Focal Utopia Update

[Editor's Note: Through a long and tedious process it was found that the Focal Utopia and Audeze LCD-4 reviewed here at InnerFidelity in numerous articles were found to be not representative of currently manufactured product. This article is therefore not entirely indicative of our current impressions of this product. To get a complete understanding of our evaluation the reader should start with this summary article and work back through the articles leading up to our current understanding.]

More Measurements
Big thanks to a headphone forum member who reached out with an offer to share his Utopia and LCD-4 for me to measure. One more for the Focal for the Utopia measurements .pdf booklet. It's A1BEHF000384 on page 10 and 11. Generally, it matches the rest of the Utopias measured with a couple small things to note. I'll reveal the LCD-4 measurements when I have enough samples for a good comparison.

Utopia sn 384 has a roughly 2-3dB channel imbalance below 400Hz. I thought this might be the pads, so I swapped the pads and remeasured. The imbalance remained, but was reduced. If you scroll through all the measurements you'll see that it is always the right channel that has a higher amplitude, however small, below 400Hz. I checked the circumaural sealed headphones in the allgraphs.pdf, and I didn't see a tendency for the right channel to be higher than the left.

170515_Blog_FocalVAudeze_Graph_384 170515_Blog_FocalVAudeze_Graph384_Padswap

Utopia sn A1BEHF000384 initial measurement top, after pad swap bottom.

I seriously doubt that the right channel bias comes from the factory, so my guess is that there might be a slight mechanical difference in the head's left and right ears may be making the seal slightly different on both sides, which, somehow, the Utopias are sensitive to.

Or, there's a slight difference in the way I mount the headphones on the head when I measure. The latter seems a bit unlikely as I monitor a 30Hz square wave as I place the headphone, and I can see the bass response there. I do try very hard to get this to match in each position; I adjust the position and fit of the cans on the head while watching the o'scope to try to make them match when I measure. I did see this imbalance with the 384 headphone while measuring, and did try with each of the 5 measurements to optimize the match.

I think I have to chalk it up to the mechanics involved between the left and right ears with the Utopias. Given the change when I swapped pads, it looks like there's about a 1dB bias towards the right channel with Utopias on my head. Knowing the vagaries of headphone measurements, my take is that sn 384 may have a real 1dB channel imbalance, with the rest of the imbalance just random chance or a difference left and right ears with the Utopias. A 1dB channel imbalance is acceptable to me in terms of product QC of a finished headphone.

Your speculation welcome in the comments.

The other thing to notice is that while the amplitude and shape of the peak at 3-4kHz matches all the Utopias (except the review pair), the peak at 6kHz is very unstable and changing with position on the head. You can also see some of this on sn 812, 310, 125 (the review pair), 65, and, interestingly, in the right channel of 971 when the pad was not properly mounted. You can also see one or two raw measurements on most of the other Utopias where this peak has a a differing measurement.

I went back to my spreadsheets where I can zoom in on this area and identify the channel and headphone position for the raw measurements and couldn't find any pattern as to when or in which ear this peak might be high or low except that it might be low slightly more often in the right ear. My spidey-sense says this particular feature is very sensitive to the shape of the pinna in the ear, and may or may not be heard by individual listeners.

Focal's Plans
As I've said before, Focal has been extremely responsive communicating with me...and with you. Here's another communication they've asked me to post:

Dear all,

We understand your numerous comments and concerns about this difference between Tyll's Utopia #125 and other Utopias.

We confirm you that all the headphones manufactured have successfully met all our requirements and complied with our strict quality control. We have saved all headphone's frequency response curves and, after a complete checking, we have no doubt that all of them are consistent (including #125). Tyll's last review on the 6 other Utopia headphones confirms it moreover. Our production is stable and such a difference of 5dB wouldn't have passed our final quality control for sure.

Concerning #125 , we are aware that the frequency response current curve of this model is apparently not normal at all. However, without this pair in hands and without having re-measured and analyzed it, we cannot jump to conclusions for the moment.

There may have been an incident between the moment in which the headphones have been controlled and Tyll's review; but once again, we need these headphones to precisely understand what happened. Which is why we are currently trying to track the itinerary of these headphones (shipment, other reviews, etc.) since its manufacture and want to ship it back in France to make new measurements.

We ask you please to wait for that crucial step so that we can explain you the reason of these differences between the quality control and the review on that specific Utopia headphones.

Thank you all for your understanding,

Best regards,
The Focal Team

That bit up there, "There may have been an incident between the moment in which the headphones have been controlled and Tyll's review..." is becoming more and more plausible to me. At this point I've begun to believe that some reviewer in Arizona or some other hot place left them in in back of his black car for a few days and they overheated. We'll see.

COMMENTS
metal571's picture

I don't own a Utopia. Um, Tyll?

thefitz's picture

Make sure you send them back to me, k?

