Symphonium Crimson - Official Thread

This is the place to discuss all things to do with the Symphonium Audio Crimson:

I’ve been listening to it for the past couple of days and I gotta say… this one is really solid.

Symphonium Crimson Measurements:
Note - this is a demo unit from the showfloor at CanJam SoCal, so it’s unclear how close this is to actual production units. If there’s any difference I’ll update this once I get one in

B&K 5128:

New data visualization method with preference boundaries:

We’re rolling out a new method of visualizing the data that includes preference boundaries (the shaded area). The bass to treble delta for the shaded region should be the same 10dB as the flat line shown above, just with contour windows for bass and treble as per the preference research this is based on (Olive).

Treat this as experimental for the moment since we’re still determining A) population average compensation for IEMs, B) the preference boundaries for IEMs may be a bit wider in the bass than for over-ear headphones, which this is currently based on.

There’s lots that goes into this, but a brief explanation is that this is being done to help prevent misinterpretations and the common misuse of headphone measurements.

A lot of people who are new to headphone measurements and even those who are more familiar with them regularly get the wrong idea when they see a measurement. One of the reasons is that preference boundaries aren’t typically included in a graph (REW doesn’t have a function for this).

This is a step towards remedying that. More to come on this topic…

GRAS data will also be posted soon.

Here’s a photo!


I was able to measure the Crimson at CanJam SoCal thanks to HBK! Here’s a measurement with the included/stock tips (Azla Sednas).


I will say, while it’s not quite for me in a few spots, I think this one is going to be very popular with folks who hear it. It’s that subtle u-shaped sound that tends to be well received. I think my quibbles are as one might expect, the lower treble is a tad much for me, as is the upper treble. But I’m currently listening to some Yo-Yo Ma on it and it sounds incredible.


Written by Listener


My interest in Symphonium is largely due to them being one of the most consistent IEM manufacturers today. Pretty much all of their recent products have been met with praise, and have become fixtures of the price brackets they occupy. Their products have even made me question my own preferences and ideas about what an IEM ought to be.

I’m almost never a fan of “bright” IEMs. However, the Helios immediately shocked me with its extraction of reverb tails, vivacious and airy treble, and near-unparalleled dynamics. As a kilobuck contender, it still sits head-and-shoulders above any of its competition for technical performance alone, though unfortunately I’ve found its lean tonality loses its novelty over time and reveals itself to be thin enough to not quite be “all-rounder” material.

The Meteor takes a completely different approach, opting for something much more warm, relaxed, and easy to listen to. It is easily my favorite IEM in the “mid-fi” price tier due to its non-fatiguing tonality, even if it’s not really a top contender at its price in any technical category (and has too much bass for my preference).

With the Crimson—their new flagship—Symphonium is seemingly counting on a sonic midpoint between their two best performing products being compelling enough to justify placing it in the flagship seat of their lineup. Does Crimson live up to the expectations placed upon it by being the next in Symphonium’s already stellar lineup? Let’s talk about it.

What we like

  • Some of the best bass you’ll find at this price point
  • Well balanced between tuning and technical performance
  • Noticeably immersive soundstage/imaging

What we don’t like

  • Treble above 10 kHz may bother some listeners
  • Midrange could use a smidge more warmth
  • Not the last word in dynamics at this price point
  • Back-order

    Regular price From $1,499

    Sale price From $1,499 Regular price $1,499.00

    Unit price /  per 

    In Stock

    Frequency Response/Tonality

    Below is the Symphonium Crimson, measured on the B&K Type 5128-B and compensated to the 5128’s Diffuse Field HRTF. The boundaries behind the measurement indicate the gradient of preference outlined by the existing speaker and headphone literature. If an area of the response sits outside these bounds, it’s possible that the average listener may find this tonal coloration to be unpalatable.


    We see that the bass sits rather nicely within the bounds of preference, which means listeners will probably find Crimson’s bass to be unobjectionable at worst, excellent at best. It has no issues with extension, and the amplitude and contour of the shelf is almost exactly where people tend to like it.

    And in listening, indeed this is probably my favorite part of Crimson’s tonality. It’s certainly richer and fuller than the bass shelf of Helios, while also not nearly as overbearing or as potentially stuffy as the shelf of Meteor.

