NAM: Neural Amp Modeler

Pretty sure NAM at present is just intended for amp DI (no cab) captures, but some people have had some luck capturing amps+cabs (and some other things).

But IMO, amp only captures are the best way anyway and offer the best trade off between convenience and options.

Also I believe it’s designed to work only at 48kHz and anything else isn’t really going to give predictable results.
 
I'm intrigued by this and started reading through the readme, etc... I'm about the least technical person in the world. There is a lot of talk about GPU - do I need to have a GPU to use the plugin or is that just to capture? I'm on a reasonably fast PC with a ton of RAM, but no GPU since that isn't my world.
 
LaMarr Bruister said:
I'm intrigued by this and started reading through the readme, etc... I'm about the least technical person in the world. There is a lot of talk about GPU - do I need to have a GPU to use the plugin or is that just to capture? I'm on a reasonably fast PC with a ton of RAM, but no GPU since that isn't my world.
Click to expand...
From my understanding GPU would be for capturing
 
LaMarr Bruister said:
I'm intrigued by this and started reading through the readme, etc... I'm about the least technical person in the world. There is a lot of talk about GPU - do I need to have a GPU to use the plugin or is that just to capture? I'm on a reasonably fast PC with a ton of RAM, but no GPU since that isn't my world.
Click to expand...
I watched this video this morning while on my few-nights-a-week insomnia trip and it answered a lot of questions for me, so may help you too...
 
Byrdman said:
I watched this video this morning while on my few-nights-a-week insomnia trip and it answered a lot of questions for me, so may help you too...
Click to expand...

Thanks…I’ll check it out this evening, sorry about the insomnia, I go through it in multi-week batches and then it disappears for months on end. Good luck!
 
LaMarr Bruister said:
I'm intrigued by this and started reading through the readme, etc... I'm about the least technical person in the world. There is a lot of talk about GPU - do I need to have a GPU to use the plugin or is that just to capture? I'm on a reasonably fast PC with a ton of RAM, but no GPU since that isn't my world.
Click to expand...

Mongillo19 said:
From my understanding GPU would be for capturing
Click to expand...

You need a GPU if you want to train *locally* - but you are free to use the colab instances if you don't and it's very easy as well. You can also drop by the FB group if you want to find captures or for more info
 
Rocco Crocco said:
So I am guessing you need a Facebook account to get other people's captures. Is this correct?
Click to expand...

No, but that's where the central group is, and most of the drive links. There are a few people working on building supporting pages for it to create more of a familiar ecosystem kind of space. Currently people have released

nam tones
neural amp modeler

And there's a team of folks working on tonehunt which you can join the discord for as well.
These are all community built projects supporting NAM. I also mod the FB group so feel free to ask me anything
 
I can feel some compression too.
If you listen to the v1.1.1 reference file you will hear sped up guitar and bass plucks and slides, maybe the sped up nature of the source has something to do with it? Just guessing.
 
So the idea of this is super cool (sorry, late to the party) and I dig it.

Popped up a bunch on my YT feed and I grabbed it and gave it a go.

The clips in here sound great; I’m a busy guy but if I get a chance and it’s not too difficult I’ll do a couple caps of my Herbert; haven’t seen too many of those in the lists I’ve seen.

Really cool someone could do this and make it freeware.
 
James Freeman said:
I can feel some compression too.
If you listen to the v1.1.1 reference file you will hear sped up guitar and bass plucks and slides, maybe the sped up nature of the source has something to do with it? Just guessing.
Click to expand...

You can actually use your own DI track to test things like this if you want to. If you're totally replacing that file you'll have to run the code manually, but you can also insert your own data in the middle. Usually it tends to reduce the ESR even further without making the model heavier (takes longer).

Steinmetzify said:
So the idea of this is super cool (sorry, late to the party) and I dig it.

Popped up a bunch on my YT feed and I grabbed it and gave it a go.

The clips in here sound great; I’m a busy guy but if I get a chance and it’s not too difficult I’ll do a couple caps of my Herbert; haven’t seen too many of those in the lists I’ve seen.

Really cool someone could do this and make it freeware.
Click to expand...

I know not everyone is on FB - but drop by the FB group if you are, there are a tone of good captures there. Also tonehive.org is going to be the best source probably, once they finish beta
 
northern_fox said:
I know not everyone is on FB - but drop by the FB group if you are, there are a tone of good captures there. Also tonehive.org is going to be the best source probably, once they finish beta
Click to expand...
I joined the FB group, and have everything going on my end, but I am doing things a little different and getting a lot of latency.

I have a Line 6 Catalyst 200 which has AI built in so I have my laptop running the plugin into USB on the amp. I have tried every setting and still getting latency. I had this when I setup Tonex CS until I changed the settings to ASIO and changed the input/output channels. I tried ASIO as well and nothing changed.

BTW, I think this is a great program and sounds better through my amp than Tonex CS. I just want to get the latency gone so I can really dig in deeper.

I posted this on the FB group to see if anyone has any suggestions, but pending post approval.

Thanks!!
 
You must log in or register to reply here.
Back
Top