<.<

>.>

Tyll Hertsens's picture
Sorry. Text changed. Metal571 is sending in some other cans for measurement and I snagged the wrong email addy. They'll go back home proper today.
cclragnarok's picture

>> At this point I've begun to believe that some reviewer in Arizona or some other hot place left them in in back of his black car for a few days and they overheated.

Another guess: A reviewer "burned in" the headphones at excessive volume levels.

--------------'s picture

Why not test the overheating hypothesis with a cheap pair of headphones?
Measure the headphones, put them in your oven for a few hours, take them out and measure them again.
Here's some inspiration for heat settings: http://heatkills.org/how-hot/

Impulse's picture

Seems like that's likely to open a can of worms. I mean I live in Puerto Rico, anything sent here is likely to spend at least 2 days being moved around in a hot USPS/UPS van thru 90F and 90%+ humidity... :/

Dan Wiggins's picture

Most headphone factories do pretty standard environmental testing which includes 100 hour soaks at 70 deg C, then check for same FR as before the soak. This includes the place that builds the Focal Utopias (from personal experience). Given this, I personally discount the concept of heat-related failure.

My guess - an overzealous application of high power to the headphones by an earlier user, and potential damage to the surround. One of the things you see when a surround breaks down is a softening of the high frequency extenion.

Only other thing I can think of - wrong damping fabric. Dropping the HF peaks by 5 dB is what I'd expect to see moving from a Nittoku 150T to a 380S - extra damping resistance. That should have been caught during production, but it would have to be measured as the two fabrics appear very similar but function very differently.

Tyll Hertsens's picture
Interesting comment. Makes a lot of sense in terms of someone just overdriving it. It would have roughly the same effect in both ears, and a damaged surround might account for the variability through the mid-range seen in the raw measurements.
Dan Wiggins's picture

They thought they were from Flips, not from Focal? So you can turn them around and use them as speakers? :)

Impulse's picture

Thx for sharing said info

tony's picture

Phew!

I've seen the Focal's diaphragm deflect from a gentle breath, it reproduces a rather full range of music from the 1 milliwatt level of energy. This might be the outer edge of Pressure Transducer technology. I've never seen any mechanical device this sensitive.

Slight variations in db at specific frequencies could be attributed to nearly any part of the system: durometer of the suspension material ( slight variations ), wiring, solder and a wide range of other specifics like humidity and atmospheric pressure.

Bob Katz, Tyll and that group Bob assembled seem to establish the idea of the human ear being even more sensitive than our instrumentation. If true, how can a manufacturer produce to levels beyond what they have tools to evaluate? ( unless they use skilled tools like Bob's panel of professionals, which they might aspire to )

The Race in On
Who can produce a pressure transducer as sensitive as the Human Ear. We have China, Europe and the USA competing. Katz seems to be leaning to Audeze but it's early.

The Benefit
Electronic Manufacturers will be able to refine their work to the point of being able to re-produce "Live" music to a much higher level of accuracy if they have an accurate transducer to guide their designers.

Tony in Michigan

Dan Wiggins's picture

As sensitive as headphone transducers are (and I've designed dozens over the last 5 years), microphones are even more sensitive! Much thinner materials, more delicate, tighter tolerances...

As far as the human ear vs. measurement gear, I am a firm believer that measurement gear is orders of magnitude (in bandwidth and dynamic range) better than the human ear. What is different is WHAT to look for. The wetware (brain) that processes stuff coming from the ear is quite a bit more sophisticated and intelligent than what your measurement gear can do.

You can measure ANYTHING that you can hear, and do it with a level of precision way beyond what the human body can do. However, knowing WHAT to measure, or what to correlate inside measurements, is the tricky part. Our ear can listen to conversations "below the noise floor" because our wetware can do signal correlation and discrimination to pick it out. We can do the same thing with measurement gear if we decide to (it's how RADAR works - oversampling to pull out correlated signals).

Designing transducers and systems really is not that hard; it's figuring out WHAT to design and what is important for the wetware processing that is hard!

tony's picture

Thanks for writing.

You have obvious insight in these matters, which makes me think that others also understand the intricacies of delicate transducers. ( I never thought of microphones ).

From a mechanical system point of view, I felt the various test reports were indicative of a Hysteresis arising from manufacturing methods.

Tyll's measurements seem ( to me ) not to be indicative or conclusive considering the wide range of listener's experiences.

But maybe they are.

This Katz listener group and Tyll might be 'tuned' ( the way a Formula 1 Grand Prix driver works at much higher levels than the Shadetree Mechanic ).

I did a Google search of Dan Wiggins to discover Periodic Audio and a Jana interview. Hmm. You mention that IEM is the future device.

Now, here you are again, shining light on a problem, helping us understand the manufacturing process and making me wonder why Focal isn't also right here pointing out various things.

I suspect this superb headphone transducer concept is being driven by the Pro Audio people striving to output better products. Everybody is upping their Game. ( the Mytec Dac claims to have 192 db of dynamic range capabilities, for gods sake ) Phew, ouch

You, Mr.Wiggins, seem to be going in that direction, can I catch a lift?

