[pearljam50000]

Hi i need a pair of flat sounding headphones for around 200$
So far it seems my only option is the beyerdynamic dt880 32ohm version
Will it work without an amp? Are there any other options? Thanks.

 

[SRKRAM]

The PSB M4U1s are neutral sounding and very easy to drive straight out of a phone.
The Sennheiser HD598cs are another good option. The highs and lows are not quite as extended as the PSBs, but the build quality is better.

 

[Me x3]

Keep in mind, a headphone like the Beyerdynamic DT880 could last you a decade or more. Best chances are you'll be getting an amp sooner or later.
If you don't listen to music at loud levels you can still use most relatively hard to drive headphones unamped. They will benefit from a quality DAC/Amp, that's granted but don't let that fact make you pick the wrong headphone. Even when I'm extremely familiarized with DACs and Amps, my advice would be to pick the right headphone (read, the one that suits your preferences the best).

Flat is a tricky concept because everyone has its own approach on its meaning.

DT880 will give you extension at both ends, more sparkle than bass quantity and relatively uncoloured midrange.
K702 will give you less sharpness and less treble extension, along with more forward upper midrange, but you get bigger stage, more layering and clearer/tighter bass response in exchange.

Both deserve a quality DAC/Amp along with quality recordings to shine, but that's the case with each and every high quality headphone / speaker setup.

 

[Alizz]

Forget the DT880 (spiky higs, bit recessed mids) and get the DT250 if you want absolute flat with a nice tight punchy bass.

 

[ProtegeManiac]

Hi i need a pair of flat sounding headphones for around 200$
So far it seems my only option is the beyerdynamic dt880 32ohm version
Will it work without an amp? Are there any other options? Thanks.

Good enough without an amp but it's not that flat. I'd look into the HD598 - here it is on the graph vs the DT880 32ohm and ATH-M50. You can see that mountain of treble on the DT880, while the M50 has those twin peaks in the bass region (and louder too). LCD-2 Rev2 as a reference for how flat it can get for a lot more money.

 

[Me x3]

Those graphs are compensated, and there's no consensued target response to speak of. Pick a different set of measurements and you'll get different results.
In other words, flat in the graph is not necessarly flat, and the way each deviation affect the perceived tonal balance is not obvious at first glance. Something that looks relatively flat can be a clearly coloured sound signature and something that looks relatively spiky can be relatively neutral. The LCD-2 for instance is a well known dark sounding headphone.
These graphs are very useful for comparative purposes, this has more/less bass/mids/treble than that. But saying which one is flatter is much trickier than just looking which one looks flatter on a certain compensated graph.

 

[ProtegeManiac]

Those graphs are compensated, and there's no consensued target response to speak of. Pick a different set of measurements and you'll get different results.
In other words, flat in the graph is not necessarly flat, and the way each deviation affect the perceived tonal balance is not obvious at first glance. Something that looks relatively flat can be a clearly coloured sound signature and something that looks relatively spiky can be relatively neutral.

Or qualitative listening tests are being conducted by ears that don't have a flat response. Not that mics totally do, but I'd sooner trust a mic than a human ear.

TThe LCD-2 for instance is a well known dark sounding headphone.
These graphs are very useful for comparative purposes, this has more/less bass/mids/treble than that. But saying which one is flatter is much trickier than just looking which one looks flatter on a certain compensated graph.

You can actually why the LCD-2 sounds like that on the graph. My ears' response must be the same as the mic, as what I hear consensuevfefe with the measurement.

Red box shows almost a flat line from 800hz down to 10hz.
Orange box shows treble, which initially does not look all that smooth, but non of it spikes above where 1000hz and barely goes above where 2000hz is.
Yellow box shows where it trails off above 800hz, and everything after that is well below where the LCD-2 is at 800hz, and that is what accounts for why some people perceive it as "dark." Not that it isn't, but the point is, the curve is closer to flat than an SR325 that some guy in his 60s might claim sounds flat to him despite looking like the output from hospital or geological equipment, because his ears already can't hear well at the upper range. I don't think anyone uses any 60yr old mics with mold build up from everyone who sang into them, hence why I'd trust a machine more than a human ear, unless I have to have the same subjective requirements.

These graphs are very useful for comparative purposes, this has more/less bass/mids/treble than that. But saying which one is flatter is much trickier than just looking which one looks flatter on a certain compensated graph.

