G3 modeling thread

I would also drop the onboard backup functionality if it could free some more space ! ^^
After all, we can backup with Axe Edit or bot.
 
With the modeling as good as it is already, my hope is that Cliff found a way to use the processing power more efficiently in the amp block to allow sharing of the dedicated amp processor, allowing more capacity for FX processing. It's all a guess at this point.
 
What I would like on the G3, is a rework of the main page of each block to make them more simple or plug'n play. There is also probably some advanced parameters that could be dropped.
 
Will there be an obligatory username and password, to get past 3 warning screens that state "WARNING, THIS FIRMWARE UPDATE WILL DESTROY YOUR PRESETS AND CAUSE YOUR MOTHER-IN-LAW TO VISIT MORE FREQUENTLY. INSTALL AT YOUR OWN PERIL. DO NOT BITCH TO ME THAT YOU HAVE TO REBUILD YOUR PRESETS, OR THE NEXT FIRMWARE WILL BE $500.00" ?
 
rodzimguitar68 said:
Will there be an obligatory username and password, to get past 3 warning screens that state "WARNING, THIS FIRMWARE UPDATE WILL DESTROY YOUR PRESETS AND CAUSE YOUR MOTHER-IN-LAW TO VISIT MORE FREQUENTLY. INSTALL AT YOUR OWN PERIL. DO NOT BITCH TO ME THAT YOU HAVE TO REBUILD YOUR PRESETS, OR THE NEXT FIRMWARE WILL BE $500.00" ?
Click to expand...

I heard yes, and you'll get a free bumpersticker saying "I survived FW 18.00"
 
G3 Modelling Technology and CPU usage

So now we all know that G3 is just around the corner and that it will be available for both the MK I & II and the XL.

That's going to be great, obviously.

But we're already at the point where just using a Layout with the most common chains of effects blocks will max out the available CPU.
Sure, there are workarounds, but all of them involve doing things like using less than the best resolution Reverbs or lower res Cab Blocks or separate Presets (vs using Scenes, for example) for effects chain Layouts where we don't always have every effect active, etc.

And I'm not talking about anything akin to some sort of an exotic Layout.
The Layout of all my current Presets is:
WAH > CPR > DRV > AMP > VOL > PIT (Detuning) > DLY2 (Mono Delay for Slapback) > DLY1 (Mono delay for Echo) > REV (High Quality) > CAB (Stereo UR)
I also have an FXL Block in parallel with the CAB Block for my power amp + guitar cabs rig.
That Layout takes me to just below 90% CPU usage.

Now some of you are probably looking at that and thinking that it IS an elaborate and/or uneconomical Layout.
But IMO that's simply not true.
The Axe-FX is supposed to be a 1-box solution for professional guitar tone and the effects I have in my chain are just the typical things that most guitar players would have in their rig (except maybe for the CAB Block).

Sure, I could use a Normal Quality REV or a Standard Res CAB Block and then I'd be in a much safer CPU range.
But those selections simply don't sound as good.

Etc., etc., etc.

So... Here's my hope for what G3 Modelling technology will involve:
The amp modelling in FW 17.04 already sounds way better than I ever thought a modeller could sound.
(But that was true of my Ultra as well when it was running FX 6. lol)
So, I'm hoping that the basic sound and technology of G3 will not be much different than 17.04 but that the amp modelling will mostly have just been re-coded somehow in order to achieve a less taxing CPU load.
If it can do that and also sound marginally better than 17.04, or not, I'll be a very happy camper.

If G3 sounds obviously better than 17.04 but involves a higher CPU load I'll be really frustrated because I might have to take a pass on it or on any other post-G3 firmware updates.

I'm going to guess though that it *will* involve higher CPU but that we'll have a choice in the AMP Block between G2 or G3 modelling per Preset.
That will be frustrating too, but I guess I'll have to learn to live with it.

Then again, like the song says, "Don't worry. Be Happy."
I'm sure that ultimately I'll be real happy with G3 no matter how it works.
 
Joe, amp modelling runs mainly on its own discrete DSP. G3 at best, could only return a few percent to the cab/fx blocks... unless the entire inner workflow is being remade, but that seems unlikely.
 
Zer0th said:
Joe, amp modelling runs mainly on its own discrete DSP. G3 at best, could only return a few percent to the cab/fx blocks... unless the entire inner workings are being remade, but that seems unlikely.
Click to expand...

A few percent is all it takes right now to take me over the edge.
 
I'm sure we'll all know soon but lets speculate because it's fun.

I would guess "G3" would refer to a major shift in the approach to modelling or it would just be firmware 18. When the AxeFx II came out with G2 modelling it unleashed a boatload of new amp aspects that were being captured and emulated. So my prediction is that Cliff and crew have probably had some algorithms that have been to cpu intensive to ever include in the current firmwares. But, they have discovered a new approach to the emulation coding to lighten the overall cpu load which in turn now allows them to implement these algorithms that were waiting for a chance to be unleashed.
 
You must log in or register to reply here.
Back
Top Bottom