Sennheiser HD650 Review (Headphone)

[brachypelma44]

Sorry for the late reply, there were some fireworks I had to attend to. Happy new year.
I tried three different EQ settings now:
1. https://www.dropbox.com/s/zr5tqw0qojom9uh/Sennheiser HD650.pdf?dl=0
Preamp: -8 dB
Filter: ON PK Fc 23 Hz Gain 2.6 dB Q 1.6
Filter: ON LS Fc 70 Hz Gain 5 dB Q 0.7 (I decreased the bass a little)
Filter: ON PK Fc 210 Hz Gain -2.2 dB Q 1
Filter: ON PK Fc 2150 Hz Gain 2 dB Q 2.2
Filter: ON PK Fc 3100 Hz Gain -1.1 dB Q 3.4
Filter: ON PK Fc 4270 Hz Gain 3 dB Q 4
Filter: ON PK Fc 5000 Hz Gain -0.8 dB Q 6
Filter: ON PK Fc 5700 Hz Gain 1.8 dB Q 3
Filter: ON PK Fc 6600 Hz Gain 0.9 dB Q 4
Filter: ON PK Fc 11000 Hz Gain 3 dB Q 1.3

2. autoEQ
Preamp: -6.6 dB
Filter 1: ON PK Fc 24 Hz Gain 6.3 dB Q 0.38
Filter 2: ON PK Fc 170 Hz Gain -3.5 dB Q 0.56
Filter 3: ON PK Fc 2002 Hz Gain 1.4 dB Q 3.08
Filter 4: ON PK Fc 4758 Hz Gain 2.4 dB Q 2.06
Filter 5: ON PK Fc 9697 Hz Gain 5.8 dB Q 2.31
Filter 6: ON PK Fc 3121 Hz Gain -1.8 dB Q 4.39
Filter 7: ON PK Fc 5359 Hz Gain 0.6 dB Q 0.61
Filter 8: ON PK Fc 12068 Hz Gain 2.9 dB Q 2.21
Filter 9: ON PK Fc 14386 Hz Gain 1.3 dB Q 2.17
Filter 10: ON PK Fc 19759 Hz Gain -12.3 dB Q 0.45

3. Amirm
Preamp: -8.5 dB
Filter: ON LS Fc 40 Hz Gain 5 dB Q 1
Filter: ON LS Fc 75 Hz Gain 2 dB Q 1
Filter: ON PK Fc 7600 Hz Gain 4 dB Q 3
Filter: ON PK Fc 13700 Hz Gain -3.5 dB Q 4

The autoEQ settings are my favourite.

Cool. I'll create a profile with them as well and do some A/B testing with them. Who knows...while I'm happy with how they sound now, I have to keep an open mind to the possibility that there may be some EQ settings that I prefer the sound of even more.

Thanks.

 

[Patrick1958]

Cool. I'll create a profile with them as well and do some A/B testing with them. Who knows...while I'm happy with how they sound now, I have to keep an open mind to the possibility that there may be some EQ settings that I prefer the sound of even more.

Thanks.

Here is another one. Measured by Rtings, PEQ calculated by Jaako, and i did some slight adjustments myself to better suit my ears
Pre-gain -9.

 

[nhatlam96]

Here is another one. Measured by Rtings, PEQ calculated by Jaako, and i did some slight adjustments myself to better suit my ears
Pre-gain -9.

Preamp: -9.6 dB
Filter: ON PK Fc 26 Hz Gain 4 dB Q 0.83
Filter: ON LS Fc 105 Hz Gain 5 dB Q 1
Filter: ON PK Fc 176 Hz Gain -3.1 dB Q 1.14
Filter: ON PK Fc 1243 Hz Gain -1 dB Q 5.65
Filter: ON PK Fc 2057 Hz Gain 0.2 dB Q 3.63
Filter: ON PK Fc 3033 Hz Gain -1.3 dB Q 5.96
Filter: ON PK Fc 4376 Hz Gain 2.5 dB Q 4.25
Filter: ON PK Fc 5521 Hz Gain -3.1 dB Q 4.13
Filter: ON PK Fc 6711 Hz Gain 2.5 dB Q 8.56
Filter: ON PK Fc 8234 Hz Gain 4.5 dB Q 2.11
Filter: ON PK Fc 11136 Hz Gain 5 dB Q 2.11

