Beyerdynamic DT 880 600 Ohm Review (Headphone)

[phoenixsong]

What is the point of using EQ with this headphone if it leads to tremendous distortion? Is the thought of listening without EQ blasphemous?

Some like myself *tend to listen at lower volume levels, hence this distortion problem wouldn't be as severe as it might be for others. Also, it boils down to preference and the sensitivity of one's hearing; a literal tradeoff between FR and distortion. EQ provides listeners with another option, so to speak

 

[amirm]

Score no EQ: 68.6
Score Armirm: 84.7 (because Armirm did not fully compensate the LF, the error regression is flatter hence the higher score)
Score with EQ: 82.3

What? I did better by eye than a computer??? Hooray!!!

 

[amirm]

What is the point of using EQ with this headphone if it leads to tremendous distortion? Is the thought of listening without EQ blasphemous?

The thought of NOT using equalization to get a more neutral response is blasphemous. Importantly, you want to listen as THD measurements are not psychoacoustically aware. 95% of the time I am able to EQ and get better results even if distortion is there. This case was not one of them.

 

[maty]

94 dBSPL is very high to the ears.

DT 880 distortion at 94 dBSPL < 100 Hz is not a problem. Much less problem at 84 dBSPL.

 

[Sharur]

The thought of NOT using equalization to get a more neutral response is blasphemous. Importantly, you want to listen as THD measurements are not psychoacoustically aware. 95% of the time I am able to EQ and get better results even if distortion is there. This case was not one of them.

Just trying to understand how this works. If you EQ down the treble spikes do you not lose the spatial qualities you referred to earlier?

 

[amirm]

Just trying to understand how this works. If you EQ down the treble spikes do you not lose the spatial qualities you referred to earlier?

Not as much in that frequency range. 1 to 5 kHz is where a lot of spatial clues are in music.

 

[Sharur]

Not as much in that frequency range. 1 to 5 kHz is where a lot of spatial clues are in music.

As they are often compared, which do you think performs better without EQ between the DT880 600 Ω and HD800S?

 

[Maiky76]

What? I did better by eye than a computer??? Hooray!!!

Well, if you could blind A/B compare and successfully prefer your EQ then you could say that...

 

[amirm]

As they are often compared, which do you think performs better without EQ between the DT880 600 Ω and HD800S?

HD800S. Not because of tonality but because of the very nice spatial qualities of the 800S where it almost separates instruments into layers.

 

[fliflipoune]

Chances are at that point you are clipping the amp already. With an amp with plenty of headroom (there aren't many) you can go quite loud without discomfort.
The only way to know this for sure is to try it on a beefier amp.

According to the amirm's review of the FiiO Q1 MK2, there is no clipping on this amp. So it should just run out of volume before that, no? Or the high impedance and low sensitivity would change things?

But I have the FiiO Q1 MK1, not the MK2, so I don't know if my model is also not clipping in the same way. It's also more powerful than the MK2 (190mW for 32ohms), so 10mW in 600ohms (Or even more? If I calculate it with the rating at 150ohms (75mW), it would be 18mW?). At 10mW, I should have Max 105dB SPL I think. I'm new at this so I might be wrong!

Yesterday I decided to check exactly at what volume I'm listening on this amp. I'm confortable at listening at position 3 out of 9 (high gain) on the volume knob. This is with -5.4dB gain for EQ. I would be surprised if the amp is clipping there, things sound just fine like on my bigger desktop amp (but I could be bad at listening for those defects), and I can get those headphones to extreme levels (for me) of sound if I go higher. Just for fun, I asked my girlfriend if she felt it was loud. She said «louder than casual listening but a little less than what I would listen my favorite song to». I also have a pair of NAD HP50, much easier to drive, and I don't think I'm listening to a higher volume level on them (position 2 on the volume knob at low gain for my preferred volume level), even if they definitely won't clip in this amp, but I cannot measure it precisely.

Maybe those headphones are still OK for people who prefer to listen at a lower volume level, even with a portable amp. It's possible that the fact that I don't like very high sound levels is not related to clipping or high distorsion, but it would be interesting to measure precisely if I'm not listening to my DT880 at a lower level than my NAD HP50 without realizing it.

 

[solderdude]

As long as it sounds fine to you it is fine.

Q1mk1 =18mW@600 ohm and is about 105dB peak which is loud enough for normal tp somewhat listening levels.
Just can't do impressively loud with tactile bass.

 

[Jwoooosh]

What is the point of using EQ with this headphone if it leads to tremendous distortion? Is the thought of listening without EQ blasphemous?

A better question would be whether or not lower distortion even matters when you can barely hear the notes in question. And if you end up increasing the volume to compensate for the lack of low end response, then you will end up with the same level of distortion anyway. Especially in the treble there are significant gains to be had through EQ and none of them would add anything in terms of distortion at a given volume level.

