Please ensure Javascript is enabled for purposes of website accessibility Jump to content

IR's and bitrate...


mileskb
 Share

Recommended Posts

This has come up before, but I'm wondering if one of the "inside" guys could address this..

 

In the Manual "48kHz, 16-bit, mono, .WAV type IRs of up to 2,048 samples are natively supported. But the Helix app allows you to import IR .WAV files of different sample rate, bit depth, length and stereo format, and the app will convert these attributes automatically before sending to the Helix hard- ware"  which makes sense.

 

 

However,   as example when recording, if my destination media is a CD, I try to record at 44.1Hz 16 bit unless I plan to do a lot of processing because 44.1kHz/ 16bit is the standard sampling and bitrate for CD's.    The reason I do this is because even though an algorithm is used to downsize from lets say 96HZ you are GETTING RID OF DATA.  

 

So my thought that I'm looking for confirmation is...  If you can get 48kHz 16bit IR's, you should use them and NOT use any other size as while the conversion loss is probably not audible...  why throw away data if you don't have to.

 

 

 

 

Link to comment
Share on other sites

Let's start with I know nothing. But as I recall, there are posts that state that the difference in fidelity, in regard to IR data high or low numbers, when used in the Helix, is negligible. Some hear a difference, but it's so close, it's worth going to the lower values, if you need more memory for your specific patch. I'm not sure if this will help you but hope it does.

Link to comment
Share on other sites

Ideally, you'd use a 48KHz IR with at least 2048 samples. Longer samples aren't a problem: they just get truncated. In that case, the only "data" that is lost is extreme low frequency response and room reflections (that probably aren't desirable anyway). A 24 bit depth would be converted to 16 but that's like downsampling a higher resolution resolution image to something your display can accommodate: information is lost but the system is still giving you the same results as if it got an input of optimal resolution.

 

TL;DR: You might be overthinking the "loss of data" issue.

Link to comment
Share on other sites

This is probably being over-thought. Just use what sounds good. Why worry about the data being tossed.

 

The question may be academic, but that doesn't mean it's irrelevant.

 

OP asked for an "insider" - I read "definitive" - answer.

 

Three people responded "I don't know, but here's my opinion".

 

I know you mean well, but some people just want to KNOW!

 

Does anybody KNOW?

  • Upvote 1
Link to comment
Share on other sites

The question may be academic, but that doesn't mean it's irrelevant.

 

OP asked for an "insider" - I read "definitive" - answer.

 

Three people responded "I don't know, but here's my opinion".

 

I know you mean well, but some people just want to KNOW!

 

Does anybody KNOW?

Theres nothing to be gained with the information.

Link to comment
Share on other sites

Thanks for telling us how to think, but we already know that you don't know. How would you know that there's something to know that isn't worth knowing?

Sorry dont share info with trolls...but thanks for the entertainment of watching you go off the deep end :)

Link to comment
Share on other sites

Will, I've posted enough on this forum that you know I'm not a troll. As well I know that you're not one.

 

I'm sorry if I sounded grumpy (would a smiley have made it better? :)), and I acknowledged your good intentions. So why are you attacking me?

 

I get that you're not interested in the answer to the question, but I am, as is (I assume) the OP.

 

Why not just step out and let someone who knows answer the question?

Link to comment
Share on other sites

...However,   as example when recording, if my destination media is a CD, I try to record at 44.1Hz 16 bit unless I plan to do a lot of processing because 44.1kHz/ 16bit is the standard sampling and bitrate for CD's.    The reason I do this is because even though an algorithm is used to downsize from lets say 96HZ you are GETTING RID OF DATA...

 

You are not getting rid of data.

You are not losing anything.

 

You are converting it. That is a BIG difference.

In theory, a well recorded song with all 24 bit/96K tracks that is then dithered/converted to CD sample rate will sound better than if you record everything at 16-bit. Also, if you are recording at the higher sampling rate and then apply compression and EQ for mastering, the compressor and EQ have more data to deal with and your final output might be better sounding. It might even be noticeable.

 

This is also why when editing video for the web, I shoot everything at 1080, even if it's going to end up at 720. I don't lose by shooting higher and then going down to the 720, but I do gain flexibility in a number of ways. Audio is similar.

 

...So my thought that I'm looking for confirmation is...  If you can get 48kHz 16bit IR's, you should use them and NOT use any other size as while the conversion loss is probably not audible...  why throw away data if you don't have to.

 

When you are loading an IR, you will probably never notice the difference between one that has been converted and one that hasn't, but I think you're right. I generally use the right one just to be sure.

