[Achlys]

To me, the strongest argument for using compensated instead of raw freq response charts is the sine illusion he references. This provides a handy interactive tool to see for yourself how it works.

One thing I may have missed though, shouldn't the compensated preference band be flatter, where you could draw a flat line across the median / average preference?

They are attempting to compensate for a flat, directionally independent head-related-transfer-function (HRTF), diffuse field, and compare the response of the headphone to preferred steady-state loudspeaker curves. It makes sense at a surface level, in theory. But it does not hold up to too much scrutiny when considering fundamental differences between headphones and loudspeakers in rooms.

 

[AaronBroad]

Not if the biggest variances in preference occur at the polar ends of the audible range, which they seem to going by the debates about bass and treble

I guess what I'm missing is, what does compensated mean.

The headphones.com measurement primer video makes the argument that compensated views eliminate the sine illusion effect. I was connecting compensated to be similar to how charts on squiglink allow you to compensate to your choice target curve. The compensated curve takes the target and flattens it to a line with no slope, which eliminates the illusion effect.

Back to the headphones.com video, I would not expect the compensated preference bands to be straight lines. The preference bands are useful in how they visualize the variation of preferences across the frequency range. (To your point, a larger variance of preference of say bass response results in a fatter preference band on the bass end of the graph.) However, I was thinking there would still be a compensated 0db flat line included in the preference band (i.e the 0db line is the median preference). The graph from DMS' review shows a tilted preference, which implies that 'compensated' has a different meaning here.

 

[Someguy420]

Same graph from video, adjusted so most of the frequency response aligns with preference. Now you guys can start arguing about dips at 5-7k, instead of the peak at 3.5k...


Just out of curiosity, is the headphone community always as toxic as this thread? Feels like Karens who didn't get enough sauce with their burger at McD and about to tear the place down. Maybe not, maybe it just seems like it.


Just to mess with your heads I made some more graphs. Keep in mind that these are the same graphs, showing exactly the same information!

MM-100: (fixed to preference)

MM-500: (already at preference)

Sundara: (fixed to preference)

MM-100 vs Sundara, no bias: (both fixed to preference)

MM-100 vs MM-500; MM-500 bias: (500 matches preference, 100 moved to match line of 500)

MM-100 vs MM-500; MM-100 bias: (100 matches preference, 500 moved to match line of 100)

MM-100 vs MM-500, no bias: (both fixed to preference)

MM-100 vs MM-500 vs Sundara, no bias: (all fixed to preference)

Had quite a bit of fun creating these. These are by no means meant to hate on the graphs from headphone show. I think they are great and a step forward! Just an experiment on how we perceive graphs and comparisons.

 

[holden4th]

Via email yesterday - the wait goes on.

We have an updated ETA for your order ATA-164615.

Unfortunately there has been a delay with our supplier and the ETA for your backorder has been changed to arrive by 24/11. We will dispatch the goods to you immediately, upon its arrival. Please let us know if you can’t wait any longer and we can get this cancelled and refunded for you.

Thank you for your patience and our apologies for any inconvenience caused.

Kind Regards,
Camille
Addicted To Audio

 

[bwardrop]

GadgetryTech review of the Mm-100 dropped a couple hours ago:

 

[RainyDog]

GadgetryTech MM-100 measurement and Super*Review MM-500 measurement.

 

[holden4th]

GadgetryTech review of the Mm-100 dropped a couple hours ago:

This review sort of confirms that I might have made the right decision in ordering the MM100. However, as I now have to wait for another month I won't find out for a while. Of interest to me was his comparison with the HiFiman XS. I've got the Ananda Stealth and apparently the XS and the Ananda share a very similar sound signature. I got the Ananda (after my LCD1 was returned after headband cracking issues) and while it's an amazing headphone in so many respects I still find myself thinking about the LCD1. A previous poster hinted that because I really like the LCD1 I'll enjoy the MM100.

 

[bonnei100]

The distributor in Norway now has moved their unconfirmed estimated delivery date to Dec. 05. The price remains NOK 5990 on their website (currently = 533 USD).

Maybe we'll be able to try them here before Christmas? But the way things have been going since the launch, probably not.

 

[Achlys]

An Apple House Sound member bought a unit for Oratory1990 to measure. Here are his results from measuring it on a GRAS45BC-10 KEMAR. Not a good look in the treble.

For comparison:

 

[blaked]

An Apple House Sound member bought a unit for Oratory1990 to measure. Here are his results from measuring it on a GRAS45BC-10 KEMAR.

Is this posted somewhere publicly yet? I don't see it in the database, and I'd like to see the EQ settings.

 

[Someguy420]

After eq'ing 3k peak, there seems to be slope \ applied from 5k>20k. Going from +4db to +1db. Pretty sure there are better ways to EQ that treble.

Edit:

Highest peak +3db, lowest dip -9db = absolute value 12db. 12db/2 = 6db. Maximum 6db of correction is needed to achieve 100% match with harman target with for ex. auto eq.

 

[Agnesss]

Is this posted somewhere publicly yet? I don't see it in the database, and I'd like to see the EQ settings.

It's not yet added to Oratory's database, but as they're my MM-100's I was sent the pdf.

After eq'ing 3k peak, there seems to be slope \ applied from 5k>20k. Going from +4db to +1db. Pretty sure there are better ways to EQ that treble.

Edit:

Highest peak +3db, lowest dip -9db = absolute value 12db. 12db/2 = 6db. Maximum 6db of correction is needed to achieve 100% match with harman target with for ex. auto eq.