Being the owner of this specific DT 880 as well as someone who doesn't listen very loudly, I can say that I never had any issues with excessive low end distortion even on very bass heavy tracks and with that area of the frequency response boosted several dB, so YMMV.

 

[quantum_wave]

It seems the last hope for Beyerdynamic in terms of measurements is the DT1990. As far as I can tell, Amir hasn't reviewed the 1990's yet, am I right?

And the Drop Beyer 177x Go. Which looks to be a 1770 with a whole lot of padding in front of the drivers.

 

[GaryH]

Score no EQ: 68.6
Score Armirm: 84.7 (because Armirm did not fully compensate the LF, the error regression is flatter hence the higher score)
Score with EQ: 82.3

Score with Oratory's EQ: 106

 

[Maiky76]

Score with Oratory's EQ: 106

Trivial: they start with a score of 88 compared to 68.
Scanning their data, I get a score of 88.8 so very close.
Optimizing my EQ on their data yields 109.3 with two biquad less...
I win?

The trough at 10k is 20dB+ and the peak at 8k is higher too on the measurements performed by @amirm with the no averaging/smoothing.
I don't know what they are doing.

Oratory Data  DT880 APO EQ Flat@HF 96000Hz
July062021-171529

Preamp: -8.2 dB

Filter 1: ON LS Fc 55.45 Hz Gain 9.87 dB Q 0.42
Filter 2: ON PK Fc 196.53 Hz Gain -3.00 dB Q 1.16
Filter 3: ON PK Fc 669.12 Hz Gain 0.85 dB Q 2.19
Filter 4: ON PK Fc 1614.97 Hz Gain 2.77 dB Q 1.03
Filter 5: ON PK Fc 4926.67 Hz Gain 6.20 dB Q 2.01
Filter 6: ON PK Fc 5714.59 Hz Gain -8.69 dB Q 4.00
Filter 7: ON PK Fc 8333.51 Hz Gain -4.54 dB Q 4.97
Filter 8: ON PK Fc 9074.09 Hz Gain 4.85 dB Q 3.51

 

[GaryH]

Trivial: they start with a score of 88 compared to 68.
Scanning their data, I get a score of 88.8 so very close.
Optimizing my EQ on their data yields 109.3 with two biquad less...
I win?

The trough at 10k is 20dB+ and the peak at 8k is higher too on the measurements performed by @amirm with the no averaging/smoothing.
I don't know what they are doing.

View attachment 139424

Oratory Data  DT880 APO EQ Flat@HF 96000Hz
July062021-171529

Preamp: -8.2 dB

Filter 1: ON LS Fc 55.45 Hz Gain 9.87 dB Q 0.42
Filter 2: ON PK Fc 196.53 Hz Gain -3.00 dB Q 1.16
Filter 3: ON PK Fc 669.12 Hz Gain 0.85 dB Q 2.19
Filter 4: ON PK Fc 1614.97 Hz Gain 2.77 dB Q 1.03
Filter 5: ON PK Fc 4926.67 Hz Gain 6.20 dB Q 2.01
Filter 6: ON PK Fc 5714.59 Hz Gain -8.69 dB Q 4.00
Filter 7: ON PK Fc 8333.51 Hz Gain -4.54 dB Q 4.97
Filter 8: ON PK Fc 9074.09 Hz Gain 4.85 dB Q 3.51

View attachment 139426

Someone asked Oratory about his score for the DT880 increasing to 88 (previously it was 67 on his pdf) and this was his answer:
https://www.reddit.com/r/oratory1990/comments/o1wqi6/_/h3nr88b So unit variation seems pretty high with this headphone. A note on your EQ profile: you shouldn't be EQing up the 10 kHz notch. As Oratory explains in his FAQ:

Why do you never use EQ to remove that drop at around 9-10 kHz on over-ear headphones?

That drop is caused by the shape of the pinna, it depends strongly on how your exact ear is shaped. It’s also very important for localization. On well designed headphones this drop is always present - and it definitely is present when listening to regular loudspeakers (because it’s created by your ears). This means that when a headphone exhibits a peak in that area (it’s often enough just to not exhibit a drop) is very often perceived as „hissy“, „sharp“ or „zingy“. Remember the Sony Z1R controversy?

If you corrected this 10 kHz notch I suspect you'd get the same score of ~106 for your EQ profile as Oratory.

 

[Maiky76]

Someone asked Oratory about his score for the DT880 increasing to 88 (previously it was 67 on his pdf) and this was his answer:
https://www.reddit.com/r/oratory1990/comments/o1wqi6/_/h3nr88b So unit variation seems pretty high with this headphone. A note on your EQ profile: you shouldn't be EQing up the 10 kHz notch. As Oratory explains in his FAQ:


If you corrected this 10 kHz notch I suspect you'd get the same score of ~106 for your EQ profile as Oratory.

