Forum Discussion

Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
8 years ago

Unrolling and used RAMs

Hello, I'm trying to understand the relationship between channels, unrolls and used RAM (M20K) For this purpose, I've created this simple program composed of three kernels: - the first i...
Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
8 years ago

So, the additional RAMs (and additional latency) are essentially due to the accumulation. The more you accumulate (increasing U), the more latency you will pay and the more block you will use.

I've noticed also that using -fp-relaxed, both the values (latency and blocks) are lower with respect to the previous case.