Do You Look at the Measurements?

[ack]

They must love you at TAS and the lot. And while perhaps not totally untrue (thanks to the marketing teams), wouldn't you put your best foot forward?

The love is reciprocal In the end, if measurements are used for _part_ of the development, it's reasonable for me to want to rely on some measurements post-production. Clearly, they don't paint the whole picture in either case, but equally clearly the "right" ones are not useless either. TAS, contrary to what they preach, as you know, have in-frequently published frequency response graphs; well, I am not surprised...

I am at ease interpreting frequency response graphs for overall tonal balance; square wave response for speed; channel voltage measurements for cartridge azimuth, after channel balance calibration with my Spectral preamp (which is brilliant in letting you do so in small 0.05dB steps); setting turntable speed by measuring a test 1kHz frequency using a very accurate Fluke device (after asserting its accuracy at that frequency with a test CD); and I wish I could do more, like measuring IMD distortion when setting cartridge SRA.

From my perspective, there is always the scientific and artisan side of things... I like and embrace both.

 

[Phelonious Ponk]

I look. I don't always understand, but I look.

Tim

 

[Soundproof]

There's too much cheating and obfuscation with stated measurements, can't trust them. Use your ears, or perform your own.

 

[treitz3]

There's too much cheating and obfuscation with stated measurements, can't trust them.

Hello, Soundproof. That is exactly why I pay them no mind. I trust my ears.

 

[Robh3606]

There's too much cheating and obfuscation with stated measurements, can't trust them. Use your ears, or perform your own.

How about when your own validate the published measurements?? Then obviously they were not cheating.

Rob

 

[amirm]

I hadn't posted in so long that the forum automatically sent me a message inviting me back. So I reset my password and decided to also post this question.

As some of you might now, I'm publisher of the SoundStage! Network. I've seen various messages about measurements here, but I have kind of a different question. Regardess of whether your understand or use the measurements, do you actually LOOK at the measurements? I'm not talking about a glance. I'm talking a meaningful look to see what's presented -- again, whether you use them or not, or fully understand them or not is not part of this.

I ask because we have our own internal metrics about how many times the measurements versus the reviews are looked at, but they don't obviously break things down into exact who it is looking. So that's why I'm asking explicitly.

Thanks,
Doug Schneider @ SoundStageNetwork.com

Welcome back Doug.

I look at measurements like a detective looks at clues. They tell me if the person measuring them knows what they are talking about. They tell me whether I know what I am talking about . They tell me what a product is doing wrong. And very often, what it is doing well. They tell me how good our test instruments are. They tell me how technology translates into solutions.

Alas, the industry does such a poor job of explaining what the measurements mean. They talk 1000X more about what is measured, than what the measurements mean....

 

[audioguy]

Eons ago I would look at such things, but the measurements, because they don't actually measure what's important to people listening with a critical ear, have virtually zero relevance.

Frank

ZERO? So if, for example, if an amplifier has an a frequency response of +/- 2 db, that means ZERO. Or a speakers impulse response shows massive ringing, that means ZERO?

Surely you jest.

 

[Soundproof]

How about when your own validate the published measurements?? Then obviously they were not cheating.

Rob

I have learned to be skeptical of stated measurements, and today it's easy to validate through the qualified experiences of others (using the net, for instance),or by performing individual listening and measurements, if necessary.
As audiophiles show incredible capacity for selective measurement acceptance, one might wonder why bother at all.

 

[fas42]

ZERO? So if, for example, if an amplifier has an a frequency response of +/- 2 db, that means ZERO. Or a speakers impulse response shows massive ringing, that means ZERO?

Surely you jest.

I'm referring to the measurements that are published by the manufacturers, which most here would agree are of extremely little value. As regards measurements in magazines, these can vary quite dramatically from reviewer to reviewer, depending upon precisely how the readings were done.

As regards the amplifier frequency response, this matters little if the speaker it's driving is swinging 2 or 3 times as much. For measurements, how about some IM distortions of low level signals in one channel while the other channel is heavily stressed driving a real load. THAT would tell you an enormous amount about its real world performance, probably will never be done ...

Frank

 

[Phelonious Ponk]

There's too much cheating and obfuscation with stated measurements, can't trust them. Use your ears, or perform your own.

I've recently learned here that specifications are not measurements, accurate, comprehensive or otherwise. My teacher didn't even seem to think they were derived from measurements. I confess, I still look at them, but as always I look at them skeptically. Fortunately everything I'm personally interested in at the moment comes from the pro audio world, where the better publications take lots of measurements and a couple of them keep deep archives of reviews online. Makes it all a bit easier to dig deep into the data I don't quite understand.

Tim

 

[Vincent Kars]

I do think measurements are very helpful.
When looking for e.g. a headphone I look at the frequency response

This measurement tells a lot.
For me it is useless to even try the Monster as it has a Monstrous bass.
But if you are a bass head this is the one to look for.
Frequency response won’t tell you the whole story but it tells you what (not) to try.

 

[microstrip]

I do think measurements are very helpful.
When looking for e.g. a headphone I look at the frequency response

This measurement tells a lot.
For me it is useless to even try the Monster as it has a Monstrous bass.
But if you are a bass head this is the one to look for.
Frequency response won’t tell you the whole story but it tells you what (not) to try.

Vincent,

You are using the headphone case, that is not a typical case IMHO.

I would not be so sure. Recently I posted the on axis frequency response of an excellent JBL speaker and it did not look nice.. Yes, Tim, it was the JBL Synthesis 1400 Array BG loudspeaker that I posted . .

I think that measurement analysis is sometimes strongly affected by the known and perceived quality of the equipment under test. Shouldn't it also be done under strict blind conditions?

It would be an interesting exercise supplying half a dozen of our members with the available measurements and specializations of known equipment and asking them to comment blind on the expected sound quality.

 

[Vincent Kars]

I've recently learned here that specifications are not measurements, accurate, comprehensive or otherwise. Tim

Depends on how you use them.
If a headphone is specked having a 600 Ohm impedance then this is not accurate because the impedance fluctuates a little with the frequency.
If you want to know if your portable is able to drive this headphone (no) this spec is helpful and accurate enough

 

[rockitman]

I'm curious which measurements "you" feel have the most relevance

In my case it is component dependent.

speaker's: frequency response, sensitivity, load impedance
power amps: power output, input impedance
pre amps: gain, output impedance
phono/tape stages: S/N ratio, gain


I pay little if any attention to THD figures.

 

[Robh3606]

I would not be so sure. Recently I posted the on axis frequency response of an excellent JBL speaker and it did not look nice.. Yes, Tim, it was the JBL Synthesis 1400 Array BG loudspeaker that I posted .

I will post a disclaimer I built a pair of them. This is actually a good example of why it is important to know what the measurement conditions are. Did you read why there was a rise in the bass response? Do you understand why the last octave is raised relative to the previous ones?? Did you read John's comments about the in-room curve and look at it?? If you look at the in-room curve they look darn good.

I based my decission to build a pair looking at JBL's anecholic measurements. I saw Stereophiles and was thrilled that they got a review. When I get a minute I will post all three measurements.

Robh