ES2a discussions \ reviews \ reasoning \ impressions

[mbrennwa]

Great info - any other insight as to why the MT might have stronger low end than the BHSE given they typically have the same PSU

I don't know, but I can speculate... I am pretty sure that both the MT and BHSE would reproduce a pure, low-frequency sine test signal in the same way. However, what you're listening to is not a sine signal, but, for example, a bass drum. The stick hitting the drum will create a signal with a very sharp rise, followed by a long, smooth "tail". The sharp rise requires the estat amplifier to pump a lot of charge into the headphone very quickly -- in other words, it needs to output a large current. If the amp struggles with this, it will distort the output, and the bass drum will not sound right.
I did some tests on the voltages and currents needed to drive estat headphones here and here. The results were not unexpected, but they also showed that something like a Stax SRM700T is far from capable of delivering enough current without distorting the output badly. Both the MT and BHSE will be much better, but maybe one is slightly better than the other.

For whatever reason, changes to the amp's feedback solution can produce audible changes that are very hard to measure in headphone FR.

Do you have a link to these tests / data?
A modification of the feedback would almost certainly change the profile of the harmonic distortion. For example, if you use a 1 kHz sine test signal, you would observe the harmonic distortion products at 2 kHz, 3 kHz, 4 kHz, etc. This test is very different from measuring the frequency response, where you'd measure the (linear!) voltage gain of the amplifier at different frequencies. Vice versa, the frequency response does not show the harmonics. It's therefore no surprise that the headphone frequency response did not show any effect related to the change of the feedback.

 

[Chefguru]

I don't know, but I can speculate... I am pretty sure that both the MT and BHSE would reproduce a pure, low-frequency sine test signal in the same way. However, what you're listening to is not a sine signal, but, for example, a bass drum. The stick hitting the drum will create a signal with a very sharp rise, followed by a long, smooth "tail". The sharp rise requires the estat amplifier to pump a lot of charge into the headphone very quickly -- in other words, it needs to output a large current. If the amp struggles with this, it will distort the output, and the bass drum will not sound right.
I did some tests on the voltages and currents needed to drive estat headphones here and here. The results were not unexpected, but they also showed that something like a Stax SRM700T is far from capable of delivering enough current without distorting the output badly. Both the MT and BHSE will be much better, but maybe one is slightly better than the other.


Do you have a link to these tests / data?
A modification of the feedback would almost certainly change the profile of the harmonic distortion. For example, if you use a 1 kHz sine test signal, you would observe the harmonic distortion products at 2 kHz, 3 kHz, 4 kHz, etc. This test is very different from measuring the frequency response, where you'd measure the (linear!) voltage gain of the amplifier at different frequencies. Vice versa, the frequency response does not show the harmonics. It's therefore no surprise that the headphone frequency response did not show any effect related to the change of the feedback.

Good thoughts. I’m going to be posting more of a review in the MT/T2 thread so let’s pick that convo up there and save this for the ES-2a.

 

[SolarCetacean]

Do you have a link to these tests / data?
A modification of the feedback would almost certainly change the profile of the harmonic distortion. For example, if you use a 1 kHz sine test signal, you would observe the harmonic distortion products at 2 kHz, 3 kHz, 4 kHz, etc. This test is very different from measuring the frequency response, where you'd measure the (linear!) voltage gain of the amplifier at different frequencies. Vice versa, the frequency response does not show the harmonics. It's therefore no surprise that the headphone frequency response did not show any effect related to the change of the feedback.

They're measurements done on a miniDSP EARS rig, and I fully acknowledge that this rig probably isn't up to snuff for measuring the finer details of things like distortion. I created 2 screencaps from the Farina sweep measurements I made in REW. The same seating was used for the measurements with the cable simply being moved from one amp to the other.

The measurements for acoustic distortion are basically within margin of error and the FRs are basically exactly overlaid on one another, despite the difference in feedback topology and the differences in perceived sound. The 727 even looks a bit cleaner despite the lack of global feedback causing, in theory, a huge increase in distortion vs having global feedback in place.

