CSS Criton 1TD-X Kit Speaker Review

[TheBatsEar]

Sure but can you say for sure from these measurments that these speakers sound bad to peoples ears?

My intuition tells me they don't sound bad, because the measurements aren't bad.

However, there isn't really much we can say with certainty beyond the measurements. @amirm uses a Klippel which calculates the speakers sound as if they where in an echoless chamber. The Klippel software also calculates an in-room response, but they don't calculate how they sound in your living room. Also, Klippel doesn't take into account what your blood sugar is or the amount of sleep you got. It's a guess.

And my guess is they don't sound bad, based on the measurements.

As for peoples ears, I can tell for sure that in a blind test most people could not differentiate these from a pair of KEF R3 or Focal Chorus 806. Most people can't tell the difference between de-clicked LP, MP3 and CD.
And by most people i mean most hobby audiophiles. Like you. Like me.

 

[droid2000]

That's the opinion of a couple of the more popular YouTube reviewers--they hold that CSS speakers in general have better low end response than similarly sized bookshelf speakers.

I don't have a lot of experience to compare with other stand mounts in my actual home, but these go far lower than my old Linn's that had an F3 of 50Hz. I only bought that subwoofer in the pic to fill a null at my MLP. Graph is 1/3 smoothing no sub.

 

[ROOSKIE]

And my guess is they don't sound bad, based on the measurements.

As for peoples ears, I can tell for sure that in a blind test most people could not differentiate these from a pair of KEF R3 or Focal Chorus 806. Most people can't tell the difference between de-clicked LP, MP3 and CD.
And by most people i mean most hobby audiophiles. Like you. Like me.

I'd agree that they likely don't sound bad and to many listeners will sound just right - meaning great. People have been enjoying 6.5" drivers paired with a 1" flat baffle tweeter in a sharp edged box in many iterations for years.
These are a pretty fine measuring example of that class of speaker.

That said if CSS wanted to they could design a wave guided version as an option and I think they should as they can likely do a great job based on this example. The cabinet would need zero changes if they used a well designed waveguide. They could just reposition the driver cutouts with the new size and just offer the variation.

I hear you on the MP3, CD testing.

In terms of the speakers I do think there are meaningful differences. I am guessing as I have not been doing blind testing.

In my sighted testing I did not find the R3 and Chora 806 sound particularly similar. Pretty different depending on one's viewpoint, more different than some other examples I have dealt with. I have not heard of CSS. Just based on the data I can't say for sure how it would play out.
CSS vs the KEF = fairly wide vs fairly narrow dispersion and then the stunning refinement with more pinpoint imaging in that KEF & the CSS is unknown to me??.
CSS vs just the Chora I am guessing would present some meaningful differences. The bass on the CSS would be stronger as the Chora in my space sounds more like a 5.25" driver vs the 6.5" it actually uses and that Focal tweeter/treble has a distinct sound at least when I am aware of it being used.
I'd just love to try this all blind.

Anyway, wondering out loud. Who knows. Who can host the 1st ASR blind test? I can't host at this time but I can bring some speaker units and some snacks.

 

[Maiky76]

This is a review and detailed measurements of the CSS Criton 1TD-X Kit speaker. I requested a sample from the company and shipped a pair to me already built in nice veneer:
View attachment 263965
I believe the kit costs US $770 including an MDF flatpack. Typical of DIY kits, you get the value add of very heavy construction likely due to (3/4?) thick MDF panels. I would say it is nearly twice as dense/heavy as any speaker costing near its its KIT price. Not much to see on the back side:
View attachment 263966

Company was a delight to work with, being highly responsive and quite patient as I got to test these speakers.

Measurements you are about to see are performed by Klippel Near-field scanner. Frequency response measurements are "anechoic." Reference axis is the tweeter center.

CSS Criton 1TD-X Measurements
Here is our usual "spinorama" CEA-2034 frequency response graphs:
View attachment 263967

The on-axis as noted is generally flat and good, sans a few dB here or there. What stands out is the directivity error around crossover point where the woofer has gotten directional but the tweeter is not. This is due to lack of large waveguide around the tweeter to match the directivities of the two drivers.

Sensitivity is less than specifications and I had to push the speaker by 4 to 5 dB to get my usual reference measurement level. Be sure to have good bit of amplification power to drive these speakers.

