FM3 VS Quad Cortex

Status
Not open for further replies.
Fair enough but I say you’re under estimating/acknowledging the realities of road use on equipment like this.
I played really bad gigs in the mid west, for many years. Some weeks, I was playing every single night, at the shittiest dive bars around. My MFC101 looked the same as it did the day I bought it...new. But, I did something smart...I always had a monitor directly in front of it, along with having it just out of drunk folks' reach, and keeping an eye on my gear... It amazes me how many stories I've read of booze getting dumped on pedal boards. We played with some really heavy bands and even their gear looked great.
 
close, 85% and up.

I was a very early adoptee I’m hoping this would be more like 60% now?
There are various ways to save CPU. Delays and reverbs chew up the most CPU -- if you set the reverb to "economy" mode you'll save a lot of CPU and the reverb will still sound great. Also, you can save CPU by using amp block gain, etc. in lieu of drive blocks.
 
close, 85% and up.

I was a very early adoptee I’m hoping this would be more like 60% now?
If you didn't have a newer firmware then there was a certain point where the CPU was getting the maxed out warning with just a few blocks.

It was fixed with a newer firmware.

I received a new FM3 about two months ago and it had happened to me, but I knew that I needed to check for a firmware update since that did not seem right to me, and sure enough, the firmware update fixed it and I was easily able to have at least 8 blocks (I didn't check to see if I could have more since 8 blocks was definitely enough for me).
 
Personally, if the idea is to have a smaller / fly rig alternative to your main af3 rig, I’d go for the fm3 so that you can use the same presets more or less. In any other case, I’d go for the fm3 anyways! 😁
 
I researched the Quad Cortex a bit because a friend was interested.

Their blocks don't have "channels" or an equivalent. So if you want 3 different Reverb sounds, you have to use 3 blocks worth of CPU. Want to switch amp sounds? Have to use scenes for that (no way to do it in Stompbox mode). Or you could use presets instead and have your sound cut out for half a second while you switch. Also, unlike the Fractal where you can assign buttons to do whatever, on the QC all the buttons are either on Preset, Scene, or Stomp, and you can toggle between those three global modes. Have fun messing with that in a live gig.

I've been with Fractal so long I've forgotten what it was like to have gear that was just straight up incapable of meeting my basic setup/routing needs (have I told you lately how much I appreciate you, Cliff?). QC is quick to flaunt its "vulgar display of power," but as the saying goes, it's not about the size but how you use it.

Oh, also the FM3 + FC-12 is the same price right now as a Quad Cortex and is actually available in addition to being better.

Edit: Appears their scenes can have different amp settings so I have adjusted my post accordingly. Still no way to switch amps in Stompbox mode to my knowledge, which is a dealbreaker for me.
 
Last edited:
Their blocks don't have "channels" or an equivalent. So if you want 3 different Reverb sounds, you have to use 3 blocks worth of CPU. Want to switch amp sounds? Have to have multiple amps set up and use Scenes to switch between them.
In the video I saw, scenes recall any change in type/parameter, so you can have different reverb/amp: 8 scenes, 8 amps.
 
Quayle’s demo has definitely been my favorite of them so far. I learned more about the unit in that one than any other vid.

Overall it’s pretty cool and at this point I can only snicker at the bold claims that were coming out a year ago and they’re now going on their, 3rd of 4th delay of the release? It’s not a hardware thing as Doug has said the units are ready to go, they’re just working on the firmware.

I don’t know enough about coding/programming, but I would think that they would have had some kind of working model of this back when they started making these original claims. Is that often a thing with software, to release the general idea as a selling point and then try to build software that will live up to those claims? In my line of work, I’m used to dealing with floor/building plans, so even if it’s just on paper, I can get an idea of what the end result will be. Software doesn’t really lend itself to this and I have no clue what the lead-up to for designing software is like.
 
I think Tom Quayle's. Also Pete Thorn. Check them...
So I checked the videos and may have missed it but didn’t see anything to the effect of running 8 amps. PT says off the top that you can run 4 amps at the same time though you can read the same thing straight off Neural’s marketing page. As Neural explicitly mentions 4 amps I’d expect that to be the upper limit.
 
So I checked the videos and may have missed it but didn’t see anything to the effect of running 8 amps. PT says off the top that you can run 4 amps at the same time though you can read the same thing straight off Neural’s marketing page. As Neural explicitly mentions 4 amps I’d expect that to be the upper limit.

4 amps or up to 8 captures
but if you want some effects down to 6 captures they said realistically
4 full amp models or 6 captures basically their version of profiles
so you could have 6 different amps
in a preset
 
So I checked the videos and may have missed it but didn’t see anything to the effect of running 8 amps. PT says off the top that you can run 4 amps at the same time though you can read the same thing straight off Neural’s marketing page. As Neural explicitly mentions 4 amps I’d expect that to be the upper limit.
Sorry, I thought you talking about "channels" (Axefx can run 1 channel at a time in each block; even scenes run one at a time for each preset). I don't read the QC manual... from the video what I get is that there are 8 "scenes" as in AXFEX... but there a different choice: QC allow each block parameters to be saved and recall. So you can have up to 8 scenes with 8 different amp (clean, rithm, lead, etc...). One at a time. Axefx allow you to select 1 channel out of four, so in a patch you could recall 4 different amp (but one a time, either with channel or scene).
Talking about block, I Know fractal can run 2 istance (1 x FM3) of amp block. QC.... I don't know! I trust Randalljax info.... :)
 
While there is a slight gap with preset changes, it is still 100% usable to change presets while performing, I did this with an Axe-Fx II for several years, and it's even faster now with the FM3.

