headphone tonality test

[solderdude]

Here is a fun test for you to try.
5 x the same excerpt of a live recording.
All MP3 VBR0 48kHz. (I know... should have been FLAC but that is 3.5x the file size) the differences have been preserved.
1 of them is the original
4 of them are the same excerpt but have undergone the exact same EQ as the frequency response of 4 headphones measured by Amir so basically simulated tonality changes of 4 headphones.
Unpack all 3 ZIP files, together they contain 5 MP3 files.
This all is under the premise that FR is the only thing that sets headphones apart.

Now the following questions:
Which sounds 'best to worst'
Which one do you reckon is the original ?

please post opinions (no one is wrong or right) or PM me.

 

[Chagall]

We just post here, I guess?

Spoiler: Results

best to worst
5 - a bit much bass,
3 - highs but lacks a little bass
2 - original but really not sure
4 - bass, but lacks highs
1 - just bass, no highs
2 or 3 original...I don't know

 

[Jimbob54]

Fun for later!

 

[muza_1]

Original: 5
Best to worst: 4, 1, 2, 3.

4 Sounds close to 5 but is missing some lower bass
1 Is also close to 5 but with a bit more mid bass
2 Sounds hollow to me
3 Is too bright

Edit to add, I used 7Hz Zero:2 and HD600 both EQed the Harman target, as for the Zero:2 I don't use the Harman target in normal everyday use, too bright for me.

 

[OnLyTNT]

For me best to worst:
3
5
1
2
4

To be honest 5-1-2 at same level for me. 3 is the most balanced and I also think this one is the original, 5 is V shaped, 1 also V shaped but a bit darker, 2 is too thin and a bit weird, 4 is definetly the worst. I used my Hexa without any EQ.

 

[Thomas_A]

DT150 HPs with DT100 pads. Difficult to guess original of course. Recordings vary as we know...

Spoiler: My vote

1 Severe exag mid/upperbass
2. subdued midrange/upper bass
3. bit too high treble, lower bass lacking a bit
4. exag midrange/upper bass, lower treble, no bass bottom
5. clarity midrange, but a bit exaggerated mid bass

Best to worst: 5,2,3,4,1
Original could be 2 or 5.

 

[MayaTlab]

Were these level matched ? Some of the tracks sounded to me like they had a lower volume.
May I also suggest to encourage all posters to share the headphones they used to reproduce these tracks, as some have already done ? I'm afraid this could suffer from the sort of issues sound demos tend to suffer from.

Anyway :
1 = too much bass, a bit boomy ?
2 = hole somewhere upper mids / low trebles (3-5kHz ?), rolled off bass
3 = flat / rolled off bass, hole 1-3kHz = Hifiman style ?
4 = rolled off bass, HD6something style mids ?
5 = original ?

I prefer 5 over the rest, but I could do with a smidge less bass.

I used a bunch of headphones to reach the same impressions anyway, AirPods Pro 2 with no EQ, a JBL Live Pro 2 with a custom EQ, and a HD58X with a custom profile that's probably not too far from Harman. I am not familiar with the original track.

 

[solderdude]

There are clearly people with a good set of ears..

Will tell what headphones were simulated tomorrow.

 

[solderdude]

Were these level matched ? Some of the tracks sounded to me like they had a lower volume.

You can't level match. They are just the original file and EQ was applied which means some frequencies are louder and other parts may be lower in amplitude which, of course, affects overall loudness.

 

[LTig]

Using an EQed DT990 pro 250 Ohm, #5 sounds best to me and is, I think, the original (most neutral sounding voice, but my aging ears still miss a bit of brightness). #1 has too much bass, the others are not neutral enough regarding the lower frequencies of the voice.

Same result using an EQed HD800, except a little less missing brightness.

 

[majingotan]

5 ( original recording and the most correct tonality per Susvara) > 1 ( warm sound / L – shaped ) > 3 (lower treble presence emphasis and recessed bass and mids) > 4 ( mid-centric , both bass and treble are rolled-off) > 2 (recessed / wonky / uneven mids)

Edit: #5 had the best *resolution* out of all samples which was the giveaway for me. This the reason I avoid digital EQ and any DSP, it takes away resolution to me

Susvara on KT88 SET amp

 

[solderdude]

So... 4 headphone tonal balances (with good seal) were simulated.


1 = DCA Stealth
2 = ZMF Caldera
3 = Hifiman Susvara
4 = Sennheiser HD650
5 = original


I got one response (via PM) which actually named the headphones I simulated correctly.

 

[majingotan]

Nice!

I said on the ZMF Caldera thread about the wonky mids

HD650 yes are mid-centric to me #4 is spot-on

Susvara on #3 doesn’t sound like that on tubes (#3 sounded more like HE1000se sounding on my system)

#1 wasn’t bad but a little bit warmer than my preference. I understand though The Stealth/Harman tuning is always preferred.

I suck at identifying which headphones on a blind test those tonal balance belong to but at least I’m aware of what kind of tonality that is being presented to me

 

[Thomas_A]

I my case I already have a slight 200 Hz boost in my DT150/DT100 so adding a fr response with additional boost in this area would be too much, thereby affecting ranking.

 

[Thomas_A]

Even if not blind, quite similar responses so far. Suprising or not?

 

[majingotan]

Even if not blind, quite similar responses so far. Suprising or not?

As Expected. Pick a reference point and any deviations from your reference point will yield the same results. No matter the headphone or IEM you'll choose, you'll end up with the same results as long as you know your favorite reference tonality

 

[IAtaman]

Edit: #5 had the best *resolution* out of all samples which was the giveaway for me. This the reason I avoid digital EQ and any DSP, it takes away resolution to me

Even on MP3?

 

[Thomas_A]

As Expected. Pick a reference point and any deviations from your reference point will yield the same results. No matter the headphone or IEM you'll choose, you'll end up with the same results as long as you know your favorite reference tonality

With those variations among individuals? The reference is of unkown tonaltiy (if novody cheated), only our ”inner reference” is used.

 

[majingotan]

With those variations among individuals? The reference is of unkown tonaltiy (if novody cheated), only our ”inner reference” is used.

The variations would be that that person does not have a full grasp of the reference tonality of their headphone or IEM. If you know how your IEM / headphone sounds from 20Hz - 20KHz, you'll easily within 5 seconds that #5 is the original and everything is a deviation from the reference tonality of your chosen IEM / headphone (even if you EQed headphones to Harman and your EQed headphones / IEMs to Harman will be your reference point and any deviations from it will be easily noticed). I posted my results before @solderdude showed the results. Clearly most responses knew how their system sounds and as such they can easily tell the original from manipulated

 

[majingotan]

Even on MP3?

Yep. And with 5% THD from my KT88 SET amp on top of that. The instruments sounded a subtly cleaner on #5 than the rest of the samples