[webspider]

The DT48 seem to be tuned to a free field response in contrast to many other headphones which are tuned to diffuse field response. Because free field sound (like things played live outside) doesn't have much bass or highs, it seems to match to their treble and lows rolling off early. This would also explain why they sound so good with binaural records and less good with normal ones.
 
Anyways, I also consider them being rather mid-centric instead of neutral. This is not necessarily something bad, but makes them sound strange at first for people who are accustomized with neutral cans that have more extension. Some styles of electronic music become almost unlistenable with them like minimal techno or dubstep. Others can be very enjoyable (like Aphex Twin). This is why I bought other IEMs, that are their opposite in terms of sounding and match more to the music I listen on the go

 

[Deep Funk]

Quote:

At this point, I find it really hard to believe that the DT48s have anything but a significantly colored frequency response, both at the dummy head and at the actual ear with iffy treble extension and rolled off bass. That doesn't even look like accurate bass in any of these graphs, they look significantly colored by lack of bass and a heavy emphasis on mids, which would definitely make sense in a field recording situation like with Nagios. I'm not sure if it is necessarily because they are neutral that Nagios likes them, it seems to me like it might be the clarity and speed in the mids, rather then neutrality. 
 
While googling for information about the DT48e, I found your Facebook page but hecking out your DT48 Facebook page didn't help much either, because it all just seemed like you were attacking Inner Fidelity and Head-Fi and playing the victim card, rather than giving out any helpful information about the DT48... I was able to get a lot of good information from the B&H Photo reviews and Dalethorn's site, though, but they seemed to really contradict a lot of the other evidence I had. I was just wondering if you guys had any personal experiences to refute these, because I'm really skeptical about buying a DT48e right now.
 
I hope I don't sound attacking or anything, I'm just somewhat frustrated by the lack of coherency in the information I'm finding.

For the midrange, try it...
 
I used to like the DT48E and after a while found the right amplification settings but I cannot consider it a keeper. For me the lows and highs on my old DT48E lack the linear response and some recordings are hit or miss. I would even prefer the K400 and I almost regret letting go of the CD900 ST. The DT48S is the DT48 that is worth keeping for me yet I should still audition the new DT48E.
 
The Facebook page, well it is K.B.I. and Dalethorn territory. I sometimes agreed with them. 
 
I can pass the old DT48E-25 to you if you want. Given I am not using it.
 
(Time to place some advertisments.)
 
 

 
 

 

[maniacal71]

Playing them out of my Vintage Receiver amp, it is neutral, DT48S, DT48A and DT48E vintage, Newer head amps add colourations and recesses the bass and highs much, nothing amiss from anything old and possibly 120Ω headphone output, tried the same from a mosfet amp with a 120Ω adapter and comes close to the sound from my Sansui R70.  Tyll's test actually imho shows the flaws in his system and not the DT48, My Nagra DT48S can even pick out circuit hum from a Sansa Fuze running on battery power during the start up. The DT48E vintage has the most changes from a very forward mid centric to a neutral on the vintage amp.

 

[SanjiWatsuki]

I find that explanation odd -- from what I've seen from measurements, most modern amps are pretty much uncolored with +- 0.1dB with rare exceptions on the super low end, which still have a pretty minor coloration of like +- 1.5dB, nothing major enough to completely change a sound signature. It's not a difficult task to make a neutral amp from a technical perspective. Do you have anything to back that up?
 
Quote:

Playing them out of my Vintage Receiver amp, it is neutral, DT48S, DT48A and DT48E vintage, Newer head amps add colourations and recesses the bass and highs much, nothing amiss from anything old and possibly 120Ω headphone output, tried the same from a mosfet amp with a 120Ω adapter and comes close to the sound from my Sansui R70.  Tyll's test actually imho shows the flaws in his system and not the DT48, My Nagra DT48S can even pick out circuit hum from a Sansa Fuze running on battery power during the start up. The DT48E vintage has the most changes from a very forward mid centric to a neutral on the vintage amp.

 

 

[webspider]

If his amp has an output impedance of 120Ω, it could indeed bend the frequency graph to a more neutral curve. You can simulate that by plugging an 120Ω adaptor between your amp (with around 0Ω output impedance of course) and the DT48.
 
