Possibility of adding more CPU

I am in no way a computer or electronics person.I just wanted to see if it would be possible to add/upgrade the CPU to the axefx maybe by building a small black box that could be connected to the axefx that would have additional chips for more processing power.

Answer A) Yes. Another AxeFxII.

Answer B) No. Don't hold your breath.

Hope that helps.
 
If someday FAS comes out with a 1RU FX only box - think FX-8 rackmount, I might consider this in the AES loop to keep the AD/DA conversion to a minimum.
Software permitting, I can imagine that adding a AES loop block could add I/O functionality without robbing the user of the option of using output 2 for local monitoring.

That could be cool and solve the issues of the power hungry.
 
I am in no way a computer or electronics person.I just wanted to see if it would be possible to add/upgrade the CPU to the axefx maybe by building a small black box that could be connected to the axefx that would have additional chips for more processing power.Rather than designing a whole new axefx3 maybe being able to add additional CPU boxes,they could probably be small and rackmountable.Kind of like adding additional hard drives to a computer.There are 48 blocks that can be used in a preset,but not enough processing power to use them.I understand most everyone is not going to use 48 blocks in one preset,but it would be nice if there was more processing power to handle the complexities that new upgrades in firmware seem to need.Like I said I am not a computer person,so I don't know if this would be physically possible to do,it just sounds like a good idea if it were .

.........wat? OP pls.
 
If someday FAS comes out with a 1RU FX only box - think FX-8 rackmount, I might consider this in the AES loop to keep the AD/DA conversion to a minimum.
Software permitting, I can imagine that adding a AES loop block could add I/O functionality without robbing the user of the option of using output 2 for local monitoring.

That could be cool and solve the issues of the power hungry.

What I was sort of saying was to sort of have that but inside a future box. Where the AES is used to connect to a whole other box, sending audio digitally to another module inside the box, may work. It would all be modular rather than parallel, ans obviously would be a future box and not current.

Now, I have no idea the roadmap of the tigersharc. If newer significantly faster models are in relatively near future, all of this speculation is a mute point.


Sent from my iPhone using Tapatalk
 
Closest analogy would be asking a surgeon to connect another brain up to your own to improve your powers of concentration and memory.

Best idea yet!

Cliff, can you add additional brains for us? I'd like one from a theory guy, maybe get on the waiting list for Tremontti's dome piece.
 
.........wat? OP pls.
Just because Cliff or anyone for that matter hasn't figured out how to increase CPU without adding latency doesn't mean it cant be done.Musicians at one time thought a modeler would never be able to sound as good as a real tube amp,Cliff proved them wrong.It isn't a matter of if it can or cannot be done,it is a matter of figuring out how it can be done.
 
Old post i know, but in defense of asking for more processing.. I have owned my axe II for 90 days and have hit the cpu limit already. Let me explain. .. Load up a preset and then install a fx return block. Plug your BASS guitar into the back of the unit, and create another "modest" fx chain. I am not complaining, I am just noticing that double the power could be used in the same box. I was going to use the front input for guitar and rear for the bass. No big deal and it can totally be done, but you have to adjust accordingly. That being said, it is amazing I can even do this at all and it work. Awesome gear for sure!
 
Load up a preset and then install a fx return block. Plug your BASS guitar into the back of the unit, and create another "modest" fx chain. I am not complaining, I am just noticing that double the power could be used in the same box.
double anything could be a benefit to anything typically. :) that said, the Axe-Fx II isn't really advertised as a solution for multiple guitars, so it's not too surprising to run out of CPU for multiple guitars. THAT said, people are getting away with a LOT, even on the lower CPU AX8. check out this thread:
http://forum.fractalaudio.com/threads/show-us-your-packed-preset.119311/

29650333962_46792d7dc2_o.jpg


perhaps you can read up on how to reduce CPU usage in your presets, as the Axe-Fx II can definitely host more than what we see in these AX8 presets.
 
I agree. It is totally enough to run guitar and bass with lots of possibilities. I just noted that when running a lot and DOUBLE at hi res, you can run out. Its not like you NEED that kind of power, but since when has that stopped anyone from wanting it? Best course of action would be to use two units and youll have a backup along with double the power. I totally LOVE my fx unit, and the ability to control guitar and bass from one pedal to one fx unit simultaneously catching guitar and bass on wireless. The power in my 3U case also includes wireless IEM. Add to it the ability to stream usb. Freaking incredible. Under my 35lb weight limit, and WAY under my setup time requirement. For live, the pa systems are usually the weak link, not the axe.
 
