Really great video, only wish it was much longer. Keep em coming.

I use Peace & Equalizer APO on my PC to tune my HiFiMan Sundara to the Harman target. It's a subtle but welcome fuller sound...

check out the Sonarworks plugin, they do EQ for different models of headphones and they have the option to do it for a particular headphone, so the headphone sample variation is taken into account

I think this is one of the most important and logical examinations of what makes headphones sound good or bad that I've ever seen.

Been doing sometthing similar to this for a long time, not with a harman curve, but trying to match headphone A to headphone B. It sometimes does wonders when done properly. I haven't done the harman curve yet because I never liked how any of the harman curve headphones sound...NAD Viso's, PSB's etc. sounded closed in and muffled to me. But I'll give it a try on the T1's and see how I like it.

So which HD800 mod do you recommended for a flatter response? Reducing that 6k boost would be great. I actually prefer that flat sound for mixing.

what happened to the part where Katz said he wanted to compare some amps ? did that ever happen ?

Too often people spend money on various headphone amps and DACs and are trying to use them like fixed EQ presets that they hope will create "synergy" with specific headphones they own, and that is just a dumb way to waste money. If you want to alter how the headphones sound, then just use sound-shaping tools like EQ to do it. In the past few years, I've EQ'd the Stax 007 MKII, Audez'e LCD-2, Denon D7000, Shure SE535, Westone 4, Audio-Technica M50, Sennheiser HD650, Parrot Zik, Noontec Hammo S, and a bunch of others. Applying parametric EQ with the intent to achieve the best accuracy/neutrality has ALWAYS made the headphones sound far better. I started out using Tyll's recommended headphone frequency response curve that is flat from 20Hz to 1KHz, then does a -10 dB slope to 20KHz. That approach served me well for a while, but then I started feeling like the bass wasn't as visceral as my mastering loudspeakers (Klein+Hummel O 300D), so I boosted the sub-bass region by about +4 dB or so, which worked really well. Later when Tyll started talking a lot about the Harman curve, I saw that it also had a similar approach to sub-bass frequencies. I'd like to know how the Harman curve relates to Tyll's old frequency response curve. I think Tyll's old curve was based on after-adjustment for the human ear's inherent resonances, while the Harman curve shown in the various graphs is the raw response and not adjusted? How would the two responses look overlaid if they were both adjusted or both raw?

Tyll, do you happen to have the values used for these curves so that we may try them ourselves? I know that Bob Katz EQ'd both the modded HD800 and LCD-X, both of which are my go to headphones depending on the recording.

Amazing !

Cheers to Autoeq!!

This video was 5 years ahead of its time. Of course that was 7 years ago so ... technically it's out of date now. But hey, it's still _kind of_ relevant! (This joke stolen/adapted from an ad for The Ben Stiller Show on Comedy Central. To be honest, this video is as relevant today as it was when it came out.)

Should you be mixing to this curve?

I would absolutely love a perfect eq compensation for the er4sr to match your reference target from Harman.:-)

I've EQ'd the T1's now using the raw frequency response from your website and the harman curve, tried to match them as precisely as possible using a 31 band EQ, but I don't like it that much for now. I just don't dig the harman curve, maybe if I listen to it for a while I'll get used to it, but at this point to me it sounds nasal and sort of echoy with an overly boosted sub bass. Yes, the soundstage somewhat increases and I guess it does sound a bit more like a pair of speakers in a room, and definitely has a bit better separation of sounds, but the sound lost some body, particularly in the low mids. Turning the EQ on / off and comparing to a raw sound of the T1's, I find that with the harman EQ it almost feels like something is missing in the sound, it's sort of "grey" and "dull", whereas with the stock sound it's more full, colorful and alive feeling, most importanly more natural sounding, but noticably more in-the-head sounding. What does the harman curve actually try to do? Make the headphones sound as neutral or natural as possible, or to make them sound like a pair of speakers in a room and trying to mimic all the changes to the sound that the space in the room creates and a headphone doesn't?

Which headphone did Bob prefer with the EQ? LCD-X or HD800?

Tyll do you have the ability to meassure the headphones as you're EQ'ing? In live time I mean, like having a headphone on the dummy head and reading out the FQ response as you use an EQ? I suggest using the 31 band EQ in Foobar2000, it's the most transparent one out there that I've played with, and it's far more precise than these slidy ones where you shape the curve by hand. It would be a good idea to read the frequency response of a headphone in live time, as it's on the dummy head, and then adjust the EQ until the frequency response curve of that headphone matches the harman curve, that would be more precise than doing it on paper, because sometimes increasing a value for 4 db on the EQ doesn't actually increase that frequency output of a headphone by 4 db, and it usually messes with other frequencies around it as well, which leads to a EQ that is not entire accurate.

0:55 A wild HiFiMan 400s leaflet appears. :^)

What software did you use to eq it?

Hi Tyll - I noticed that the software used for the EQ is EQuilibrium. I am able to use EQuilibrium with Audirvana Plus. Have you got the saved EQ settings for the HD 800 please? My HD 800s are anax modded also.

But your HD800 is modified

i think you need a better Teeth doctor :-)