I thought the same, and asked Oratory to try out an EQ I had made (obviously with no intention of discrediting Oratory's EQing ability, I haven't even had the chance to listen to them yet!):

Preamp: -6.2 dB
Filter 1: ON LSC Fc 57 Hz Gain 2.3 dB Q 0.710
Filter 2: ON LSC Fc 105 Hz Gain 3.2 dB Q 0.710
Filter 3: ON PK Fc 240 Hz Gain -1.6 dB Q 0.920
Filter 4: ON PK Fc 1038 Hz Gain -3.0 dB Q 1.540
Filter 5: ON PK Fc 2002 Hz Gain 5.2 dB Q 2.510
Filter 6: ON PK Fc 2873 Hz Gain -8.4 dB Q 1.230
Filter 7: ON PK Fc 4598 Hz Gain 5.1 dB Q 2.620
Filter 8: ON PK Fc 7600 Hz Gain -1.7 dB Q 3.000
Filter 9: ON HSC Fc 3000 Hz Gain 5.2 dB Q 0.710
Filter 10: ON HSC Fc 10000 Hz Gain -5.2 dB Q 0.710

The measured frequency response of which can be seen below,

I would recommend decreasing the region from 2-5kHz a bit, but even so this EQ manages a predicted preference percentage of 96.57%. Lastly, it's worth mentioning that Oratory confirmed that the phase angle is around 0° when time-of-flight is subtracted.

 

[renji1337]

Preamp: -6.2 dB
Filter 1: ON LSC Fc 57 Hz Gain 2.3 dB Q 0.710
Filter 2: ON LSC Fc 105 Hz Gain 3.2 dB Q 0.710
Filter 3: ON PK Fc 240 Hz Gain -1.6 dB Q 0.920
Filter 4: ON PK Fc 1038 Hz Gain -3.0 dB Q 1.540
Filter 5: ON PK Fc 2002 Hz Gain 5.2 dB Q 2.510
Filter 6: ON PK Fc 2873 Hz Gain -8.4 dB Q 1.230
Filter 7: ON PK Fc 4598 Hz Gain 5.1 dB Q 2.620
Filter 8: ON PK Fc 7600 Hz Gain -1.7 dB Q 3.000
Filter 9: ON HSC Fc 3000 Hz Gain 5.2 dB Q 0.710
Filter 10: ON HSC Fc 10000 Hz Gain -5.2 dB Q 0.710

The measured frequency response of which can be seen below,

I would recommend decreasing the region from 2-5kHz a bit, but even so this EQ manages a predicted preference percentage of 96.57%. Lastly, it's worth mentioning that Oratory confirmed that the phase angle is around 0° when time-of-flight is subtracted.

The EQ you created sounds better than Oratory's to my ears.

 

[squibble]

It's not yet added to Oratory's database, but as they're my MM-100's I was sent the pdf.

I thought the same, and asked Oratory to try out an EQ I had made (obviously with no intention of discrediting Oratory's EQing ability, I haven't even had the chance to listen to them yet!):

Preamp: -6.2 dB
Filter 1: ON LSC Fc 57 Hz Gain 2.3 dB Q 0.710
Filter 2: ON LSC Fc 105 Hz Gain 3.2 dB Q 0.710
Filter 3: ON PK Fc 240 Hz Gain -1.6 dB Q 0.920
Filter 4: ON PK Fc 1038 Hz Gain -3.0 dB Q 1.540
Filter 5: ON PK Fc 2002 Hz Gain 5.2 dB Q 2.510
Filter 6: ON PK Fc 2873 Hz Gain -8.4 dB Q 1.230
Filter 7: ON PK Fc 4598 Hz Gain 5.1 dB Q 2.620
Filter 8: ON PK Fc 7600 Hz Gain -1.7 dB Q 3.000
Filter 9: ON HSC Fc 3000 Hz Gain 5.2 dB Q 0.710
Filter 10: ON HSC Fc 10000 Hz Gain -5.2 dB Q 0.710

The measured frequency response of which can be seen below,

I would recommend decreasing the region from 2-5kHz a bit, but even so this EQ manages a predicted preference percentage of 96.57%. Lastly, it's worth mentioning that Oratory confirmed that the phase angle is around 0° when time-of-flight is subtracted.

I didn't try Oratory's settings, but I quite like yours! I just dropped the 4.5kHz peak a little to make distorted guitars less grating. Excellent job! Now I'm curious how this EQ curve stacks up against the one I tuned by ear.

 

[Someguy420]

Some more eq stuff.

Auto-EQ results (Harman 2018):

Preamp: -6.44 dB
Filter 1: ON LS Fc 105.0 Hz Gain 5.2 dB Q 0.70
Filter 2: ON PK Fc 26.6 Hz Gain 0.6 dB Q 2.57
Filter 3: ON PK Fc 176.2 Hz Gain -1.6 dB Q 0.41
Filter 4: ON PK Fc 1054.5 Hz Gain -3.3 dB Q 1.57
Filter 5: ON PK Fc 1965.8 Hz Gain 3.1 dB Q 4.08
Filter 6: ON PK Fc 3006.3 Hz Gain -8.6 dB Q 1.76
Filter 7: ON PK Fc 4807.4 Hz Gain 8.2 dB Q 1.09
Filter 8: ON PK Fc 5885.0 Hz Gain -2.6 dB Q 3.75
Filter 9: ON PK Fc 9086.8 Hz Gain 6.2 dB Q 1.46
Filter 10: ON HS Fc 10000.0 Hz Gain -4.8 dB Q 0.70