Two mod jobs: HE560 and JAR600

Sorry I haven't posted much. I needed to take a break to avoid burn-out. Also, I wanted to rush to make top 1000 leaderboards on Diablo 3 this new season. As far as last night, I was around 750ish on the Witch Doctor Spirit Barrage build. (They are squishy at low Paragon, meaning less than 1200, so it's been a challenge). Look me up as purrin in the Americas region if you still play this ancient game.



I don't think I will get to everything today, maybe in the next few days. What we have up are the following:

1. @rhythmdevils modded HiFiMan HE560
2. @rhythmdevils modded Fostex T50RP
3. @ext1's JAR HD600

First up is the HE560. I took these to work and listened to them for days at my work desk. Here is the funny thing: I thought they were modded HE5s with the fast transient response, tight sound, and clarity. Well, that is until I realized that I wasn't getting that last little bit of low level information which I knew the HE5s were capable of extracting. Then I noticed a touch, just a bit, barely noticeable plasticky timbre... I took the headphones off, examined them, and then realized the gray HE560 letter on the black. Haha!

(FYI, I do not know if the HE560 have been reformulated since I last reviewed them). Normally with the HE560, I would choose to use the Focus-A pads otherwise they would come off as bright. The issue with the Focus pads is that there wasn't very much depth and the ears can sometimes touch the baffle. They also kind of close off the sound. @rhythmdevils opted to use these nice squishy thick pads which offer plenty of room and thus a more open sense with more air. Normally, the use of more "spacious" pads would make the FR issues much worse, but somehow @rhythmdevils pulled it off. Note that there was a small bit mid-treble emphasis, which was easily corrected with the twist of a knob on the Loki EQ.

I'll let him speak to the mods that he performed - I've known him for a long time and he takes a fastidious iterative approach using only his ears - balancing many factors that go into what constitutes good sound. In a nutshell, these modded HE560 almost had me fooled. They are clearly a two steps, maybe three steps up from the typical entry level HFM (HE400S, HE5XX, Sundara, etc.). Heck, better than the Ananda too. However, I have to admit that I am not big fan of the new tech HFMs and still prefer the old school models (even if they are reformulated). I think it's the damping. There's a weird thing going on with the new stuff. @rhythmdevils calls it undampened. I've called it "bouncy-bouncy". Some people will notice it, others will not.

@rhythmdevils modded HE560
Frequency Response
Flat Plate Coupler Compensated


@rhythmdevils modded HE560 (BLU/RED) vs OG HE560 (WHT)
Frequency Response
Flat Plate Coupler Compensated


Note: My prior FPC measurements from years ago were never compensated, thus HE560 which sounded slightly bright never appeared so measurements. The compensated results now give us a better understanding.

 

Heh, no thanks. I like these pads better. Also, OG HE560 measurements with Focus A pads are available somewhere on this site. They do tame the upper mids and highs, but at the expense of air and space.

--

Let's now check out CSDs. There was a discussion about the use of CSDs to assess the "bouncy bouncy" disease of most modern orthos. It's a characteristic I am OK with for entry level orthos, but I think it drives some old guard ortho guys nuts. For some good background reading, see here: https://www.superbestaudiofriends.o...k-ages-of-ortho-tech.10824/page-2#post-344111

Side thought: We really need to move away from ONLY frequency response measurements. They tell us so little. Experiment: surgically digital EQ an HD800 and HE6 so they have the same response. Guess what: the HD800 still sounds like an HD800 and the HE6 an HE6.

@rhythmdevils modded HE560
CSDs - not compensated



We still see an "ortho" wall between 3kHz and 3.5kHz. Before and after this two valleys, which recover somewhat quickly. This may suggest the dips are not as severe the the FR. There is low level fairly short lived ridge at 8kHz that goes with the peak here (the one I had to EQ down). Otherwise very clean with no sight of the bouncy-bouncy ridges.

So how does the fare against the OG HE560?

OG HE560 CSD

 

LOL. I didn't read the spoiler. (Probably on purpose). For what it's worth, I did not measure first. My component chain during the listening period was Pixel 2 -> Modi 3+ (OPA1656) -> special Continuity Magni (not available in stores and never will be, unless...), which is probably brighter than yours. I should have used a multibit DAC.

I think you did a great job toning down the plasticky timbre. Took me a while. Also, no bouncy bouncy, which is a relief! I'll get some distortion numbers (I expect them to be good) and maybe burst attack and decay envelope.

 

Unknown since those CSDs are from OG and RD modded the V4. However, I don't think the HE560 has changed all that much since V1. Also, HFM has been getting worse with respect to the under dampened behavior. It's sort of been their new sound: higher efficiency, thinner diaphragms, smaller magnets = bouncy bounce.

Take Sundara for instance. The sonic effect is less traditional ringing (peaks, although that could be it too). It's hard to describe this underdampened sound. It can be loosy goosy, not tight, bouncy bounce, smeared poop, etc.

 

@rhythmdevils modded HE560
Harmonic distortions vs Frequency at Various SPL

Note: getting some environmental pollution from 20-30Hz (construction a few blocks away)

Measurements look good overall. H3 and H4 barely move with increasing SPL. Only H2 goes up, and only a bit. Performance in the lows is very very good, as expected for a midrange ortho. H2-H4 creeps up slowly past the midrange. Would have liked to see better here; however, overall levels are still low.

Conjecture: could the slightly elevated distortion plateau at all orders (at least H2-H4) from 1kHz upward, even at low levels, account for the "plasticky" timbre in the highs? Could be wrong and we would need more data, particularly more headphones with plasticky timbre. There is typically a small rise (noise) with this microphone and measurement setup; however, it doesn't account for this much.

 

@rhythmdevils modded HE560
Burst Envelope


More or less expected for a good ortho. Decay is good. A bit of overshoot at the first cycle for the attack on 1.5kHz and 4.8kHz which correlates with fast transient attack. There is a secondary bump at cycle 2.5 for 4.8kHz attack. Normally a bump at the cycle 2 would indicate glare, sharp, or prickly treble timbre. However this may be masked by the initial attack. Also cycle 2.5 is slightly past cycle 2. By all indications, a good competent sounding ortho with respect to transient attack and decay.

BTW, the 3700X is MUCH faster than a 1700. I never bothered optimizing my code.

Wow, that's a half day of work! Fostex will have to wait until later.

 

You didn't hear this from me:
https://www.h-ear.audio/2021/03/22/...-list-of-2021-products-going-into-production/'

Biggus Ampius. Be careful what you wish for. That announcement was early on this year. Lots of stuff delayed - supply side issues - around the world - every company. (I was finally able to snag a AMD RX6700). And it's what I said: historically, we never know. It can be months. It can be years.

I plan on buying x2 Tyr. And then hook them up to either RAAL SR1a (actually will let @zerodeefex do that) or iFi iESL to power Koss ESP950s. Just because I can. Just to be an asshole.

 

Yeah, I'm using those puppies with the RAAL. What kind of absurd asshat would I be if I didn't?

 

RD modded HE560v4
4kHz tone at 102dbSPL
FFT size 32768


HD650
4kHz tone at 103dbSPL
FFT size 32768

 

Update: The Fostex did not survive shipping. We will move the JAR600 next.