Dynamic vs Estat Round 15

Focal's just another manufacturer making great drivers like Audio-Technica, Fostex, and Koss and sticking them in middling headphones that only sound good enough to move off the cans off of a stock room shelf. Focal's again priced out of their intended market just like with the Spirit closed headphones which sounded okay but were more fragile than Beats and you never saw them on TV broadcasts or pictures of studios with boxes of a dozen of them like HD 280/25, DT 770, AKG K240/271, or the Fostex T40/50 line as who the hell would pay 400 dollars for a closed headphone when a 100 dollar one does the same job and then you can pick up a good open headphone or pair of powered monitors and have your tool and something that actually sounds great for less?

The Elear and Utopia are just priced way too high again. You have a big speaker manufacturer pricing their headphones at DIY in a warehouse planar/electrostat audio jewelry and mediocre sound prices. Let's be real here. This is what Audeze is. Audeze outsources industrial design to a BMW subsidiary, has the Chinese guys who make Matrix designing their amps, and if the Sine is consistent sounding in anyway I doubt Audeze does anything other than assemble it in their warehouse factory. At least Hifiman has reasonably priced open headphones to sell to people who don't want to wire an HE-6 to a speaker tap. Focal doesn't. They have 2x overpriced Beats/studio can "competitors" and overpriced open headphones 3-5x the street price of the models they want to compete with. I don't see these doing as well for Focal as car speakers. Unless they just want to take Audeze/Hifiman quick cash from rich idiots and manufacture these in small quantities just large enough to pay for a mold and R&D but not large enough to have great quality on. This strategy didn't really work for STAX in long run as they were barely afloat for decades and got 100% bought for almost nothing by a big Chinese speaker maker.

EDIT: Edifier bought Stax for 1.5 million in cash. Damn. At least that's better than Harman purchasing the bankrupt AKG and finding out the only things of value they had were land, buildings, and molds used to make obsolete gear that could be sold for scrap.

 

Stax has zero texture or sustain so that's like saying the human colon isn't 1/10th as clean as it would be as if you gave it an undiluted bleach enema with a Super Soaker for an hour.

 

Someone had a bad day today ...

 

Best description of the Stax sound I've heard.

 

Sustain is in the recording, not the transducer (unless you like ringing). What you call 'texture' is probably grain or distortion, which is fine to prefer.

 

The sustain is on the recording but unfortunately the estats are really good at removing that information (not resolving it). The texture is not a grain or distortion, it's the inability for transducer to get out of the way and render the textural information of what is being reproduced. The 009 applies a sterile and plasticky texture to everything so in that sense it is does have a texture. The sound of plastic is fine to prefer though

 

Maybe this should go to another thread.

Ya, sorry but this makes no logical sense. So you're saying the headphone can present the note at say -5dbfs but can't present information (sustain) at say -30dbfs? If that's the case there'd be almost no information presented unless you think the transducer has internal logic to ignore guitar sustain. Electrically it also makes zero sense and would be very easy to test by seeing what dB level the diaphragm starts to vibrate at as you increase voltage going to the stators. Physically is also doesn't make sense as they have the lowest inertia of any transducer. Quite the reach honestly.

Fact of the matter is that 'sustain' isn't a demanding thing to resolve at all, and I'd wager even the most primitive version of the DT48 could present that info, along with any other HP on the market. Is this really what you guys have been calling resolution all this time?

The grain/distortion is measurable, still waiting on the 'smoothing over info' measurements. It makes much more sense that you guys are used to the false additions presented from the higher distortion dynamics and have conflated the lack of audible distortion on stats with 'missing info', 'smoothed over info', etc.

 

Why do Sign of the Southern Cross and When the Levy Breaks sound hilarious on Stax? Relatively uncompressed double kick rolls or heel to to toe hits in older recordings sound like you ran them through some additional compressors and then used some bad plugins on Stax. Why do electrostats and planars completely ignore pinch harmonics and the Haas effect for the most part? Bob Vigna from Immolation is a nice guy despite looking like Anton Lavey so why does Stax want to f**k him over? Why do Stax not accurately reproduce the tones of different instrumental micings, guitar amp cabs, and even analog tape? You know most of the grain on old recordings is from the tapes and old CDs used to warn you that about in the booklet? That's gone on Stax beyond tape problems. Getting rid of grain is getting rid of detail like heavy-handed digital noise reduction of film grain on poorly mastered Blu-rays. I prefer to watch and listen to humans rather than the Kardashians.

Here's the sound of Stax.

 

That's exactly what we're saying. Estats in general have worse low-level resolution and plankton retrieval than good dyanmics. Hence the reason they don't pick up sustain and other related instrument dynamics.

