IRs 2048 vs 1024

[Dshow]

Hi,

there is the possibility to load IRs with 2048 and 1024 sample.

What is the difference between those two, meaning does it have a big impact on the tone if 1024 sample is used?

I have tried both and feel that there is not that much difference.

 

What sample was used when Line 6 created the cabs that are in helix?

 

Thanks :-)

[duncann]

There likely won't be much difference. I remember trying in a DAW cutting off the samples until I noticed a difference. I don't recall the sample number where that was, but it was low, much lower than 1024. And the first thing to go is the low end.

[Dshow]

Thanks duncann,

I always try 2048 first. If I run out of DSP then I change of 1024 but as said very little difference.

[jaeger28]

2048 samples are 42milliseconds at 48khz. 1024 is half of that. So basically if you have a 400ms IR then it will be truncated after 10% on the bigger setting.

 

Think about capturing a cab with a real amp. Imagine hitting a very short chord ant then palm muting it. The cab will resonate beyond your palm stop. If you only get 42ms of that, it might be short to an exten that is a bit unrealistic or static. In the 1024 setting it is half of that.

[Dshow]

2048 samples are 42milliseconds at 48khz. 1024 is half of that. So basically if you have a 400ms IR then it will be truncated after 10% on the bigger setting.

 

Think about capturing a cab with a real amp. Imagine hitting a very short chord ant then palm muting it. The cab will resonate beyond your palm stop. If you only get 42ms of that, it might be short to an exten that is a bit unrealistic or static. In the 1024 setting it is half of that.

Thanks - good explanation!

[jaeger28]

You know the 1024 setting has still more of the IR than for exampe my standalone torpedo live which only uses 18,7 ms from externally loaded IRs.

[kensington]

Use your ears.
Personally I don't hear any difference, so I go for the 1024 version and save dsp for other fun stuff:)

[duncann]

Found the post I made, in the FAQ thread, where I used Reaper to truncate an IR and this is the number I came up with: ~15ms. Anything below that is where you'll start to notice a difference, with the low end vanishing first. Now, I didn't try every IR in existence, but I did try a few, and every one, for the most part, had the same threshold.

[Dshow]

Use your ears.

Personally I don't hear any difference, so I go for the 1024 version and save dsp for other fun stuff:)

 

Yeah that's what I do. I usally do not like that many effect in my sound so in most of the time I can go with the 2048 but I really don't notice that much difference.

[silverhead]

I think it's similar to the looper running at full speed or half speed. I really can't notice a difference but I use full speed unless I need more time on the loop. Similarly, I would use 2048 IRs unless I run out of DSP. Switching to 1024 is the first dsp-freeing move I'd make.