Yes. For comparative purposes. Which is what is being done here. While you can't really quantify what is "flatter" in this context that does not mean there is absolutely no criteria, when there is - the fewest (at each frequency, ie X-Axis) and less severe deviations (in dB, ie Y-Axis). Using that criteria for example the HE400i, HD600, LCD-2, and HD650 for example are the closest to a smoothest curve, but that does not mean they have the same deviations and thus does not mean they should sound the same, only that compared to some other headphones these have fewer deviations.

 

[watersedge1234]

The Shure outings are options. 840/940 represent the spectrum relatively equally.

Audeze Sines can be had for or a little more than $200 these days. Very flat with a little fun up top.

 

[kadrian]

Thinksound on1/on2 are slightly V-shaped sound but they are really great.

 

[Me x3]

Or qualitative listening tests are being conducted by ears that don't have a flat response. Not that mics totally do, but I'd sooner trust a mic than a human ear.

You can actually why the LCD-2 sounds like that on the graph. My ears' response must be the same as the mic, as what I hear consensuevfefe with the measurement.



Red box shows almost a flat line from 800hz down to 10hz.
Orange box shows treble, which initially does not look all that smooth, but non of it spikes above where 1000hz and barely goes above where 2000hz is.
Yellow box shows where it trails off above 800hz, and everything after that is well below where the LCD-2 is at 800hz, and that is what accounts for why some people perceive it as "dark." Not that it isn't, but the point is, the curve is closer to flat than an SR325 that some guy in his 60s might claim sounds flat to him despite looking like the output from hospital or geological equipment, because his ears already can't hear well at the upper range. I don't think anyone uses any 60yr old mics with mold build up from everyone who sang into them, hence why I'd trust a machine more than a human ear, unless I have to have the same subjective requirements.

Yes. For comparative purposes. Which is what is being done here. While you can't really quantify what is "flatter" in this context that does not mean there is absolutely no criteria, when there is - the fewest (at each frequency, ie X-Axis) and less severe deviations (in dB, ie Y-Axis). Using that criteria for example the HE400i, HD600, LCD-2, and HD650 for example are the closest to a smoothest curve, but that does not mean they have the same deviations and thus does not mean they should sound the same, only that compared to some other headphones these have fewer deviations.

It's not about mics vs subjective impressions. It's about mics not being able to measure the perceived sound signature due to the extra variables involved when using headphones instead of speakers. That's why most headphone measurements are compensated according to a particular criteria instead of being what the mic captured.

Here's the same graph but without the smoothing applied:

And here's what the mic really captured:

Of course we can say HD600 and DT880 Pro are flatter than DT990 Pro or SR325i. That's very evident. But on the other hand, saying LCD-2, HD600 and HE400i are flatter than DT880 or K712 is completely arbitrary. We all know DT880 is slightly bright, as much as we all know HD600 slightly raised mid bass, raised upper mids and recessed treble. You can prefer one or the other, but saying HD600 is flatter than DT880 (as fact) is not a wise move. It's not that simple. Maybe the raise in the treble (DT880) on the compensated plot should be there, maybe not. DT880 might boost the treble as much as HD600 cuts somehwere else. FR analysis is a complex matter. On top of that the shape of our body is different (from person to person) so what's flat for person A is not flat for person B.

I don't think subjective impressions should be disregarded.
Praising flat a headphone that almost everyone hears markedly dark (LCD2 or HD650) is not a great idea from my perspective. Looking relatively flat under 800Hz and relatively flat over 800Hz (few dBs lower in level) on certain compensated and smoothed graph means very little in absolut terms.

Here you have K712 compared with the "flatter" HE400i (both measured on the same rig)

Pink is K712, green is HE400i

Is the HE400i significantly flatter? I really don't get it. I wouldn't rush saying K712 is flatter either.
That's why I like the word neutralish and I would apply it to both of these headphones.

What I would like to emphasize is that headphone measurements mean very little in absolute terms. Change the measurement rig or the compensation applied and you'll end up with a very different plot.

I've been comparing Innerfidelity measurements with Jude's measurements and you can spot differences over 6dB in certain parts of the treble region, so what's flat on a graph is spiky on other. You can read on Innerfidelity about how Audeze's own calibration differs from Tyll's and that leads to different conceptions of what's flat and what's not. Nighthawks are made based on a different theory of how flat should be measured, and so on...

 

[ProtegeManiac]

It's not about mics vs subjective impressions. It's about mics not being able to measure the perceived sound signature due to the extra variables involved when using headphones instead of speakers. That's why most headphone measurements are compensated according to a particular criteria instead of being what the mic captured.

Here's the same graph but without the smoothing applied:

And here's what the mic really captured:

And yet, subjective evaluations of what the LCD-2 sounds like "consensuses" more with the compensated graph, ie, many find it dark, some find it warm-ish just off neutral. That's why I use the other graph more rather than hand somebody the latter graph where they'd expect a lot of treble and not get it.

