesla is mostly being used to do research(or am i wrong) can't it be used for games?
are they better performer than nvidia's new gaming cards as gaming cards( GTX 290)? or tesla is better at performing complex calculations and not at gaming gfx?
anyways i should have mentioned that i am not taking about magnetic flux intensity.
back to topic, i have read tesla is mostly being used to do research(or am i wrong) can't it be used for games?
are they better performer than nvidia's new gaming cards as gaming cards( GTX 290)? or tesla is better at performing complex calculations and not at gaming gfx?
That was obvious from your first post. But since I had not read up yet on Nvidia's new tehnology and someone else had already diverted attention to the name, I simply offered an explanation as to where NVidia got the name.
dOm1naTOr
Wise Old Owl
its the other way around. U cant use it in games while it is intended to divert those resources which usually are used for gaming into general purpose computing. nvidia says their newer gaming cards can be part of tesla...i think it can do general purpose computing on such setups.
nvidia's dedicated physics is a fine example for this. with a tri SLI, u can choose 2 GPU's to do the pixel crunch and the other to do the physics number crunches. That is one card act as a dedicated physics processor.
Physics calculations are somewhat general purpose type.
This is the only use of tesla[strictly speaking> dedicated Physx] for gamerz.
- Status