Dynaudio (Evoke) + ExpertPro 220

[zoran_pavlovic]

Napisao sam ceo carsav, ali reply nije snimljen
Zvao sam Laleta, poslace mi telefon pa se cujemo

[David A]

… Though I thought the Roon is just some kind of cosmetical thing, I read that it can improve (don't know how) the way Tidal sounds? 
…

Roon can do the first unfold for MQA encoded streams soit effectively gives you most of what you can get with MQA. A Tidal 24/48 stream will be unfolded to 24/96 before it is sent to your Devialet. The Devialet has no MQA functionality whatsoever so it is incapable of performing either of the 2 MQA unfold processes.

Roon also has its own DSP processing engine and provides options including parametric EQ for content streamed from Roon and I understand that it's possible to create similar sorts of filters to those for Devialet's Sweet Room and use them in Roon.

If you're using Roon and ir's your only source, then Roon's DSP functions are possibly superior to Sweet Room. Roon let's you apply DSP adjustments without creating filters so you can measure your room and simply apply parametric EQ using Roon's graphic EQ interface and make adjustments on the fly to fine tune the result. That isn't possible with Sweet Room but Sweet Room has the advantage of working with all music sources the Devialet can accept whereas Roon's DSP functionality only works with music you stream from Roon. I haven't tried or heard either of Roon's DSP functionality or Sweet Room in use so I can't comment on the relative effectiveness of either.

[Delija]

The major problem with Roon DSP is inability to measure the results - and results can in no way be 100% predicted.
There are a lot of trials and errors in the process - even if you exclude that perfect measured FR curve doesn't sound the best and combination of measurements and listening has to be used.

Not to mention that using DSP introduces artifacts into sound - you do fix something, but you lose the ultimate transparency, resolution and dynamics. It's all a big compromise.

I limit DSP (PEQs / room correction) usage only to subwoofers.

[David A]

You can measure the results in the same way that you measure the room in the first place in order to decide what adjustments you're going to make. You've git exactly the same problem if you use Devialet's Sweet Room.

Yes, there can be a lot of trial and error in the process but the only way to reduce that is by using an amp which includes technology like Dirac and even then you're probably going to have to use different software to measure the results.

DSP is an acronym for Digital Signal Processing. Sweet Room and Dirac both use DSP but don't rely on a parametric equaliser function. Roon can also use filters like those which Dirac and Sweet Room use, you don't have to use Roon's parametric equaliser which also uses DSP but is a separate function. Yes, DSP can introduce artifacts into the sound but so do any digital process you apply including the basic operations of a DAC. Anything you do with a digital signal can introduce artifacts just as anything you do with an analog system can introduce harmonic and/or intermodulation distortion and noise. Digital or analog, there is no way to completely avoid problems occurring in the signal path. Every thing you do to or with the signal has a cost. You can't avoid some level of DSP with a Devialet because no matter what resolution your input digital signal is, it's going to be upsampled to 392 kHz, analog signals are going to be converted to digital and upsampled, and then it's going to go through the DAC and other processes associated with the digital part of the analog digital hybrid amp and the Devialet delivers extremely transparent sound so it's possible to use DSP with minimal costs. If you really believe that DSP introduces artifacts which are detrimental then you should be avoiding amps with digital inputs and/or room correction functionality because it's impossible to avoid some level of DSP with such amps.

[Delija]

You can measure the results in the same way that you measure the room in the first place in order to decide what adjustments you're going to make. You've git exactly  the same problem if you use Devialet's Sweet Room.

How can you measure response with applied Roon PEQs using REW? That is mandatory, because simulation of final result is not good enough, as I already wrote.
On the other side, you can do that easily with Sweet Room or any other external DSP - between source and Devialet or between Devialet and external amp.

Speaking of DSPs, of course it is used for different purposes, but there are always audible artifacts when FR is changed - even if it is limited only to lower frequencies, e.g. below Schroeder frequency. No matter if PEQs, Dirac, Anthem ARC, Room Perfect, Trinnov or some other room correction algorithm is used. With some more than with others. It can be easily heard in systems with high level of transparency, resolution and dynamics and with proper audio material. Whether the benefits of room correction are greater than the drawbacks made by these artifacts - it's up to a person / system / room. In my case, I choose path as clean as possible on main speakers and do corrections only on subwoofers. One thing I cannot avoid - using high pass filter on main speakers, but I find Devialet's HP filter pretty good - its effects on the rest of the frequency range are really minor and in this case benefits are much bigger than drawbacks.


Long story short - usage of (automatic) room correction or manual PEQs in high level Hi-Fi systems should be avoided if possible. Speaker (and listener) positioning and room treatments should be priority.