This is getting off-topic from the ES-2a, but the conversation was that the effect of amps and amp-side distortion isn't always easy to see in standard measurements.

 

[mbrennwa]

They're measurements done on a miniDSP EARS rig, and I fully acknowledge that this rig probably isn't up to snuff for measuring the finer details of things like distortion.

Ok, I see. The dynamic range of the harmonics relative to the fundamental is somewhere at 40-50 dB in this measurement. That's really not enough when it comes to amplifier distortion. You'd want something like 90 dB or maybe even more. I don't think you'll get there with an acoustic measurement. You'd need to measure the electronic output from the amplifier.

 

[number1sixerfan]

So I've had the ES-2a for two days now, with two lengthy listening sessions.. with the current planned to go well into the evening. Pretty easy to grasp things. My listening impressions are pretty favorable thus far. It is* a nice improvement from the ES-1a. Feels a bit more refined and better positioned than I believe the original was (I thought it had clear elements from the 009 and 007 but didn't really outperform or perform on the same level without quirks as either).

There is only one drawback thus far and it's that it has a bit of a laid back/slightly recessed midrange. It's not egregious and certainly not a dealbreaker, but would've preferred just a slightly more forward sound. Outside of this, every other aspect has been pretty impressive and balanced. Enjoying them a lot actually.

Giving it more time before jumping into a lot of specifics and jotting down thoughts tonight. Will share more over the weekend.

 

[haearn]

I've just seen this on the ES Lab website:

Notice to US customers
In response to the new tariff policy implemented by the US government, arrangement on shipping to the US has been revised as follows:
From May 1 2025, orders to the US will be shipped with tracked Airmail, in which import tax will be fully payable by ES Lab. The estimated delivery time will be 9-13 working days.
We also have a limited stock of ES-2a that has cleared customs and is ready to ship to your location from within US.
Our products can be purchased directly from our online shop. Feel free to contact us via website inquiry page or email for all sorts of inquiry.
Respectfully,
ES Lab

source:
https://www.eslabhk.com/news-and-notice

 

[number1sixerfan]

Sharing a post with early in-depth thoughts on the ES-2a.

 

[hutzelmeier]

Today I've listened to a few tracks together with @mbrennwa, using his ES2a, X9000, May DAC and T2, and my X9000. The good news is that we could not hear a difference between his and my X9000

My initial impression with the ES2a was "wow", very similar to the X9000, but like an X9000 with loudness switched on. After a while I thought it is simply warmer, I could not really hear a treble peak that some people reported here. The X9000 sounded somewhat thin in comparison. I could immediately understand why people do like the ES2a; for example I found the soprano of Christa Ludwig (Mahler, Das Lied von der Erde) really beautiful.

I then switched to Mahler 5, Dudamel and Berlin Phil: the soundstage was not really like I know it from, say, the tenth row in the concert hall, which is how I experience the X9000 (very natural soundstage, very high spatial resolution). Instead, the ES2a melted everything closer together; strange, because the soundstage was very wide. I also found the violins to be more realistic on the X9000.

We then listened to Bubbles by Yosi Horikawa – a perfect track for the X9000, which clearly assigns well-defined places for the different sounds in the width and depth of the soundscape. With the ES2a, this track had much less of a wow-effect for me; again, the sounds flowed more into each other, so to speak. Perhaps this is what some people describe as more natural about the ES2a soundstage?

We then listened to Norah Jones, Don't Know Why – beautiful voice with the ES2a, but the X9000 simply gave a more clearly defined image of her voice. The 2022 remaster of Sultans of Swing that impressed me so much with the X9000: a very good listen with the ES2a, but it didn't seem to matter much whether this was the new remaster or not.

So in summary I thought the two headphones were more similar than dissimilar but still gave a quite different representation, so I can understand why for some people it might be worth owning both (especially since the ES2a offers such good value for money). However, for the moment I will certainly not switch to the ES2a. But I know that much of what I described above is unfair since I've listened to the X9000 for more than a year so far and only one hour to the ES2a.

