A Confused streaming system - Mutec / SOtM Ultra

[PeppaPig]

I am not quite understand how the external clocker like MC3 can help to improve sound quality of Devialet.

When the sound is reclocked by Mc3 and then output to Devialet, the Devialet will processed it using its own internal DAC and clocker again, we cannot bypass it right?

Before the advent of the CI and Pro upgrades iirc a lot of us were unhappy with the sound quality of the USB input. AES/EBU sounded superior. Mutec came out with the MC3+USB, which as it’s name implies, has a USB input and also has an AES/EBU output. The Mutec also reclocks the incoming signal before conversion. With a decent cable, you a great digital connection.
I understand that, since the CI upgrade, USB performance has improved, although these increases are probably across the board!

The main advantage though is the ability to feed the Mutec with a 10MHz clock input, as @Confused has posted about.

All inputs go through the DAC(s).

Hi Axel, so after CI upgrade, do we still need MC-3 in the middle of Devialet and the MAC Mini, is it still providing huge improvement?

I thought everything goes into Devialet internal DAC and gets reclocked again, plus the Devialet has a internal buffer, I don’t know why we still need a external clock, I think the improvement should be very minor or zero, correct me if I am wrong.

[Confused]

A much happier update today.  I tried some more experimenting with Roon's parametric EQ today, and now I can get it to do exactly what I failed to do yesterday.  A simple case of being on the learning curve, doing a bit of reading, and discovering what the "Q" parameter does.  The situation is not totally satisfactory though, I can achieve an almost perfect curve, move the microphone a few millimeters, and then the curves drift significantly.  I have also tried REW using the microphone pointing upwards, which is something that has been posted about elsewhere on the forum.  Based on subjective listening, this does seem to have yielded the best results, but again, move the microphone a bit, and different settings are needed.  I can produce fine looking curves, but the results are clearly a bit arbitrary.  I think I have a lot to learn, and I plan to do a bit of reading up before spending much more time on this.  The good news is that I now have a small collection of EQ's saved on Roon, and a couple of them seem to really work.  I am quite sure that what I have done today is a very long way from the optimum, but the good news is that that with a couple of the EQ's I can now sit and listen totally untroubled by my earlier presence region woes.  Some tracks that were bothering me a little since introducing the SOtM kit now sound sublime, indeed, I have just been listening to one track that has been bothering me a little since upgrading to the 1000Pro, and this sounds better than ever.  Plus, I have only tried correcting from about 4kHz upwards, there is much more I can do here.  I am sure that I can optimise this further, but meanwhile, I am delighted by the results.

Also, many thanks for the recent posts, suggestions and comments.

@David A - Some very wise words in your post!  In fact, there is only one point I do not fully agree with, you mention "There could be several reasons but I strongly suspect that part of it is simply that you're noticing the presence band more because of the aspects of the sound that you're paying attention to"  In part, you are correct, I have most definitely been paying attention to treble and the presence band, no doubt about it.  But this is missing the point, prior to installing the SOtM kit, I was not paying attention to treble or the presence band.  It is a case of after installing the SOtM kit, the treble and presence band become subjectively more prominent, and this is why I started to pay so much attention to it.  Yes, this might be a bit of a vicious circle, but I remain intrigued as to why improving resolution of the HF makes it subjectively more prominent.  Oh, and there is no shortage of bass with the Blades!  Yes, the peaks and nulls could be tidied up, but the Blades are showing decent levels down to 10Hz, which is adequate.  That said, now that I have entered the room correction rabbit hole, tidying up the LF end of things has to be a good idea, something for the future.  I absolutely agree with your point about making changes slowly though.  My own rule is never change more than one thing at a time.  If you can A/B test the change, do so, this provides a good reference.  Then, live with the change for at least a month before changing anything else, it is not until you have lived with something for a while that you really understand it.  In the case of my recent changes, I had my reasons for not following my own rules, in hindsight, this was a big mistake!

@Rama - I think I have written so much in some of my posts that the key points get lost a little.  So one point to clarify, at first I thought I had some HF harshness, but after much experimenting, it became clear that the issue was not harshness, just that the HF subjectively sounded too prominent.  You mention speaker cables, so for the record I can advise that I am using Chord Epic Reference cables, which I bought at a very good price when buying the Blades.  If I am honest, I have heard the Blades with these, and with hugely expensive Chord Sarum cables, and couldn't really tell a blind bit of difference!  It is not something I have experimented with much, maybe I should be more open-minded, but I kind of feel that if the cables are of a large gauge and made of copper, then it is more or less job done.  The Chord Epic Reference cables do look good though, which is a bonus.

