Seven Focal Utopias Measured

[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.]

Okie dokie. After having measured seven Focal Utopias, some of them numerous times, two with serial numbers lower than the review pair, it now feels like I have enough information to draw some conclusions and move forward. Here's the .pdf booklet with all the measurements. First, the evidence so far.

The following two plots are the averaged raw response of the seven headphones for the left and then the right ear.



As you can see, with the single exception of the pair I reviewed (orange trace), the Focal Utopias measure very, very similarly. I'm not terribly surprised at seeing some difference in bass response—getting a reliable seal on the test head is difficult, and even the smallest wrinkle in the pads will give a small leak and a change in bass response.

I'll also note that Bob's shootout pair do seem to be a little different in the left ear (red trace) between 10kHz and 20kHz.

Some folks in the forums mentioned some differences in the impedance response of the headphones. I gave it a look and sure enough, there is some moderate differences. Here's the plot of all the impedance responses.


The primary driver resonance for most of the headphones is centered around 60Hz and averages about 350 Ohms. The review unit (orange) is the lowest of the group both in frequency and amplitude at 50Hz and 315 Ohms respectively. I'll note that Bob's shootout unit (red) is shifted 10Hz higher in frequency than the rest, and seems to be a little narrower than the others.

I'm not too worried about the amplitude of these readings as here again a small leak can effect the amplitude of the impedance. For example here's a plot that shows the impedance curve with the headphones on the head, and then taken off the head and measured in free space. (At this point in the tester it's just the voltage drop across the headphones, but it is the data used to calculate the impedance plot.)


As you can see the impedance peak is significantly higher (probably in the neighborhood of 600 Ohms), but it doesn't change in frequency. So, the amplitude differences can be chalked up to slight differences in seal, but the frequency shift and change in Q (the quality of the resonance seen in how narrow or wide the peak is relative to its base) of my review sample and Bob's headphones are a bit bothersome...though it wanders around a bit for the others as well.

And a couple of random observations: The Utopia with 310 in the serial number does apear to have a channel imbalance of about 3dB starting below 500Hz. The owner was notified.

When installing the ear pads on the Utopias, you push five little locking nubs that click into place. At one point I though maybe the variations in my pair may have come from someone before me (these were PR agency loaners that go to multiple reviewers) may not have pushed all the nubs in place. So I did do a set of measurements of headphone 971 with the forward upper nub not seated. It's the second to last measurement in the .pdf booklet. You can see it makes the bass roll off quite a bit harder; doesn't effect the height of the 3kHz peak; and does cause the impedance peak to go above 400 Ohms...which none of the Utopias did with earpads properly in place.

And last, I noticed the raw measurements of the review pair (sn 125) varied a lot more than the other headphones. This observation is reinforced when looking at the same headphone with the Elear pads.



None of the other headphones came close to having this much variation. And seeing the same characteristic with differing pads makes me think it wasn't just a measurement error. This seems like the clearest evidence to me that there was something substantially different with that pair.

My Conclusions
It seems evident to me that the pair I got for review was defective. I've seen some posts on forums where people speculate that I got a pair that were tampered with to suite my warmer taste. I find this idea completely implausible. I just got one of a handful of Utopias that were in a pool of headphones at the PR company and got which ever one was available when it got to my turn.

It's far more likely, it seems to me, that either the headphone was bad from the factory, or was damaged while making the rounds to reviewers. Because the problem matches in both ears, I suspect the former rather than the latter. I suppose the headphones may have undergone some extreme temperature and/or humidity exposure during shipment at some point that would have effected both ears the same, but that's a bit of a stretch.

I tend to think there was a bit of a manufacturing glitch that didn't get caught. It happens. Even the best manufacturers in the world, with the best quality assurance practices, aren't absolutely perfect. And I will point out that other than the review pair all the other headphones measure very similarly. Even though there is the one with a channel imbalance in addition to the odd reviewed pair, I tend to see this otherwise highly matched sampling as evidence of really good manufacturing practices.

What's Next?
I've been in contact with both Focal and Audeze. Focal has told me the shootout pair and the review pair will be tested. I've asked to get the review pair back for re-measurement. I've worked with Focal in the past and they have always been extremely responsive and level headed. They've asked me to post this:

Dear all,

We have read Bob ‘Kat’z Corner: Episode 16’ article and your numerous comments about it.

We are currently in close touch within Focal company to understand the origin of these differences between frequency response curves. To do so, the Focal company has decided to get these headphones back in order to analyze, control and measure them. However, we would like to be precise that these headphones are not defective. Indeed, each and every pair of headphones is controlled and measured before being sold and all of them comply with their strict quality control. Moreover, Focal has saved all headphone’s frequency response curves and they ensured us that the concerned models are consistent with their requirements.

Focal and Innerfidelity teams are trying to figure out the differences and we will keep you informed as soon as we more info.