Here's where I get CPU limited:
- can't use 2 amps, 2 cabs with preamp, 2 reverbs with high quality settings and a multi-comp on the end
- hard to build in organ-type scenes with 2 pitch blocks, filters, and rotaries, on top of a patch with normal guitar sounds

This last wouldn't be a big deal if I could use my scenes setup on MFC-101 and have an IA dedicated to an organ patch with tap again to get back to the previous patch. I play rhythm and need to cover organ-type sounds often, and this is a challenge. My compromise now is just to use a basic Rotary block. But it's a compromise.

It's always going to be easy to find ways to max out whatever resources are available. With that said, overall I'm very satisfied with what I have. Just a few patches with very different scenes where it's a struggle to work around CPU. On the positive side, I often find a simpler chain just plain sounds better.

But a patch that has scenes with many variations (not just parameter tweaks but diff blocks) can quickly hit the limits.
 
Last edited:
I've never needed Ultra res cabs and High quality Reverb. So much more CPU available with those off and really no difference in sound during performance. Don't let your mind convince yourself of things you aren't really hearing!
 
Here's where I get CPU limited:
- can't use 2 amps, 2 cabs with preamp, 2 reverbs with high quality settings and a multi-comp on the end

Use X/Y?

hard to build in organ-type scenes with 2 pitch blocks, filters, and rotaries, on top of a patch with normal guitar sounds

That's feasible, I have a few of those.
But one can always come up with preset designs which require more CPU than available.

This last wouldn't be a big deal if I could use my scenes setup on MFC-101 and have an IA dedicated to an organ patch with tap again to get back to the previous patch. I play rhythm and need to cover organ-type sounds often, and this is a challenge. My compromise now is just to use a basic Rotary block. But it's a compromise

The MFC lets you assign an IA switch to a Program Change.

http://wiki.fractalaudio.com/axefx2...1#MFC-101:_turning_an_IA_into_a_preset_switch

Or you can program an Alternative or Backtrak Preset switch.
 
I admit to not having invested enough time to figure out why I can't get the Backtrack preset switch working correctly. I hit a roadblock that may be related to me using bank size 0. Not enough clear documentation to definitively figure it out without my own experimentation. I'll get there eventually....

I used to use X/Y for amp parameter changes, but now I use scene controllers which frees up X/Y for diff amp types. That problem solved. i need two reverb blocks for some patches because I like to have a tremolo and reverb with mix controlled by a pedal for Born on the Bayou type sounds to quickly go wet/clean, and then at end of chain I still want some reverb for basic ambience. Cabs going with one block in stereo is probably not a big deal instead of two separate blocks... And 2 cab blocks each with stereo settings I've enjoyed with 2 separate monitors but it loses definition in a mono mix- at least in my amateurish attempts- all of which means I don't really need 2 independent cab blocks.

I can get my organ sounds working in same patch as guitar, but I have to use normal instead of high in reverb, and normal instead of high in cab block preamp.
I like the preamp sound too much (adds a nice thickness) to outright disable it.
And might have to give up multi-compress on end of chain depending on what else I'm trying to jam into the patch.


All of these issues are really minor whines in the context of an amazing product. But I say them here to give voice to how more CPU is always useful. If anyone is trying to decide whether to buy the product, just buy it! I whole-heartedly recommend it and my feedback would be badly taken out of context if it dissuaded anyone.
 
  • Like
Reactions: yek
I understand. But suppose the Axe-Fx III comes with 3 DSPs, then people will still find a way to cramp so much in the preset that it'll run out of juice.
 
I understand. But suppose the Axe-Fx III comes with 3 DSPs, then people will still find a way to cramp so much in the preset that it'll run out of juice.

Yes, but in the meanwhile, it will be possible to do more with one preset. Which is good.
 
Solution: run an ax8 or fx8 in front of the axe fx. Not sure if that has already been discounted for one reason or another but seems like it could work. Another work around: get a strymon big sky to run your reverb, then all the cpu hungry reverb will be outboard. If you have to have it in the same box, take the strymon guts and instal them in the axe fx box.
 
Back
Top Bottom