@PeppaPig - To add to Axel's comments, when you feed the Devialet via AES/EBU, the Devialet has to extract the clock signal from the source feed, so the better the clocking in the source, the better the sound, or at least that is the theory.

@cereal killer - Are you suggesting I go back to AIR?

[ogs]

@PeppaPig you do not have to understand. Just read and accept, or even better, try it for yourself. As you can see from my signature I also use a Mutec. The AES/EBU input of the Devialet is very good. Using the Mutec to reclock the stream before entering the Devialet improves sound quite markedly. I guess you'd have similar improvement on one of the RCA inputs as well, but the AES is better. Sound quality from a Devialet amplifier (this goes for Phantoms too) depends on the quality of the input signal. A reclocker enhances purity of the stream.
The Devialet does not have a built in reclocker. It ought to have one. Especially on the CI board as this is a NEW design, but Devialet missed the boat here.

[cereal killer]

@Confused

Yes, if you can configure your NAS the same way. I’ve found it to make a defined improvement.

[Axel]

I am not quite understand how the external clocker like MC3 can help to improve sound quality of Devialet.

When the sound is reclocked by Mc3 and then output to Devialet, the Devialet will processed it using its own internal DAC and clocker again, we cannot bypass it right?

Before the advent of the CI and Pro upgrades iirc a lot of us were unhappy with the sound quality of the USB input. AES/EBU sounded superior. Mutec came out with the MC3+USB, which as it’s name implies, has a USB input and also has an AES/EBU output. The Mutec also reclocks the incoming signal before conversion. With a decent cable, you a great digital connection.
I understand that, since the CI upgrade, USB performance has improved, although these increases are probably across the board!

The main advantage though is the ability to feed the Mutec with a 10MHz clock input, as @Confused has posted about.

All inputs go through the DAC(s).

Hi Axel, so after CI upgrade, do we still need MC-3 in the middle of Devialet and the MAC Mini, is it still providing huge improvement?

I thought everything goes into Devialet internal DAC and gets reclocked again, plus the Devialet has a internal buffer, I don’t know why we still need a external clock, I think the improvement should be very minor or zero, correct me if I am wrong.

I haven’t tried USB direct into the Devialet, but I hear definite improvement in SQ with the Mutec in the chain.

I cannot comment on the influence of an external clock, as I don’t own one.
If you search for posts on the Mutec, there are plenty to be found.

This is going off topic though....

[PeppaPig]

Thanks both of you!!

I have learn something here!!!

I always think all DAC has a clocker built in and always re-clock the incoming signal.

So this logic apply to power supply as well? Use a power conditioner/regenerator will improve Devialet sound as well?

By the way, that AES cable has to be 110Ohm right?

[David A]

…

@David A - Some very wise words in your post!  In fact, there is only one point I do not fully agree with, you mention "There could be several reasons but I strongly suspect that part of it is simply that you're noticing the presence band more because of the aspects of the sound that you're paying attention to"  In part, you are correct, I have most definitely been paying attention to treble and the presence band, no doubt about it.  But this is missing the point, prior to installing the SOtM kit, I was not paying attention to treble or the presence band.  It is a case of after installing the SOtM kit, the treble and presence band become subjectively more prominent, and this is why I started to pay so much attention to it.  Yes, this might be a bit of a vicious circle, but I remain intrigued as to why improving resolution of the HF makes it subjectively more prominent.  Oh, and there is no shortage of bass with the Blades!  Yes, the peaks and nulls could be tidied up, but the Blades are showing decent levels down to 10Hz, which is adequate.  That said, now that I have entered the room correction rabbit hole, tidying up the LF end of things has to be a good idea, something for the future.  I absolutely agree with your point about making changes slowly though.  My own rule is never change more than one thing at a time.  If you can A/B test the change, do so, this provides a good reference.  Then, live with the change for at least a month before changing anything else, it is not until you have lived with something for a while that you really understand it.  In the case of my recent changes, I had my reasons for not following my own rules, in hindsight, this was a big mistake!
…

Maybe I didn't make that point as well as I could.

When we make changes it's the differences that tend to grab our attention. If something changes, it often tends to "stick out like mad" and for some time after the change every time  we sit down to listen we notice and pay attention to that change. It's hard to avoid because we're hearing something we aren't used to hearing and until we become used to hearing things the way they now are we keep paying attention to the change and not paying as much attention to the things we're used to, the things which didn't change.

The things which highlight increased resolution are usually in the presence region and while we're paying attention to that increase in resolution we aren't going to be giving the same amount of attention to the bass and lower mid-range but we are going to be more aware of what's going on in the presence range even when it doesn't relate to resolution. We may not be aware of those things as much as we are of the increased resolution but we're probably still more aware of those things which are closer in frequency to where we're noticing the increase in resolution than we are of things happening in the lower mids and bass.

