[MalVeauX]

Quote:

How is a graph not reliable? I'd say a review is much less reliable than a graph. A graph is as objective as possible (if it's accurate, of course), it doesn't tell you if it's good or bad, it doesn't give you an opinion, it gives you values. You can say it isn't really as important as a review, since that one relates more to what the buyer wants (how does it actually sound), but that doesn't make it more reliable. Human opinion changes, and therefore is unreliable =) 

Heya,
 
I appreciate what you're expressing, however graphs do not take into account the enclosure of the headphone. There's a lot more to what is actually heard than what is measured by a set of microphones. Some headphones look like they should be bassy according to a graph, yet are not. Others don't even look that bassy, yet are out of this world bassy. Comparing on a graph, seeing a few dB difference in the response, shouldn't result in out of this world differences, when you can do that on an equalizer and not get that same difference with the same jump in dB on a similar frequency range.

Relating a graph to a headphone that you've heard however is quite useful for discovering whether or not the graph is at all close to what is perceived to form a better relationship of information for the next person.
 
Very best,
 

 

[oldshoe99]

How is a graph not reliable? I'd say a review is much less reliable than a graph. A graph is as objective as possible (if it's accurate, of course), it doesn't tell you if it's good or bad, it doesn't give you an opinion, it gives you values. You can say it isn't really as important as a review, since that one relates more to what the buyer wants (how does it actually sound), but that doesn't make it more reliable. Human opinion changes, and therefore is unreliable =) 

This is a classic question and I would add to Malveaux's comments thus: If the graphs and other test results covered everything, they would be reliable. OTOH, if the reviews in words described everything perfectly, they would be reliable too. So it's good to read as much as you can, and try not to get stuck in any one train of thought.

 

[kn19h7]

why don't just put them on and listen?

 

[wind016]

Quote:

Heya,
 
I appreciate what you're expressing, however graphs do not take into account the enclosure of the headphone. There's a lot more to what is actually heard than what is measured by a set of microphones. Some headphones look like they should be bassy according to a graph, yet are not. Others don't even look that bassy, yet are out of this world bassy. Comparing on a graph, seeing a few dB difference in the response, shouldn't result in out of this world differences, when you can do that on an equalizer and not get that same difference with the same jump in dB on a similar frequency range.

Relating a graph to a headphone that you've heard however is quite useful for discovering whether or not the graph is at all close to what is perceived to form a better relationship of information for the next person.
 
Very best,
 

Exactly. The graphs for open, closed, or semi-closed headphones always seem to be different from my experience. Nothing is ever 100% consistent from what I perceive but you get an idea of where frequency peaks are which is pretty important if you're treble adverse or something like that. The most important to me personally are treble peaks and distortion. I don't like either.

 

[LizardKing1]

Quote:

why don't just put them on and listen?

Because I, and probably the OP as well, don't have warehouses filled with demo headphones. In the very best of chances, I can try on a Sennheiser HD595. I have not even ever seen an Ultrasone, and a Grado is just a mythological object around here.
 
 
Quote:

Heya,
 
I appreciate what you're expressing, however graphs do not take into account the enclosure of the headphone. There's a lot more to what is actually heard than what is measured by a set of microphones. Some headphones look like they should be bassy according to a graph, yet are not. Others don't even look that bassy, yet are out of this world bassy. Comparing on a graph, seeing a few dB difference in the response, shouldn't result in out of this world differences, when you can do that on an equalizer and not get that same difference with the same jump in dB on a similar frequency range.

Relating a graph to a headphone that you've heard however is quite useful for discovering whether or not the graph is at all close to what is perceived to form a better relationship of information for the next person.
 