Tony in Michigan

Emersob's picture

Could definitely be heat. I've damaged a pair of headphones by leaving them on top of desktop hp/dac units that gets pretty hot for hours/days at a time. My stupidity + Mytek Brooklyn heat signature = Broken Hifiman 560. Although very convenient, desktop DACs don't make good HP stands.

berita poltik's picture

wow, nice article

berita poltik's picture

wow, nice article

may i copy it?
but i will refer to your site

btw, if you have time, you can visit me back :D
Berita Politik | Berita Terkini | Berita Unik

GearMe's picture

I'm sure the 'investigation' results of the sound signature variance for Tyll's review Utopias (2016 Innerfidelity Product of the Year/World's Best Headphone) are useful in a QA and process improvement sense.

That said, it seems as if the more measurements are taken of additional sets of Utopia's, the more it appears that the review sample is an 'Outlier'.

So...since the other Utopias measure consistently, why not simply write up a new review on the real Utopias and let the chips fall where they may?

Argyris's picture

"So...since the other Utopias measure consistently, why not simply write up a new review on the real Utopias and let the chips fall where they may?"

I would say at this point this is necessary. The review sample IF received is clearly not representative of the Utopia in general, and to leave things as they are could prove misleading. I'm not saying IF has done anything wrong; Tyll reviewed the headphone he was sent, and there was absolutely no reason to suspect that, for whatever reason, it was significantly different from other samples. I also don't see any evidence to suggest that Focal did anything wrong, either. It's a blameless incident, but nonetheless it happened, and it needs to be addressed.

In the short term, I think a note should be placed on the Utopia review, linking to the last few articles investigating this issue. In addition, the headphone should be suspended from the WoF pending re-evaluation with a more typical example, since it's clear that the headphone buyers will be getting is significantly different from (and arguably significantly worse than) the sample that made the WoF.

TFB's picture

My first pair of Utopias had a channel imbalance where the right ear sounded louder than the left. I sent it in for warranty and got a brand new pair. The new pair also has the exact same issue. The right ear is louder by 1db (adjusted in Jriver to +1DB to left sounds balanced). Do not discount some factory production factors at play where the sound in the right ear is always louder when there is an imbalance.

amartignano's picture

I think that one interesting thing is that the probably damaged Utopia sounded suprisingly good and topped the Wall Of Fame, while correct Utopias appear to tend to be brightish.

GearMe's picture

.

100VoltTube's picture

I also find this intriguing.

If we can figure out what caused this anomaly, we may be able to emulate it to improve (or damage in a way that causes subjective improvement) other Utopias.

GearMe's picture

For those folks that bought Utopias without getting to demo them, this would definitely help them if they were looking for the 'Review' sound signature instead of the normal signature.

Personally, I can't fathom buying a $4K headphone period. However, if I did, I'd sure have demoed it first! I certainly can't imagine having to mod a $4K headphone to get it to sound good. Strike that -- for $4K, you should get awesome...not just good...or worse yet 'cold', 'analytical, or 'brightish'

I bought my HD800's for less than a grand knowing full well all of their 'unique' qualities and also knowing that I wouldn't mod them. That said, rolling tubes allows me to 'mod' them a different way.

Visigoth's picture

That's a ridiculous statement. I live in the Caribbean, there isn't a Utopia to "test" in a 1000 mile radius. So even though I want to spend $4000 on the Utopia, there's no way for me to demo it. I therefore have to rely on reviews like these to determine if they really are the best in the business - and IMHO they certainly are.

sszorin's picture

In the light of finding out that the Utopia headphones are clearly not what they at first appeared to be, your statement that these headphones are "the best in business" is a ridiculous statement.

Beagle's picture

We could measure a few Utopias after a year and/or several thousand hours use.

Rthomas's picture

Tyll, thank you for making the effort to do this. I have heard that Focal will not repair second hand Utopias for the second owner even if he is willing to pay.

Could you please check with them if this is the official policy?
I would also like to know how much a single driver replacement costs for this headphone.

It took me 30 seconds to find out from Sennheiser that they charge $340 for single or double driver replacement. I called Focal in the UK but they did not know.

These two pieces of information would be helpful for all of us considering used Utopias.

Thank you

castleofargh's picture

you found 2 pairs of utopia with signature variations and somehow we must leave not stone upturned until we get the truth.
OK... but why?
even keeping the very same protagonists, I can remember you or Bob Katz making mention of some pair of stax not sounding as pleasing as some other pair of the same series. and same for one or 2 audeze models. just your measurement list shows more or less significant variations as soon as you put 2 or more headphones of the same series on it. yet I don't remember you moving to defcon1 for such a reason before, so what's so different this time?

GMP100's picture

Maybe I am getting paranoid by all this but shouldn't you also check if the elear was defective?

100VoltTube's picture

I agree that the Elear should also be tested again. I heard an Elear at a meet a little over a month ago, and my impressions were considerably different from yours.

Pages

X