Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

CST Hardware for Body Simulations

Status
Not open for further replies.

mcg2016

Newbie level 3
Joined
Apr 8, 2016
Messages
3
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
28
Hi everybody!

We are going to purchase CST for antenna and propagation simulations in body environments. We are considering GPU computing so we are algo going to buy a workstation for running CST.

Our official Nvidia provider has recommended us the GPU Titan X since CST computes in Time-Domain in single precision. However, CST says that they don't support this specific GPU so they recommend us the Testla K40 (5 times more expensive than Titan), which works in double precision. Nvidia provider says that K40 will probably works slower than Titan since it has a lower number of CUDA cores (but double precision capabilities, this is the reason for its price).

I would need some help to clarify which could be the best one, since there is a significant different in price between them and we want to save as much money as possible.

Thanks!!
 

hi,
there are many GPUs out there work faster but for your application (CST) you need to buy cards which supported by CST. no point buying cards not supported even though its faster since its wont help CST.
 

hi,
there are many GPUs out there work faster but for your application (CST) you need to buy cards which supported by CST. no point buying cards not supported even though its faster since its wont help CST.

Thank you for your response. CST warned us about that. However, our CST sales person told us that not supported means that "they weren’t tested" so they "cannot assure/guarantee that they will perform 100% well. It’s not a matter of the GPU card being good or not, it is just a matter of us not taking responsibility" in case we are "not happy with its performance on the context of CST simulation."

What do you think about? Anyone has a simmilar experience?

Thanks!
 

GPU cards are expensive! if you have money to take risk, you can buy the new GPU and try. but i wouldnt take the risk.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top