[opnly bafld]

Noted and appreciated Lin.

Dan

Maybe I shouldn't have replied in kind.
Oops, I'm too late also.

Lin 

[planet10]

I'm surprised you're publicly attempting to question the validity of such works

I am not. Just the opposite, this is a landmark work with a lot to teach us. We may thou be arguing interpretation.

I am passing completely thru the book for the 2nd time. And have read some parts already many times. Toole dismisses in room measurements, and the metric that he (& Olive) used to correlate what people hear to what is measured takes 70 measurements.

One thing that stands out, is that Toole says over & over again, is that the ear + brain is way more capable thean a mic + analyzer.

dave

[planet10]

I would agree that it could be made flat, and flat is often (usually?) better than "not flat," but does flat == good?

Back to Toole & Olive (and Geddes too). On axis measures are not that useful, at least not without support from off axis measures.

dave

[Saurav]

Personally I think this is a false dichotomy. It doesn't have to be all measurements or all subjective. It seems foolish to ignore the broad spectrum between those extremes.

Ultimately, for most of us, the only person that needs to be satisfied is ourselves. If you can do that by ear, great. It's not that hard or expensive to get some basic measurement capabilities. I've often tuned something by ear, thought I had it sounding pretty good, then took some measurements, made some simple adjustments, and had to admit the end result was better. And vice versa - sometimes I'll listen to the theoretically better setup, but then go back to something I enjoyed more.

[DanTheMan]

Toole dismisses in room measurements, and the metric that he (& Olive) used to correlate what people hear to what is measured takes 70 measurements.

One thing that stands out, is that Toole says over & over again, is that the ear + brain is way more capable thean a mic + analyzer.

dave

Dr. Toole is an engineer and far more critical than the DIY guy can ever afford to be.  He's correct that anechoic measurements ARE better.  You just can't get that resolution in a room.  However, if you look at how closely in room measurements (if done correctly) correlate with anechoic, the difference isn't worth worrying about--at least to me.  Fractions of a decibel aren't my concern.

The ear and brain may well be more sensitive, but that sure doesn't mean they'll be better for designing a better speaker.  I suspect Harman or any speaker company wouldn't buy all that expensive testing and evaluating gear if the ears were a better design tool.

Dan

[DanTheMan]

Back to Toole & Olive (and Geddes too). On axis measures are not that useful, at least not without support from off axis measures.

dave

That we agree on.

[DanTheMan]

Personally I think this is a false dichotomy. It doesn't have to be all measurements or all subjective. It seems foolish to ignore the broad spectrum between those extremes.

Ultimately, for most of us, the only person that needs to be satisfied is ourselves. If you can do that by ear, great. It's not that hard or expensive to get some basic measurement capabilities. I've often tuned something by ear, thought I had it sounding pretty good, then took some measurements, made some simple adjustments, and had to admit the end result was better. And vice versa - sometimes I'll listen to the theoretically better setup, but then go back to something I enjoyed more.

I agree with this as well.  Now Dr, Geddes would disagree, but he's going to be considerably better at measuring than I am.  Listening to what I've measured has certainly helped me realize what to look for in measurements. 

[markC]

There is a lot of talk about measurement, which is very usefull for the average diy'er and even more so for the professional designer. Although, not very practical for the average amatuer, imo. Published data is a great starting point for choosing drivers, etc. But, how many of us are going to spend as much or more on equipment to measure what we hear than the equipment that produces what we hear? It's just not practical.
Mesurements may be the most accurate, but the room, the gear and the ears have the final say for the average diy'er. A good computer program along with trial and error go a long way to achieve affordable high quality performance.

[rjbond3rd]

Dr. Toole is an engineer and far more critical than the DIY guy can ever afford to be.

Dan, plenty of DIY guys happen to be engineers

[lowtech]

But, how many of us are going to spend as much or more on equipment to measure what we hear than the equipment that produces what we hear? It's just not practical.

Assuming you already have a computer, a calibrated mic and preamp will set you back $100.  There is excellent, free measurement software avaliable.

Mesurements may be the most accurate, but the room, the gear and the ears have the final say for the average diy'er.

No amount of room treatment will fix a poorly (but easily corrected) crossover design.  OTOH, a $100 measurement system can help you address room issues once your speaker is performing properly.  Seems like a wise investment to me.

[DanTheMan]

I'm looking for an inexpensive RTA.

Do you have an iPhone?

[DanTheMan]

There is a lot of talk about measurement, which is very usefull for the average diy'er and even more so for the professional designer. Although, not very practical for the average amatuer, imo. Published data is a great starting point for choosing drivers, etc. But, how many of us are going to spend as much or more on equipment to measure what we hear than the equipment that produces what we hear? It's just not practical.
Mesurements may be the most accurate, but the room, the gear and the ears have the final say for the average diy'er. A good computer program along with trial and error go a long way to achieve affordable high quality performance.