And you still have that relatively flat response from 800hz down to 10hz regardless of which graph you're looking at.

Of course we can say HD600 and DT880 Pro are flatter than DT990 Pro or SR325i. That's very evident. But on the other hand, saying LCD-2, HD600 and HE400i are flatter than DT880 or K712 is completely arbitrary. We all know DT880 is slightly bright, as much as we all know HD600 slightly raised mid bass, raised upper mids and recessed treble. You can prefer one or the other, but saying HD600 is flatter than DT880 (as fact) is not a wise move. It's not that simple. Maybe the raise in the treble (DT880) on the compensated plot should be there, maybe not. DT880 might boost the treble as much as HD600 cuts somehwere else. FR analysis is a complex matter. On top of that the shape of our body is different (from person to person) so what's flat for person A is not flat for person B.

I don't think subjective impressions should be disregarded.
Praising flat a headphone that almost everyone hears markedly dark (LCD2 or HD650) is not a great idea from my perspective. Looking relatively flat under 800Hz and relatively flat over 800Hz (few dBs lower in level) on certain compensated and smoothed graph means very little in absolut terms.

Here you have K712 compared with the "flatter" HE400i (both measured on the same rig)

Pink is K712, green is HE400i

Is the HE400i significantly flatter? I really don't get it. I wouldn't rush saying K712 is flatter either.
That's why I like the word neutralish and I would apply it to both of these headphones.

What I would like to emphasize is that headphone measurements mean very little in absolute terms. Change the measurement rig or the compensation applied and you'll end up with a very different plot.

I've been comparing Innerfidelity measurements with Jude's measurements and you can spot differences over 6dB in certain parts of the treble region, so what's flat on a graph is spiky on other. You can read on Innerfidelity about how Audeze's own calibration differs from Tyll's and that leads to different conceptions of what's flat and what's not. Nighthawks are made based on a different theory of how flat should be measured, and so on...

First off, I wasn't stating it as "absolute terms." Just that one can rank them based on how relatively smooth the graph is.

Go look at that very example and see how much more often and how much more severely the K712 deviates from 1000hz. It drops at 3500hz more than the HE400i does at 2000hz, rises more at 2000hz more than the other does at 3000hz and 4000hz, and rises farther at 7000hz than the other does at 9000hz (which also tends to be more audible since it's lower down in the treble region).

And again this isn't even to be taken as absolute - I've said time and again the problem is that no headphone really is flat to begin with and what people think is a "sound signature" is in reality more like "most acceptable compromise for the response given other parameters (driver size, sensitivity, impedance, chassis/earpad design, etc)." But one can still set some kind of hierarchy as to what will likely sound more neutral (ie taller 7000hz peak vs 9000hz peak).

The way you interpret what I'm saying is like if I say, "let's raid Lehman Bros and get their records," and you start accusing me of starting another Kristalnacht.

 

[Me x3]

And yet, subjective evaluations of what the LCD-2 sounds like "consensuses" more with the compensated graph, ie, many find it dark, some find it warm-ish just off neutral. That's why I use the other graph more rather than hand somebody the latter graph where they'd expect a lot of treble and not get it.

And you still have that relatively flat response from 800hz down to 10hz regardless of which graph you're looking at.

First off, I wasn't stating it as "absolute terms." Just that one can rank them based on how relatively smooth the graph is.

Go look at that very example and see how much more often and how much more severely the K712 deviates from 1000hz. It drops at 3500hz more than the HE400i does at 2000hz, rises more at 2000hz more than the other does at 3000hz and 4000hz, and rises farther at 7000hz than the other does at 9000hz (which also tends to be more audible since it's lower down in the treble region).

And again this isn't even to be taken as absolute - I've said time and again the problem is that no headphone really is flat to begin with and what people think is a "sound signature" is in reality more like "most acceptable compromise for the response given other parameters (driver size, sensitivity, impedance, chassis/earpad design, etc)." But one can still set some kind of hierarchy as to what will likely sound more neutral (ie taller 7000hz peak vs 9000hz peak).

The way you interpret what I'm saying is like if I say, "let's raid Lehman Bros and get their records," and you start accusing me of starting another Kristalnacht.

Not exactly. What I say is this: you don't know if the treble peak should be like the one on HE400i or like the one on K712 in order to sound flatter. In the very same way you don't know what flat means on each and every different compensated graph. Take a look at Sonarworks compensations, Harman compensations, Innerfidelity compensations, what's flat on graph A from 10Hz to 800Hz, is not flat anymore on graph B which uses a different target response as compensation. On top of that there's a compensation that can be made according to the shape of your body and it will most likely differ from the models used to make other compensations. This is just the tip of the iceberg.