Near-field response shows a resonance from the port/cabinet and and another from the woofer outside of its range:
View attachment 263969

Early window reflections have a bit of deviation due to the directivity error:
View attachment 263971

But interestingly enough sum up well with on-axis to produce a good predicted in-room frequency response:
View attachment 263972

As predicted from the spin, there is directivity error:
View attachment 263973
View attachment 263974

The wider response can be a benefit or not, depending on your room situation. Vertical response is typical for a 2-way speaker so stay at tweeter axis relative to your ear height:
View attachment 263975

Distortion performance was very good at 86 dBSPL and 96 dBSPL with latter above low bass:
View attachment 263976

View attachment 263977

Impedance is above average which makes it an easier load for an amplifier (requires less current):
View attachment 263981

Finally here are CSD/Waterfall and step responses:

View attachment 263978

View attachment 263979

CSS Criton 1TD-X Kit Listening Tests and Equalization
Note: my allergies are acting up and my ears are rather plugged up!

First impression was fairly positive with my female vocal tracks. After a bit though, I found the sound a bit bright so applied different EQ filters and landed on these:
View attachment 263982

Fully correcting to on-axis response in upper bass/mid-range was challenging as it improved things on some tracks, but made others boomy. So I had to back off as you see. Bass response was very good but if you cranked it up too much, it would gradually get distorted. I played with a high pass filter but at the end, I didn't like it. Proper correction in bass region requires room measurement to make sure one knows the interaction with the room modes. Other that, the sound was excellent especially on vocals, instrumental and techno music.

I was impressed by the dynamics of the speaker. That woofer has fair amount of excursion compared to what we typically see in speakers in this size. Yes, as noted, it tries to play too low of a frequency and can get distorted. Other than that though, this speaker can play loud, really loud! Despite only using one speaker, I was able to fill a very large space with very enjoyable sound.

Note that I was driving the speaker with 400+ watts of power so as noted, you need to give it fair amount of juice (if you want to listen loud).

Conclusions
The CSS Criton 1TD-X design has benefited from some good design choices such as a woofer with impressive power delivery with good on-axis and predicted-in-room frequency response. There are minor imperfections here and there but at higher level, it achieves very good fidelity. Subjective experience with a bit of filtering was very good. As a kit, it will give you the satisfaction of building something you can take pride in.

Let me also repeat the willingness of the company to have its speaker subjected to our rigorous testing where others run away fast. Or don't even respond. These things score big points in my book.

I am going to put the CSS Criton 1TD-X Kit speaker to my recommended list.

----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Here is my take on the EQ.

Please report your findings, positive or negative!

The following EQs are “anechoic” EQs to get the speaker right before room integration. If you able to implement these EQs you must add EQ at LF for room integration, that is usually not optional… see hints there: https://www.audiosciencereview.com/...helf-speaker-review.11144/page-26#post-800725

The raw data with corrected ER and PIR:

Score no EQ: 5.3
With Sub: 6.9

Spinorama with no EQ:

• Poor directivity
• Port!
• Typical bookshelf with no waveguide.

Directivity:

Better stay at tweeter height or just above.
Horizontally on axis.

EQ design:

I have generated two EQs. The APO config files are attached.

• The first one, labelled, LW is targeted at making the LW flat
• The second, labelled Score, starts with the first one and adds the score as an optimization variable.
• The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.

Score EQ LW: 5.4
with sub: 7.1

Score EQ Score: 6.4
with sub: 8.1

CSS Criton 1TD-X APO EQ LW 96000Hz
February132023-113501

Preamp: -2.5 dB

Filter 1: ON PK Fc 38.35,    0.00,    1.25
Filter 2: ON PK Fc 84.39,    -1.92,    1.16
Filter 3: ON PK Fc 361.68,    1.47,    0.97
Filter 4: ON PK Fc 643.09,    -1.93,    5.63
Filter 5: ON PK Fc 1737.88,    -1.14,    1.85
Filter 6: ON PK Fc 3430.73,    1.85,    0.92
Filter 7: ON PK Fc 5039.17,    -0.75,    6.67
Filter 8: ON PK Fc 9359.17,    2.34,    1.83
Filter 9: ON PK Fc 14696.76,    -2.96,    3.95

