Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Reduce resource usage of Resample instantiations #46

Open
vk2seb opened this issue Sep 29, 2024 · 1 comment
Open

Reduce resource usage of Resample instantiations #46

vk2seb opened this issue Sep 29, 2024 · 1 comment
Labels
enhancement New feature or request

Comments

@vk2seb
Copy link
Member

vk2seb commented Sep 29, 2024

The Resample DSP component instantiates quite large FIR filters for upsampling beyond 4x, this can quickly eat FPGA area and is heavily used for scope plotting (interpolating lines from points).

We can reduce the area and time footprint quite a bit by adding different internal filter types to Resample, for example one which uses a CIC or Boxcar filter for the primary anti-aliasing filter.

Maybe useful: https://tomverbeure.github.io/2020/09/30/Moving-Average-and-CIC-Filters.html

@vk2seb vk2seb added the enhancement New feature or request label Sep 29, 2024
@vk2seb
Copy link
Member Author

vk2seb commented Oct 13, 2024

Now that we have a true polyphase resampler that infers BRAM correctly, this is essentially done: #60

I was also able to reduce the tap count multiplier from 8 to 5 and get similar performance (now the new default).

However, having some different internal filter types may still be useful, particularly as Resample is used for Stroke plotting, where aliasing is not such a big deal. It may be worth saving some multipliers there.

@vk2seb vk2seb changed the title Reduce FIR size in Resample interpolators Reduce resource usage of Resample instantiations Oct 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

1 participant