Avantone CLA-10 (Yamaha NS-10M Clone) Review

[lowkeyoperations]

Yesterday, I stumbled over a comparison video of two flat-measuring monitors, the Neumann KH80 DSP vs the Genelec 8020D. I know that we should never judge how a speaker sounds over an in-room recording like that, but as long as the microphones are not moved as is the case with the recording, we can at least hear that the two speakers don't render the sound in the same even if both of them are considered to be very neutral reference studio monitors.

This is the measurements of the speakers by the person making the video:

View attachment 305689


Even if both of these two studio monitors measure almost the same with fairly small visual deviations, I find them to sound very different from each other. In the measurements, we can see that the Neumann speaker has just a little more energy in the range of 250Hz to 1kHz and just a little less energy in the range of 6kHz to 10kHz, and those small differences are enough to make the representation distinctly fatter-sounding in the low-mid frequency area and less "sparkly"-sounding in the top-end.

If a mixing engineer was handed the task to mix a track exclusively on both of these two flat-sounding reference monitors with the same sounding end result as the goal, and without the chance to check the mixes on other speakers for translation or that any comparisons were done to reference tracks during the mix, I would say that the outcome of the two mixes would sound fairly different from each other.

But if these two monitors were just used for certain mixing tasks as the NS-10 was mostly used, and the mix was checked on other monitors during the production, as full-range speakers during the mixing stage and another set of full-range and flat-measuring speakers at the mastering stage, then it is a good chance that the resulting product (the track) that we as the consumers hear, will sound pretty much the same no matter if the Neumann KH80 DSP or the Genelec 8020D was used at some stage of the production.

Here is the comparison video:

Neumann KH80 DSP vs Genelec 8020D || Sound & Frequency Response Comparison

This is one of the problems.

Say the standard is 20Hz-20KHz +-2dB

You could have one approved speaker that is -2db in the low mids and +2db in the highs.
Another might be +2db in the low mids and -2db in the highs.

They are going to sound different.

And the listener might be flat at those two points but +2db in the bass region and -2db in the mids.

All three speaker systems comply with the standard. They all have a different sound.

The standard says if you have one pair of complying speakers everyone will hear the same thing?

 

[goat76]

Didn't sound hugely different to me. Also didn't sound quite level matched. Close, but not the same audibly. I noticed they matched level with an SPL meter. Sorry, this is almost surely not good enough. It can be done, but the way most do it is not correct. Also the microphones are pretty good, but cardioids. The low end is going to be rolled off versus the reference track. So not representative of what you would hear in person on the low end.

I hear a huge difference in the representation of those two speakers. The Genelecs sound way brighter and give noticeably less body to the instruments, and the Neumanns sound a bit dull at the top end and a little bit too fat and "boxy" in the low midrange.

As I said, we can't judge by a recording like this how these two speakers really sound in-room if we were there, and which one we personally would find sounding more "right" than the other outside such a recording. But what the recordings can tell us, as they are both recorded at the same time without moving any of the microphones, is that these two speakers don't sound the same and there is a fairly distinct difference in the sound representation. At least to me, there is a very audible difference and I'm sure many others can hear that too.

If that recording was truly a good representation of how these monitors sound like, and I used these two monitors exclusively for mixing over a long period of time and got acclimated to their sound representation, and only relied on my inner reference of what I thought would sound well-balanced and "right". The mix on the Genelec monitors would most likely lead me to lower the energy a bit in the top end, and give the instruments in the mix a boost in the midrange for increasing the amount of "body" to them and the overall mix. With the Neumanns on the other hand, the adjustments would have gone in the opposite direction and I would have increased the amount of energy in the top end and lowered the energy in the low-mids to make the overall mix sound a bit less "boxy". The end result of the two mixes would most likely sound fairly different from each other.