CSS Criton 1TD-X APO EQ Score 96000Hz
February132023-113501

Preamp: -2.5 dB

Filter 1: ON PK Fc 39.74,    0.00,    1.39
Filter 2: ON PK Fc 81.18,    -2.13,    0.85
Filter 3: ON PK Fc 356.87,    1.21,    0.85
Filter 4: ON PK Fc 629.90,    -1.84,    3.80
Filter 5: ON PK Fc 2121.62,    -1.00,    2.48
Filter 6: ON PK Fc 2457.60,    0.87,    2.00
Filter 7: ON PK Fc 4906.61,    -1.32,    3.57
Filter 8: ON PK Fc 9794.10,    3.62,    3.02
Filter 9: ON PK Fc 14039.46,    -3.52,    1.68

Spinorama EQ LW

Spinorama EQ Score

Zoom PIR-LW-ON

Regression - Tonal

Radar no EQ vs EQ score
Nice improvements

The rest of the plots is attached.

 

[lurkera]

The wider response can be a benefit or not, depending on your room situation.

Did this translate into a wider sounding soundstage compared to similar sized speakers? I recall it being mentioned here that this measurement sometimes does.

 

[droid2000]

Did this translate into a wider sounding soundstage compared to similar sized speakers? I recall it being mentioned here that this measurement sometimes does.

Don't sweat details. The room will play a larger difference in sound than those measurements.

It's instructive to notice that no one posted better measurements for an alternative speaker for the same price despite many claims

 

[Paweł L]

My question regarding drivers freq response: where that 5kHz resonance comes from. Looks a bit scary and it doesn't appear on the manufacturer data. Somehow I feel CSS could have done better job with filter the 'junk' above 3 kHz. But the complexity of the filter increases and beginners might have problem assembling the crossover.

 

[PeteL]

My intuition tells me they don't sound bad, because the measurements aren't bad.

However, there isn't really much we can say with certainty beyond the measurements. @amirm uses a Klippel which calculates the speakers sound as if they where in an echoless chamber. The Klippel software also calculates an in-room response, but they don't calculate how they sound in your living room. Also, Klippel doesn't take into account what your blood sugar is or the amount of sleep you got. It's a guess.

And my guess is they don't sound bad, based on the measurements.

As for peoples ears, I can tell for sure that in a blind test most people could not differentiate these from a pair of KEF R3 or Focal Chorus 806. Most people can't tell the difference between de-clicked LP, MP3 and CD.
And by most people i mean most hobby audiophiles. Like you. Like me.

OK, that's possible, although for LPs I'd be surprised since they are mastered differently, but may I ask why you decided to adress this to me specifically? I haven't done the blind tests you mention, so I can't comment, but what does it have to do with what i said?

 

[TonyJZX]

at $2,300 it faces very very tough competition

to me it sounds like they wanted to build the best they could in a 7.5" driver box but really... once you ask $1,000 let alone $2,300 then you may as well not limit yourself to that size driver...

 

[Toni Mas]

My question regarding drivers freq response: where that 5kHz resonance comes from. Looks a bit scary and it doesn't appear on the manufacturer data. Somehow I feel CSS could have done better job with filter the 'junk' above 3 kHz. But the complexity of the filter increases and beginners might have problem assembling the crossover.

Note that the on axis response is pretty flat too, and as usual the xover has been optimized to measure so.
The problems above the xover point is fue to the way this speaker interacts with the room and its boundaries, and is the result of the change of directivity around the xover point. The radiation of the 7' woofer is narrowing very early, from 1khz aprox, while the small diameter tweeter is almost omnidirective up to over5k.

This results in an excess of energy in this área that boosts the in room response in this área, and though the direct sound (tied to on axis anechoic response) is flat, the perceived tonal balance of the speaker cannot be smooth. You get with this speaker and any similar speaker (7' W+ 1'T) the typical "tweeter Bloom" which is not a problem of the drivers but a side effect due to the design.

 

[TheBatsEar]

what does it have to do with what i said?

Not much, I was mostly just rambling into the universe. I guess triggered by you questioning who had actually listened to the speakers:

Wondering how many of those "many others that have commented here" heard this speaker. Me what I read is that the one who did recommends it.

To me subjective evaluation is meaningless. Ok, so @amirm liked them. That doesn't mean i like them, as i will not hear what he heard. However, the measurements on the other hand give me a pretty good picture that these don't sound terrible, and that is far more valuable to know.