Over time as we become used to the change and expect to hear the difference it brought as part of the sound, we pay less attention to it and start to get back to attending more to the music than to the sound. When you do that I suspect you'll find that you aren't as aware of the presence range as you are now because you will be spreading your attention more evenly over whole of the sound you're hearing.

[Confused]

@David A - Thanks for the post and clarification.  I do understand exactly what you are saying, in fact, it looks like we are saying pretty much exactly the same thing, although there is maybe a slight change in emphasis between how we are expressing things.  So we are in agreement.  Of course, I have the advantage over other posters on the forum, as I can actually hear the things I am describing.  It is for this reason that I keep mentioning the fact that although the mR and SOtM kit measure the same in terms of frequency response, the SOtM kit really does sound much brighter, a curious phenomena!  It is true though that this has propelled the HF issue to the front of my conciseness, so I am definitely paying far too much attention to it, exactly as you describe.  It is a vicious circle I think, but it will be interesting to see how this one plays out over time, as I get more used to things.  One thing I have realised is that I have been subconsciously concentrating on HF ever since adding the 1000Pro.  One example of this is reading my own notes that I made when trying HQPlayer filters, the majority of the notes relate to things like HF and presence band performance, with some notes mentioning bass performance (not as many as for HF), with no mention of mid range at all.  So here I can see you are absolutely correct.  It is fascinating how the HF issue was kind of lurking in my subconscious for a while, but the SOtM kit managed to propel it firmly into my conscious side of my thinking.  Genuinely fascinating, especially when you consider the SOtM kit has made zero difference to measurable frequency response.  It shows how powerful psychology is with audio.

As an interim update, I have also been thinking a bit about my efforts with room correction.  I can see now that trying to correct for the "sweat spot" with the microphone in one specific location does not work, because if you shift the microphone a couple of inches (roughly the distance between my ears), the peaks ant troughs move, and the correction goes out of sync.  (in the HF region at least)  I can see now why some room correction methods record frequency sweeps with multiple microphone positions and then average the results.  I think that if I do this it will give me a more accurate / practical / usable idea of any issues with room / speaker interaction and what I should be correcting.

As I said before, I am an absolute novice with respect to Room correction, but learning fast (learning the hard way though).  If anyone has any tips for someone at this early stage of playing with RC, or links to good reading material, this would be appreciated.   Something very useful, but relatively simple, is probably what I need at this stage!

Edit:  A further question to the room correction experts out there.  I have a UMIK-1 microphone which I have been using for my measurements with REW.  I note that there is a suggestion that when measuring with REW the microphone should be pointing directly upwards, rather than horizontal / towards the speakers.  Is this correct?  Are there uses when horizontal or vertical is preferable?  I set up the UMIK-1 for use in REW a couple of years ago and added the specific calibration file.  I cannot remember if this was a "90°" calibration or not.  What is the recommendation?

[alandbush]

Edit:  A further question to the room correction experts out there.  I have a UMIK-1 microphone which I have been using for my measurements with REW.  I note that there is a suggestion that when measuring with REW the microphone should be pointing directly upwards, rather than horizontal / towards the speakers.  Is this correct?  Are there uses when horizontal or vertical is preferable?  I set up the UMIK-1 for use in REW a couple of years ago and added the specific calibration file.  I cannot remember if this was a "90°" calibration or not.  What is the recommendation?

I recall that the vertical orientation was recommended for multi channel with the horizontal orientation recommended for a stereo layout. That may have had something to do with the type or quality of the microphone in the past.
I don’t believe there is a great deal of difference with either orientation with the UMIK1. Probably no more than you are experiencing with small changes to the location of the microphone.

[thumb5]

Edit:  A further question to the room correction experts out there.  I have a UMIK-1 microphone which I have been using for my measurements with REW.  I note that there is a suggestion that when measuring with REW the microphone should be pointing directly upwards, rather than horizontal / towards the speakers.  Is this correct?  Are there uses when horizontal or vertical is preferable?  I set up the UMIK-1 for use in REW a couple of years ago and added the specific calibration file.  I cannot remember if this was a "90°" calibration or not.  What is the recommendation?

I'm not an RC expert by any stretch of the imagination, but I also have a UMIK-1 and have been puzzled, not to say confused, by the same thing.  Last time I looked into it, I found a thread on CA (apparently since deleted) that said:

Well... it's the mic's calibration file that prescribes the orientation, 0° for an horizontal positioning or 90° if the calibration has been done for a mic's vertical positioning.  If one happens to have both (i.e. miniDSP's UMIK-1) then for Stereo systems it's a choice that depends on various factors (the vertical positioning will get slightly more of the room vs. direct radiation).

I hope that helps.