But if the mix is heard on a number of other loudspeakers during the mixing stage, and then later on by a new set of ears by the mastering engineer at the mastering stage of the production, it can be assured that the mix's overall balance is corrected and there is no longer such a big risk that they will sound vastly different when they reach the end consumer.
The same goes if the NS-10 was used at some stages during the production and if some artifacts caused by that speaker get through to the next step in the production chain, it will most likely be addressed before it reaches the consumers. It all depends on if the NS-10s were even used for any frequency-related tasks during the mixing stage.

 

[Deleted member 21219]

This is one of the problems.

Say the standard is 20Hz-20KHz +-2dB

You could have one approved speaker that is -2db in the low mids and +2db in the highs.
Another might be +2db in the low mids and -2db in the highs.

They are going to sound different.

And the listener might be flat at those two points but +2db in the bass region and -2db in the mids.

All three speaker systems comply with the standard. They all have a different sound.

The standard says if you have one pair of complying speakers everyone will hear the same thing?

Poor standard. DSP can get more accurate than +/- 2 dB.

And I listened to the video. I heard nothing different with just a casual listen. If your hearing is so good that you hear appreciable and meaningful differences between these two speakers, then you certainly don't need anything like the NS-10 to make comparative decisions.

Jim

 

[Blumlein 88]

I hear a huge difference in the representation of those two speakers. The Genelecs sound way brighter and give noticeably less body to the instruments, and the Neumanns sound a bit dull at the top end and a little bit too fat and "boxy" in the low midrange.

As I said, we can't judge by a recording like this how these two speakers really sound in-room if we were there, and which one we personally would find sounding more "right" than the other outside such a recording. But what the recordings can tell us, as they are both recorded at the same time without moving any of the microphones, is that these two speakers don't sound the same and there is a fairly distinct difference in the sound representation. At least to me, there is a very audible difference and I'm sure many others can hear that too.

If that recording was truly a good representation of how these monitors sound like, and I used these two monitors exclusively for mixing over a long period of time and got acclimated to their sound representation, and only relied on my inner reference of what I thought would sound well-balanced and "right". The mix on the Genelec monitors would most likely lead me to lower the energy a bit in the top end, and give the instruments in the mix a boost in the midrange for increasing the amount of "body" to them and the overall mix. With the Neumanns on the other hand, the adjustments would have gone in the opposite direction and I would have increased the amount of energy in the top end and lowered the energy in the low-mids to make the overall mix sound a bit less "boxy". The end result of the two mixes would most likely sound fairly different from each other.

But if the mix is heard on a number of other loudspeakers during the mixing stage, and then later on by a new set of ears by the mastering engineer at the mastering stage of the production, it can be assured that the mix's overall balance is corrected and there is no longer such a big risk that they will sound vastly different when they reach the end consumer.
The same goes if the NS-10 was used at some stages during the production and if some artifacts caused by that speaker get through to the next step in the production chain, it will most likely be addressed before it reaches the consumers. It all depends on if the NS-10s were even used for any frequency-related tasks during the mixing stage.

Poppycock. Anyone using NS10s these days is full of it. Put down the crack pipe.

 

[Robbo99999]

Oh, they absolutely "think about the room." Problem is they farm out the work to an acoustician that is following old school strategies that are simply not backed by modern science. Look at the work that Keith Yates did for us in those CFD simulations and compare it to credentials and tools that Pro acousticians deploy. It is night and day.

Ultimately though, if they had done a great job here, we wouldn't have the problem we have. Or see reports like that Genelec with all the variabilities. And they would proudly talk about the response of their rooms.

Instead what permeates is "I know what I am doing, how dare you question me. Here are all the albums I have produced." I know because that is how the last discussion went that I had on GS forum. In a discussion of lossy audio for example, the engineer had zero knowledge of the technology and just made up stuff about what AAC compression did (and in an interview no less). No amount of explaining the reality of that coming from someone who spent a decade managing development of the same (me), would register with him. Somehow they think ability to produce music gives them true knowledge of audio science and engineering by osmoses.

And oh, this notion of "how it translates" was constantly used to justify using whatever monitors they had with claims that using Genelecs won't do it.