Thank you all for your understanding.


Audeze will be sending me some LCD-4 headphones. There is an enthusiast who is ending me both his LCD-4 and Utopia for another measurement. You'll be seeing a post like this one about LCD-4 measurements in the near future.

Once I have in my hands a Focal Utopia and Audeze LCD-4 that appear to be representative of the manufacturing standard, I will compare them in order to form a more accurate impression of their performance and then write up a post. Then I'll send them off to Bob Katz for a second opinion from him. And then I'll make any adjustments needed to the Wall of Fame.

I'd love to hear your thoughts and suggestions in the comments!

[Editors Note: See this post for Focal's comments.]

metal571's picture

Crazy stuff with that one-off pair you got there. It's almost like your review unit was in Audeze tuning and the rest of production is targeting diffuse field.

Beagle's picture

....that Tyll's review pair were worn in, drivers had some miles and the pads softened? This might account for the FR difference.

Maybe's picture

Decreasing the earpads thickness usually increases the energy at 3kHz rather than decreasing it. In addition Bob's pair definitely also experienced some wear so it seems more likely that Tyll's unit was simply defective.

Serious's picture

I think we shouldn't think of it as a decrease in the upper midrange/lower treble, but rather as a boost in sensitivity below 3kHz. Something must be very different about that one Utopia.

Maybe's picture

Thanks for the correction! I guess I was speaking relatively, or atleast that's what I like to believe.

I honestly can't explain what's up here. Everything from the weird positioning variability to the lower Fs with lower Q.

Did one of the reviewers put it in an oven or something? That could melt the nylon dampening mesh making it less permeable causing acoustic short circuiting to be less extreme in turn giving the headphone more energy below 2kHz. Would also damp the resonant frequency.
The mesh over the driver could then be "dense" enough to filter out that 6kHz spike...

5dB variance from target is something that anyone during production should have noticed. And it even channel matches decently, how (un)lucky can you be?
I'd really like to give Focal the benefit of the doubt but this is a little too convenient.

Ste's picture

If I were to buy a 4 grand headphone I wouldn't really care why it has minor (or even microscopic) flaws: I would want what I paid for. It doesn't quite look that way with the Utopia...

dustdevil's picture

Well then your not the utopia's target customer ;)

Ste's picture


isquirrel's picture

Focal advertise in their specs sheet that the impedance of the Utopia's 80 ohms, looking at your graphs I am confused as this does not seem to be the case. Can you tell me what is the average impedance of the Utopia's please?

isquirrel's picture

Focal advertise in their specs sheet that the impedance of the Utopia's 80 ohms, looking at your graphs I am confused as this does not seem to be the case. Can you tell me what is the average impedance of the Utopia's please?

isquirrel's picture

Focal advertise in their specs sheet that the impedance of the Utopia's 80 ohms, looking at your graphs I am confused as this does not seem to be the case. Can you tell me what is the average impedance of the Utopia's please?

hanshopf's picture

You say you cannot explain different measurements and impressions, but at the same time you are sure, that everything is ok? You know what? I do not believe you.

I bought my Utopia after having read Tyll's review. It sounded very brightish and lacked bass. I got a second pair from my dealer, which sounded very slightly darker, but still was too bright for what Tyll wrote about their reference sound. I even tried them with Elear pads, which made everything worse. I wrote to customer service, but never got an answer.

Finally I decided to sell them and went to a shop to listen to other headphones. I found that LCD-4 sounded muffled and definitely not as good as Bob Katz had described them. BUT: I got to listen to two other Utopias, one of which sounded clearly darker than my own sample. The next day I compared three Utopias in two shops with the same DAC/Amp and the same music, two of them side by side. I have got a relatively small head and was surprised to find, that these Utopias differed slightly in size, the darker sounding sample also having stronger clamping pressure.

Conclusion: I listened to four pairs, the range from the brightest to the darkest sounding one being remarkable and clearly bigger than should be tolerated. Different size gave different clamping pressure. Probably this contributed to the different sound I perceived (and not only me: the dealer wanted to know what is going on and kept my sample overnight and came to the same conclusion).

All this trouble after having paid 4000€. And Focal stubbornly denies that there is a problem.

Beagle's picture

Different clamping pressure changed the sound? Whodathunkit?

thefitz's picture

The "Wall of Fame" pair is a good 5db less hot in the upper midrange, and it got put on the Wall of Fame. Now we're calling it defective. What's going on here? If this god-pair of Utopias were indeed a mistake, shouldn't the Utopia be handily removed from the WoF?

thefitz's picture

Since this god-pair of Utopias rocked the headphone world, maybe the wall of fame should be re-assessed and "cold, analytical" cans like the HD800S will resonate anew.

amham's picture

There seems to be a disconnect with the "excuses" made for these very expensive headphones. Am I the only one who believes $4000 is an obscene amount of money for headphones not to mention that they may be defective or exhibit significant variation? At this price point, and the substantial profit margins along the way, there is no excuse for product variation/defects. It was once, very recently, unthinkable that headphones should cost close to $2000. People, a large segment of the worlds population is starving...

