AITR

My novice understanding is that the reverb block (or any other simulated reverb) is not going to be able to replace the quality of fullrez IR reflections which are "real" in the sense that they are captures of the actual sounds that surrounded the cab at recording. So given this, it does not seem it can be done in a separate block. Was reading through the post you mention but it had turned into such a mud slinging match I never did get to the crux of if/why Jay feels/felt that IR reflections could be as well or better replicated outside of an IR.
It would probably help to read about convolution based reverb.

it’s all about the length. A short IR is not long enough to characterize a room or space. But a 1024 sample IR is enough to characterize a guitar cabinet.

Since I’m the past we did not have support for real-time convolution of seconds long IRs - then yeah it made sense if you want to provide some room/space simulation to do it through the Reverb block.

Now with this new feature you will have the choice of both reverb through convolution with long length IRs or the tools you already previously had

Since this feature is about supporting much much longer Impulse responses, it makes sense to put it in the blocks that already do convolution (the cab and the IR player blocks). Could it be done in a completely separate block? Yeah it could. Could it be done in the reverb block? Yeah it could. But what does that really change for you? It would just be a different block.

Why does MKI need to “give up something”? Simple, longer IRs take more storage which the MKI version had already partitioned to user IR slots.
 
Thanks man. My room captures are all 500ms, but I’ve got some ideas for larger and longer room sounds to try to FullRes. If they sound cool, I’ll include them in future packs.

I mainly use and enjoy your 4x12 Mesa OS and 4x10 Ampeg packs' close miced mixes.

Do you see any benefit for these mixes with FullRes? What are their average length?
 
I mainly use and enjoy your 4x12 Mesa OS and 4x10 Ampeg packs' close miced mixes.

Do you see any benefit for these mixes with FullRes? What are their average length?

It’s up to you and depends how you wanna mix for a particular song, project. How can someone else answer this for you? Do you want your room/someone elses in your mix/sound or not? Up to you...soon the door is opening for it.
 
If you prefer a more dry, closed mic'd sound, then FullRes cabs will likely be of little use to you. The extra length is really only useful for longer room reflections.
 
Since it’s ment for recording, imagine it as a room mic channel. You can record it to a separate track and mix or even mute later on. You don’t have to commit to it.

It is giving us many more options for recording, even for live, for inear mixes etc.

and in my opinion this is also smth which is getting Axe Fx closer to having a convolution reverb block!!! I am pretty sure it will happen... not too soon probably.

Is it meant for recording though?

The problem here is that if you want to use a reverb to glue instruments together, it's better to use the same reverb for all of them, IMO - that will kind of create an impression that they are in the same space.

With this reverb hardcoded into an IR, how do you achieve that? It'll be from some room where you can't do anything else, you don't even know where it is. And you can't add reverb to signal already containing reverb, I'm afraid. Neither can you adjust the mix to lower reverb level. Or length. Or anything.

Guess that's the whole point of close micing - to get rid of room influence and add it during mixing.
 
I would be more than happy to give up user 2 slots for this, if it is possible.
It requires a bank of 1024 existing IR slots to get 32 FullRes... That means you would need to give up 32 slots for ONE FullRes IR.

Regardless, from previous comments it's a full bank or nothing as far as I can tell.
 
I mainly use and enjoy your 4x12 Mesa OS and 4x10 Ampeg packs' close miced mixes.

Do you see any benefit for these mixes with FullRes? What are their average length?
All of the files in a YA pack are 500ms, so using any of the Mixes and Single mics with FullRes will probably just accentuate some low end, so you don’t need to use FullRes with those. The Room mics will sound much better with FullRes though, so that will be fun to try.
 
1+ for using the AXE3 MK1 user IR banks for this new feature in one way or another. My poor knowledge on the unit’s memory distribution unleashes my imagination: Is there a way to add support for longer IRs in both user banks and leave it up to the user to manually load the desired IR length? In that case a single FullRes IR will take the space of four regular IRs or so, leaving the decision to the end user.

The two clips Cliff provided are so inspiring and bringing the Axe to a whole new level, that it would be a shame for MK1 to be left out, especially with all that processing power available!
 
@FractalAudio :
Will there be a way to import wavs with 1,3 s length to the syx format for users of 3rd party IRs like already existing?

We would be able to provide our complete portfolio, but preparing the files would only make sense if it is possible for users to do this.

Edit: I have read, that a new cab lab will do this, question answered)
 
Last edited:
I was not advocating for it to be done in another block in my posts above (think maybe you mistook another post
for mine - no matter), and was understanding that fullres quality reflections really could not be done outside of an IR. However, if you say that, yes it could be done in a separate block!, then I do think there would be benefits to separating it out ie: 1. in a separate block, I could have parameters to control it (mix, level, ...), 2. I can put the separate block where I want it - with a full on reverb baked into the IR, I either have to have my reverb before fx like modulation which I don't want, or put the cab block at the end which I don't want it - these two items, and maybe others, would be compromises I may accepts if the baked in IR reflections sound so much more amazing. May be a moot point though as I'm not sure the kind of IR reflections were talking about are feasably segregated into a separate block. Not saying this is not going to be a great feature/option - I think it is - just want to really definitively understand this area better including the pro/con of different ways of doing it.

The good news is that I think you'll be able to do that with the IR Player block.
 
I agree with you on this. I kind of touched on it a few pages ago, and got blasted for being ungrateful, so it's nice to see someone else think objectively about it as well.