    Crimson’s bass occupies what is essentially a perfect midpoint between the two, combining Meteor’s expansive and enveloping warmth and size, with Helios’ trademark tactility, speed, and bounce. While I’d say Helios gets the edge in texture in the bass, it does so while having a much leaner and less inviting presentation overall. And while Meteor does win in terms of raw quantity, I always felt Meteor had a bit too much bass for its own good anyway.

    Crimson’s bass is big, rich, lumbering, but somehow never slow.

    Interestingly, I noted early on that the bass of Crimson actually sounded rather similar to the bass shelf of Harman 2018 (for over-ear headphones) but transposed to an IEM. It’s almost eerily normal sounding in that it works with basically any music I throw at it with no complaints. I assume the bass here will be a highlight for most people for this reason.


    The midrange offers the first potential speed bump for IEM listeners, as it’s perhaps slightly thin overall. The tilt between lower and upper midrange is just a bit too counter-clockwise tilted for me, being rather close to the un-tilted Diffuse Field target (which on the above graph would be represented by a flat line) until about 3kHz.

    However, it must be said: Most IEMs are significantly thinner and "colder" here. As such most IEMs are much worse offenders than Crimson in this region, especially since it opts to undershoot around 3-4kHz, which helps bring the midrange to a more headphone-like tilt overall. Crimson is just warmed-up enough to not sound sucked out, and it offers competent support to the fundamentals of vocals and snare drums even if it’s not quite as warm as I personally prefer.

    Even with that preference, I think it’s likely that a more warm-tilted midrange paired with Crimson’s large bass shelf could make things too cloudy. How this area is tuned has a lot to do with how we hear the main relationship between fundamental frequencies and overtones in a piece of music, and it’s arguably the hardest area of the tonality to get to cohere with an IEM’s bass and treble.

    Even if Crimson’s midrange is a smidge lean for my taste, it works very well with Symphonium’s bass tuning choices here, and still gets a solid passing grade from me given it rarely presents issues to me and still beats many IEMs in its class for this aspect.


    I’ve said in basically every one of my prior reviews that the Treble section is where IEMs and headphones go to die… and unfortunately Crimson doesn’t escape this.

    I want to be clear that the above 5128 measurement of the Crimson was taken after almost 3 weeks of having Crimson at home for testing, so the following impressions are not at all influenced by the measurement.

    Upon receiving Crimson and listening to my reference tracks, I immediately knew I had a big problem.

    For what it’s worth, I could see much of Crimson’s treble being fine for most listeners. It has a slight flare in a similar spot to the Thieaudio Monarch Mk2 (5-7kHz), but curtails the boost just enough to stop the overall presentation from being a dry, grainy mess like I hear Monarch Mk2 to be. This elevation wasn’t immediately noticeable to me, as it was actually masked by the main issue here: a large 12kHz peak that no amount of tip rolling (and I did a lot of tip rolling) was able to make tolerable for me.

    Now, to be clear, I already know a few people for whom this elevation isn’t an issue. Treble is very user-specific, having a wide range of anatomical as well as preferential variance. Plenty of people will likely hear this peak and appreciate the sense of “detail” it brings to smaller cues like vocal intricacies, minor percussion changes, and reverb tails—the latter of which is something even I can say I like about it at times.

    But for me, it’s more distracting than anything else. Vocal sibilance and breath are impossible to ignore, and it borders on painful at times. Cymbals are splashy and ringy, being pushed much more forward and closer to my eyes than is comfortable. Snare drums at times even sound like they only use a bottom mic instead of blending between the “chains” on bottom head and the stick attack & shell resonance of the top head.

    Unfortunately, very few things I listened to seemed to escape the uncomfortable sizzle from Crimson’s 12kHz peak, and so this ends up being yet another case where a flagship IEM takes a risk by boosting its treble tuning—ostensibly to increase perceived technical performance—that falls completely flat for me. While again, to others this still may not only be palatable, but preferred, I still have to dock major points from Crimson for this choice.

