[Skylab]

Quote:

Originally Posted by Snacks /img/forum/go_quote.gif I gave up trying to understand graphs a long time ago... Quick someone give a quick Graph 100 course.

Frequency response charts like the ones above have the range of human hearing, usually from 20 Hz to 20kHz (Hz = Hertz, or cycles per second), on the horizontal axis, and volume level, in dB or decibels, on the vertical axis. Do as you go from left to right on the horizontal axis, the sound goes from the very low bass to the very high treble, and on the vertical axis the sound is louder at the top and quieter at the bottom.

Then the "line" is the frequency response of the headphone, and what you see is how, at a given and constant signal input level, the headphone will produce different sound pressure levels at different frequencies. This is the degree to which the headphone deviates from flat frequency response.

A couple of notes:

Measuring headphones is both hard, and controversial, so the charts/graphs you see for headphone frequency response are at best rough estimates, and are based on certain measurement assumptions inherent in the given measurement techniques.

By way of comparison, 99.99% of the CD players made will provide 100% ruler flat frequency response, deviating often by less than .01db at the MOST, and yet most of us believe that CD players sound different from each other, so certainly not all differences in sound quality can be explained by frequency response anomalies anyway.

 

[b0dhi]

Great website pkshan. I wish I could read Japanese. I don't suppose you've done any measurements on the TakeT H2?

 

[rsaavedra]

Quote:

Originally Posted by D_4_Dog /img/forum/go_quote.gif HD800 on the other hand.... that 5 KHz peak looks painful but then again I wonder how something like the 325i would look

It does look like the 800 would sound on the brighter side of neutral, especially reading the first graphs in this tread.

Quote:

Originally Posted by pkshan /img/forum/go_quote.gif hd650 vs hd800

That graph above shows a deviation of +7 dB between the peak around 5-6 kHz and the bass at 100 Hz. That would correspond to a bright headphone (graph-wise at least), same kind of graph that for example the Beyer DT880 would have.


The Headroom graph for the 800, however, doesn't show anything near that much difference between the bass and the 5-6Khz response of the 800, and would not correspond to a bright headphone, but to a rather neutral one. (This graph difference is probably due to the smoothing out that Headroom applies to their curves)

 

[wnmnkh]

No, it is not that headroom smoothed out the graph, they compensated the graph for human's actually preceiving.

Here are the raw graphs for both HD800 and HD650....

 

[Graphicism]

Quote:

Originally Posted by rsaavedra /img/forum/go_quote.gif (Maybe this similarity has been mentioned before in the forums, but I'm just realizing it now)

I think this proves graphs means absolutely nothing, because the SR60 sounds NOTHING AT ALL like the D5000s.

 

[donthuang]

That "waterfall" picture maybe more interesting than the FR picture.

This is is the first time i seem this in a headphone measure.

According the picture,seems HD800 do better resonance control than HD650 in mid

So that "the Sennheiser veil" is no more problem to HD800?

 

[wnmnkh]

That's for decay, less waterfall better the decay is. As you see, HD800 has less waterfall than HD650.

 

[Snacks]

Quote:

Originally Posted by Skylab /img/forum/go_quote.gif Frequency response charts like the ones above have the range of human hearing, usually from 20 Hz to 20kHz (Hz = Hertz, or cycles per second), on the horizontal axis, and volume level, in dB or decibels. By way of comparison, 99.99% of the CD players made will provide 100% ruler flat frequency response, deviating often by less than .01db at the MOST, and yet most of us believe that CD players sound different from each other, so certainly not all differences in sound quality can be explained by frequency response anomalies anyway.

Thanks Skylab...now where is that bookmark button?!

 

[b0dhi]

Quote:

Originally Posted by wnmnkh /img/forum/go_quote.gif No, it is not that headroom smoothed out the graph, they compensated the graph for human's actually preceiving. Here are the raw graphs for both HD800 and HD650.... ...

Very interesting. I didn't know HeadRoom applied perceptual correction to their graphs. How did you get these raw graphs?

 

[ABathingApe]

Quote:

Originally Posted by Graphicism /img/forum/go_quote.gif I think this proves graphs means absolutely nothing, because the SR60 sounds NOTHING AT ALL like the D5000s.

These are estimates. And of course as with any sort of test, these only present how the circuits, more specifically the drivers, respond to inputs of a specific frequency, and not how a human would perceive them. That only shows you that drivers are half the battle in terms of creating sound. Just like a concert hall, without good acoustics, even 50 cent can sound like crap...oh wait...

 

[RedBull]

Quote:

Originally Posted by pkshan /img/forum/go_quote.gif decay timing hd650 hd800

Cool graph pkshan! thanks, this is new information for me.

From the 2 above, does it mean on the bass region HD800 is a little bit faster and at the 5KHz HD800 a lot slower than HD650. This is surprising, because I hear exactly the same from my ear, vocal decay is quite long, maybe it's one of the reason HD800 seems to have much wider soundstage. Am I correct?

 

[rsaavedra]

The cummulative spectral decay of the HD800 at 1 Khz is very clean.

Yet at 5KHz the HD800 shows a pretty bad decay, in fact worse than the HD650's as mentioned. Anyone feeling anything weird in the upper midrange / lower treble with the HD800? (This is probably hard to answer I know, because it's not related to frequency response or brightness, decay is a different feature)

 

[hanius]

Thank You ^

 

[donthuang]

Thanks fuchinove san, he seems watch this thread too, and update the waterfall plot picture of AKG K701.

That picture very interesting to me,and somehow confused.i even thought that K701 is a "fast" and "clear" headphone.

You can see this new picture in his site. ttp://fuchinove.chillout.jp/

He also answer why his FR picture different to Headroom's picture, his FR picture is the same as headroom's "raw" data, but headroom's
use some free-field or diffus-field EQ to correct . Another reason that is His dummy head and equipment different to Headroom's.
(He use his head as "dummy head" ? )

 

[RedBull]

Quote:

Originally Posted by rsaavedra /img/forum/go_quote.gif The cummulative spectral decay of the HD800 at 1 Khz is very clean. Yet at 5KHz the HD800 shows a pretty bad decay, in fact worse than the HD650's as mentioned. Anyone feeling anything weird in the upper midrange / lower treble with the HD800? (This is probably hard to answer I know, because it's not related to frequency response or brightness, decay is a different feature)

Yes, I feel it the very first time I auditioned it @ Sennheiser, the vocal seems take longer to 'stop' then what I expected. I keep mentioning this on every occasion in head-fi but seems not everybody hear it, only one agree with me. I refer this as 'hollow' sound in another post. Now that I see this waterfall graph for the first time, now I know why .... technically.

But man, HD650's 1K is really dissapointing. It look like a faulty headphone