I would like share my latest blog post and hopefully generate some discussion on perhaps next steps.
What started out as a post about step response, later evolved to a larger discussion around the notion that a driver can be "fast", and if this is something that could be measured.
https://josephcrowe.com/blogs/news/step-response
What started out as a post about step response, later evolved to a larger discussion around the notion that a driver can be "fast", and if this is something that could be measured.
https://josephcrowe.com/blogs/news/step-response
Whilst I dislike the term 'fast' I believe it could also be termed 'fast decay' or 'low energy storage'. A steep rise time response to a waveform is relatively simply to achieve; an equally quick decay far less so as frequency decreases. Since the original term describes a purely subjective aural observation, we would have to redefine it in scientific terms before trying to measure it.
While ‘fast’ can be described in terms of bandwidth and cutoff slopes, I missed out the point of your blog. Which measurement setup for a 5kHz square wave would suffice to make valid measurement results?
No, you really don't. 🙂
If you're going to go down the audiophile rabbit hole on this, you need to define "fast" in very precise terms. Putting quotation marks around it doesn't cut it.
Dave.
Joseph, this is 43 pages of discussion regarding this topic. Jeff Bagby had a lot of interesting input about it.
Fast vs. Slow Subs: Myth Or Reality
I wish to propose a quantitative yet intuitive (very old?) measure of transcient response distortion, namely input-output wave-envelope area difference (i.e. belonging to one not both) divided by area common to both.
Example here: https://www.diyaudio.com/community/...go-for-3-way-then.391639/page-10#post-7167908
Example here: https://www.diyaudio.com/community/...go-for-3-way-then.391639/page-10#post-7167908
Last edited:
Well, "Fast" as it's used here is a Subjective term.
I think it's best described as perceived transient clarity, not objective transient clarity (though the two might have some correlation).
All-else-equal it's relatively well known that a lower Mms driver tends to sound a bit "Faster". The same is also true of a more efficient driver. (..though an actual "all-else-equal" is extremely rare, so it makes a really good comparison almost impossible).
What do you mean 'redefine it in scientific terms'?
I know that's the general consensus in the more technical community. I understand that I am re-opening a debate that is largely considered case closed. Technically a square wave should be identical between two different drivers that have the same frequency response, as far as I understand. So I'm curious why my results show differently.
I describe it as a more immediate sound. I think most would understand in the audiophile community that particular attribute.
Thank you, I'm likely in alignment with it. I have a pretty in depth knowledge of FFT and measurement. I knew the standard test suite would not reveal anything different between the tweeters, and I know the current science on the matter is that increased moving mass introduces a roll off, so if the driver has bandwidth then it is considered "fast".
Last edited:
99% when "fast" is used in conjunction with speaker sound, it has nothing to do with actual rise time of the produced signal. Rather, its absence of resonances in the playing environment (aka: room) and a flat FR.
As long as two different speaker both reach 20k FR within a dB, they are equally fast.
//
PS. popcorn for everyone 😉 DS.
PPS: serious speaker designer don't use the term "fast" DS.
PPPS: everything is measurable - it's the interpretation that is lacking 😉 DS.
As long as two different speaker both reach 20k FR within a dB, they are equally fast.
//
PS. popcorn for everyone 😉 DS.
PPS: serious speaker designer don't use the term "fast" DS.
PPPS: everything is measurable - it's the interpretation that is lacking 😉 DS.
I figured you would, "fast" in regards to woofer and tweeter behavior is definitely just a word to describe what we (think we) are hearing.
It’s early in my testing and I don’t really have a test metric developed, if any. Regarding square wave, perhaps more sqaure looking? 🙂 I understand the bandwidth limitation is mostly affecting the square wave shape. But two drivers with matched responses should have identical square wave shapes no?
Not according to the detailed explanation offered to me in the thread I cited. "Identical" fequency response had to include complete phase response, group delays and all other distortions (for FFT) to guarantee identical transcient response; just the "top line" curve won't do (and no manufacturer would ever publish those awful truth).
To be frank - I don't think you will get anywhere with this. The scientific aspect was clear 100 years ago - if not earlier. What are you expecting to add?
This is just gerilla advertisement - no?
//
This is just gerilla advertisement - no?
//
As a long-time "purist audiophile" (explained in thread) and not-so-long-time-but-very-productive "diy" I do not believe so-called subjective and objective criteria are in conflict, only certain people are. I think "fast" is how close reproduced sound (perception) comes to the real thing we all should be familiar with (but some turn deaf ears).