Questions about IRs

[SymphonicDischord]

So I've been utilizing IRs with some of my presets, I've tried both the 1024 and 2048 versions but I really don't notice any discernible difference from a sound perspective. I'm hoping someone can explain what if any difference there is and if there is any benefit to using the more DSP intensive 2048 samples.

 

Thanks in advance.

[duncann]

I think you already answered your question. Sonically, there is going to be very little, if any, difference, as you've already discovered. The advantage of using a 1024 sample IR is less DSP use. I don't think there is any discernible advantage to using a 2048 sample IR, but if DSP isn't a problem, then I would use the 2048 block just because it's there.

[spikey]

Some can hear the differences but for me its hard to hear the difference too, so I just trust in the fact that it sounds good to my ears, and that a higher sample rate IR should sound better than a lower one. Why? From what i can gather and from what I've read and seen on a graphic of a sample IR, the number of  "points" that were sampled in making the IR have more at a higher sampled rate, than a lower one does. Whats that mean to us? Well, The more "points" on the graph of a higher rate IR sample adds a better detail (digital description) of the sample (sound quality) that was "copied" or sampled. Thus a better copied sound. There are others that can describe this better, but for me the bottom line in the end result on any IR/patch, is how it sounds to "you".

[shanecgriffo]

i'm sure i can hear those gaps between the samples on the lower rate one 😆

[spikey]

If you can NASA needs you bud LOL...  :lol:

[JLondon]

I have a thought (and that's all it is) in the form of a question.  Just because one setting may sound "different" to some, does said difference necessarily equate to "better"?  In other words - are those, that can, actually perceiving a difference in "quality" vs. just hearing a difference in "sound"?  If the latter is the case, couldn't the difference be compensated for with the tweaking of a few knobs?  I'm of the opinion, for instance, that the hundreds of file options you are bombarded with when you download an IR is overkill to begin with.  I do respect all the trouble they go through to make them, but I suspect a lot of the options could be eliminated simply with the tweaking of a few knobs on the Helix itself.   By the time we adjust an IR's Hi/Low cuts, mix, etc, (as we all do) we most likely just converted the IR into an IR that already existed as a file option.  This is absolutely true of all the effect/amp blocks in the Helix itself.  In reality the Helix is not a magic box with a chip (or whatever) in it for every/effect5 amp featured.  In most cases one amp or effect can be made to sound exactly like another of its own type if you tweak it just right.  So my point in all this, I suppose, is that unless you are being honest with yourself and hearing an actual drop in "quality" rather than a difference in "sound" - then the choice is subjective and I'm going with the "whatever sounds best - do it" camp.

[SymphonicDischord]

Spikey thanks for the indepth explanation, I guess you could kinda say it's like tube sniffing for IR aficionados then😉. I supposed I'll just take the advice of duncann and if I have the DSP available, I'll use the 2048 setting.

[jeremyn]

The length of the IR is all about the bandwidth/detail of filter response. The longer the IR, the higher the resolution of detail that can be captured and 'played back'. For real world cabinets, the extra detail of longer IRs is mostly noticeable in the lowest frequencies (i.e. < 150Hz), or for speaker/cabinets that have very noticeable resonant peaks.

 

Many real world speaker/cabinets already have a fairly smooth characteristic response in those areas, and increasing the resolution (ie. length) of the IR won't do much. Same for real cabinets that have been well designed for a smooth response across their output spectrum (ie. no deep/sharp nulls anywhere in their frequency response). Those cabs won't sound much different (if at all) with a longer IR.

 

A rig where the combined (guitar/amp/speaker/cab/mic/EQ) doesn't have much of any low frequency response, probably isn't going to benefit from a longer response, whereas a rig with a lot of low end detail will. Also, a super clean trebly guitar/amp/speaker/cab with a lot of chimey resonance, may benefit somewhat from a longer IR as some of that 'chime' is due to narrow resonant peaks in the speaker.

 

And of course, an IR capture of a far mic'd cab (which includes some 'room' response) will benefit simply from the longer tail of the longer IR (although even 2048 points isn't anywhere near enough to properly capture most room reverberation).