Very best,
 

Right, I hadn't thought of that. Good point. As far as I know, HeadRoom uses ear-shaped enclosures for their microphones to make sure it relates to what we hear, and they even use a program to equalize the resulting graph to what is perceived by human inner ears. I'm not sure about Inner Fidelity. Anyway, my point still comes across: a graph has objective measured values, while a review has subjective points. One isn't better than the other, but I like including both in my decision-making process. Also, even if you have, let's say, a flawed microphone that increases consistently the 7kHz frequency in a range of 100Hz by 3dB, every graph that microphone produces will have that exact same flaw. So while that wouldn't be accurate, like you mentioned, I would still consider it reliable. Maybe relating a graph to a hearing experience isn't all that good, but if you're comparing graphs of different headphones it's a great method.

 

[oldshoe99]

Freq response graphs have helped me, so no objection there. But those square waves and stuff I dunno. Technically they're informative, but given Tyll's comments about them and contrasting that with his views on sound quality, I don't find a useful correlation.

 

[Acix]

The graphs show true info, but is only a half of the story.

 

[MalVeauX]

Heya,
 
Graphs don't even tell half the story. They can in fact be completely misleading. The recording method and all the titration they do doesn't get all headphones represented equally. Some graphs literally sound like they look (I think the AD700 is a good example of that). But others are so completely wrong that you have to wonder how they bothered to keep that graph. Not to completely derail into a "frequency response graph thread", but look at this:
 
SRH940 vs K701 vs DT770
 

 
Yet we know they sound really, really different on the low end and high end. Right? Look at that!
 
Graphs are pretty hit or miss.
 
Very best,

 

[Acix]

As I said... the graphs are not wrong, it show you basic measurements in decibels, what they don't show you is the sound stage, depth, transparency/clarity/definition speed and dynamic. 

 

 

 

[oldshoe99]

As I said... the graphs are not wrong, it show you basic measurements in decibels, what they don't show you is the sound stage, depth, transparency/clarity/definition speed and dynamic. 

 

 

That's a lot of 'don'ts' Acix.

 

[Acix]

Quote:

That's a lot of 'don'ts' Acix.

 
lol, in away the don'ts balanced out...on the graph side you have more than 22khz.

 

[LizardKing1]

Quote:

As I said... the graphs are not wrong, it show you basic measurements in decibels, what they don't show you is the sound stage, depth, transparency/clarity/definition speed and dynamic. 

Actually, they should show you something very important which is the sound signature. However, looking at MalVeauX's graph, I can't believe the AKG K701 even compares to the DT770 in the low end.

 

[wind016]

 
Quote:

As I said... the graphs are not wrong, it show you basic measurements in decibels, what they don't show you is the sound stage, depth, transparency/clarity/definition speed and dynamic. 

 

 

 
What about impulse response? Distortion? Square wave? CSD waterfall?
 
Why are we only focusing on the FR graphs?

 

[NamelessPFG]

In sound quality terms, the only way to know for sure whether a headphone is for you is to audition it yourself, under the conditions you would use it in. Too bad that most of us don't have the opportunity without paying up first, especially for more exotic equipment such as Stax, so we have to take a gamble and hope we like it when it arrives on our doorstep.
 
But there's more to a headphone than sound quality. As MalVeauX pointed out, build quality is another consideration. You want your headphones to remain in one piece, right? You'd probably also like them to not make creaking noises as they flex around a bit on your head during sudden movements, or have a "microphonic" cord (a cord that, if bumped, will transmit the vibrations right into the headphone enclosure and your ears).
 
Then there's comfort. What point is there in having the best-sounding headphones in the world if you can't stand wearing them for more than a few minutes? Ideally, they should all but disappear on your head, but still sit tight enough that rapid head movements won't cause them to fall off-a tricky balance to accomplish, especially since everyone has a different head size.

 

[MalVeauX]

Heya,
 
Some day, we will have the technology to generate a singularity right in our ear, and vibrate it with a laser to generate magical fantastical soundscapes. They will mount on our shoulders, or possibly just hover around us like floating crowns that are on fire.
 
I cannot wait.
 
Very best,