Do you think It is possible to run NAM? I can try to build from source, but what about compatibility with pisound card? Do you think we can hope to have prebuilt module sooner or later?
What is NAM ?
The bass/guitar amp modeler everyone is talking about:
LV2 plugin is here:
@ladron, you are the creator of NAM, yes? Would you be willing to make a pull request to our builder repo adding your NAM plugin? Here is a short guide on achieving that - Platform: LV2 Plugins · patchstorage/patchstorage-docs Wiki · GitHub. If you have any questions, we will do our best to answer them!
I’m just one of the contributors to the NAM project - but the LV2 is my repo.
As @redcloud says, for modep you would need a way to send models to the device. I believe the mod ui already has a method for this (for things like cabinet IRs). It may just work - I haven’t tried it, though, as I’m not currently using the mod stuff.
What is the cpu usage like for nam on pi at the moment? Is anything other than feather models usable?
I’m running full “standard” models on RPi4. It takes a large percentage of the CPU, but I still have headroom for a cabinet IR and some lightweight effects.
Oh that’s awesome! I’ll look into doing a pr for file handling for mod this week for you, if you’d like? I can also make you a stock standard ui from mod-sdk as a placeholder. At least then it won’t be a sardine tin
I’m assuming this is you?
Yeah that’s me. UI to follow in a bit
Did anyone do an RTL (Roud Trip Latency) test while using NAM on an RP4/5 and PIsound?
I am asking to check if this setup would be viable for a live scenario. I only need to run NAM with a Cab IR. No more effects are needed.
@Deltabootis welcome to the Blokas community!
I use a RPi4/PiSound setup live every week. I typically use the AidaX amp modeler but also have used NAM (with the “Buffered” option set to On) and have personally never experienced any latency issues.
Thank you @jtemple967
Nice! I am excited to try such a setup.
If you have a rpi4 you might be able to run a simple setup with a buffer size of 64. That should give you a sub 3ms latency.