You haven't made the case for how good dynamic transducers have enough distortion to create all of this audible "grain" that you always talk about, so I'd say the burden of proof is just as much upon you if you want to play that card.Works both ways buddy. I can just as much tell you that the grain IS on the recording and that your setup isn't picking it up, it's not a reach at all. Do you really think that music is supposed to sound like some textureless plastic concoction of bleh? That's the more unnatural coloration here.

When the majority of senior members and a lot of others are picking up how the estats just aren't quite as resolving as moving coil stuff, maybe you should be the one to reconsider your proposition seeing as there is no more of a "measurable" or "objective" basis for it.

 

Um, I don't really have a horse in this race. Nor a leg to stand on. Running out of idioms...

But, intuitively, this seems like a pretty bulletproof argument to me.

 

On the grain issue, I think the problem here is that we don't have proof to point out whether or not the grain exists...

However, I think there are a few clues that may point to a middle ground:

1) Some dynamic drivers (especially earbuds, IEMs, and such small drivers...) inherently reproduce "grain" that other dynamic drivers that are either considered better or bigger in size do not. This is apparent and is measurable either via distortions in the midrange region and/or wiggles in frequency response. I think it is generally true that better dynamic drivers are LESS LIKELY to reproduce grains than lesser drivers.

2) Some songs inherently have a grain/tizz/weird texture "issue" either due to the recording equipment (this means software compression or lack thereof does not matter), or due to the compression algorithm being used (most MP3s would sound weird in the treble, for instance). In this case, I think a truthful audio source should reproduce some sort of tizz/grain/weirdness in the recording. I have heard this with a Stax headphone that's mated to a "lesser amplifier" (basically an energizer connected to a speaker amp), so it's not like e-stat cannot reproduce it. However, the same Stax headphone on a "better amplifier" (SR-009 on Liquid Lightning, to be more precise) did not seem to be able to reproduce this. I'm inclined to believe this behavior is therefore dependent on the amplifier/driving unit rather than on the technology of the transducer itself. Or at least, the amplifier/driving unit may be contributing to this "problem".

3) Smoothness and absence of this "grain" in some cases is enjoyable depending on the taste of the listener. For instance, I personally would enjoy something like the Stax SR-007 Mk.I because it does not spotlight sibilance and other weird treble issues in some recordings. Most dynamic headphones I have tried on or owned DO reproduce these sibilance and weird treble issues REGARDLESS of whether the treble region is emphasized or not. Even when a dynamic headphone has the same FR as the SR-007. So it is not merely a frequency response issue. This may belong to a separate discussion, but this is just to say, maybe there is a middle ground, and neither e-stat nor dynamic is truly accurate in their reproduction of the recording in question, and it is entirely down to taste whether a person would like to hear smoothness (plasticky timbre?) or a bit of sharpening (not plasticky).

4) This is probably a weird thing to consider... but how do we know for sure that whatever "grain" is being reproduced is actually "accurate" to the grain that's on the recording itself? We know the tonality of the headphone would have an effect on this...

Maybe a "perfect" transducer would be able to do both... i.e.: reproduce "grain" that's "smooth".

But it is clear there are two trains of thoughts here, and I'm going against the "general opinions" here to go with what Milos is saying: perhaps e-stats may offer something equally or more "truthful" than whatever dynamics are doing. Absence of something is just as "inaccurate" as wrongful reproduction of it after all, right?

 

None of what you mention is difficult for anything but the slowest headphones to reproduce. I shouldn't even acknowledge a post with this much hyperbole but I'll humor you for now. Curious to know what Stax you have experience with and in what setting. Regarding the idea that 'grain is on the recording', see below.

Again, this simply isn't true. You have a measurement rig, I gave you a very easy way to test if this is true or not. I think you've mentioned in the past that your normal listening level is ~90-95 dB, yes? So have the headphones play a full scale signal of your choice at your reference listening level, then perform the measurement again at -10dB, -20dB, -30dB, etc. If the mic captures the signal at those lower levels then the headphone is resolving it and thus a recording detail at that given level below reference can be reproduced. It should be self evident that the stats will pass this test without breaking a sweat, but that's why the 'can't resolve the hard-to-reproduce mystical sustain' is so silly to me. Again, headphones don't have a brain that tells them 'let me be able to capture this minute level detail like the air that gives you an idea of the size of a recording space but ignore anytime this guitar sustains a note at a much higher volume level'.

Actually, maybe Tyll or somebody with a low noise floor setup can run a similar test and see when a particular headphone starts deviating from the expected output. That's the only way I can think of right now to test headphone detail as a sin of omission rather than commission (see below).