 

[bobbooo]

Sorry for the late reply, there were some fireworks I had to attend to. Happy new year.
I tried three different EQ settings now:
1. https://www.dropbox.com/s/zr5tqw0qojom9uh/Sennheiser HD650.pdf?dl=0
Preamp: -8 dB
Filter: ON PK Fc 23 Hz Gain 2.6 dB Q 1.6
Filter: ON LS Fc 70 Hz Gain 5 dB Q 0.7 (I decreased the bass a little)
Filter: ON PK Fc 210 Hz Gain -2.2 dB Q 1
Filter: ON PK Fc 2150 Hz Gain 2 dB Q 2.2
Filter: ON PK Fc 3100 Hz Gain -1.1 dB Q 3.4
Filter: ON PK Fc 4270 Hz Gain 3 dB Q 4
Filter: ON PK Fc 5000 Hz Gain -0.8 dB Q 6
Filter: ON PK Fc 5700 Hz Gain 1.8 dB Q 3
Filter: ON PK Fc 6600 Hz Gain 0.9 dB Q 4
Filter: ON PK Fc 11000 Hz Gain 3 dB Q 1.3

2. autoEQ
Preamp: -6.6 dB
Filter 1: ON PK Fc 24 Hz Gain 6.3 dB Q 0.38
Filter 2: ON PK Fc 170 Hz Gain -3.5 dB Q 0.56
Filter 3: ON PK Fc 2002 Hz Gain 1.4 dB Q 3.08
Filter 4: ON PK Fc 4758 Hz Gain 2.4 dB Q 2.06
Filter 5: ON PK Fc 9697 Hz Gain 5.8 dB Q 2.31
Filter 6: ON PK Fc 3121 Hz Gain -1.8 dB Q 4.39
Filter 7: ON PK Fc 5359 Hz Gain 0.6 dB Q 0.61
Filter 8: ON PK Fc 12068 Hz Gain 2.9 dB Q 2.21
Filter 9: ON PK Fc 14386 Hz Gain 1.3 dB Q 2.17
Filter 10: ON PK Fc 19759 Hz Gain -12.3 dB Q 0.45

3. Amirm
Preamp: -8.5 dB
Filter: ON LS Fc 40 Hz Gain 5 dB Q 1
Filter: ON LS Fc 75 Hz Gain 2 dB Q 1
Filter: ON PK Fc 7600 Hz Gain 4 dB Q 3
Filter: ON PK Fc 13700 Hz Gain -3.5 dB Q 4

The autoEQ settings are my favourite. The HD650 are not a bad headphone, just comparatively to the Clears, those problems become more obvious. Price to performance wise it's 10/10.

I notice your most preferred out of these three is the one with the highest preamp gain (i.e. the loudest). Just a small increase in overall level can be subjectively judged as sounding better. It's best to level-match when switching between EQ profiles by setting the preamp gain on all of them to the same as the lowest value (-8.5 dB in this particular case). The same goes for comparing with and without EQ (by e.g. making a flat 'EQ' profile with only the preamp set, matched to the actual EQ profile preamp level.)

 

[nhatlam96]

I notice your most preferred out of these three is the one with the highest preamp gain (i.e. the loudest). Just a small increase in overall level can be subjectively judged as sounding better. It's always best to exactly level-match when switching between EQ profiles by setting the preamp gain on all of them to the same as the lowest value (-8.5 dB in this particular case).

Good catch. I accounted for that I tried 4 different EQ settings now, the autoEQ one seems to be my preference. I also use the autoEQ for Focal Clear.

EDIT:
Hold up, am I doing this right?