 

[jody2k]

Sharing a post with early in-depth thoughts on the ES-2a.

Thank you for the in depth impression, confirms for me upgrading from sr-009 to ES-2a would be more of a sidegrade, I think i'll save up for a x9000.

 

[icnaiwhimor1975]

Thank you for the in depth impression, confirms for me upgrading from sr-009 to ES-2a would be more of a sidegrade, I think i'll save up for a x9000.

x9000 is a very controversial headphone
not everyone likes this too spread out sound and artificially expanded soundstage

 

[mbrennwa]

My initial impression with the ES2a was "wow", very similar to the X9000, but like an X9000 with loudness switched on. After a while I thought it is simply warmer, I could not really hear a treble peak that some people reported here.

Agree! I do not hear too much treble with my 2a, which does not have the "filters" installed.

We then listened to Bubbles by Yosi Horikawa ... We then listened to Norah Jones

In my opinion, these were the two tracks that revealed most of the core differences between the 2a and the X9000. Bubbles showed how the imaging is more resolved and precise on the X9000 than on the 2a. Norah Jones' voice was very natural und relaxed on the X9000, while it sounded slightly strained on the 2a.

the Es-2a was perfectly able to keep things in their own lane. It's not as much of an ingredient separator the way the x9k turns a burger into bread, meat and a piece of lettuce.

That's a very nice way to describe the difference between 2a and X9000 -- but it's worth mentioning the burger still remains in one piece on the X9000.

x9000 is a very controversial headphone
not everyone likes this too spread out sound and artificially expanded soundstage

I don't think the X9000 soundstage is very expanded. If anything, I actually feel the 2a sometimes throws a slightly larger (but a more blurred) image than the X9000.

 

[number1sixerfan]

Thank you for the in depth impression, confirms for me upgrading from sr-009 to ES-2a would be more of a sidegrade, I think i'll save up for a x9000.

No problem . I think it's possibly an upgrade to the 009 overall, but it will be really personal when comparing to the x9000. The x9000 is a bit more hyper detailed and open, but on the spacing aspect sometimes artificially so. I'd be interested in comparing them head to head one day. But yea, I suspect the x9000 overall is just a bit better possibly.

 

[benjifx19]

At the end of the summer it will be time for me to get my next headphone and im now considering this as the top choice

My favorite headphone at CanJam the last 3 years has been the X9000. I wont have the budget for that but the ES-2A even with a tariff price hike would be within reach.

I dont mind the treble not being as forward as my planars do that well, I am willing to sacrifice some of that for more clarity overall

 

[Chefguru]

After two weeks with the Es-2a: it leans a little bit brighter than my preference but still within the range of what I enjoy. I’ve tried no mesh , stock mesh and even a 007 mesh ; ultimately the stock mesh is my preference and it mostly tames the highs. For those worried about needing to modify - no need.

My only issue with the sound signature would be that the mids feel a little bit recessed compared to the sr-omega , sr-lambda NB or Sr-007. The Es-2a more closely matches the x9000 in that regard. Is that a trade for a larger soundstage? Perhaps.

The comfort of the headphone is superior to other Stax / on par with the sr-007. The other night I actually found myself listening more to the es-2a than the sr-omega simply because the sr-omega touching my ear was bugging me after a while.

The Es-2a sounds very good out of the BHSE, and even better on the megatron since it emphasizes the mids further. It’s my favorite in-production headphone. While not as technical as the x9000 it is far more enjoyable since it’s not quite as recessed and compressed sounding.

Using all amps at my disposal I would rank the in-production TOTL as:
1) ES-2a
2) x9000
3) SGL sr
4) sr-009s
5) L700
The crbn would be #4 above 009, so I assume crbn 2 would be somewhere above it.

Including out of production headphones I do still prefer the 007mk1 , sr-omega and Sr-lambda NB but all 3 have their own issues not including that if a driver dies there is no replacement part.

 

[number1sixerfan]

The Es-2a is damn good. That's all