The 'grain' is simply a term I use to describe what this 'thing' sounds like. I used to think it wasn't distortion, but as time goes by the more I'm convinced it's actually the low level distortion we're hearing. The correlation is too strong between the amount of 'grain' a headphone has and the amount of distortion it has. It may not be the THD in its entirety: it could be specific orders of distortion or more audible in a particular frequency range. Here's an example plot of headphones with varying amounts of grain, from most to least:

airplane buds >> DT48 >>>> ibuds >>>>>>>>> midfi Beyer >>> HD6x0 >>> HD800 >>>>> typical planar >>>>>>>> typical stat >> real life

It should be clear now why the grain I'm talking about isn't 'on the recording'. I'm not talking about tape hiss, clipping, or other signs of poor recording here. The grain is over the entire spectrum and is present on every single recording, no matter how pristine, no matter the medium, no matter the upstream gear. What's more plausible: A) the grain is inherent to the transducer, and the worse distortion transducers tend to correlate with how much grain is heard, or B) the grain is 'on the recording' and it just so happens the DT48, airplane buds, etc are monsters at resolving this detail on the recordings, with the senns significantly lagging behind in that resolving ability, and planars and stats lagging even further behind in that ability? Be logical here, the answer is apparent.

Again, it may not perfectly track with THD, but it's really close I'd say. I should note that years ago I didn't hear the grain. It wasn't until I had my first stat and didn't listen to dynamics for a long time until I heard it. Now I can spot it instantly, it's the first thing I hear. It doesn't take magic ears to hear this. My ears will only get worse with time, the reference simply increased. There are many stat freaks who know exactly what I'm talking about when I say 'grain', and likely many planar enthusiasts too. If your reference is a much higher level distortion device you probably won't hear it; I know I didn't.

A nice appeal to authority, but I have to say you might be living in a dream world if you think this is the case. First, what do you mean by 'senior members'? Most people joined SBAF around the same time so I'm assuming you're either talking CS or HF. If you're talking CS officers, I'm pretty sure shipsupt, muppetface, anetode, tari, ceetee, and others I'm forgetting not only would disagree with you that 'stats aren't resolving' but would say the complete opposite. Outside of Tari, I've talked to each of them quite a bit in private or off the forums and have a good enough idea of their general audio opinions that I can assure you this is the case. The only CS officers I've heard say this are you and Marv, and maybe Voldemike but I don't recall. If you're talking about senior HF members, again I can tell you this is far from the case. Stats are practically synonymous with speed, clarity, and detail on HF, HC, and the rest of the audio world. Always have been. In fact not once did I ever hear somebody say the opposite until Marv a couple years ago in the shoutbox, so really not sure who you're alluding to here.

A fundamental difference in how we see detail/resolution: you, and I think most people, look at detail/resolution as what a headphone is omitting. While I do think this is true to some extent (mainly based on inertia and the motor design), I'm starting to think it's only a small part of the story. As I've said a few times, a headphone doesn't have the ability to pick and choose what it wants to leave out of the recording. The signal goes to the driver and a resulting magnetic/electrostatic field is generated every time. There are variables that have some effect after this. The vast majority of what I see people refer to as 'resolution' (knife on a plate at a live recording, a random cough, etc) are BS to me. Those sounds are way too loud for the transducer to not respond to. Not only that, but you're telling me the sound can be 'resolved' by even the most ancient mics (since they're on the recording) but magically can't be resolved by a modern headphone, when a mic is basically a headphone in reverse?

I'm starting to think the majority of detail/resolution is a result of what a headphone is committing. I'm mainly talking about distortion here. The distortion is everything added by the transducer that wasn't there in the signal. The less of this you have, the more you're able to 'see through' to the signal ie more resolution. For the most part I think there's a pretty tight correlation between resolution and distortion, but distortion isn't the entire story. Sometimes an emphasized or boomy bass can really mask detail, for instance. Take the Audeze LCD-2 vs HD800: the HD800 is a more detailed headphone, but a huge reason for this IMO is that the Audeze has this thick, almost monotone bass that kinda covers everything. In this case the lesser detail retrieval from the LCD-2 is still a sin of commission.

The easiest way I've found to hear a drastic difference in resolving ability is to listen to bass notes played by an acoustic instrument on different headphones. Something like a HEK or 007, etc absolutely crush just about any dynamic in the ability to present the bass notes with differentiation instead of your typical one-note bass deal. I've found that this ability is very much in line with the amount of distortion a headphone has in the bass, as long as the headphone doesn't have boomy bass ala LCD2, and it's no wonder I hear such a huge difference here as that's the spectrum where distortion varies the most among different headphones.

zealot out

 

ROFL

Ali

 

Get back to me with a coherent and condensed argument

 

I win. Again.

Me getting OJ to see the light and leave his primitive technology behind:

 

I'm too wired right now to respond to your nonsense. Give me 24 hours of prep like you had.

The whole line-by-line nonsense is not my preferred style, but I will just as easily debunk your missteps.