Preamp: -6.6 dB
Preamp: -3 dB
Filter 1: ON PK Fc 24 Hz Gain 6.3 dB Q 0.38
Filter 2: ON PK Fc 170 Hz Gain -3.5 dB Q 0.56
Filter 3: ON PK Fc 2002 Hz Gain 1.4 dB Q 3.08
Filter 4: ON PK Fc 4758 Hz Gain 2.4 dB Q 2.06
Filter 5: ON PK Fc 9697 Hz Gain 5.8 dB Q 2.31
Filter 6: ON PK Fc 3121 Hz Gain -1.8 dB Q 4.39
Filter 7: ON PK Fc 5359 Hz Gain 0.6 dB Q 0.61
Filter 8: ON PK Fc 12068 Hz Gain 2.9 dB Q 2.21
Filter 9: ON PK Fc 14386 Hz Gain 1.3 dB Q 2.17
Filter 10: ON PK Fc 19759 Hz Gain -12.3 dB Q 0.45

Preamp: -9.6 dB
Filter: ON PK Fc 26 Hz Gain 4 dB Q 0.83
Filter: ON LS Fc 105 Hz Gain 5 dB Q 1
Filter: ON PK Fc 176 Hz Gain -3.1 dB Q 1.14
Filter: ON PK Fc 1243 Hz Gain -1 dB Q 5.65
Filter: ON PK Fc 2057 Hz Gain 0.2 dB Q 3.63
Filter: ON PK Fc 3033 Hz Gain -1.3 dB Q 5.96
Filter: ON PK Fc 4376 Hz Gain 2.5 dB Q 4.25
Filter: ON PK Fc 5521 Hz Gain -3.1 dB Q 4.13
Filter: ON PK Fc 6711 Hz Gain 2.5 dB Q 8.56
Filter: ON PK Fc 8234 Hz Gain 4.5 dB Q 2.11
Filter: ON PK Fc 11136 Hz Gain 5 dB Q 2.11

 

[bobbooo]

Good catch. I accounted for that I tried 4 different EQ settings now, the autoEQ one seems to be my preference. I also use the autoEQ for Focal Clear.

EDIT:
Hold up, am I doing this right?

Preamp: -6.6 dB
Preamp: -3 dB
Filter 1: ON PK Fc 24 Hz Gain 6.3 dB Q 0.38
Filter 2: ON PK Fc 170 Hz Gain -3.5 dB Q 0.56
Filter 3: ON PK Fc 2002 Hz Gain 1.4 dB Q 3.08
Filter 4: ON PK Fc 4758 Hz Gain 2.4 dB Q 2.06
Filter 5: ON PK Fc 9697 Hz Gain 5.8 dB Q 2.31
Filter 6: ON PK Fc 3121 Hz Gain -1.8 dB Q 4.39
Filter 7: ON PK Fc 5359 Hz Gain 0.6 dB Q 0.61
Filter 8: ON PK Fc 12068 Hz Gain 2.9 dB Q 2.21
Filter 9: ON PK Fc 14386 Hz Gain 1.3 dB Q 2.17
Filter 10: ON PK Fc 19759 Hz Gain -12.3 dB Q 0.45

Preamp: -9.6 dB
Filter: ON PK Fc 26 Hz Gain 4 dB Q 0.83
Filter: ON LS Fc 105 Hz Gain 5 dB Q 1
Filter: ON PK Fc 176 Hz Gain -3.1 dB Q 1.14
Filter: ON PK Fc 1243 Hz Gain -1 dB Q 5.65
Filter: ON PK Fc 2057 Hz Gain 0.2 dB Q 3.63
Filter: ON PK Fc 3033 Hz Gain -1.3 dB Q 5.96
Filter: ON PK Fc 4376 Hz Gain 2.5 dB Q 4.25
Filter: ON PK Fc 5521 Hz Gain -3.1 dB Q 4.13
Filter: ON PK Fc 6711 Hz Gain 2.5 dB Q 8.56
Filter: ON PK Fc 8234 Hz Gain 4.5 dB Q 2.11
Filter: ON PK Fc 11136 Hz Gain 5 dB Q 2.11

Yes, there's no need for two preamps on the first profile though, you can just set one preamp to -9.6 dB for each profile (including the other two you originally posted if also comparing to them).

