News
 

Bookmark and Share

(10) 

At the Intel Developer Forum 2011, Intel Corp. showcased a system running a new version of its real-time “Wolfenstein” ray-tracing demo. In order to render the game scenes in full HD (1080p) resolution and with additional visual effects, Intel had to use a 2-way Xeon system with eight code-named Knights Ferry compute accelerators.

The up-to-date Wolfenstein game is rendered through a real-time ray tracer with several special effects that haven’t been possible before in games with such an accuracy. The latest Wolfenstein game demo also includes such effects as depth of field, HDR bloom, inter-lens reflection, smart anti-aliasing.

Las year Intel rendered Wolfenstein demo using a “cloud” of four Intel's Knights Ferry/Aubrey Isle compute accelerate cards in 720p (1280*720 resolution). This year Intel decided to install eight KNF compute cards and render the demo in 1080p (resolution). As ray tracing is a highly parallel application it can therefore take very good benefit of the many cores that are in a single chip on the Knights Ferry board.

The demonstration is a proof-of-concept of real-time ray-tracing as well as many Intel core (MIC) architecture. However, the show off does not have a lot of practical meaning since real-time ray-tracing will hardly be used in the consumer applications for many years.

The Knights Ferry has 32 x86 cores clocked at 1.2GHz and featuring quad-HyperThreading. The unit, aimed at PCI Express 2.0 slots, has up to 2GB of GDDR5 memory. The chip itself has 8MB of shared L2 cache, which is quite intriguing by itself since highly-parallel applications do not require a large on-chip cache.

According to Intel, eight KNF compute boards (on 2-way Intel Xeon X5690 [3.46GHz, 6 cores, 12MB L3 cache] with 24GB DDR3 RAM at 1333MHz) achieved 7TFLOPS performance in SGEMM [single precision general matrix multiply] operations, which is not a truly high result.

Tags: Intel, Knights Ferry, Aubrey Isle, Larrabee, MIC, GPGPU

Discussion

Comments currently: 10
Discussion started: 09/14/11 07:02:10 PM
Latest comment: 09/16/11 09:43:04 AM
Expand all threads | Collapse all threads

[1-5]

1. 
AMD Cayman can produce 2.7 TFLOPs of single precision performance
2.7 Times 8 = 21.6 TFLOPs
2 2 [Posted by: blazzin  | Date: 09/14/11 07:02:10 PM]
Reply
- collapse thread

 
So you need to write a ray-tracer to 6990 to prove that Intel is wrong.
1 0 [Posted by: Tristan  | Date: 09/15/11 02:29:05 AM]
Reply
 
SGEMM is not about ray-tracing.

Apparently, actual horsepower of KNF is a well-kept secret.
0 0 [Posted by: Anton  | Date: 09/15/11 06:36:58 PM]
Reply

2. 
Intel is so far behind AMD and Nvidia in Graphics that even there fab powerhouse cant help.
3 0 [Posted by: vid_ghost  | Date: 09/15/11 02:18:55 AM]
Reply

3. 
AMD Cayman can produce 2.7 TFLOPs of single precision performance 2.7 Times 8 = 21.6 TFLOPs


That is theoretical peak performance.The same applies for CPUs.
Achieve xTFLOPS in BLAS is not theoretical peak, but actual performance I presume.
If you look at CUBLAS numbers i.e., C2050, which in double precision has around 500GFLOPs of peak performance (is comparable to Cayman in other words), in DGEMM it manages around 160GFLOPS which is already pretty high.

Look here:
http://www.microway.com/p...050-Fermi-Performance.pdf
0 0 [Posted by: dkarko  | Date: 09/15/11 02:25:00 AM]
Reply
- collapse thread

 
The smallLuxGPU demo shows that the AMD GPU gets near theoretical TFLOPS when raytracing under OpenCL.

http://www.anandtech.com/...hd-6970-radeon-hd-6950/23
1 0 [Posted by: user99  | Date: 09/15/11 12:29:14 PM]
Reply
 
I see your point but I don't see flops there. I see them saying that it is closer to theoretical than civ5 (probably based on the fact that it performed worse on civ5).

Anyway Intel saying they achieved that in some ray-tracer says nothing. KNF will have to prove itself in more wide problems in GPGPU
0 0 [Posted by: dkarko  | Date: 09/16/11 01:09:45 AM]
Reply
 
Fair Enough
0 0 [Posted by: blazzin  | Date: 09/16/11 09:43:04 AM]
Reply

4. 
Intel tries to compete in graphics,but will never succeed. Amd will mop Intel right up in graphics.
1 0 [Posted by: twainamd  | Date: 09/15/11 09:31:45 AM]
Reply

5. 
Intel wants the HPC market, where 80% is NVidia and 20% is AMD now. However, they and AMD, need SDK to get the market. AMD hopes are for Open CL to take traction.
2 0 [Posted by: Ananke  | Date: 09/15/11 12:26:45 PM]
Reply

[1-5]

Add your Comment




Related news

Latest News

Monday, July 28, 2014

6:02 pm | Microsoft’s Mobile Strategy Seem to Fail: Sales of Lumia and Surface Remain Low. Microsoft Still Cannot Make Windows a Popular Mobile Platform

12:11 pm | Intel Core i7-5960X “Haswell-E” De-Lidded: Twelve Cores and Alloy-Based Thermal Interface. Intel Core i7-5960X Uses “Haswell-EP” Die, Promises Good Overclocking Potential

Tuesday, July 22, 2014

10:40 pm | ARM Preps Second-Generation “Artemis” and “Maya” 64-Bit ARMv8-A Offerings. ARM Readies 64-Bit Cores for Non-Traditional Applications

7:38 pm | AMD Vows to Introduce 20nm Products Next Year. AMD’s 20nm APUs, GPUs and Embedded Chips to Arrive in 2015

4:08 am | Microsoft to Unify All Windows Operating Systems for Client PCs. One Windows OS will Power PCs, Tablets and Smartphones