Information

Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!



Discussion

Discussion on Article:
Intel’s Next-Gen Xeon Phi “Knights Landing” to Deliver Up to 3TFLOPS of DP Performance.

Started by: mudi1 | Date 07/04/13 08:34:32 PM
Comments: 10 | Last Comment:  07/09/13 07:47:48 AM

Expand all threads | Collapse all threads

[1-3]

1. 
intel products always looks good on paper remember ivy-bridge.
Future is FUSION-
AMD FirePro™ S10000(28nm 2D) delivers up to 5.91 TFLOPS of peak single precision and 1.48 TFLOPs of peak double precision floating point performance, compared to Nvidia Tesla K10 that is capable of up to 4.58 TFLOPS of peak single precision and 190 GFLOPs double precision peak floating point performance and even better than latest k20x(28nm 2D) and xeon-phi(22nm 3D) in double precision performance the best thing with s10000 is unified design.
AMD can do even better than xeon-phi with ARMv8 modified cores so lower TDP, lesser die size(one of the the most important to get denser server)
i m sure intel will use 3D memory stacking to achieve that performance in lower TDP, at that time AMD will also have access to 14nm-XM and for 3D memory AMD had that running successfully since 2011 in labs.

actually this all is waste coz of AMD hetrogenous APU/SoC. Scientists said with help of HSA APU/SoC it's possible to make petascale supercomputer in 1/3rd of power consumption which is beyond reach of xeon-phi and tesla or Fire-pro .
AND FOR ALL INFORMATION IN 2013 AMD WILL LAUNCH A SOC OF 25GFLOPS/WATT WHICH IS EVEN HIGHER THAN 2015 XEON-PHI AND NVIDIA TESLA BUT WE CAN'T BUY THAT AMD SILICON IT'S ONLY FOR SOME PARTNERS.
8 6 [Posted by: mudi1  | Date: 07/04/13 08:34:32 PM]
Reply
- collapse thread

 
ooohhh good one.. go get your paycheck...
3 5 [Posted by: amdzorz  | Date: 07/05/13 08:06:52 AM]
Reply
 
25 GFLOPS/W in 2013? Yeah, right.
0 0 [Posted by: lol123  | Date: 07/09/13 07:47:48 AM]
Reply

2. 
15Gflops/w and for 3TFlops - it is approximately 200W. Wow. This is going to be really hot and power hungry
0 2 [Posted by: Ravindra Munvar  | Date: 07/04/13 09:27:03 PM]
Reply
- collapse thread

 
Really? The k20x does only 1.32TFlop at 235W, and that's already the most energy efficient computational tool out there. 3Tflop for 200W means Phi is 2.5x more energy efficient!
3 0 [Posted by: basroil  | Date: 07/04/13 10:44:00 PM]
Reply
 
Lets see what AMD/Nvidia can come up with till 2015, but I like how Intel is trying to put on pressure
4 0 [Posted by: Rollora  | Date: 07/05/13 12:14:21 AM]
Reply

3. 
show the post
0 4 [Posted by: s23e7h4kf936hklnf7y8b  | Date: 07/05/13 06:02:54 AM]
Reply
- collapse thread

 
LMAO, so Knights Corner (available at stores today) won't actually support its performance rating of 1 DP TFLOPS until 2014? The 22nm process tech, the GDDR5 memory interface and the PCIe3 form factor and bus are available today, but the product won't actually come with - so to speak - teraflops until next year? Features and performance figures are given in two different colums for space reasons, not because they arrive in different time frames. Notice how Knights Corner and Knights Landing occupy two years in the diagram because the same release stretches two years whereas for the CPUs a new either tick or tock is released each year.
0 0 [Posted by: lol123  | Date: 07/05/13 07:16:57 PM]
Reply
 
Looks like you don't know how to read a graph!
0 0 [Posted by: basroil  | Date: 07/06/13 12:02:11 AM]
Reply

[1-3]

Back to the Article

Add your Comment