Using low impedance outputs however makes them sound midrange heavy. Calling the output of a well built amp colored is of course stupid, but if people prefer technically less well built amps because they make their headphone sound better, they should maybe try equalizing or even switching to another headphone.

 

[SanjiWatsuki]

The DT48E has a surprisingly flat impedance curve, though, and Dalethorn has noticed an increase of his bass with his low impedance output sources on Headphiles, possibly due to the electrical damping factor. I don't think it's an impedance matching issue.
Quote:

If his amp has an output impedance of 120Ω, it could indeed bend the frequency graph to a more neutral curve. You can simulate that by plugging an 120Ω adaptor between your amp (with around 0Ω output impedance of course) and the DT48.
 
Using low impedance outputs however makes them sound midrange heavy. Calling the output of a well built amp colored is of course stupid, but if people prefer technically less well built amps because they make their headphone sound better, they should maybe try equalizing or even switching to another headphone.

 

 

[maniacal71]

Quote:

I find that explanation odd -- from what I've seen from measurements, most modern amps are pretty much uncolored with +- 0.1dB with rare exceptions on the super low end, which still have a pretty minor coloration of like +- 1.5dB, nothing major enough to completely change a sound signature. It's not a difficult task to make a neutral amp from a technical perspective. Do you have anything to back that up?
 
 

 

 
I own them 3 different DT48 don't I? The DT48E vintage does change the most when used with an impedance adapter, or through the vintage amp. Also seems more transparent and not so forward mid centric, Just like the Nagra and the DT48A. The colorations are in the FR if the DT48E, caused by the impedance of the headphone output, tried them on different amps and even direct from the dap, but with the vintage amp, all my vintage headphones including my Wharfedale DD1 and YH1 even sounds their best. Don't believe the measurements, buy the DT48 and do your own personal tests, if you don't like them sell them and get the HD250 Linear II.
I'm trying to help you guys, with what my experiences have been with them. The Nagra DT48S and DT48A don't change much but they benefit at the highs and lows.
 
To be rid of all the troubles, get a DT48A and run a mosfet amp. To see how an impedance adapter might work, check out Jan Meiers explanation on impedance optimization here, http://www.meier-audio.homepage.t-online.de/tipstricks.htm
 
 

 

[shamu144]

You raised some very valid concerns, and I am afraid noone will really be able to give you definitive answers... I think the DT48 is a very tricky headphones when it comes to actual measurments vs subjective listening impressions. 
 
The low end roll off observed from this russian site and from this Japanese paper (actual ear response) is obviously driven by a very poor seal (see the standard deviation for actual ear response shown at the bottom between 5 and 10 dB) . I do not think I have seen any comments or impressions that support this massive roll off below 300 Hz in this thread.
 
The coupler response from this Japanese paper matches Tyll measurements and might very well represent the true behaviour of the headphone, under perfect seal conditions. This will not be the case at all when you wear it. So your actual ear response should be somewhere between those 2 curves.
 
Have a look below at my subjective comparison (loudness intensity matched) of the DT48A (reference 0 at all frequencies) with the Sennheiser HD250 Linear II (one of the best low end response I have heard to date by far). Yes, the DT48A low end is rolled off, but it only starts at 70Hz.
 

 
Quote:

Hey DT48 fans, I have question. I've been thinking about buying high-end headphones and the DT48E is one of those on my list of cans to check out, but I've been really bothered by the correlations between the published frequency response curves. I see the DT48A's Beyer hand-drawn frequency response picture tossed around a lot, but it doesn't jive with the the reviews I see or these graphs.
 
 

 
It seems like the actual-ear response from these measurements are very similar to the above Russian measurements with the significant drop-off of bass at 300hz.  Furthermore, the dummy-head response looks almost exactly like the Inner Fidelity Tyll graphs that you all seem to blame on coming from an improper source, like a Macbook Air. With these correlations, though, I just can't imagine that a source like a Macbook would color the response so much from neutrality. 
 

