IK Multimedia TONEX

Training is done on CUDA or RT cores when you have an NVIDIA GPU, and it is done on the integrated GPU on Apple Silicon. But Apple Silicon just does not compete to the raw power of a dedicated GPU; so I'm not surprised at all that it still takes people around 30minutes on a non-dedicated card.

If you want an idea of the raw processing difference, you could do worse than look and compare using the Blender benchmarks:
1756394829685.png


It really isn't surprising that Apple are being trounced when it comes to training models.
 
That’s not great, but what were you seeing before? I have an M3, but so far I have only done captures on my desktop PC which has a potent GPU.

I don’t know, because I didn’t select the higher quality training. I’ll do it over again with that, but it was about an hour, IIRC.

Are you guys using the Tonex capture hardware or capturing it another way?

Right now, I’m cheating. Routing it all through the FM9, and capturing a highly edited Fractal model.

It really isn't surprising that Apple are being trounced when it comes to training models.

The upside of using an Apple is that I don’t have to go back to using a Windows machine, when the training is done. LOL

Honestly, now that there’s a que function, it really doesn’t matter too much. The only downside is that I can’t reamp the file, and then bring it home. I’ve only got a four year old Windows laptop to bring to the shop, right now.
 
30mins on the lower quality capture doesn’t seem like an improvement at all to me. I’ve been running the high quality ones on an m1 mini in ~40mins.
 
Can someone confirm the improved v2 models loads normally on the Tonex Pedal and the Tonex One? Silly question, but I wasn't able to find that info online. I'm really looking forward to testing the new improved tone models!
 
I wish/hope they come up with something that allows for a clustered workflow, and/or the ability to make a capture, and then transfer it to a beefier machine later.

It turns out they did! You can use a laptop to capture a bunch of training files, and then transfer those files to another machine to do the heavy processing. I just saw this in a video by someone who I will not mention here but his initials are something like JS.
 
This is going to depend on if Helix models the mid pot as linear or audio. The log/audio mid pots on a BF amp need to be around 9 to give 6.8k of resistance to match the amps without a mid control. I see a lot of people trying to run just below 7 to match the 6.8, but that will give you less gain and less mids, unless it is a linear pot, or a model of a linear pot.

Yeah, I know Line 6 has said to set the knob at 6.8 at some point or another, but I don't remember if that was for all of the relevant Fender amps, or just for a specific model. Either way, it was just as a starting point / reference, and I was finding pretty consistently that the mid knob had to be quite a bit lower on Helix to match the captures. Interestingly enough, I remember people here (I think) saying that to match the Helix Deluxe to the UA Dream, the mids had to be pushed quite a bit.

And of course these tonestacks are so interactive; I think some of what I was hearing was the mid range sitting a little differently on the Helix models, which I guess is related to the bass and treble adjustments I had to make to match the high and low end.

Speaking of which, the other difference was the high end on Tonex tended to be obviously attenuated when comparing with IR disabled. I think this is the nature of Tonex captures though? Not sure, but it was also pretty consistent in the comparison, though didn't matter much in a real world usage situation, except that I usually needed to push the treble down on the Helix slightly.

Anyway, it's apples to oranges in a lot of ways, but I was mainly surprised that I was feeling better about the Helix Fenders and how they reacted than a few weeks ago lol
 


Some really good sounding tones so far in this playthrough. I feel like I’m hearing a much more lively top end on these caps, more or better harmonic content up top.


I might be missing it, but does he say when the direct capture versions of these v2 amp packs will be available on his site? Or is that waiting for the new website too?

I know it'll take a while to recapture individual amp packs, but I'd be pretty happy with what's in the "ToneX v2 BoutiqueAmp KIT II", except I'm not really interested in the amp + cab captures.
 
Okay, thank you! It would be cool to get direct versions of these packs at least, but I totally get it.

From his post on TOP:

"There already are a lot of questions about the timeline - there is none and i will give no dates. I said from the get go that this is gonna take a while and is gonna happen slowly. But i assure you, i will stick to it. :)"
 
Gotta say, these updates to training make it 100000000x more enticing to make models. It's been a while since I dived into making big batches of ToneX models, but these updates make it totally viable again. I had been asking for many of these features from the beginning and they're actually making a way bigger difference than I could have imagined. The process is so much more streamlined now.
 
From his post on TOP:

"There already are a lot of questions about the timeline - there is none and i will give no dates. I said from the get go that this is gonna take a while and is gonna happen slowly. But i assure you, i will stick to it. :)"

And here’s more info from Amalgam’s second newsletter about v2:

 
From his post on TOP:

"There already are a lot of questions about the timeline - there is none and i will give no dates. I said from the get go that this is gonna take a while and is gonna happen slowly. But i assure you, i will stick to it. :)"

Makes sense!

My only hope was that he captured direct versions at the "same" time as doing these amp + cab versions for these packs, but I get that's still double the work.
 
Back
Top