I did link to in one of my posts to a twelve year old post with a discussion about this with Jay Mitchell.

I'm just finding it odd that we were told for the whole duration of the product lines that longer IR's weren't needed, and to use reverb to fill in the missing pieces.

So why is this new feature added to the IR block now, and not the Reverb block, or its own block, and may or may not fit on the mk1, but if you buy the mk2 it will definitely fit.

That older discussion
Because the reverb block is an algorithmic reverb and doesn’t do IRs afaik. So it’s either the cab block or the IR player.
Irrespective of what block is used, you need the extra memory to store these long IRs. That’s where the mk2 has more nvm.
 
As I understand it, these long IRs are essentially room captures. It would be best to not have the cab IR baked into this sound. Cab IR from the regular direct short IRs and then an ability to add the room IR to another block. There are lots of other high quality room IRs from top recording studios already captured.
I think the IR player is the best place for this right before hitting the outputs. I don’t think I want a delay or chorus or tape drive after the cab block to work on the reverberated signal.
 
Because the reverb block is an algorithmic reverb and doesn’t do IRs afaik. So it’s either the cab block or the IR player.
Irrespective of what block is used, you need the extra memory to store these long IRs. That’s where the mk2 has more nvm.
I believe that the Bricasti M7 is algorithmic, the reverb all others are judged against.

There are a few close competitors with Bricasti that did some fancy IR capturing of the M7 and are fairly close, but there's still a gap in quality between the them. Seventh Heaven, Reverberate 2 are the two big ones. I think they use something called Fusion IR or something like that. I don't remember exactly as its been a while since I researched them all. But the point still stands. Algorithmic reverb is currently the gold standard, not IR based.

To me it just makes sense to have it as a standalone block for primarily the reason already mentioned. You don't want to time base effect your reverb.
 
What intrigues me about this change is having the room reverb I normally use the reverb block for, replaced by a far better room reverb baked into the IR. From what I can gather reading about this, no separate reverb effect could approach the quality of the true reflections captured
into an IR.

My apprehensions are:
  • I have no parameter control at all - with a separate reverb block, I have loads.
  • I have no placement control as I do with reverb block (do I really want this much reverb before blocks like modulation?).
  • few variations are available yet.

Maybe I still don't have my head properly wrapped around this having abandoned ir reflections for the most part some time ago in favor of cpu miserly 512s and some room in the reverb block.
You've touched on some of the reasons why convolution reverb isn't automatically "better" than a digital reverb engine. Yes, it can be more true to the room, but there are things you can do with a good reverb algorithm that you can't do with a room capture. Things like crafting a reverb that's lush, but gets out of the way of the dry tone.

It's just another tool in the arsenal. A sweet and unique tool, but not always the right tool for the job.
 
I believe that the Bricasti M7 is algorithmic, the reverb all others are judged against.

There are a few close competitors with Bricasti that did some fancy IR capturing of the M7 and are fairly close, but there's still a gap in quality between the them. Seventh Heaven, Reverberate 2 are the two big ones. I think they use something called Fusion IR or something like that. I don't remember exactly as its been a while since I researched them all. But the point still stands. Algorithmic reverb is currently the gold standard, not IR based.

To me it just makes sense to have it as a standalone block for primarily the reason already mentioned. You don't want to time base effect your reverb.
Correct. Algorithmic reverb is considered to be superior to convolution reverb because algorithmic reverb can be perceptually tuned.

From a perceptual standpoint the ideal reverb is not found in typical reverberant spaces. See the work done by Griesinger, et. al. on perceptual reverberation.

As usual this whole topic has been confounded with people chiming in their opinions on topics of which they have little knowledge or experience.

The point of FullRes IRs is to add some room sound to your recordings and/or headphone playing. To do this you would mix conventional IRs with FullRes IRs.

You can ALSO use FullRes IRs to do short-to-medium convolution reverbs. If you have some reverb IRs you can convert them to FullRes IR format and load them in the IR Player block and use that as a reverb.
 
Correct. Algorithmic reverb is considered to be superior to convolution reverb because algorithmic reverb can be perceptually tuned.

From a perceptual standpoint the ideal reverb is not found in typical reverberant spaces. See the work done by Griesinger, et. al. on perceptual reverberation.

As usual this whole topic has been confounded with people chiming in their opinions on topics of which they have little knowledge or experience.

The point of FullRes IRs is to add some room sound to your recordings and/or headphone playing. To do this you would mix conventional IRs with FullRes IRs.

You can ALSO use FullRes IRs to do short-to-medium convolution reverbs. If you have some reverb IRs you can convert them to FullRes IR format and load them in the IR Player block and use that as a reverb.
I appreciate the clarification.

Sounds like a gnarly addition then, and I look forward to it. 🤙
 
Correct. Algorithmic reverb is considered to be superior to convolution reverb because algorithmic reverb can be perceptually tuned.

From a perceptual standpoint the ideal reverb is not found in typical reverberant spaces. See the work done by Griesinger, et. al. on perceptual reverberation.

As usual this whole topic has been confounded with people chiming in their opinions on topics of which they have little knowledge or experience.

The point of FullRes IRs is to add some room sound to your recordings and/or headphone playing. To do this you would mix conventional IRs with FullRes IRs.

When you put it like that, sounds like this wouldn't be necessarily any better than using the reverb block, just different. Rec Studio C on 100% has suited me well for years.
 
Back
Top Bottom