Besides, @amirm has not recommended the WiiM Mini, so his subjective opinion is ignored in this household.

for LPs I'd be surprised since they are mastered differently

They are? I thought they are mostly pressed from a CD "master" in the last 25 years or so?

at $2,300 it faces very very tough competition

The kit is $775 without MDF. You could easily build and veneer a MDF box in your kitchen so you are below $900.

 

[Rick Sykora]

While I think we are getting better at using measurements to predict how a speaker will sound, I would only use to determine whether I might purchase to listen to it and then decide whether to keep. In my mind, any review is about improving your odds for a successful purchase. Because of room interactions (and perhaps some other intangible qualities), you should audition speakers in your intended setting. While eq can be helpful, unless the speaker can be corrected or is a really great value, I would avoid any speaker that you need to eq to correct aspects that the designer should have corrected originally.

A DIY speaker design needs to pass a few more key tests in my opinion and that is partly because it cannot be returned and has a reduced resale market:

1. I suggest it needs to be a better value (without including the value your time) than a comparable commercial speaker.
2. It should measure significantly better than what you might purchase off the shelf.
3. You want something custom that is not found in the mass market (design, form factor, aesthetic or other).

I realize that most experienced DIYers already apply some comparable test before they invest. Am hoping this post helps some of the others who are less experienced or pondering whether a DIY speaker is worthwhile.

 

[TonyJZX]

rick has clarified things as a the 'house' builder

i look at this kind of like the kit car industry... its quirky... the potential customer is also quirky

typically these companies are factory five, superformance of south africa and the caterhams of the uk

some people just want a turnkey experience like the Corvette C8 or a $2,300 built speaker or the Revel equivalent

some people are happy to spend some hobby time building the base mdf kit at $900

some people are happy to fashion their own cabinets and just need the cone parts for $700

as rick is suggesting if you go to the trouble of building a $700 or $900 kit then your expectations is that it betters conventional speakers up to 1.5x the cost (lets be reasonable)

if the kit you finished is just comparable to a mass market speaker of the same cost then what did you waste your time for? unless you like the build experience?

i would also agree that kit speakers should offer something not found in conventional boxes

as far kit cars goes, you are getting a unique ownership experience... what drives like a Lotus 7?

in speakers you would want some kind of strange design, like transmission line, that would justify its existence

another way i look at it is that its come to the point that LABOR is now a very low cost component... a typical speaker made in china has very low labor costs

and so why is MY LABOR competing against this???

i would actually prefer a $2,300 speaker is made in china so more money goes into box and components rather than labor

 

[sarumbear]

Creative Sound Solutions Criton 1TD v2 Bookshelf Speaker Review

noaudiophile.com

Read about and heard these...............

One this is for sure CSS, under new management, are doing a big marketing push to the DIY fans.

 

[Paweł L]

Note that the on axis response is pretty flat too, and as usual the xover has been optimized to measure so.
The problems above the xover point is fue to the way this speaker interacts with the room and its boundaries, and is the result of the change of directivity around the xover point. The radiation of the 7' woofer is narrowing very early, from 1khz aprox, while the small diameter tweeter is almost omnidirective up to over5k.

This results in an excess of energy in this área that boosts the in room response in this área, and though the direct sound (tied to on axis anechoic response) is flat, the perceived tonal balance of the speaker cannot be smooth. You get with this speaker and any similar speaker (7' W+ 1'T) the typical "tweeter Bloom" which is not a problem of the drivers but a side effect due to the design.

I understand quite well what you just wrote. My question is the woofer's resonance around 5k (huge spike), which shows separate frequency response for the woofer and tweeter in Driver Components Near Field . Looking at the woofers freq resp it looks as if there wasn't enough attenuation, unless the near field measurement by Klippel is not what we would see in regular sweep at the distance of 0.5 m. The waterfall plot in spectral decay doesn't show obvious 5k related resonance, so it must be Near Field related artifact. I understand that the crossover is around 2kHz looking at the tweeter's freq resp which would be usual choice regarding 7" woofer and 1" dome tweeter.

 

[Toni Mas]

My question is the woofer's resonance around 5k (huge spike), which shows separate frequency response for the woofer and tweeter in Driver Components Near Field . Looking at the woofers freq resp it looks as if there wasn't enough attenuation, unless the near field measurement by Klippel is not what we would see in regular sweep at the distance of 0.5 m. The waterfall plot in spectral decay doesn't show obvious 5k related resonance, so it must be Near Field related artifact. I understand that the crossover is around 2kHz looking at the tweeter's freq resp which would be usual choice regarding 7" woofer and 1" dome tweeter.