 

[nhatlam96]

Yes, there's no need for two preamps on the first profile though, you can just set one preamp to -9.6 dB for all profiles (including the other two you originally posted if also comparing to them).

Okay good reminder, yeah I did that previously. I only posted the original preamps.

 

[Tachyon88]

The HD6xx is ultimately why I have a 2 channel system now. I like them stock and or with a +6db 40hz shelf. The ot1990 settings didnt do it for me, however I really like the ot1990 EQ with the HD800s. I will have to experiement with some of the EQ settings though.

 

[nhatlam96]

The HD6xx is ultimately why I have a 2 channel system now. I like them stock and or with a +6db 40hz shelf. The ot1990 settings didnt do it for me, however I really like the ot1990 EQ with the HD800s. I will have to experiement with some of the EQ settings though.

have you tried ot settings on both and compare them at the same time? when I do ot settings on clear and hd650, they have similar sound signature, but the technicalities and driver sound signature are discernible.

 

[Tachyon88]

have you tried ot settings on both and compare them at the same time? when I do ot settings on clear and hd650, they have similar sound signature, but the technicalities and driver sound signature are discernible.

No, not at the same time. Its been a while since I tried the ot1990 settings with the hd6xx though.

 

[bobbooo]

No, not at the same time. Its been a while since I tried the ot1990 settings with the hd6xx though.

Note Oratory periodically updates his EQ profiles after measuring and averaging the frequency response of additional units with previously measured units of a particular model, so his current HD650/6XX EQ profile may be an improvement for you, if your particular model has a frequency response closer to the latest units he's measured (I think he updated the HD650/6XX a month or two ago). Due to unit and placement variation, statistically the more units (and so in turn the more positionally varied reseats) he measures, in general the more likely it is that the averaged frequency response will be closer to the one you hear with your particular unit and your (naturally somewhat randomized) placement over your ears every time you put them on, and so increasing the likelihood of his EQ profiles getting closer to you hearing their intended target.

 

[JIW]

First track: Time: 2:54 to 3:06. Mid sharpness on the vocalist for the HD650.
Second track: Time: 3:40 to 4:00. Mid sharpness on the vocalist for the HD650.
Third tack: Time: the whole track. Lower resolution, half of the detail, worse bass, uneven balance of the tracks (in comparison to Focal Clear, otherwise it's fine headphone)

While listening to the second track using HD600, I found it sharp and the vocals sounded strained throughout. Level was set such that a 0 dBFS sample is at a level of about -6 dBV giving about 97 dB SPL according to manufacturer specification. According to YouTube's measure of loudness, it is 7 dB louder than YouTube's playback volume, i.e. YouTube lowered the level by 7 dB before playback. Even this way, it sounded too loud to me.

Afterwards, I have downloaded the audio. Cutting the first 8 seconds of noise and the last 7 seconds of silence, the RMS level is -10.70 dBFS in the left channel and -10.63 dBFS in the right channel, while peak levels are in excess of digital maximum - 2653 and 3046 three subsequent samples at positive or negative digital maximum and additionally 4512 and 4590 two subsequent samples at positive or negative digital maximum for the left and right channels, respectively. In other words, it is heavily clipped - about 25 and 27 times per second on average.

Thus, peaks were over 90 dB SPL and average level was only around 79 dB SPL. Listening to the downloaded version, I set the levels such that a 0 dBFS sample is at a level of about -13.5 dBV giving about 89.5 dB SPL peaks. Peaks were over 89.5 dB SPL and average level was again at around 79 dB SPL and it was already quite loud.

However, I typically listen to music with similar RMS 2-5 dB louder without any discomfort. The problem I have with this track is the increased energy in the upper midrange due to the high distortion.

The waveform speaks for itself. The most dynamic part seems to be the telephone ringing at the end.