 

I think a lot of it can't be explained by distortion. For example I find the UERM to sound much more resolving than most (if not every) dynamic IEMs and yet it has audibly much higher distortion.

Yes, there are things that correlate well with distortion and the main thing I find is what I like to call clarity and tbh seems to be mostly what you are describing. I find estats have by far the most clarity in any transducer type. Do I care enough about it to prefer the planar presentation? No.

I find every measurement on the internet I could find is too limited by noise to really show the differences in distortion for very good transducer types. No, not Tyll's either. That's just the limit of a small diaphragm mic inside an ear canal. FWIW I would say the clearest transducer I have heard is the HE90, above the SR009. I think I read that a Senn employee said that the HE90 measured with lower distortion than the SR009.

I also find stats to typically sound faster but artificially soft. I'm not quite sure why stats don't seem to handle transients with the same kind of explosiveness that I hear in real life. This depends on the particular transducer design. Maybe more rigid stator design makes stats sound more dynamic? Like Quad ESL vs MiTec maybe?

I also don't think stats necessarily sound faster than every dynamic driver by design. The Accuton diamond midrange and tweeters sound very clear and fast and IMO more resolving than stat speakers.

So how do I define resolution? To me it is how much information that was on the recording gets reproduced by the transducer and how it gets reproduced. I do think that even very old microphones are fundamentally much closer to ideal than even today's headphones.
Now that does not mean that every headphone will struggle with ambient cues and 'filling the soundstage with subtle information' as much as the Ether I heard and as you said, most of the coughs are so loud that you could likely hear them on Apple EarPods, but there is a certain level of subtle information that you hear only on very good amps and headphones. This is also not a frequency response thing as the HD800 never loses resolution, no matter how much you reduce the treble with mods.

Now, the whole grain vs absence of grain topic is a very difficult one. I think that the HD800 itself does add a slight amount of grain, but to me this is a more realistic textural coloration than STAX phones. In terms of texture, lack of coloration and 'inherent driver sound' I would say the Accuton diamond drivers are the closest to ideal I've heard yet and I found them a quite significant improvement over the ceramic ones. I don't really have all that much experience with the different Accuton drivers but I would place the HD800 driver about equal to the ceramic Accutons but with a very different, harder to describe coloration. Other may disagree here and it's possible that I might change my opinion on this when I have Accuton drivers in the house. Anyway, I'm talking too much about Accuton drivers again, but I hope that I made my point clear. It's essentially a matter of personal preference to me and I happen to prefer dynamic drivers most of the time.

 

Hell, take 48. The only thing this thread is missing so far is suspense.

Bob Vigna from Immolation is a nice guy despite looking like Anton Lavey.

 

We're not talking about driver speed but about proper sustain and decay which electrostats don't have. Uncompressed heel to toe hit won't fully decay. Have someone do a double kick roll or a heel to toe hit in real life right in front of you on real drums. The best recordings with the more natural drum tones can't be reproduced by planars and electrostats properly as those transducers decay artificially quickly.

SR hundred something series including the new SR-207, SR-009, maybe the SR-007 but they all run together on both the Stax Energizers and the BHSE with the low end stuff sucking and being worse than cheap Koss.

Of course the grain is on the recording. Why do different micing, recording, and mixing techniques and mediums have varying amounts of grain? Why is this all gone on STAX? Why does early German thrash metal from the mid 80s sound grainy as shit and then suddenly they have great all-analog productions with some of the best metal/rock producers in the world (Flemming Rasmussen and Harris Johns) by the end of the decade with no grain except that from the tape? Have you ever miced a guitar amp? Do you know how easy it is to get scratchy shit irrespective of how their picking sounds in real life?

Do you know there are many records that use worse analog tape for one track of a multi-track recording and this is easily audible on stuff not STAX? Have you ever listened to Dark Side of the Moon, Aerosmith, or Rust in Peace on the HD 800 or your typical brittle near fields that need toilet paper without low pass trim, filtering, or EQ?

Why do overmixed, quantized, digitally reamped, sample-replaced, and compressed to hell productions have no grain? Why does Burzum sound more real than Beyonce even if Burzum sounds like shit and committed 23 more stabbings than Beyonce did? You know how Nirvana and other 90s rock or death metal bands are kinda noisy and grainy but wonder why the drumming is so surgically sterile sounding? As it was fake! Dave Grohl and most of these drummers sucked so drum samples were pasted onto the tape with an Akai sampler and a calculator. Why did Slayer sound so much more lifelike (and grainier) than Def Leppard despite the reverb? Def Leppard was manually assembled by Mutt Lange in the studio note by note! Mutt Lange wouldn't even let them play a chord; he made them play the notes separately and then pasted them next to each other!

STAX make everything sound like Def Leppard.