    Technical Performance

    This is where Symphonium’s previous flagship Helios had gone above and beyond its price tag in the past, so I won’t lie: my expectations for Crimson’s technical performance were very high.

    I’d say in terms of dynamics, Crimson and Helios are on roughly equal ground, though they present rather differently. Crimson is more like a closed back headphone, where its dynamics manifest mostly in how hard the mid-bass of kick drums can hit you. Helios, by comparison, is more about portraying pure contrast between all elements and areas of the music, and thus its dynamics are a little more universal instead of just being relegated to the bass.

    Crimson being a warmer tune than something like the Helios means—to my ear—Helios is still an overall more detailed, texture-oriented, and “fast” IEM. Helios has a lean, unencumbered midrange that brings texture and transient attack forward at any cost.

    However, it’s pretty widely agreed that the "cost" in this case is timbral naturalness, as basically everyone has agreed that Helios is a pretty thin sounding IEM.

    Crimson decisively takes a step forward in the realm of timbre, and wrangles back some tonal versatility and warmth that Helios clearly lacks. Unfortunately, in doing so it has to take one step back in technical performance, and sacrifice some of Helios’ nimble celerity for what is ultimately a significantly more well-rounded and listenable presentation.

    That being said, the delta between the two in resolution and dynamics certainly isn’t massive by any means, but I do think it’s likely many people may notice that Crimson is a little less “tech-forward” than Helios.

    Luckily, there’s still at least one thing Crimson does better than Helios and Meteor, and it’s actually the thing I’d say Symphonium’s IEMs have unilaterally done well up to this point: soundstage.

    I really, really don’t care about soundstage in IEMs and headphones. I regard it as an illusory perception borne of tonal coloration (among other things) that is rarely going to be experienced the same way between listeners.

    But it’s hard not to give credit where credit is due: Crimson is probably one of the most spacious and enveloping IEMs I’ve heard. It actually mixes my favorite parts of Helios and Meteor’s technical performance in this regard, combining Meteor’s effortlessness in layering and size with Helios’s eerily discernable reverb trails.

    I’d say even if Crimson mostly plays on the overall level of technical performance of the Helios, the presentation it goes for is more akin to the Meteor. It prioritizes being expansive and fun, with large but well-layered sonic images.


    Cards on the table: purely as a function of personal taste, Crimson is unfortunately my least favorite of Symphonium’s current IEM lineup.

    Meteor is still my favorite, as the warm midrange, relaxed treble, smaller shell, and appealing price still makes it one of my favorite IEMs on the market. Helios is still the most technical overall, and for that reason it has a place in my heart above most other kilobuck IEMs that simply don’t come close in terms of dynamics and texture.

    While yes, Crimson is tuned better than Helios, it loses some of the technical “X factor” that makes me adore Helios in spite of its tonal flaws. And while yes, Crimson is more technical than Meteor, it’s not tuned as well as Meteor for my taste. So I’m left feeling like Crimson, as a set of compromises chosen to maximize the best parts of its predecessors, just wasn’t made for someone like me.

    But I think this all depends on perspective, and really comes down to what the listener indexes for. Instead of the framing I just gave, I could very well see someone saying that Crimson—both tonally and technically—occupies a rather comfy spot between Helios and Meteor. It’s almost certain that a lot of people are going to find Crimson much more compelling than I do for this reason.

    After all, audio is a market of compromises. If we’re going based on well-roundedness, there’s a solid argument to be made that Crimson is the most well-rounded offering that Symphonium has yet released.

    For me though, I can’t help but feel like Symphonium was so close to nailing this, but just took a risk with that 12kHz elevation that—to put it generously—didn’t pay off for me.

    If you want a more technically-proficient take on the Meteor and are willing to try a more aggressive tuning, or want a more relaxed and palatable take on the Helios and are willing to sacrifice a smidge of raw technical performance, Crimson rather indisputably hits these marks.

    As such, I think it deserves its place in the flagship seat as it most completely embodies the values and philosophy that’s underpinned Symphonium’s products so far.

    But if you—like me—are more treble sensitive or want a warmer, darker, thicker option, I think the Nightjar Singularity might end up being a better pick for that set of needs.