If you are going to compare headphones not by comparing how well they reproduce the particularities of the source material but whether the reproduction of the source material sounds good, make sure the source material actually sounds good.

I couldn't dream of seriously judging headphones by how good they make Nattens Madrigal by Ulver sound - although I like to listen to it every now and then.

 

[nhatlam96]

The mids became a little too sharp and the background instruments too quiet.

First track: Time: 2:54 to 3:06. Mid sharpness on the vocalist for the HD650.

Guys I did a mistake here!!! When I compared those two, I did not use the equivalent EQ profiles.
I fixed this now by using the autoEQ settings on both headphones and volume matched them with my best abilities.
the mid sharpness is actually more noticeable on the Focal Clear, because the Clear reproduces the sharpness "clearer"
That's only a problem on very loud volumes though. If you want to fix this, adjust the frequencies at ~3.5k.
Otherwise all my other points are still valid, like:

Third tack: Time: the whole track. Lower resolution, half of the detail, worse bass, uneven balance of the tracks (in comparison to Focal Clear, otherwise it's fine headphone)

The technical side of things, like soundstage, imaging, microdynamics, dynamics, detail, resolution etc. were also of lower quality (in comparison to Focal Clear).

----------------------

Thus, peaks were over 90 dB SPL and average level was only around 79 dB SPL.

If I understood that right, then the second track of mine was compressed in the dynamic range? loudness wars etc.

The problem I have with this track is the increased energy in the upper midrange due to the high distortion.

What is the cause of this? Badly mastered?

If you are going to compare headphones not by comparing how well they reproduce the particularities of the source material but whether the reproduction of the source material sounds good, make sure the source material actually sounds good.

Yeah, I only use music that I often listen to, which should be the most important part. There will be tracks that aren't "perfectly" mastered, but the differences were consistent among all tracks I have listened, including Daft Punk flac files. What you think about the First track? It's a live recording.

I couldn't dream of seriously judging headphones by how good they make Nattens Madrigal by Ulver sound.

Is this also compressed in the dynamic range? Starting from 1:55 it gets very loud.

 

[JIW]

Nice write up, never have read something like this. If I understood it right, then the second track of mine was compressed in the dynamic range? loudness wars etc.

What is the cause of this? Badly mastered?

Yes, compressed but also clipped or at least limited too close to digital maximum values such that lossy encoders can overload and produce distortion. The level is not too atypical. The hi-res master of Lorde's Pure Heroine has an RMS of -11.88 dBFS - only about 1 dB lower - yet I could listen to it comfortably more than 1 dB louder.

At what stage of production this happened, I cannot tell you but mastering probably played a significant part.

Anyways, for something without any compression.

It does get quite loud in sections, however.

I listen to it at a level such that a 0 dBFS sample is at a level of about 1.5 dBV giving about 104.5 dB SPL peaks. RMS level is -28.29 dBFS while peak level is -0.98 dBFS. Thus average level is only about 76 dB SPL while peak SPL is 103.5 dB.

 

[JIW]

Yeah, I only use music that I often listen to, which should be the most important part. There will be tracks that aren't "perfectly" mastered, but the differences were consistent among all tracks I have listened, including Daft Punk flac files. What you think about the First track? It's a live recording.

Is this also compressed in the dynamic range? Starting from 1:55 it gets very loud.

The first track took too long to load for me to bother listening to. Daft Punk's Random Access Memories has very good production.

Since you also addressed the point about the HD650 sounding harsher than the Clears, I don't really have any more to question about your assessments. To me it read like the sharpness was particular to the HD650.

The Ulver album is infamous for its high loudness in part due to its low dynamic range - -7.56 dBFS RMS with 0 dBFS peaks - but also its emphasis on the upper midrange. I have to admit, it was a bit of a tease.

 

[nhatlam96]

Anyways, for something without any compression.

Time: 6:20 to 6:52: Obvious differences: The Clear hits a lot harder, the soundstage and open space feeling is also bigger. Smaller noticeable differences, where I had to focus more: more detail on the cymbals, better instrument separation.
Similiar findings also here at 33:33 to 34:00.

