Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Register Log in

Latency trade off in IP core

Status
Not open for further replies.

spman

Advanced Member level 4
Joined
Aug 15, 2010
Messages
113
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
2,061
Hello,

What is the trade off of adjusting latency when implementing an IP core such as floating point with coregen?
How does latency affect the core? and where does latency have usage?

Thanks
 

ads_ee

Full Member level 6
Joined
Oct 4, 2012
Messages
328
Helped
87
Reputation
176
Reaction score
87
Trophy points
1,308
Location
San Diego
Activity points
3,469
adding latency in the floating point core will result in inputs generating the corresponding outputs after a larger number of clock cycles.

Latency is a way to improve performance by pipelining a design so that each stage has less logic between registers and therefore has less delay between registers. This results in outputs for a given set of inputs showing up later by the number of additional pipeline stages. As long as the operation that is pipelined doesn't have feedback you will be able to generate output data at every clock cycle.
 
  • Like
Reactions: spman

    spman

    points: 2
    Helpful Answer Positive Rating

spman

Advanced Member level 4
Joined
Aug 15, 2010
Messages
113
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
2,061
So latency doesn't have effect on performance (clock frequency, delay, ...) of the pure floating point core. right?
 

ads_ee

Full Member level 6
Joined
Oct 4, 2012
Messages
328
Helped
87
Reputation
176
Reaction score
87
Trophy points
1,308
Location
San Diego
Activity points
3,469
Latency stages are likely to improve performance due to the reduction in logic between the registers that are added.

Latency won't affect data throughput performance of the core at a given clock frequency.
 

Status
Not open for further replies.
Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top