To top it off, one of them registered on a throw-away mail service and sent me a death threat in email claiming because I talked about Genelec, he is going to come over and go after my wife and I in gory detail. He said some things that made it easy to identify who he was on GS forum which turned out to be a very senior member and in that argument with me. Here is a bit of what he wrote:
View attachment 305637

So please excuse me if I don't much weight about industry claims of knowing what they are doing.

Oh my God, that's a ridiculous thing to have received by someone, they've totally lost the plot! Also proves that they're so emotionally tied in with their process of working that they wouldn't be open to any new ideas or ways of improving - for this one individual that is.

 

[Robbo99999]

Yes, I know translation from headphones to speakers is tricky.

But if the mix was made on ‘standard’ headphones and the song then listened to on ‘standard’ headphones, the theory is that the result would be improved.

I think that needs testing. At the moment it’s a concept. And it would only be headphone to headphone ‘certified’ in this case. I don’t think you could do headphone to speaker successfully unless it’s something like VSX.

I think that the Slate VSX headphones are a good example that might be a way forwards.
Anyone listening to those hears the same thing. And they emulate rooms with crosstalk and measurements of monitors in their actual rooms. You can mix reverb on these way better than normal headphones.

I can listen to mixes done in Mike Deans room on my headphones emulating Mike Deans room.

I can also listen to Masters done by Howie Weinberg on my headphones emulating his mastering suite.

That’s the concept isn’t it? That I can hear the same thing they did when they mixed and mastered?

The way headphones work from individual to individual, even the same headphone used on different people will sound a bit different, due to HRTF and headphone coupling differences between different physical anatomies of different individuals, so unfortunately headphones will never be a completely stable reference unless headphone DSP gets to the point where it can measure and account for different HRTF and different coupling, which is not possible without the individual doing lots of in ear mic measuring of calibrated speakers in a room - along the lines of Symth Realiser and Impulcifier Project. So headphones can't be a great reference ever, certainly not currently - although the Harman Headphone Curve is current best starting point average, but you can't rely on it being a stable standard like Anechoic Flat Speakers.

 

[goat76]

Poor standard. DSP can get more accurate than +/- 2 dB.

And I listened to the video. I heard nothing different with just a casual listen. If your hearing is so good that you hear appreciable and meaningful differences between these two speakers, then you certainly don't need anything like the NS-10 to make comparative decisions. You've just shot holes in your own thesis.

Jim

If you can’t hear meaningful differences in the overall representation of the sound between those two monitors, then I can’t see how you can have such strong opinions that all those songs you find sounding bad are caused in the mixing stage of the production.

As you are one of those who think that many mixes are badly done, maybe you can give some examples of that as Amir seems to avoid the question. And to be clear, I want you to point out the problems that you think are related to the actual mixing stage, not the problems related to the mastering stage of the production.

 

[lowkeyoperations]

Didn't sound hugely different to me. Also didn't sound quite level matched. Close, but not the same audibly. I noticed they matched level with an SPL meter. Sorry, this is almost surely not good enough. It can be done, but the way most do it is not correct. Also the microphones are pretty good, but cardioids. The low end is going to be rolled off versus the reference track. So not representative of what you would hear in person on the low end.

any roll off in the low end is only going to make the difference in the low mids and low highs easier to hear. listen to the high end of the snare and low end of the hats in any of the songs. its quite easy to hear the difference. its also pretty easy to hear the slight fullness of the lower mids in one compared to the other. Also listen to the kick in DEEP. The boosts in the low mids gives the kick more weight and body.

 

[lowkeyoperations]

Poor standard. DSP can get more accurate than +/- 2 dB.

And I listened to the video. I heard nothing different with just a casual listen. If your hearing is so good that you hear appreciable and meaningful differences between these two speakers, then you certainly don't need anything like the NS-10 to make comparative decisions.

Jim

Thankyou for your perfectly illustrative post Jim!

You heard "nothing different with just a casual listen". Suppose your experience is not uncommon on hifi playback, given Blumlein 88 also thought it "didnt sound hugely different". So with a very limited data set we might say that a 2db boost from 250-600 hz combined with a 2db cut from 5k-10kHz "doesnt sound hugely different" or is "nothing really different on consumer playback systems. At least for casual listening.

As Amir has pointed out, the people who come to asr are likely to value sound quality more highly than the general public, so if the difference is nothing worth worrying about to you, then it follows that most casual listeners wont be bothered either.

So the standard of +/-2dB is probably accurate enough for you.

 

[lowkeyoperations]

The way headphones work from individual to individual, even the same headphone used on different people will sound a bit different, due to HRTF and headphone coupling differences between different physical anatomies of different individuals, so unfortunately headphones will never be a completely stable reference unless headphone DSP gets to the point where it can measure and account for different HRTF and different coupling, which is not possible without the individual doing lots of in ear mic measuring of calibrated speakers in a room - along the lines of Symth Realiser and Impulcifier Project. So headphones can't be a great reference ever, certainly not currently - although the Harman Headphone Curve is current best starting point average, but you can't rely on it being a stable standard like Anechoic Flat Speakers.

I think you might find the work that Stephen Slate is doing with the VSX headphones interesting. Its only 2 channel operating as an audio plugin with calibrated matching hardware headphones, for 1/10th of the price. It uses eq to adjust the sound to your ear canal size. Its pretty cool. It uses BRIR and has modelled multiple studios with their monitors in place, running trinnov.

I have a feeling that any standard is likely to be proprietary.

 

[Robbo99999]

I think you might find the work that Stephen Slate is doing with the VSX headphones interesting. Its only 2 channel operating as an audio plugin with calibrated matching hardware headphones, for 1/10th of the price. It uses eq to adjust the sound to your ear canal size. Its pretty cool. It uses BRIR and has modelled multiple studios with their monitors in place, running trinnov.

I have a feeling that any standard is likely to be proprietary.

I did a quick google of that and read their website, they didn't show the research & how they did it. I'm not confident that it will be a solid implementation. I don't think you can really use headphones currently as a standard in music creation, not unless you go down the route of Smyth Realiser and have used in ear mics in your own ears to record the frequency responses of an actual "reference" studio, and then you do in-ear mic measurements when wearing your headphone, and then the Smyth Realiser processes it all together to recreate it - so that process takes into account both your HRTF (from the 1st process) and also the headphone coupling aspect (from the 2nd process). I'm not convinced on that Stephen Slate VSX stuff & can't find additional information after a quick google.

 

[lowkeyoperations]

If you can’t hear meaningful differences in the overall representation of the sound between those two monitors, then I can’t see how you can have such strong opinions that all those songs you find sounding bad are caused in the mixing stage of the production.

The differences are there.

This from the DEEP track on that video. Orange is Neumann, Blue is Genelec. 2 bar loop of each playing the same section of music. Electronic music is good for that type of comparison. Its repetitive and there is no difference in human playing of instruments. This is in Ozone Match EQ. The white line is the eq curve that would match one to the other.

Now if you were sitting in a room, moving speakers, starting the music again etc, then that all adds to the complication. But none of that is here. Its just a straight up audio file being played back. The frequency spectrum between the 2 is different. But even for some hifi enthusiasts its difficult to hear " a meaningful difference".

I did a quick google of that and read their website, they didn't show the research & how they did it. I'm not confident that it will be a solid implementation. I don't think you can really use headphones currently as a standard in music creation, not unless you go down the route of Smyth Realiser and have used in ear mics in your own ears to record the frequency responses of an actual "reference" studio, and then you do in-ear mic measurements when wearing your headphone, and then the Smyth Realiser processes it all together to recreate it - so that process takes into account both your HRTF (from the 1st process) and also the headphone coupling aspect (from the 2nd process). I'm not convinced on that Stephen Slate VSX stuff & can't find additional information after a quick google.

I owned Sennheiser HD650s for about 5 years and understand from experience all of the problems you describe. I really struggled using them for music creation for all those reasons.
The VSX headphones solve some of the problems and are a real improvement over standard or even eq'd headphones.

I'm sure the Smyth Realiser is amazing! It looks great.

 

[PeteL]

Oh, they absolutely "think about the room." Problem is they farm out the work to an acoustician that is following old school strategies that are simply not backed by modern science. Look at the work that Keith Yates did for us in those CFD simulations and compare it to credentials and tools that Pro acousticians deploy. It is night and day.

Ultimately though, if they had done a great job here, we wouldn't have the problem we have. Or see reports like that Genelec with all the variabilities. And they would proudly talk about the response of their rooms.

Instead what permeates is "I know what I am doing, how dare you question me. Here are all the albums I have produced." I know because that is how the last discussion went that I had on GS forum. In a discussion of lossy audio for example, the engineer had zero knowledge of the technology and just made up stuff about what AAC compression did (and in an interview no less). No amount of explaining the reality of that coming from someone who spent a decade managing development of the same (me), would register with him. Somehow they think ability to produce music gives them true knowledge of audio science and engineering by osmoses.

And oh, this notion of "how it translates" was constantly used to justify using whatever monitors they had with claims that using Genelecs won't do it.

To top it off, one of them registered on a throw-away mail service and sent me a death threat in email claiming because I talked about Genelec, he is going to come over and go after my wife and I in gory detail. He said some things that made it easy to identify who he was on GS forum which turned out to be a very senior member and in that argument with me. Here is a bit of what he wrote:
View attachment 305637

So please excuse me if I don't much weight about industry claims of knowing what they are doing.

Disgusting. Report that to the police, this is criminal, horrifying. Hope to one day we can get judiciary accountability for this type of stuff. Take care.

 

[Deleted member 21219]

Thankyou for your perfectly illustrative post Jim!

You heard "nothing different with just a casual listen". Suppose your experience is not uncommon on hifi playback, given Blumlein 88 also thought it "didnt sound hugely different". So with a very limited data set we might say that a 2db boost from 300-600 hz combined with a 2db cut from 5k-10kHz "doesnt sound hugely different" or is "nothing really different on consumer playback systems. At least for casual listening.

As Amir has pointed out, the people who come to asr are likely to value sound quality more highly than the general public, so if the difference is nothing worth worrying about to you, then it follows that most casual listeners wont be bothered either.

So the standard of +/-2dB is probably well and truly accurate enough for you.

..... In a situation where the room response varies from a reference of 82 dB at 67 Hz to 69 dB at 260 Hz (70 dB for the Genelec) back up to 79 dB at 260 Hz (78 dB for the Genelec) before a relatively smooth response from 1.2K onward, yes. That's +/- 6 dB, not +/- 2 dB. That sort of response masks (or swamps) a great deal, especially regarding the 2 dB cut in the high treble. Remember that I'm old and have hearing loss. That's why I said, "If your hearing is so good ......."

And yes, I said "a casual listen". That's what it was. I am just one person, and I've noted in previous posts that, as I said, I'm old. There are literally millions of people out there who have MUCH better hearing than I do. So I am not the average listener .... not at ASR, and not in the general public, either. Maybe I would be the average listener among that segment of the population who use assisted living, but that's about it. (Yes, I really am that old. @Blumlein 88 is no spring chicken, either.)

I have no idea as to the location of the room in which this recording was made, but if this was a room in a studio and this was the finished response at the workstation using DSP, then the so-called "circle of confusion" not only lives on, but it's getting worse! If anything, this indicates that the industry needs standards for the room response as well as standards for the response of the speakers to achieve consistency.

But to get back to the original, core issue ......

If you can hear the differences in response between these two speakers so clearly, especially in-room, any argument that the NS-10 or the CLA-10 clone are necessary (or even useful) is specious. (Compared to these two speakers, the NS-10 is like the Incredible Hulk hitting a person over the head with a lamp post.) That's the original, core issue. That is the issue that @amirm brought up in this review.

Indirectly, you have proven him correct.

Jim

 

[Zensō]

Although headphones can be useful for balancing levels accurately, being anechoic they can give a misleading rendering of reverberant effects. Utilizing a pleasing amount of reverb in a headphone mix will often result in an excessively wet mix as heard over loudspeakers.