 
At this point, I find it really hard to believe that the DT48s have anything but a significantly colored frequency response, both at the dummy head and at the actual ear with iffy treble extension and rolled off bass. That doesn't even look like accurate bass in any of these graphs, they look significantly colored by lack of bass and a heavy emphasis on mids, which would definitely make sense in a field recording situation like with Nagios. I'm not sure if it is necessarily because they are neutral that Nagios likes them, it seems to me like it might be the clarity and speed in the mids, rather then neutrality. 
 
While googling for information about the DT48e, I found your Facebook page but hecking out your DT48 Facebook page didn't help much either, because it all just seemed like you were attacking Inner Fidelity and Head-Fi and playing the victim card, rather than giving out any helpful information about the DT48... I was able to get a lot of good information from the B&H Photo reviews and Dalethorn's site, though, but they seemed to really contradict a lot of the other evidence I had. I was just wondering if you guys had any personal experiences to refute these, because I'm really skeptical about buying a DT48e right now.
 
I hope I don't sound attacking or anything, I'm just somewhat frustrated by the lack of coherency in the information I'm finding.

 

 

[shamu144]

I also believe the DT48A is tuned to a FF response, however, this does not imply a roll off in the frequency response of the headphone neither in lows or highs as evidenced below:
 
A Free Field frequency response design goal (open ear canal for full size headphones, bottom/right chart):
 


Quote:

The DT48 seem to be tuned to a free field response in contrast to many other headphones which are tuned to diffuse field response. Because free field sound (like things played live outside) doesn't have much bass or highs, it seems to match to their treble and lows rolling off early. This would also explain why they sound so good with binaural records and less good with normal ones.

 

 

[shamu144]

I just believe folks are obsessed those (digital) days with frequency response when there are other fundamental aspect to the music reproduction (why phase response get so little coverage in places like this, this is beyond my understanding).... Frequency response is conditionned by so many variables and our psychoacoustic mental FR maps are all messed up ! Don't expect to find the perfect linear headphone, because such a thing is only an utopy IMO.

 

[shamu144]

Agree. Not a night and daty difference though, and the resistor will introduce some distorsion as well if not of high quality. But definitely worth a try before ditching the DT48E. Or just go for the A model, which doesn't suffer this.
 
Quote:

If his amp has an output impedance of 120Ω, it could indeed bend the frequency graph to a more neutral curve. You can simulate that by plugging an 120Ω adaptor between your amp (with around 0Ω output impedance of course) and the DT48.
 
Using low impedance outputs however makes them sound midrange heavy.

 

 

[webspider]

Uhmm, yeah, the FF graph doesn't go below 200 Hz.

 

[shamu144]

 
You are correct. But I haven't seen any mention of this necesarry roll off in the low end for a FF as compared to a DF. Do you have any background information you could point me towards ?
 
Quote:

Uhmm, yeah, the FF graph doesn't go below 200 Hz.

 

 

[webspider]

Obviously my argumentation was about that when headphone graphs after a certain part goes wild (upper highs especially), you can't really use that part for prediction of a headphones treble sound. If I suspect a bass roll off, a not so far going frequency graph will neither prove or make my assumption incorrect. So yes, if these two points make the top and low end unpredictable, it remains uncertain whether the roll off is really due to strictly going for free field response or just because the driver itself can't go that far.
 
I tend to believe the latter. The amount of detail compared to other closed headphones is for me just a result of less tuning.

 

[maniacal71]

Sanjiwatsuki, please read the Latest review about the DT48 Audiometric version http://headfonics.com/2011/11/beyerdynamic-dt-48a-the-audiometric-headphone/ hope this does not break any rules here. There is a more active DT48 thread over there, you can find your answers too. You can actually still purchase this item new from B&H, all you will need is a pair of new E model earpads as the Audiometric version comes with semi hard rubbers which are uncomfortable for long periods of use. http://www.bhphotovideo.com/c/product/313912-REG/Beyerdynamic_101060_DT48A_00_Audiometric_Headphones.html
 
The DT48s that I have actually play very well from and Ipod or a Macbook, On the Sansa when played at loud volumes starts to jitter due to current demands.