That's one of the reasons why I think guys who can take measurements have the responsibility to post them for the others who want to make a design out of said speaker.  I also think the measurer should also post responsibly--IOW as accurate measurements as possible and state their limitations.  Too bad I can't do that with the B200 graph I posted as it's not mine.  Still it seems to correlate with what users hear b4 tweaking and with what Visaton says about it and its use.

I spent $270 on my measuring rig w/free software.  I only spent that much b/c it does double duty as a home studio.  You could get away with cheaper for sure.  It's the best money I've spent on my stereo as it's improved it more than any other device I own.

Oh, on a side note.  Only use an RTA for measuring/EQing bass.  If it can't do gated measurements, above the transition frequency will not be accurate.

Dan

[rjbond3rd]

Hi Dan, I wish!  I was thinking of this: http://www.trueaudio.com/

[planet10]

Personally I think this is a false dichotomy. It doesn't have to be all measurements or all subjective. It seems foolish to ignore the broad spectrum between those extremes.

Ultimately, for most of us, the only person that needs to be satisfied is ourselves.

Bears repeating

dave

[planet10]

Fractions of a decibel aren't my concern.

It seemed he was much more concerned with the measurements frequency resolution...

dave

[DanTheMan]

Hi Dan, I wish!  I was thinking of this: http://www.trueaudio.com/

As far as I know, that one won't do except for bass.

It seemed he was much more concerned with the measurements frequency resolution...

dave

Was there a reason to repeat what I said?

Dan

[planet10]

Was there a reason to repeat what I said?

I didn't. You were talking about differences in the y axis, i in the x-axis.

dave

[DanTheMan]

I didn't. You were talking about differences in the y axis, i in the y-axis.

dave

Nope.  Reread my post.

Dave, please read my posts before you attempt to refute them.  It just makes having a meaningful discussion much easier.  If you don't understand what I'm saying, just ask.

Dan

[Saurav]

As long as we're being pedantic...

'Fraction of a decibel' is resolution in the amplitude (y) axis. Dave was talking about resolution in the frequency (x) axis, even though he said 'y' in his post. For example: your measurements may be accurate to 0.1dB, but you're only taking one measurement every octave. That's high resolution in the Y axis, low resolution in the X axis.

So maybe you guys are in fact talking about the same thing, but you're not typing what you're thinking

[DanTheMan]

As long as we're being pedantic...

'Fraction of a decibel' is resolution in the amplitude (y) axis. Dave was talking about resolution in the frequency (x) axis, even though he said 'y' in his post. For example: your measurements may be accurate to 0.1dB, but you're only taking one measurement every octave. That's high resolution in the Y axis, low resolution in the X axis.

So maybe you guys are in fact talking about the same thing, but you're not typing what you're thinking

No, not at all. I'll repost my post to make this easier.  I was talking about both axis.  I knew what Dave meant.  First the X(resolution) then the Y(fraction of a db).  Input and result relationship.  Dave chose to only quote part of my post and replied as if that's all I said.

Oh, and what you are saying is why I said it's good that anechoic measurements are less useful for bass.  Or rather why it's good we have to do bass measurements in room otherwise getting useful measurements would be difficult.  You can get good resolution in the treble gated in room, but not in the bass.  It still takes a lot of effort to get clean time domain performance in room so that you can take good measurements.

Here's my previous post: 

Dr. Toole is an engineer and far more critical than the DIY guy can ever afford to be.  He's correct that anechoic measurements ARE better.  You just can't get that resolution in a room.  However, if you look at how closely in room measurements (if done correctly) correlate with anechoic, the difference isn't worth worrying about--at least to me.  Fractions of a decibel aren't my concern.

The ear and brain may well be more sensitive, but that sure doesn't mean they'll be better for designing a better speaker.  I suspect Harman or any speaker company wouldn't buy all that expensive testing and evaluating gear if the ears were a better design tool.

Dan

I could add to that, that I'm not even worried too much about a dB or 2!  Not that it's not audible, just compared to what most of us are used to listening to(when looking at polar data), it is still great.  Designing a speaker is sort of chipping away the stone.  Once the larger problems are gone, you can worry about the small ones.

One other thing I suppose I should add so others reading this thread don't get misled.  Dave's interpretation of Dr. Toole's beliefs are sort of out of place in this thread.  Dr. Toole thinks my measurement protocol is not good enough for designing a speaker, I'm sure he wouldn't advocate designing a speaker by ear!  Let's be rational here, how many of you can draw an accurate, high resolution FR graph?  I can't.  It's important to discuss things in their context.  Otherwise we'll add new elements to the confusion.

Dan