  • Upvote 1
Link to comment
Share on other sites

If a maker of IRs offers higher bit rates, it stands to reason that their original content was created that way, and any lower rate versions were created by converting those.

 

So it's really not a question of whether to convert or not, it's who do you want doing it. Arguably, if they're pros, they're likely to do it well. If not, who knows, and who knows how well Helix does it.

Link to comment
Share on other sites

that's like downsampling a higher resolution resolution image to something your display can accommodate: information is lost but the system is still giving you the same results as if it got an input of optimal resolution.

Actually video has a "feature" that when you take a high amount of data and put onto a smaller screen or output size, it actually appears to improve.  It doesn't actually improve, but it looks like it's sharper.  but i digress

 

 

 

 

 

You are not getting rid of data.

 

You are not losing anything.

 

You are converting it. That is a BIG difference.

 

In theory, a well recorded song with all 24 bit/96K tracks that is then dithered/converted to CD sample rate will sound better than if you record everything at 16-bit. Also, if you are recording at the higher sampling rate and then apply compression and EQ for mastering, the compressor and EQ have more data to deal with and your final output might be better sounding. It might even be noticeable.

 

but I think you're right. I generally use the right one just to be sure.

 

 

True enough.   However for example, voice-over for things like Computer Based Training are done at lower resolution because it just sounds better for that, but yes..  for mastering it's nice to have the extra bits for the compression and eq to use.  However, if you aren't mastering, for whatever reason, then recording at the same rate it will be played back "may" be better.  

 

For those who think the IR's that are presented at multiple bit rates are recorded at the highest then downsized, I would not make that assumption at all especially if it's the same people who do room IR's.   I would guess, as they state, they are "recorded" at the different bit rates.   I hate to rain on the general party of "bigger and faster is always better" but in audio, it isn't. 

 

I'm gonna stick with the "fewer conversions is better" thought process.  I admit, the likelihood of actually hearing the difference is unlikely, but no need to do a conversion that need not be done.

 

And while intended to no one in particular some of the responses reminded me of the famous quote...

"Better to remain silent and be thought a fool than to speak and to remove all doubt."

Link to comment
Share on other sites

Will, I've posted enough on this forum that you know I'm not a troll. As well I know that you're not one.

 

I'm sorry if I sounded grumpy (would a smiley have made it better? :)), and I acknowledged your good intentions. So why are you attacking me?

 

I get that you're not interested in the answer to the question, but I am, as is (I assume) the OP.

 

Why not just step out and let someone who knows answer the question?

Its all good man. I apologize as well. I honestly dont look at names a lot so quite often i DONT know, but after you mentioned and i thought a bit, i realized that you make your share of contributions. I just didnt see a reason to jump on me. Its all good though. 

Link to comment
Share on other sites

We're good.

 

Actually, I agree with you that it's much to-do about nothing. I sure can't tell the difference between loading an ir at 48/16 and loading the same ir at 96/24 and letting Helix convert . Considering what I've read and other contributions made to this discussion since it started, I believe that I can now say with absolute certainty that it is definitely better to know than not to know, even if what I now think I know is that no one REALLY knows!

Link to comment
Share on other sites

Oh boy.

 

So much has been written about this subject... i don't want to be an a$$, but I wouldnt bring this up had the OP not shouted about "DATA".

Keep in mind that your IRs are of guitar cabs, that have a huge high-frequency roll off. Oh, and that the human brain and almost all of humanity's cumulative technology in physical sensors FILTER OUT ALMOST ALL THE DATA IN THE UNIVERSE, perpetually.  ;) Then consider how sound changes when you tilt your head 5 degrees, or the humidity changes 5 percent.  :)

 

After all that try a digital sample rate converter to convert the IR from 48khz to 44Khz and see if you can hear the distortion on the resultant white-noise "pop" of an IR file shorter than 50ms - and listen (don't look) for the noise at around 40Hz @ -80dbFS  and also perhaps 20khz @ -80dbFS.  

And of course, always use your ears before trusting any brand name, specification, review or recommendation. All that matters is that it sounds good.

/rant off

Link to comment
Share on other sites

After all that try a digital sample rate converter to convert the IR from 48khz to 44Khz and see if you can hear the distortion on the resultant white-noise "pop" of an IR file shorter than 50ms - and listen (don't look) for the noise at around 40Hz @ -80dbFS  and also perhaps 20khz @ -80dbFS.  

 

 

I knew a guy who would swear up and down that emailing an MP3/WAV file changed changed how the file would sound.

  • Upvote 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...