Since you also addressed the point about the HD650 sounding harsher than the Clears

yeah, about that... I made a mistake and it's actually the other way around. It's also only about the mids and with verrrryyyyy loouuud volumes.

When I compared those two, I did not use the equivalent EQ profiles.
I fixed this now by using the autoEQ settings on both headphones and volume matched them with my best abilities.
the mid sharpness is actually more noticeable on the Focal Clear, because the Clear reproduces the sharpness "clearer"
That's only a problem on very loud volumes though.

 

[Francis Vaughan]

I take a soft approach to equalization already. As you see there are very few filters. They are starting points and something you are welcome to tune. No amount of averaging or messing around in measurements will result in anything objectively better.

As a general principle in all equalisation the above contains some very important truths.

Something I like to emphasise is that every one of the peaks and troughs has a physical reason for being there. Some of them are compensatable for in a useful manner, and some are not. There seems to be a lot of merit is trying the smallest number of parameters for each apparent deviation on the basis that you are attacking a single physical phenomena with each. This is especially true when you have some idea for the reason for the deviation. That can guide your choice of parameters.

Sometimes they overlap. Two peaks could be one wide peak due to one cause, and a dip from a second cause. Or it could be two separate peaks each with its own underlying cause. Start simple.

Averaging can mean that you are simply chasing a single underlying physical phenomenon that is shifting around. The average is simply never going to work. The numbers you see will not correspond to any actual physical reality, and applying them will correct for something that isn't there, whilst leaving the actual problem intact, thus making things worse.

 

[Robbo99999]

As a general principle in all equalisation the above contains some very important truths.

Something I like to emphasise is that every one of the peaks and troughs has a physical reason for being there. Some of them are compensatable for in a useful manner, and some are not. There seems to be a lot of merit is trying the smallest number of parameters for each apparent deviation on the basis that you are attacking a single physical phenomena with each. This is especially true when you have some idea for the reason for the deviation. That can guide your choice of parameters.

Sometimes they overlap. Two peaks could be one wide peak due to one cause, and a dip from a second cause. Or it could be two separate peaks each with its own underlying cause. Start simple.

Averaging can mean that you are simply chasing a single underlying physical phenomenon that is shifting around. The average is simply never going to work. The numbers you see will not correspond to any actual physical reality, and applying them will correct for something that isn't there, whilst leaving the actual problem intact, thus making things worse.

Just to quickly reference your last sentence about averaging making things worse, Oratory averages his measurements (apart from poor seal issues on the bass) and his EQ's are all an improvement to the stock sound of my different headphones.....so it doesn't "make it worse". If you have inherent variance in your measurement system then averaging from multiple measurements is a good way to see the main trend differences (and also some specific characteristics) between 2 sets of data. I was gonna link a post I wrote earlier on this topic, but I think it was deleted by admins along with a whole lot of other people's posts too......it was on the topic that taking multiple headphone measurements and averaging them is a more accurate/representative approach for a headphone rather than relying on a single measurement that might otherwise be an unrepresentative outlier, EDIT: found the post, it hadn't been deleted (https://www.audiosciencereview.com/...00i-review-planar-headphone.18544/post-617570). I'm of the viewpoint that averaging headphone measurements is a more useful approach in creating a measurement curve for equalisation purposes.

 

[amirm]

Just to quickly reference your last sentence about averaging making things worse, Oratory averages his measurements (apart from poor seal issues on the bass) and his EQ's are all an improvement to the stock sound of my different headphones.....so it doesn't "make it worse".

??? Did he provide the no averaging EQ so you could determine that?

 

[amirm]

If you have inherent variance in your measurement system then averaging from multiple measurements is a good way to see the main trend differences (and also some specific characteristics) between 2 sets of data.

Not so here. Let's assume I know how to mount the headphone to get the most optimal response. I then random take the headphone off and put it on haphazardly as I think Tyll did. Now averaging these will get you worse results than the first one that was optimally done.

The variance here is not purely random so filtering could very well take away useful precision.