Current Trend of Commercial IRs: More Balanced and Polished

The difference Dave is the phase shift of EQ and the post recording compression that "mix ready" IR files give you are at least one generation away from the warts and all capture.

As a user, you can EQ and compress to make a track with an unprocessed IR fit in a mix. No one can undo those moves in a "mix ready" IR because no user knows what manipulations the producer did in the name of making the file "mix ready".
Sure, but the question isn't really whether users can undo the maker's predone processing. It's more about whether they can get to something they like relatively quickly, and feel good about the process that got them there.

They can't "undo" your mic choices or room either.

Personally I don't go for noticably processed IRs myself. I just don't think not post-processing takes the maker's artistic choices out of the picture.
 
I audition IRs, don’t care if they are post processed or not, they either sounds good with relatively small amount of tweak or need too much EQ work.

Comparing old OH IRs to their new ones in the Evolutionary pack, there is an obvious difference, and generally positive. However, i am not sure if that’s post processing, and they still all have/carry obvious OH sound signatures, eg their low mids are not excessively boomy as their IRs but it’s still a bit too busy in those frequencies yet not sounding as full As it should.

I tried to dial presets with the same amp and similar settings (not exactly the same settings, eg same gain slightly different EQ depending on the IRs) to compare IR from/among different venders over a relatively long period of time. So I know my preference fairly well and stable.

Personally, I feel IR should aim for the guitar to sound as good as possible in isolation, NOT in the mix. The latter is the job of mixing when post processing tracks.
 
If the IR pack provides some additional, polished mixes which include some EQ/compression tweaks then ok, great, I can choose to use them or not, but, I don't think the main aim of a given IR is to make the guitar sound good in a mix or not - rather, when I choose a specific IR, I expect it to represent a specific Cab Mic'd in a specific environment in a specific way with specific tools essential to do the capture. Once there is additional non optional baked in secret sauce IR polish elements beyond the realm of those elements essential to the capture process, and done only to improve the finished guitar sound, then, as a purchaser of IRs, I'm not as sure of what I'm buying or why I'm buying it. Also, if all IR vendors were to bake in non-optional eq/compression tweaks, then those of us (most of us probably) who do not have access to real cabs and related basic capture elements like mics, mic pre's etc (+ related skill sets), will no longer have as accurate a reference of what a given Mic'd cab specifically sounds like without the post polish.

One solution to this is to just make "mix ready" IRs additional/optional. Then we have the choice, though I may consider using mix ready versions with caution if I don't know specifically what the polish is. As Fractal users I think we tend to want amp models with no unknown polish not characteristic of the original amp baked in that can't be overridden as can be the case with other modellers - for me at least , I tend to want this same approach in the Cab modelling.
 
Last edited:
I tend to go straight for the 57 and a ribbon (160, 121, etc.) on individual IRs, and mix them to suit my sound, and the mix it will sit in. The manufacturer's person/people mixing the pre-mixed ones are, undoubtedly, good at it, but (as was mentioned elsewhere, above) they aren't in my chair mixing my amps/effects, so it's hit or miss whether it will suit what is needed. I have a bunch of IRs, from York, Mikko, Ownhammer, Cab.IR, and the old RedWirez ones that I have had around since I got into using an IR player (MixIR3 or something like that) in 2015/6 or so.
 
I have been using impulses for many many years since the first recabinet pack was released and guitarhack posted his first IRs on the internet.
Now I know that I plain hate what I call "position IRs"...
What do I care about a "cone/dustcap/edgecone/cone" IRs? That means nothing...what if the perfect sweet spot for that mic was an 1/2" inch away but for consistency the developer decide to make them all the same.
Maybe the sm57 sounded best at X position but the MD421 sounded like sh*t at that position but you already have your own recipe to kinda automate the capture process and everything has to be the same position.

Nowadays I just want good sounding sweet spots and in every single cab, speaker and microphone the sweetspots will be different (and also greater or lower in quantity)

I think we went from totally RAW in the very beginning to a couple of years lately of everything processed to hell and back and now we are almost back to the point were most devs just use the mic position to adjust the "natural" eq of the IR.

I really like what mikko, NeuralDSP and amplitube5 are doing with their full 3D cab sims were you can move the mic anywhere but for it to be even remotely realistic (and decent sounding) you need an absurd amount of IRs.

I don't like very ultra processed mix-ready (how I hate that term) IRs but I like and use premade blended mics mixes a lot
and most of the time I dont go out of that folder because it is my opinion that if I "need" to go out of the pre-made blends/mixes (not want....NEED) to get a decent sounding guitar sound there is something wrong with the IRs or the developers ear/taste.

Also I like something that a colleague once said when testing new IR packs
"I ignore everything and go directly to the sm57 IRs because if I cant get a good sound or at least 80% there with just the SM57 IRs which is the base for most modern guitar sounds then the ir pack is just not up to par."
 
Last edited:
I really like what mikko, NeuralDSP and amplitube5 are doing with their full 3D cab sims were you can move the mic anywhere but for it to be even remotely realistic (and decent sounding) you need an absurd ammount of IRs.
Yep, that's the approach I have been advocating for Fractal to dab into for a while now.

A large amount of IRs, sure, but each IR is tiny and the large amount of IR could/should be shot and collected by an automated process, e.g. robots move mics around systematically etc.
 
Back
Top Bottom