My point is, you are not stuck with only a single preset of 4 channels.

Y'all need to stop with the limited "must stay within a single preset," mindset IMHO.
 
My fpga experience from a different hobby (MiSter) has taught me that there’s a LOT you can do with code optimization to the point where chip specs are important, but harder to compare. To that end I’ll stay cautiously optimistic that the QC will be able to do some great things as long as they continue to work on the algorithms, improve the UI, and shake out the bugs. My impression probably won’t solidify for about a year after physical release.
 
4 amps or up to 8 captures
but if you want some effects down to 6 captures they said realistically
4 full amp models or 6 captures basically their version of profiles
so you could have 6 different amps
in a preset
I can see 6-8 captures as they are less resource intensive. I took the upper limit of 4 to be for models. You’ll know my bias when I say I prefer a white-box to a black-box approach but it does like an interesting device. I do find some of the reviews a little ‘gushy’ to my taste. When TQ talked about scenes as though they were something new and wondrous I had to check my impulse to retch. But then I guess that’s what he’s being paid for.
 
My impression probably won’t solidify for about a year after physical release.
There’s a lot of stuff still to be sorted and new things will crop up once it’s released into the wild. My stick-a-finger-in-the-air guesstimate was 12-24 months. My experience with s/w-h/w projects was to take my initial estimate and about double it. There are just too many variables that fall outside your control.
 
In the video I saw, scenes recall any change in type/parameter, so you can have different reverb/amp: 8 scenes, 8 amps.
Looking at the manual a little more carefully, I think you’re right. I’ve edited my post accordingly. Thanks for pointing that out.

Having amp changes tied to Scenes is still a dealbreaker in my book. The Fractal system of having channels and having the scene control the channel is vastly superior.
 
So I checked the videos and may have missed it but didn’t see anything to the effect of running 8 amps. PT says off the top that you can run 4 amps at the same time though you can read the same thing straight off Neural’s marketing page. As Neural explicitly mentions 4 amps I’d expect that to be the upper limit.

I can see 6-8 captures as they are less resource intensive. I took the upper limit of 4 to be for models. You’ll know my bias when I say I prefer a white-box to a black-box approach but it does like an interesting device. I do find some of the reviews a little ‘gushy’ to my taste. When TQ talked about scenes as though they were something new and wondrous I had to check my impulse to retch. But then I guess that’s what he’s being paid for.
4 parallel amp blocks makes sense, their 4 lines map to 4 CPU cores. So one amp per core would check out. More captures if you can fit 2 captures per core. And then each of those amps/captures can have a different setting and type per scene with 8 scenes. That should cover most use cases except the guys that like to stompbox style toggle their amp channels and effects.


Quayle’s demo has definitely been my favorite of them so far. I learned more about the unit in that one than any other vid.

Overall it’s pretty cool and at this point I can only snicker at the bold claims that were coming out a year ago and they’re now going on their, 3rd of 4th delay of the release? It’s not a hardware thing as Doug has said the units are ready to go, they’re just working on the firmware.

I don’t know enough about coding/programming, but I would think that they would have had some kind of working model of this back when they started making these original claims. Is that often a thing with software, to release the general idea as a selling point and then try to build software that will live up to those claims? In my line of work, I’m used to dealing with floor/building plans, so even if it’s just on paper, I can get an idea of what the end result will be. Software doesn’t really lend itself to this and I have no clue what the lead-up to for designing software is like.

Coding is a notoriously hard discipline to do estimates in. Especially if you're estimating things you haven't done before. Building your companies first hardware modelling platform would have a lot of "things you haven't done before". I'm sure they had prototypes, I'm just they had an idea about the architecture and now it would come together, they may even have done the thinking about how to make it all work. But software is also one of those that the more exactly you try to plan it then either the slower it goes or the further off the rails it goes. It's a very iterative process so often you're just estimating "how many iterations do we need before we have enough features and polish to be able to release it?" and hopefully enough margin for error that when some of those iterations for awry you're not ending up way off base.
 
4 parallel amp blocks makes sense, their 4 lines map to 4 CPU cores. So one amp per core would check out. More captures if you can fit 2 captures per core. And then each of those amps/captures can have a different setting and type per scene with 8 scenes. That should cover most use cases except the guys that like to stompbox style toggle their amp channels and effects.




Coding is a notoriously hard discipline to do estimates in. Especially if you're estimating things you haven't done before. Building your companies first hardware modelling platform would have a lot of "things you haven't done before". I'm sure they had prototypes, I'm just they had an idea about the architecture and now it would come together, they may even have done the thinking about how to make it all work. But software is also one of those that the more exactly you try to plan it then either the slower it goes or the further off the rails it goes. It's a very iterative process so often you're just estimating "how many iterations do we need before we have enough features and polish to be able to release it?" and hopefully enough margin for error that when some of those iterations for awry you're not ending up way off base.

Thanks!

After reading that, I don’t feel as bad for busting Neural’s balls in discussion of the QC. They hyped this up pretty big in the beginning and in another group of friends, we’ve been snickering at their marketing for a while before the QC was announced. Great plug-ins and Doug did great with the Darkglass stuff and I‘m of the mindset that the more cool gear out there, the better it is for us guitarists, so my ball busting is in jest and I could definitely see myself owning a QC down the road. (Well, even QC buyers are seeing themselves with a QC down the road :D)
 
Status
Not open for further replies.
Back
Top Bottom