    Thanks so much for reading. If you have any questions about this article, feel free to start a discussion on our forum below or ping me in our Discord channel, which is where you can find me and a bunch of other headphone and IEM enthusiasts hanging out. Until next time!

    This is a companion discussion topic for the original entry at
    1 Like

    The new way of showing the measurements seems a bit unusual…and a bit small to make out the details of what means what on the chart. Can prefetence regions simply be added to the usual plots from the b&k?

    That’s effectively what this is, just as a more complete way of showing it.

    You can see here the 10dB tilt from before. One of the problems we noticed with the static tilt representation is that people continued to treat it as a “reference target” in the strict sense. As in, “headphones should match this line”, and indeed people make this mistake all the time with the Harman target as well. The research specifically indicates different preference groups, hence the preference boundaries.

    We’re working on a way to show the raw comparison as well, but we can’t just show default raw measurements, since they’re already being misused all over the place. We routinely see raw graphs from the 5128 be compared to raw graphs from other measurement systems, and that is a categorically incorrect reading of the data.

    And its not necessarily the fault of those misunderstanding this. We’ve even seen well-known and respected manufacturers make this same mistake, or try to compare 5128 data against the Harman target. It’s perfectly understandable since these measurement systems are billed as being the “most accurate” measurement systems. While they are more accurate than what we had in the past - at least with respect to the acoustic impedance of a human ear - they are also different ‘heads’.

    This means the 5128 and GRAS 43AG for example have a different HRTF, and require compensation to the DF standard in order to be compared. That’s the reason for this new view of the data.

    Where within the shaded area would the average listener preference be? The middle? …with even distribution of grey above and below? Or is it one of the solid or dotted lines?
    It seems the chart could use a little more y-axis resolution for easier reading perhaps.

    So… the point is more that the average listener could fall anywhere in this range, depending on which preference cluster they fall in, the Harman paper on segmentation goes into this. We’re currently experimenting with gradients a bit to see if we can show some of the clustering.

    But for bass to treble delta around 10dB (what most people prefer) would be roughly indicated by the static line there. Keep in mind though, that’s not to say any single individual would prefer exactly that line, merely that this is indicative of the bass to treble delta that is preferred by most. We all have different HRTF as well, and so it has to be shown as a range rather than a static line.

    Intetesting how the average listener, at some frequencies, would be at the very edge of the preference region? Surprising. Evenly distributed would seem more intuitive.

    Again, this is precisely the kind of reading of the graph folks need to get away from, and why it actually makes sense to not have that static line there in the long run. I imagine that most people are bound to interpret it exactly the way you just did (so it’s not your fault!).

    It’s not that the average listener prefers a static 10dB line, merely that its the bass to treble delta that is most preferred. So for example, Harman OE 2018 that we all know well uses a bass shelf rather than a slope, but for both of them, its a bass to treble delta of around 10dB.

    Edit: Think of the preference region as a more sophisticated view of the 10dB slope. I’ve only added it here to help explain that this is actually the same as before, just with preference boundaries.

    Time will tell.
    I suspect people that misinterpret the traditional chart by way of the optical illusion(instead of simply seeing the y-axis difference) might be more challanged with this approach. Respectfully.

    Oh they’re definitely finding this change challenging haha. We knew this was going to incur some frustration among those who are used to the old style of reading graphs, but in the words of our science enthusiast Blaine, the people need to learn!

    This was never going to be an easy transition, but we also knew we had to do something about the status quo, since not only do people more commonly than not misunderstand what the graphs mean (even among those who are familiar with reading graphs), but the traditional visualizations of headphone measurements have up until now been an incomplete view of the data.

    That’s not to say that the older graphs are ‘wrong’ or anything, merely that they don’t include key information from the research and end up by their default view misleading people who haven’t read the segmentation paper for example.

    I should also note, upon seeing the video we did on this, Dr. Olive voiced his support for showing the preference boundaries, as did Oratory1990, who has been a significant contributor to what we’re doing here. He actually has another key point that we’re keen to include eventually, which is how to better visualize the range of positional variation. Right now we’re trying to balance all of this with visual clarity, since many of the things we want to add risk making things more cluttered, but showing headphone positional variation is another important part of it.