Interesting, they used averaging (as I expected from the data) and their initial score was 67 which is also very close to the 68 I got from @amirm data.
If you check the initial EQ I published, there is no compensation for the 10k notch.
I sometimes do compensate a bit for it but ONLY when the bandwidth is wider than what I would expect and never to full extend.

The points I was trying to make might have been lost, sorry:
- The data used for the EQ design is (obviously) critical and initial scores 68/88 discrepancy is very significant, so comparing the EQed scores as you did is IMO irrelevant .
- One can practically always find a way to EQ differently and one-up the EQ that is tested regardless of the suitability/sanity of the "score based improvements". So yes, if I did not compensate for the 10k notch, the score would have been about 106 (after re-optimization).
I did not do so in the original EQ because although I knew it would help the score, it might not have yielded the best performance...

 

[solderdude]

Yes, DT880's vary a lot in midbass response depending on model (there are many DT880's) and production year.
Averaging them out ensures that when using 'averaged EQ' that the final result may be too warm or too cold sounding and most likely not accurate.
It would have made more sense to just make a pdf of each measured headphone (like Tyll did and Crinacle does).

 

[Robbo99999]

Trivial: they start with a score of 88 compared to 68.
Scanning their data, I get a score of 88.8 so very close.
Optimizing my EQ on their data yields 109.3 with two biquad less...
I win?

The trough at 10k is 20dB+ and the peak at 8k is higher too on the measurements performed by @amirm with the no averaging/smoothing.
I don't know what they are doing.

View attachment 139424

Oratory Data  DT880 APO EQ Flat@HF 96000Hz
July062021-171529

Preamp: -8.2 dB

Filter 1: ON LS Fc 55.45 Hz Gain 9.87 dB Q 0.42
Filter 2: ON PK Fc 196.53 Hz Gain -3.00 dB Q 1.16
Filter 3: ON PK Fc 669.12 Hz Gain 0.85 dB Q 2.19
Filter 4: ON PK Fc 1614.97 Hz Gain 2.77 dB Q 1.03
Filter 5: ON PK Fc 4926.67 Hz Gain 6.20 dB Q 2.01
Filter 6: ON PK Fc 5714.59 Hz Gain -8.69 dB Q 4.00
Filter 7: ON PK Fc 8333.51 Hz Gain -4.54 dB Q 4.97
Filter 8: ON PK Fc 9074.09 Hz Gain 4.85 dB Q 3.51

View attachment 139426

Don't correct the 10kHz massive dip! That's a peculiarity of the measuring apparatus, pretty much all headphones have some kind of sharp dip there, so correcting that dip would end up in you boosting that area when it doesn't want any boosting.

EDIT: typed this before I read GaryH's reply to you.....Oratory's response that GaryH quoted is the one to pay attention to........ah, so the 10kHz dip happens in our own ears too, as well as the measuring apparatus.

 

[GaryH]

Interesting, they used averaging (as I expected from the data) and their initial score was 67 which is also very close to the 68 I got from @amirm data.
If you check the initial EQ I published, there is no compensation for the 10k notch.
I sometimes do compensate a bit for it but ONLY when the bandwidth is wider than what I would expect and never to full extend.

The points I was trying to make might have been lost, sorry:
- The data used for the EQ design is (obviously) critical and initial scores 68/88 discrepancy is very significant, so comparing the EQed scores as you did is IMO irrelevant .
- One can practically always find a way to EQ differently and one-up the EQ that is tested regardless of the suitability/sanity of the "score based improvements". So yes, if I did not compensate for the 10k notch, the score would have been about 106 (after re-optimization).
I did not do so in the original EQ because although I knew it would help the score, it might not have yielded the best performance...

Well even for Oratory's original DT880 measurements with a pre-EQ score of 67, his post-EQ score was 99, a significant 17 points higher than your algorithm's

Yes, DT880's vary a lot in midbass response depending on model (there are many DT880's) and production year.
Averaging them out ensures that when using 'averaged EQ' that the final result may be too warm or too cold sounding and most likely not accurate.
It would have made more sense to just make a pdf of each measured headphone (like Tyll did and Crinacle does).

He'd probably show that individual data for you if you asked him. However, unless you have measured many units of each production year and found a significant common pattern, you've likely just seen unit to unit variation. As Oratory says:

I've measured over a dozen units of the DT770/880/990.
Whatever differences there are between the individual impedance versions, they are smaller than the average difference between two single units of the same model.
Or in other words:
If you get a DT770-250 and a DT770-80, they will sound a bit different.
If you get a DT770-250 and another DT770-250, they will also sound a bit different.
The difference between ten DT770-250 is larger than the difference between the average DT770-250 and the average DT770-80.
be very careful about reviews that only compare one unit of each model. Those reviews are comparing units and not models, so any differences they report are likely caused by unit variation, not by the models actually being different.
Except for the 32 Ohm variant of the DT770, that one is specifically tuned to have more bass.