• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

Impulse responses- load highest hz or hz specific to your device?

krfoss

krfoss

Supporting Member
Nov 29, 2007
3,039
4,505
Orange County, CA
#1
I’m new to impulse responses and have a boss gt1000core and a shift line Olympic. Both can run IRs.

The pack I bought has a bunch of hz options. Should I be running the highest available, or another option?

Any help would be great.
 
#2
What is the system sample rate of each device? Using an IR captured at a higher rate than the system sample rate will result in some content above Nyquist being removed by the system’s anti-aliasing filter (assuming there is one). If there is no such filter it’s likely that frequencies above Nyquist simply aren’t reproduced. Since an IR is a wholly linear process it’s unlikely that aliasing would be introduced. That said, sometimes systems will behave strangely as you approach Nyquist, so my inclination would be to match the sample rate of the IR to the sample rate of the system where you’re hosting it. Alternatively, and perhaps counter intuitively, you could just use the lowest sample rate as it’s guaranteed to not have content above Nyquist. This all assumes that each unit can do a good job converting a sample rate different than its internal sample rate if presented with a mismatch.
 
Last edited:
  • Like
Reactions: dbsfgyd1
#3
silky smoove said:
What is the system sample rate of each device? Using an IR captured at a higher rate than the system sample rate will result in some content above Nyquist being removed by the system’s anti-aliasing filter (assuming there is one). If there is no such filter it’s likely that frequencies above Nyquist simply aren’t reproduced. Since an IR is a wholly linear process it’s unlikely that aliasing would be introduced. That said, sometimes systems will behave strangely as you approach Nyquist, so my inclination would be to match the sample rate of the IR to the sample rate of the system where you’re hosting it. Alternatively, and perhaps counter intuitively, you could just use the lowest sample rate as it’s guaranteed to not have content above Nyquist. This all assumes that each unit can do a good job converting a sample rate different than its internal sample rate if presented with a mismatch.
Click to expand...
Uhh, yeah.. I think I got all that…LOL!!
 
#4
silky smoove said:
What is the system sample rate of each device? Using an IR captured at a higher rate than the system sample rate will result in some content above Nyquist being removed by the system’s anti-aliasing filter (assuming there is one). If there is no such filter it’s likely that frequencies above Nyquist simply aren’t reproduced. Since an IR is a wholly linear process it’s unlikely that aliasing would be introduced. That said, sometimes systems will behave strangely as you approach Nyquist, so my inclination would be to match the sample rate of the IR to the sample rate of the system where you’re hosting it. Alternatively, and perhaps counter intuitively, you could just use the lowest sample rate as it’s guaranteed to not have content above Nyquist. This all assumes that each unit can do a good job converting a sample rate different than its internal sample rate if presented with a mismatch.
Click to expand...
I think this Nyquist is poking his nose in matter that isn't his business...

But yes, use the IR files with that same sample rate and bit depth as your system.
 
#5
Ritter667 said:
I think this Nyquist is poking his nose in matter that isn't his business...
Click to expand...
That would be Harry Nyquist. It's more correctly called Nyquist-Shannon to denote the proof of the sampling theorem by Claude Shannon, but everyone just calls it Nyquist. You can't talk sample rates without considering Nyquist, or his nose.

As for bit depth; That will be fixed by the WAV used to create the IR and ultimately doesn't matter. If the IR is 16-bit and the host runs natively at 24-bit, the 16 will work just fine. It's a difference of 144db versus 96db above the noise floor, both of which are significantly greater than is necessary for this application.
 
#6
silky smoove said:
That would be Harry Nyquist. It's more correctly called Nyquist-Shannon to denote the proof of the sampling theorem by Claude Shannon, but everyone just calls it Nyquist. You can't talk sample rates without considering Nyquist, or his nose.

As for bit depth; That will be fixed by the WAV used to create the IR and ultimately doesn't matter. If the IR is 16-bit and the host runs natively at 24-bit, the 16 will work just fine. It's a difference of 144db versus 96db above the noise floor, both of which are significantly greater than is necessary for this application.
Click to expand...
I mean, I've talked about sample rates quite a lot without ever considering Nyquist. It's far less confusing to say "use the same sample rate as your projects" vs overloading people with jargon. If someone cares to learn why sample rate is important they can go on to do so.
 
Last edited:
#7
joebingo said:
I mean, I've talked about sample rates quite a lot without ever considering Nyquist. It's far less confusing to say "use the same sample rate as your projects" vs overloading people with jargon. If someone cares to learn why sample rate is important they can go on to do so.
Click to expand...
Why would you not consider Nyquist? That’s like talking about a bass and not considering strings, notes, or any other core concept.
 
  • Like
Reactions: Passinwind
#8
silky smoove said:
Why would you not consider Nyquist? That’s like talking about a bass and not considering strings, notes, or any other core concept.
Click to expand...
Because I don't need to in order to get the results I want.

It's more similar to talking about the bass guitar as a mechanically driven, tension modulated wave-system governed by the laws described by d'Alembert, Fourier, and Helmholz. Very useful information for a small amount of people.

Don't even get me started on those who play the instrument without considering Rayleigh
 
Last edited:
#9
You seem very worked up over a response where I provided an answer to the OP that covered two likely situations, and additional considerations that may have been useful to the OP. If they don’t understand something, they can ask a follow up. If they know more than you’re assuming it will have been useful information. If they only want the short answer, it was provided in my response. Dumbing/watering down responses because a third party commenter doesn’t like it is not the path forward for meaningful discourse on an Internet forum.

If you find my posts that objectionable there’s always the ignore button.
 
#10
sometimes it's difficult to read the difference between light hearted silliness and being very worked up I guess, sorry for not including emojis.

Are you open to my point about overloading people with information they're not ready for being a suboptimal way to teach?
 
#12
silky smoove said:
Not really in this context. You’re making assumptions about the OP and ignoring my prior points. Provide the info and if clarification is needed let people ask for it.
Click to expand...
OP asked about Hz and didn't use the term sample rate. He didn't reply to you, two other people did reply to you but only to make jokes about how needlessly complicated your reply was. The assumption was safe.
 

Latest posts

Top Bottom
Back