Help with my Audeze LCD2 Frequency chart

[ADU]

This is what Ora's LCD-3 measurement looks like with diffuse field compensation btw...

And here is Audeze's DF plot of the same headphone from your link above.

They are not identical. But they are somewhat similar, except that Ora's graph droops down more in the mid-treble.

The above two curves could also be used to calculate a translation curve for the Audeze measurements, so they could be more easily compared to Ora's other diffuse field compensated headphone plots.

 

[ADU]

DF compensated Audeze LCD-3 shown above vs. a few other headphones I sometimes like to use as general benchmarks. The Audeze is the red curve.

The above plot suggests that this particular Audeze LCD-3 headphone is too flat in the sub-bass, and could come up at least a few dBs around 50 Hz. It is also too withdrawn in the upper mids and low treble. And should come up at least a few dBs in the 4 to 5 Khz range, and around that area. (Ignore the dip in the Apple APM in green in that range.)

The Audeze also looks like it might be a bit too bright in the upper treble, and could maybe come down a few dB there.

The above plot was created with Ora's graphing tool btw.

 

[ADU]

Instead of translating or adapting Audeze's headphone plots to make them more compatible with Oratory's graphs, what you could do is try to go in the opposite direction instead. And create an EQ correction curve for a headphone like the LCD-3, based on graphs like the one above (or using some other group of headphones for your target). And then apply that EQ correction curve to Audeze's diffuse field plot of the LCD-3. And then use the EQ-corrected Audeze LCD-3 plot as a rough target for some of Audeze's other diffuse field headphone plots. Including the OP's LCD-2 graph.

 

[Phoney]

Please be mindful of the 8-10khz area that is not really being consistenly measured by @oratory1990, due to the rig that he uses. On the LCD-2F for example (and possibly the classics), oratorys EQ boosts this area by ~4-5db. Audezes measurments told me that there's a peak around 8250hz, which rises to about 3db on average above the harman curve, on the 2Fs. This makes for a total peak of 8db above harman in the 8-9khz area. I used his EQ for the 2F, which seems very accurate apart from the highest frequencies, not realising what was wrong. Something made the sound really unpleasant after some time, and I ended up stopping the session after 3 hours because of discomfort inside my ears. After this I had this horrible low frequency tinnitus, which has dropped in intensity 1 week later, but now acts as a reactive tinnitus which is very annoying, and I can't listen to headphones yet because my ears will hurt. I'll probably be fine after some time, but I believe this may have been triggered by that huge treble spike, because the rest of the volume was not loud. I'm also sensitive to treble spikes, and usually never listen to treble spikes above harman, so I was not used to this 8db peak.

When I recieved the measurments from Audeze, I realised what had happened. I was able to fix the peak with EQ, and all of a sudden the sound was very nice and comfortable, and even after bringing down this peak by 7-8db the treble now seemed natural and at normal levels, also not recessed. This kind of confirmed what had happened. Ofcourse, I should've probably realised this earlier, but I'm not a very experienced listener and I thought that oratory would correct obvious mistakes by listening tests. I also thought that this 7-8db peak after his EQ is unlikely to occur, despite this area not being measured.

Lesson learnt, if you use oratory EQ, it may be smart to see if you can find other frequency response measurments that might measure the 8-10khz area and compare them to harman, especially if oratory is boosting that area. Atleast make sure you can test this area in some way with your headphones after applying his EQ, and not just blindly trust them. There is also unit to unit variations.

 

[ADU]

Not sure I'm followin everything you're sayin above, Phoney. But I think if you try to use the Harman curve as your target for Ora's graphs, you could easily end up boosting the area around 9-10 kHz too much.

Most neutral headphones will have a fairly pronounced dip in that range on Ora's plots. And if you try to boost that dip to match the levels of the Harman target in that area, you will undoubtedly end up adding an unnaturally bright peak/sibilance, which shouldn't really be there.

There is a note about this somewhere on Ora's instruction pages, which was pointed out to me by some folks on head-fi awhile ago. (I think it was markanini, but not sure.) And this is one of the reasons I prefer to use the frequency responses of other neutral headphones as a guide for my EQ adjustments with Ora's graphs, rather than Harman.

Fwiw, you can see this rather large dip at about 9-10k on most of the headphones in the graph below. As well as another fairly sizable dip in the 15 kHz range.

Both of those larger dips should be left intact, as well as probably some of the smaller oscilations in the treble. And generally speaking, the only thing that should be adjusted are the overally levels in that range.

If the headphone's response lacks the dip at 9-10k (and/or 15k), then that could also be an indication that it has a sibilance in that range, which might need to be lowered.

 

[Phoney]

Not sure I'm followin everything you're sayin above, Phoney. But I think if you try to use the Harman curve as your target for Ora's graphs, you could easily end up boosting the area around 9-10 kHz too much.

Most neutral headphones will have a fairly pronounced dip in that range on Ora's plots. And if you try to boost that dip to match the levels of the Harman target in that area, you will undoubtedly end up adding an unnaturally bright peak/sibilance, which shouldn't really be there.

There is a note about this somewhere on Ora's instruction pages, which was pointed out to me by some folks on head-fi awhile ago. (I think it was markanini, but not sure.) And this is one of the reasons I prefer to use the frequency responses of other neutral headphones as a guide for my EQ adjustments with Ora's graphs, rather than Harman.

Fwiw, you can see this rather large dip at about 9-10k on most of the headphones in the graph below. As well as another fairly sizable dip in the 15 kHz range.

Both of those larger dips should be left intact, as well as probably some of the smaller oscilations in the treble. And generally speaking, the only thing that should be adjusted are the overally levels in that range.

If the headphone's response lacks the dip at 9-10k (and/or 15k), then that could also be an indication that it has a sibilance in that range, which might need to be lowered.

Yes I'm aware that his measurments will show a dip in that area, that's why I'm warning people about it. Audeze sent me the real average frequency response of the LCD-2F, and I found out that it has a peak at around 8250hz that rises to about 3db above harman. Also the 7-8khz area is much higher in Audezes measurments, about 1.5db below harman. Oratorys EQ looks like this:

Oratorys measurments pretty much matches Audezes, apart from a 1.5db peak above harman at around 800hz which is present in Audezes measurments, and also apart from the 7-9 khz area where there is a HUGE difference that leads to a bunch of siblance using oratorys EQ. I'm aware of why his measurments fail to identify this area, I'm just warning people about it, especially for people that buy this particular headphone. In reality, Harman looks more like this: