Jump to content

Synergistic Research XOT Carbon review and testing.


Recommended Posts

20 minutes ago, sir sanders zingmore said:

I guess my point (or one of them at least) is that before you rush off and try to discover which measurements correlate with air or focus or soundstage depth, shouldn't you first establish that the differences you are hearing are actually caused by the XOTs that you just plugged in?

I agree with this.  In my earlier comments I did not mean to imply A-B tests, or any other means of trying to establishing a causal link between XOTS and what is heard, are not worthwhile.  Rather I was  just pointing out some issues to be aware of.  

Link to comment
Share on other sites



On 21/09/2020 at 2:25 PM, sir sanders zingmore said:

2) The XOTs somehow improve the system even if they aren't plugged in.

So good! (No smiley face emoji because it could be open to interpretation).

 

The bottom line is that you heard great qualities of your HiFi system and you heard those qualities with and without XOTs plugged in? Is that an accurate summary?

 

It is great that the owner of XOTs is being as generous as they are in providing others an opportunity to test the XOTs. Honestly, it is great!


I appreciate the write ups that people have offered of their experiences with XOTs. This includes all of the write ups.

  • Like 1
Link to comment
Share on other sites

2 hours ago, RussB said:

In the case of audio, we can’t directly measure air, focus, soundstage depth, as far as I know, but we can measure impedance, capacitance etc.


There are measurement results that correlate to the subjective qualities that are perceived. I think background knowledge does make a difference to how easily and accurately one interprets results. Some of that interpretation relies on knowing initial things like experimental aim, and knowing the overall experimental structure.

Experiments are fairly involved. Skipping details here is a 'rough' summary:
1. Know what the experiment aim is

2. Understand measuring equipment and measurement techniques 
3. Setup the experimental procedure to be valid and appropriate for the aim

4. Setup the experiment in line with the procedure

5. Take the needed measurements as per the procedure

6. Structure the results in a form that enables interpretation 

7. Look over the results and determine if any of the experiment needs to be repeated as there may be errors in any of above steps (including the procedure) or it may be great to confirm the data. Ideally no repeat is necessary 

8. Undertake an in-depth analysis of experimental data

9. Place the results of analysis into the context of the aim

10. Write up the findings using select data

11. Edit the report for accuracy in interpretation and accuracy in communication

12. Publish the report

 

By the time a third party reads the report there is a lot to critique. The peer review process is spans all parts of the experiment. When it comes to research there may be many experiments and all of them, including the interpretation within context, need to be analysed. So a fair bit of background knowledge goes in to understanding experimental and research results. Even more knowledge goes in to linking the recorded data to subjective descriptions.

 

There are shortcuts but they are not always accurate.

 

The sentiment that presenting data and/or analysis of data to someone could be a not productive exercise is accurate. The aim of communicating as simply and as accurately as possible is always a good aim. It does become tricky when one person's experience of 'soundstage' (or any subjective quality) differs to another person's experience. Then accuracy may start going out the window.

Link to comment
Share on other sites

17 hours ago, RoHo said:

Most well respected speaker designers I have read about, and learned a lot from, would include listening as a necessary part of the design process.

Indeed.... but the context of the listening is extreme important.

 

If I have to choose between something and something else.... and they are both desirable..... then it can come down to a subjective choice about which is done.    This doesn't happen a lot though.

 

Another part of it realy comes down to levels (ie. over al frequency response) in room.    There is no universal solutions to this, with room size,, position, and listening axis/distances, all varying among users.....  sometimes this ends up in a subjective choice/compromise too..... although with decent quality "room correction" systems these days (I hate that term) - it can be calibrated for each user, and the desinger can be left to focus on dsigning for performance, and not designing to a specific "use case".

 

 

17 hours ago, RoHo said:

This process is something like "measure, adjust , listen, repeat...".  Do you think the listening part is not really necessary?

I think it is a great source of confusion for designers.... and can often lead to suboptimal results.

 

17 hours ago, RoHo said:

I guess most of these designers also have their own "aural goal"

That is the problem in a nutshell.

 

As I said.... sometimes, it's a necessary evil to make a choice ..... but for the most part it is best avoided.

 

I can tell you that many famous speakers were not designed exactly the way you say they were ..... even though sometimes the marketing department or even the designers themselves make it seem like that's what they do.    Of course, they all spend lots of time listening to things..... as there is a lot of "choices" to make in a practical speaker which can't do or be everything.

 

17 hours ago, RoHo said:

Also once one becomes more experienced in these matters wouldn't that translate to their "ear" being more accurate?

I can understand why people woud think so.   I think almost the opposite is true.

 

What you find is that if you A/B something..... you might form some optinions, etc...... but if A or B is actually "better" with regardss to "less errors" in the speaker response ..... Then you will find that less errors version is actually better sounding (even if you "preffered" the other way").

 

Sometimes you don't have the luxury of a clear choice (ie. one is known better), and subkectivity can't be avoided, but I think it's less common than most people would assume.

  • Like 1
Link to comment
Share on other sites

51 minutes ago, davewantsmoore said:

 

 

What you find is that if you A/B something..... you might form some optinions, etc...... but if A or B is actually "better" with regardss to "less errors" in the speaker response ..... Then you will find that less errors version is actually better sounding (even if you "preffered" the other way").

 

Sometimes you don't have the luxury of a clear choice (ie. one is known better), and subkectivity can't be avoided, but I think it's less common than most people would assume.

I guess what I'm talking about is the ability to recognize frequency response aberrations from listening.  Vocals are an example of what I'm trying to get at.  I recently re-designed the passive crossover of my speakers partly to try to improve the sound and partly as a learning exercise.  One of the main things I wanted to address was obvious harsh sibilants with female vocals and a less obvious hardness in male vocals.

Flattening the treble response improved the female vocals but the issue with male vocals was harder for me to pin down as I didn't really know what frequency was responsible for the issue.  Lou Reed is one of my go to's for this as his recordings almost always have a very clear, live sound and, of course, I think he rocks.  So I reckon now I can tell how Lou should sound.  No lower midrange glow, no excessive sibilants or hardness on consonants.  Does it sound right or not?  Is the voice like a real voice or does it have some colouration?  Recognising this is the skill I'm talking about. 

Anyway eventually after several iterations of stuffing around I put in a notch filter at about 1kHz as there was a bump in the FR there.   This did make Lou sound more real.  Would a more skilled/experienced/musically talented  person have heard Lou and identified the problem frequency straight away?  I like to think so.   

Don't get me wrong, I realise that if I had the option to use some method to automatically optimise the FR then I would have got an equally good or better result.  But that's not the point here.

Link to comment
Share on other sites



  • Volunteer
13 minutes ago, RoHo said:

Anyway eventually after several iterations of stuffing around I put in a notch filter at about 1kHz as there was a bump in the FR there.

A great example of correlating measurements with what you heard.

But wouldn't it have been much simpler to just correct the FR bump in the first place rather than going through "several iterations of stuffing around"? 

Link to comment
Share on other sites

18 minutes ago, RoHo said:

Flattening the treble response improved the female vocals but the issue with male vocals was harder for me to pin down as I didn't really know what frequency was responsible for the issue.  Lou Reed is one of my go to's for this as his recordings almost always have a very clear, live sound and, of course, I think he rocks.  So I reckon now I can tell how Lou should sound.  No lower midrange glow, no excessive sibilants or hardness on consonants.  Does it sound right or not?  Is the voice like a real voice or does it have some colouration?  Recognising this is the skill I'm talking about. 

Anyway eventually after several iterations of stuffing around I put in a notch filter at about 1kHz as there was a bump in the FR there.   This did make Lou sound more real.  Would a more skilled/experienced/musically talented  person have heard Lou and identified the problem frequency straight away?  I like to think so.   

Don't get me wrong, I realise that if I had the option to use some method to automatically optimise the FR then I would have got an equally good or better result.  But that's not the point here.

When it comes to speakers that sort of tuning is much easier to do using instrumentation.

 

Any designer that offers products for more than just themselves need to take into account the fact that customers will have a large variance in their hearing. The more the designer favours their personal bias the smaller their customer set.

 

There is also plenty of information on what is generally preferred by people. Its one of the reasons that a lot of headphones have a V shaped frequency response. Its not everyone's cup of tea but its a cup of tea that appeals to many.

 

Edit:

Something you could have done:

  • Take a sample of recording where you wanted to change the frequency response (e.g. a portion of audio with singing and very little other sounds)
  • Perform frequency analysis of that sample via an instrument or computer
  • Perform frequency analysis of the sample post your circuit (crossover, etc) using an instrument or computer
  • Modify the circuit to alter the frequency response to your desired state, using the above frequency analysis as a guide
  • Perform frequency analysis of the sample post current circuit using an instrument or computer
  • Listen to the sample & to some recordings with the new circuit in place
  • If not satisfied with the outcome go back to step 4 using the current circuit and keep repeating the process

 

It is important to know what you have changed, how much change you introduced, what else has it effected, etc. Remember that not everyone will enjoy the changes you enjoy and at times you may need to strike a middle ground.

Edited by gwurb
Link to comment
Share on other sites

33 minutes ago, sir sanders zingmore said:

A great example of correlating measurements with what you heard.

But wouldn't it have been much simpler to just correct the FR bump in the first place rather than going through "several iterations of stuffing around"? 

The FR showed several "areas on concern" and my primary goal was getting the crossover area smooth, particularly the phase which was really difficult.  I thought this might fix the particular problem but it didn't.  It wasn't as if there was one single big bump in the FR, unfortunately.

Link to comment
Share on other sites

3 minutes ago, RoHo said:

The FR showed several "areas on concern" and my primary goal was getting the crossover area smooth, particularly the phase which was really difficult.  I thought this might fix the particular problem but it didn't.  It wasn't as if there was one single big bump in the FR, unfortunately.

https://www.dpamicrophones.com/mic-university/facts-about-speech-intelligibility

 

There wouldn't be a single frequency or hump relating to voice reproduction 'issues'.

Link to comment
Share on other sites

4 minutes ago, gwurb said:

https://www.dpamicrophones.com/mic-university/facts-about-speech-intelligibility

 

There wouldn't be a single frequency or hump relating to voice reproduction 'issues'.

Exactly.  I read this sort of information which didn't really help my situation.  That's why I'm wondering if a more experienced/knowledgeable type might have listened for a few seconds and said " you've got excessive activity in the 1 kHz band".  And now I have this experience and am hopefully wiser would I recognize a similar issue in the future, just by listening? 

Link to comment
Share on other sites



1 hour ago, RoHo said:

I guess what I'm talking about is the ability to recognize frequency response aberrations from listening.

Indeed.

I find when I compare actual frequency reponse differences, with the ones people think they hear... it's often quite different.   Also, only doable in fairly wide/q bands.

 

... but, errors which are lower-q..... even though they might not be detectable "by ear" ..... can be very noticable when fixed.... but not in a specific sense, just in a "generally better" type of sense.

 

1 hour ago, RoHo said:

Vocals are an example of what I'm trying to get at.  I recently re-designed the passive crossover of my speakers partly to try to improve the sound and partly as a learning exercise.  One of the main things I wanted to address was obvious harsh sibilants with female vocals and a less obvious hardness in male vocals.  Flattening the treble response....

I see crossovers, and the flatness/balance of the overall response as two very different things.   A crossover is either perfectly summing, or not..... but this doesn't have to affect the overall balance of the speaker.    The two should be designed, and optimised, totally seperately.

 

The way some speakers are designed (ie. the methods used) ... that option isn't available.

 

 

1 hour ago, RoHo said:

improved the female vocals but the issue with male vocals was harder for me to pin down as I didn't really know what frequency was responsible for the issue.  Lou Reed is one of my go to's for this as his recordings almost always have a very clear, live sound and, of course, I think he rocks.  So I reckon now I can tell how Lou should sound.  No lower midrange glow, no excessive sibilants or hardness on consonants.  Does it sound right or not?  Is the voice like a real voice or does it have some colouration?  Recognising this is the skill I'm talking about. 

I understand.   What I am saying it that "that skill" can easily least to a less than great speaker.    When you design the equipment to make the signer sound like you think he should sound..... you likely design the speaker to distor the original signal.

 

The paradox is that even thouhg you end up in place where you say "I've got it sounding just the way I want it... and everything is great"..... what I have found is that if we audition "technically better"  (ie. close to the original signal, ie. "less errors") that it often turns out that we prefer the latter.

 

..... or, so go a little deeper.   What we find is that the design aspects of the speaker are better in the "less errors" version...... but there is something missing from the room setup or "overall EQ" (very related to the room setup).

 

Tangling the two up, is a problem IME.

 

1 hour ago, RoHo said:

Would a more skilled/experienced/musically talented  person have heard Lou and identified the problem frequency straight away?  I like to think so.   

I would very much have liked to think so too..... I found it was not really quite that straightforward.

 

There is more to be gained by objectively reducing (the right) errors .... than there is on flavouring to prefrence (in equipment design)

 

1 hour ago, RoHo said:

use some method to automatically optimise the FR then I would have got an equally good or better result.

I'm not to sure about that in general ..... "automatically optimise the FR" is very difficult.    It can often be better than an unoptimised speaker.... but it gets a bad name for a reason too.

 

Acoustic issues (ie speaker directivity, and placement wrt listener and room) can only be fixed well with acoustic solutions.

 

Keep in mind I'm not just suggesting anyting is a simple as "just use a speaker correction box and it will be perfect".    This cannot sucessfully undo the larger and much more fundamental distortion in the speaker, or a clipping amplifier, or munted gain structure, etc.

Link to comment
Share on other sites

24 minutes ago, davewantsmoore said:

This is exactly the sort of thing I'm talking about....

 

It's a very bad way to design a good playback system.

It's a very good way to design a system that the designer subjectively enjoys at the time of adjustment :) 

 

Remove the adjustment for personal preference and it's a way to design a system. I would aim for reference/accuracy of reproduction, or what you may be referring to as reducing errors. Start with a baseline, understand the baseline, build your circuit, understand how your circuit alters the baseline, adjust the circuit to have the least deviation from baseline possible, address the next part of the system, keep going until all parts are finished.

 

Need to keep track of changes...

Edited by gwurb
Link to comment
Share on other sites

18 minutes ago, sir sanders zingmore said:

Somewhere in the deep distant past I have a memory of this thread being about something else entirely.

 

We need the current holder of the XOTs' to quit mucking around and give us his impressions … I'm lookin' at you @Luc 

The design discussion probably needs to move to the related science topic.

Link to comment
Share on other sites



1 hour ago, sir sanders zingmore said:

Somewhere in the deep distant past I have a memory of this thread being about something else entirely.

Has it ever been any different?:)

 

As far as hurrying @Luc along, he needs time to catch the vibe  of these things. These long considered and seminal opinions are heavily influential in forming an ongoing holistic understanding about these devices and their attendant gestalt.

Link to comment
Share on other sites

Just now, rantan said:

As far as hurrying @Luc along, he needs time to catch the vibe  of these things. These long considered and seminal opinions are heavily influential in forming an ongoing holistic understanding about these devices and their attendant gestalt.

You mean see through the bullshot and see these for the snake oil rip-off scam they are.

  • Like 1
  • Haha 1
Link to comment
Share on other sites

1 hour ago, rantan said:

Has it ever been any different?:)

 

As far as hurrying @Luc along, he needs time to catch the vibe  of these things. These long considered and seminal opinions are heavily influential in forming an ongoing holistic understanding about these devices and their attendant gestalt.

Breaking in new speakers first and a general rearrangement of the system is hampering any chances of a critical listen at the moment.

  • Like 1
Link to comment
Share on other sites



Just realised my review of the XOT's  did not get posted to this thread as it as buried in the behemoth that is the tweaks thread.  so here it is for your reading pleasure:)

 

I was fortunate to have the chance to try the XOT xover transducers. Thanks @Cafad for the opportunity.  I did have a go at finding some sort of measurable property of these mysterious tubes and can confirm the same outcome as others that I was unable to measure any effect on an audio signal. 

I then connected them up and had a listen.  My impressions are that they increased the size of the instruments and the space they were in to create a more enveloping, 3d, and pleasant but more interesting experience.  The effect was immediate and evident to me. Upon removing them the system sounded flatter, shrunken in scale with less detail and more hash.  Compared to other tweaks like Mad Scientist Discs and ground plane enhancers (https://www.diyaudio.com/forums/parts/102180-groundside-electrons.html) they have a much bigger impact.  

  • Like 4
Link to comment
Share on other sites

23 hours ago, gwurb said:

the designer subjectively enjoys at the time of adjustment

This is the entertainment industry, so don't let anyone prevent that enjoyment.  ;) 

 

... but tweaking to preference is best done on equipment that is designed well.... not on equipment which is designed by tweaking to preference.

 

  • Haha 1
Link to comment
Share on other sites

Had a brief listen this evening as I settle the system back in after some parts of it have had a long siesta.

 

I chose to play one track over and over again. 1 minute 30 seconds into a track I know very well.

Drop needle down, race back to seat. Get up remove XOT's, race back to seat and listen. Repeat 10 times.

 

IMG_20201003_202951482.thumb.jpg.7878424ee4053179f2f4cb766953fadd.jpgIMG_20201003_203107753.thumb.jpg.52d2a4ae868dd8d8a398be2ae0916a87.jpg

No thoughts just yet.

Will try to test further tomorrow.

No thoughts? Well I thought I did hear something then I didn't then I was maybe...

So I'll try again tomorrow

  • Like 2
Link to comment
Share on other sites

If you're having trouble hearing anything Luc, the track I heard the most difference on was Enya's "Only Time".   A negative effect, the soundstage was pushed forwards all over the place with instruments getting it the way of the vocals and each other, but an effect none-the-less.  

 

https://www.youtube.com/watch?v=7wfYIMyS_dI

 

There are two recorded versions, the link above is the one I'm talking about.

 

Link to comment
Share on other sites

  • 2 weeks later...

Hi All.     Terry O’Neil  brought these XOT devices into the factory today ! What can I say ! They work .

better 3D soundstage and detail , going back makes the sound a little and 2D and flat .  I’m going to try and measure what these things are doing when I get some time .     Cheers Mike Lenehan 

  • Like 5
Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...
To Top