File-It-Under: hardware listing By: D.F.Skinner!~
Coming from a guy like myself who is a hardcore hardware junkie,this is a rather late but never forgotten entry detailing the spec sheet on the GeForce GTX Titan graphic card.This is just not any ole run of the mill graphic card,but its a well constructed,rich in features and cost $1,000,whoa.Yes,this card cost more than an entire computer system altogether.When a graphic card like this come upon the market,it needs to me mention on this site.The question that begs to be answered is this graphic worth the $1,000 bucks?Lets go in depth with the features and specs that's embedded on the GTX GeForce Titan graphic card.
First Off About Those Specs
The heart and brains of the entire graphic card is centered around the GK110 GPU which is comprised of,are ready for this,2688 cores,7.1 billion transistors with a die size that measures around 551 mm. The GTX is leaps and bounds in both the size of the graphic card which measures 10 inches length, and performance.
Lets go deep under hood
Graphic processing clusters 5
Streaming Multiprocessors 14
Texture Units 224
ROP Units 48
Base Clock 836 MHz
Boost Clock 876 MHz
Memory Clock 6008 MHz
L2 Cache Size 1536k
Total Video Memory 6144 GDDR5
Memory Interface 384-bit
Total Memory Bandwidth 288.4 GB /s
Form Factor Dual slot
Connectors 2 x Dual Link DVI
1 x HDMI
1 x Displayport
Power Connectors One 8-pin and One 6-pin
Power Supply Recommendation 600 Watts
When it comes to raw power the GeForce GTX Titan is a beast, its based on the power of the GK 110 GPU,inside the GPU there is a total of 15 SMX units ,14 of these units are enabled for a grand total of 2688 shaders and 224 texture units.The second best feature within the GeForce GTX Titan is definitely the memory,it has not 1 or 2 or even 3, but there is 6 GB GDDR5 video memory,that's right 6 GB.
Add up a large Cuda core count (which is 2688), 7.1 billion transistors and a hefty 6 GB of video RAM,NVIDA takes throne as the worlds most powerful graphic card.
What are Cuda Cores?
As you know,the balk of the graphic cards workload is done on the GPU and its programmable,unlike the CPU that's currently sitting in your computer it has a predetermined areas for certain task,or function.A CUDA core can be programmed to do any task the programmer wants it to do.Like for an example,video editing software can use just about all of its cores to optimize for video editing.Video games uses cores to process textures and some physics calculation.To put it simply,the programmer can tell the GPU what task to perform.
Source Credit: "HotHardware,PC Perspective,AnandTech"