True, though the inverse is also true. Since >80% of music is now consumed on headphones, logic would dictate that mixing on headphones is likely to result in an end product that is better optimized for today’s dominant listening paradigm.

 

[PeteL]

On a previous post I referred to a Floyd Toole article but the link is broken. I can't edit so I'll post it again since I think it is quite interresting to the discussion i find and it's open access.

E-library page - AES

AES E-Library ← Back to search

www.aes.org

 

[Joseph Crowe]

@amirm Was your subjective listening evaluation done with the speakers mounted on the bridge of a mixing console as a stereo pair? I've read that the boundary reinforcement of the physical mix console provides the correct response through the mid-bass. Also, near field <1m reduces distortion by virtue of a lower listening level. There are published response graphs available for the NS10 where anechoic measurement was done with the mix console as part of the test setup.

 

[PeteL]

..... In a situation where the room response varies from a reference of 82 dB at 67 Hz to 69 dB at 260 Hz (70 dB for the Genelec) back up to 79 dB at 260 Hz (78 dB for the Genelec) before a relatively smooth response from 1.2K onward, yes. That's +/- 6 dB, not +/- 2 dB. That sort of response masks (or swamps) a great deal, especially regarding the 2 dB cut in the high treble. Remember that I'm old and have hearing loss. That's why I said, "If your hearing is so good ......."

And yes, I said "a casual listen". That's what it was. I am just one person, and I've noted in previous posts that, as I said, I'm old. There are literally millions of people out there who have MUCH better hearing than I do. So I am not the average listener .... not at ASR, and not in the general public, either. Maybe I would be the average listener among that segment of the population who use assisted living, but that's about it. (Yes, I really am that old. @Blumlein 88 is no spring chicken, either.)

I have no idea as to the location of the room in which this recording was made, but if this was a room in a studio and this was the finished response at the workstation using DSP, then the so-called "circle of confusion" not only lives on, but it's getting worse! If anything, this indicates that the industry needs standards for the room response as well as standards for the response of the speakers to achieve consistency.

But to get back to the original, core issue ......

If you can hear the differences in response between these two speakers so clearly, especially in-room, any argument that the NS-10 or the CLA-10 clone are necessary (or even useful) is specious. (Compared to these two speakers, the NS-10 is like the Incredible Hulk hitting a person over the head with a lamp post.) That's the original, core issue. That is the issue that @amirm brought up in this review.

Indirectly, you have proven him correct.

Jim

I don't think anybody said that the NS-10M has a correct response. We all agree with Amir on that.

 

[Robbo99999]

I owned Sennheiser HD650s for about 5 years and understand from experience all of the problems you describe. I really struggled using them for music creation for all those reasons.
The VSX headphones solve some of the problems and are a real improvement over standard or even eq'd headphones.

I'm sure the Smyth Realiser is amazing! It looks great.

Yeah, fair enough, but I can't have confidence in the VSX system because it's not accurately described how it's created along with the associated research, and we see all sorts of somewhat lacklustre DSP headphone solutions all the time, but the Harman Headphone Curve research done by Sean Olive & Co is the one that has the most scope & validity, and even that can't be claimed to be used as a stable/great standard to create a mix & recording (for the reasons I mentioned in that earlier post). So headphones just can't come close to Anechoic Flat Speakers as a standard to be used for creating a mix/recording.

 

[Zensō]

Yeah, fair enough, but I can't have confidence in the VSX system because it's not accurately described how it's created along with the associated research, and we see all sorts of somewhat lacklustre DSP headphone solutions all the time, but the Harman Headphone Curve research done by Sean Olive & Co is the one that has the most scope & validity, and even that can't be claimed to be used as a stable/great standard to create a mix & recording (for the reasons I mentioned in that earlier post). So headphones just can't come close to Anechoic Flat Speakers as a standard to be used for creating a mix/recording.

I’m my experience, using Harman or Sonarworks as a starting point, then making small adjustments by ear to match a headphone to an anechoic flat speaker in a good room is a viable approach.