Ste's picture

No, you're not the only one.

Beagle's picture

$4000 for headphones is ridiculous beyond the pale.

arnaud's picture

Hi Tyll,

As you may recall, these headphones driver have no spider and just a very light / thin surround around the cone.

The test you did with the headphone "loaded" (on the dummy head) vs. open, looking at the first mass/spring/damper resonance is interesting as it implies that the effective damping is not just due to the surround but also the rest of the system (airflow through the perforated baffle, pads and other leaks).

Is the sharper / higher peak impedance result when the headphone is tested "open"?

Assuming that effective damping is being affected by the pads and also knowing the pads have large influence on the response in 1kHz up (e.g. shift in tonality when switching from elear to utopia pads), I wonder if the pads are the main culprit in your skewed sample results.


JVG's picture

Interesting idea. Tyll, perhaps it would be worth choosing one of the consistent samples and measuring it with a few different sets of pads, including ones that you intentionally compress? It's just so strange that that one sample would follow basically the same frequency response curve but with the lower end shifted upward 5db in both channels equally, and would basically exhibit no other measured performance difference besides the impedance curve. That doesn't seem to me to point toward physical damage or defect as much as either a discrepancy/defect in the electrical structure of the original review or an acoustical/resonance difference, which given the (I assume) consistent structure of the housing, would have to be caused by the pads or possibly the clamping force. I wonder if any changes have been made to the pad design since your early review sample. If so, then these earlier-production units could still sound and measure consistently with the current-production units if they all have the current-production pads.

Tyll Hertsens's picture
Nice to hear from you, Arnaud!

Yes, the higher peak is in open air. I kind of suspect the pads, but I did do that experiment with the not-properly-installed pads and it didn't seem to effect the area of concern. (3kHz peak.) Dunno.

arnaud's picture

Hi Tyll, I suspect the pads installation issue breaking the seal are way more influential on the first resonance (80Hz, seen in the impedance plot) than at 3kHz up (seen in the microphone response).

What seems to control the 80Hz resonance Q factor are the leaks in the earcup cavity:
1. Earpads leakage (sound transmitting through the earpad)
2. Earcup baffle leakage (sound transmitting through the perforated earcup baffle)
3. Additional leakage (intentional or not such as in case of improper pads installation)

The driver surround obviously also provides damping at you've shown in the open measurement but comparatively less than the other mechanisms above.

Now, the SPL response at the microphone at ~1kHz and up should be governed by the following:
1. Earpads absorption (sound waves not reflecting off the earpad surface)
2. Earpads leakage (sound waves transmitting through the ear pads)
2. Earcup baffle leakage (sound transmitting through the perforated earcup baffle)

While the earpads sound absorption is negligible <100Hz, it can approach 100% at 3kHz or such. Depending the on the pads perforation, the foam (and how compressed it is when installed etc.), I expect very significant variations in absorption performance.

As was posted earlier, I also suspect Focal will be able to make your review sample sound like other production units once refitted with production pads...

Now, I may be completely off, (as we "armchair designers" often are lol), but it would be nice if Focal can get back to you after completing their review of the "defective" unit.


Chiumeister's picture

In light of the latest test results, is the Focal Utopia still on the Wall of Fame?

Tyll Hertsens's picture
Yes. I will be doing another comparison with the LCD-4 after which I will decide if it needs to be moved. It's still a good sounding headphone, but it is different than the one I experienced.
JE's picture

Utopia should go out, HD800/HD800S should go back to WOF.

Audiognome's picture

Wow such expensive headphones and you have problems like this.

GearMe's picture

Hmmm, 6 match, 1 doesn't. What a coincidence! ;-)

Conspiracy theories aside, whether "It's far more likely, it seems to me, that either the headphone was bad from the factory" or "damaged while making the rounds to reviewers" matters very little because The Innerfidelity Product of the Year/World's Best Headphone was an Outlier -- and the other 6 measure consistently.

The question now the REAL Utopias still deserve these banners (or even WOF status)?

donunus's picture

Now with this out of the way, can you say that the average Utopia sounds better? Or is your defective Utopia better for you. It seems to me that the brighter Utopia is not going to be my cup of tea.

Yepp's picture

Definitely the Utopias tested are what Tyll's probably prefer(?)

But this smells very funky to me. Suddenly the 4000$ "awesome" headphone is actually quite bright sounding....

Tyll should take the Utopia from WoF and put HD800S there where it belongs.

Reticuli's picture

Doesn't the variation in impedance plot prove this manufacturing variance is not simply the cushions or seal?