1. I simply don't understand those Klippel nearfield measurements. If were done with another software ( I use Arta) I would say they are not very reliable and simply help giving some hints. But I have no idea about how the Klippel stuff works. Personnaly I only do nearfield measurements to get the response a low and mid low frequencies. Above that, I do the same you say: gated semi anechoic measurement at 0.5-1m and i merge both to get te global response.

2. It is true that like most similar muscular midwoofers this CSS show a huge breakup mode around 5khz which requires at sharp crossover with a notch intended to deal with that bump, usually with a classical LR24 target. But without more info about the crossover and no other measurement of each way than the nearfield one it is impossible to say.

3. The Xover seems to do its job fairly well on axis, but not so well off axis, and then again, it is impossible to know what is going on. The directivity issue will surely impact, but there might be something else from the woofer side.

 

[sejarzo]

It's instructive to notice that no one posted better measurements for an alternative speaker for the same price despite many claims

Depends on what is meant by "same price." Kit, or a fully finished speaker?

The Ascend Acoustics Sierra LX is $1500/pair, substantially less than the finished CSS. 6"W/1"T, no wave guide, F3 ~ 50 Hz. The graphs below are from the Ascend site.

It's also returnable under their 30 day satisfaction guarantee if it doesn't work for your room. Unsure what a finished pair of CSS would go for on the used market if you put in all the labor to build and then decided they weren't your cup of tea. The guy who put together the CSS, liked the sound, but wasn't happy with the fit of the baltic birch cabinets and what it would take to finish them given the problems was willing to sell them to me for $500.

 

[TheBatsEar]

I simply don't understand those Klippel nearfield measurements. If were done with another software ( I use Arta) I would say they are not very reliable and simply help giving some hints. But I have no idea about how the Klippel stuff works.

It will end in tears but let me try to explain what i think the Klippel rig does:

A robotic arm with a microphone measures at a lot of points on a sphere (or two spheres?) around the speaker. The distance to the speaker is never larger than that from the microphone to the middle of the sphere, say 1m. Sound needs 0.3ms to travel 1m. By ignoring anything coming from the microphone that takes longer than 0.3ms, you basically have ignored any room reflections.

You have now a anechoic measurement, but you measured hundreds, maybe even thousands of points around the speaker, instead of usually just one or a few points. The software can now calculate how the sphere looks in 2m distance, 3m distance and so on.

From this data you can calculate with confidence how the speaker sounds in a room, if you have solved the math, as Klippel did.
Since @amirm is refusing to measure and model our rooms, he uses a "default" room provided by Klippels software for the in room prediction.

Please correct me where I'm wrong.

 

[Toni Mas]

It will end in tears but let me try to explain what i think the Klippel rig does:

A robotic arm with a microphone measures at a lot of points on a sphere (or two spheres?) around the speaker. The distance to the speaker is never larger than that from the microphone to the middle of the sphere, say 1m. Sound needs 0.3ms to travel 1m. By ignoring anything coming from the microphone that takes longer than 0.3ms, you basically have ignored any room reflections.

You have now a anechoic measurement, but you measured hundreds, maybe even thousands of points around the speaker, instead of usually just one or a few points. The software can now calculate how the sphere looks in 2m distance, 3m distance and so on.

From this data you can calculate with confidence how the speaker sounds in a room, if you have solved the math, as Klippel did.
Since @amirm is refusing to measure and model our rooms, he uses a "default" room provided by Klippels software for the in room prediction.

Please correct me where I'm wrong.

Yes but the nearfield response cannot be taken that way. It is not anechoic nor semi anechoic, it needs to be taken with a close mic measurement, <1cm from the driver, an is no longer valid above a frequency limit depending on the size of the driver, and is mainly useful to get the response at low frequency only, hardly valid above 500hz.

 

[sejarzo]

Yes but the nearfield response cannot be taken that way. It is not anechoic nor semi anechoic, it needs to be taken with a close mic measurement, <1cm from the driver, an is no longer valid above a frequency limit depending on the size of the driver, and is mainly useful to get the response at low frequency only, hardly valid above 500hz.

Take a look at this and explain why you think Dr. Klippel has it all wrong.

https://www.klippel.de/fileadmin/klippel/Files/Know_How/Literature/Papers/Klippel_Holographie_Poster.pdf