Back to HE400i vs K712, someone can come and say hey the K712 is flatter. Look the bass is more extended on K712, and the treble is more extended as well, and the midrange remains flatter up to 2.5kHz instead peaking at 1kHz and then going down. Comparative is fine, but saying which one is flatter in this context is quite arbitrary. You say, "K712 drops at 3500hz more than the HE400i does at 2000hz" pick a different compensation where K712 don't drop that much at 3500Hz and you'll get flatter K712 and more boosted HE400i (in the 3500Hz region).

Although we agree it is possible to establish some kind of hierarchy as to what will likely sound more neutral, I don't agree it can be taken that far.
We will surely agree that LCD-2 is flatter than this: https://www.innerfidelity.com/images/FlareAudioReferenceR1.pdf

But when differences are more subtle and considering there's no possible (absolut) frame of reference, I wouldn't say this is flatter than that.
I can not say whether HD600 is flatter than LCD-2 for instance.

Keep in mind one more thing, what's neutral at 60dB average, is not neutral at 85dB average.
The headphone's frequency response will remain the same at 60dB and 85dB, but our hearing will perceive a sound signature variation along the lines of the equal loudness contours.

Last but not least, there's huge variance among recordings and this plays a big role with regards to what we perceive as natural sounding.
If I recall correctly you like albums from Nightwish, Epica and Kamelot. Most of those albums sounded very well on the HD600 when I had them. I often consider getting an HD600 back for them. I wouldn't pick HD800 for those genres as it can sound unnatural and boring to my ears, despite sounding excellent with different recordings.

 

[ProtegeManiac]

Not exactly. What I say is this: you don't know if the treble peak should be like the one on HE400i or like the one on K712 in order to sound flatter. In the very same way you don't know what flat means on each and every different compensated graph. Take a look at Sonarworks compensations, Harman compensations, Innerfidelity compensations, what's flat on graph A from 10Hz to 800Hz, is not flat anymore on graph B which uses a different target response as compensation. On top of that there's a compensation that can be made according to the shape of your body and it will most likely differ from the models used to make other compensations. This is just the tip of the iceberg.

And what I'm saying is this: everything I've listened to that has a sharp treble were easily confirmed by what those graphs show. Which is why I use those.

Last but not least, there's huge variance among recordings and this plays a big role with regards to what we perceive as natural sounding.
If I recall correctly you like albums from Nightwish, Epica and Kamelot. Most of those albums sounded very well on the HD600 when I had them. I often consider getting an HD600 back for them. I wouldn't pick HD800 for those genres as it can sound unnatural and boring to my ears, despite sounding excellent with different recordings.

That's because the HD800 has a treble peak and you get auditory masking, ie, you hear more of that treble more than anything else.

 

[Me x3]

That's recording dependant.
Sharp treble can be found in real life and is one of those things the headphone should be able to represent.
Some metal winds, some violins, hard hitting cymbals, etc. aren't smooth or soft, especially at close distance.

With regards to HD800 sounding unnatural and boring to my ears with those particular recordings, I think the headphone is too tight/lean for them. And too bright as well, but it should be noted what kind of recordings we are talking about. We aren't talking about real life carefully mic-ed. We are talking about strongly produced/tuned recordings. As much as I've enjoyed most of their songs "The Phantom Agony" has a Dynamic Range = 5.

Not saying HD800 is perfect, don't get me wrong. But I prefer to put these things in (what I consider to be) a proper framework.
I really don't think the HD800 was engineered to make strongly compressed metal recordings sound good.

 

[sikki-six]

Using HD600 and LCD-2(.1 pre-fazor) as my main pairs in my home set I find the two pretty close generally in frequency response. The most obvious differences are bass extension, upper mids (small spike vs small null) and overall treble energy (HD600 is actually the darker pair).

With EQ, the Audezes get closer to my idea of a neutral response, because the 600 doesn't take sub-bass boost very well at all. It gets kind of sloppy and boomy when boosting more than 1-2db. The bass is still far off of the clean punch you get from LCDs without any tuning. I actually like that 3-4k "spike" of HD600 - it brings out vocal presence and distorted guitars very nicely. This can be found by EQ'ing the Audezes pretty well too.

In general, HD600 is more middy and some might say boring, but the differences are not huge between the two. Without EQ the Audezes are more fun, maybe slightly more colored too.. BTW, there's a some grain in the Senn-treble vs. LCD.