[David A]

@Delija

You can measure the response with applied Roon PEQs in exactly the same way you measured it without the PEQs. Roon itself gives you no way to measure its output, you have to use your own mic and software, and you need a test track of some kind. If you use something like a white or pink noise track stored in your Roon library as your test signal for your initial measurements without PEQ, then you can also play that track again with the PEW adjustments active and measure the result in the same way as you made your initial measurements.

I agree with you that speaker and listening position setup should be priority, and I have a preference for physical room treatments, my room is treated with physical treatments and has no electronic room correction, but I do believe that electronic correction processes are a useful option for those who don't want to use physical treatments and/or who cannot set their speakers and listening position up optimally because of constraints on the use of the room. Yes, there there can be audible effects but they may well be smaller than the audible problems which your digital processing corrects and if that is the case, then DSP is a useful and effective technique and for many it is going to be the technique of choice for quite valid reasons. This isn't a black and white question, DSP is not something to be avoided at all costs. Like everything else, it is something to be used judiciously when appropriate to the circumstances and any problems it causes are often less detrimental than the problems it fixes.

[Delija]

All measurement software generates sound signal (frequency sweep) and measure the result - you cannot just use some "test track" with white/pink/brown/whatever noise.

Have you ever used REW?

[David A]

I haven't used REW but I do seem to remember first setting up an analog graphic equaliser something like 40+ years ago using a vinyl test disc with a suite of test tones at third octave intervals and making my measurements with a hand held sound pressure level meter. I made the initial set of measurements with each band of the equaliser set to 0 dB, adjusted those settings based on my measurement results, played the same suite of test tones again and remeasured, and found it rather easy to tell whether I'd made an improvement and whether I needed to refine my adjustments a bit further. Way back then REW wasn't even a twinkle in its software designer's eye.

It is actually possible to measure the frequency response of a system without using software that generates its own test tones. You just need a way of measuring actual sounds in your environment and I'd suggest that there are a lot more sound measurements made every day of environmental sounds than there are of test tones generated by the software used to make the measurement. People working in health and safety, as I did for some years prior to my retirement, are actually interested in the frequency spectrum and sound pressure levels of noise in the workplace for hearing protection purposes and they aren't interested in the slightest in the levels of test tones generated by their measurement tools because the one sound you can bet that people in the workplace aren't being exposed to is test signals from someone's measurement system. If it's possible to measure noise from machines operating in a workplace, it's certainly possible to measure things like white noise or specific frequency tones played on an audio system from a CD or from audio files on your own hard drive.

Have you ever tried using something other than test tones generated by software like REW? That shouldn't be possible if you're right and all measurement software uses its own test signals but strange as it may sound, there are people who want or need to measure things like the frequency response of actual real life sounds for reasons other than speaker design or correcting the frequency response of audio systems and there is software which allows them to make such measurements. There's absolutely no reason why that software can't be used with signals like white or pink noise or a suite of specific test tones at specified frequency intervals, and before and after measurements are actually quite easy with those options.

Audio measurement options are nowhere near as limited as you seem to think.

[Confused]

I have actually used REW, used with both frequency sweeps and pink noise played via Roon.  In REW you can generate a sweep file and save this as a WAV file to play in Roon.  The file uses a timing reference tone as a trigger to commence the measurement in REW.

Arguably better than the above, you can use the RTA measuring system which requires pink noise, which similar to the above can be generated in REW as a WAV file.  I only recently realised this method even existed after  @Soniclife posted about the "moving mic" method in another thread.  See link below.  (Thanks for this one Soniclife, a great tip, and it shows the power of this forum, we can all learn from each other for the greater good)

https://devialetchat.com/Thread-Sweet-Ro...1#pid96941

https://www.youtube.com/watch?v=6Riuwqzj...e=emb_logo

I have not enough time to fully explore what can be done with the above, but I have got to the early stages of experimenting.  Last weekend I was able to play pink noise in Roon, and play with the settings in the Roon PEQ and see the influence on the REW RTA in room response curve in real time.  I tried a similar thing with SAM, I could adjust the SAM% on the remote, and actually see the low end of the frequency curve rising and falling.  This alone was educational, as I could see that the main influence of SAM was operating at a frequency below my largest room node peak, so SAM was actually leveling off the very lowest bass versus the adjacent peak, rather than directly adding to it.  Fascinating stuff.  I will be honest here, it did need quite a bit of time experimenting with the various REW RTA settings before I could get this to work satisfactorily, and I am still learning.  That said, this does look like a very powerful tool, you can adjust SAM, adjust the Roon PEQ, anything you want, and see the results in REW RTA in "real time".

As an aside, playing with the above does serve to highlight the room issues, and also highlight how tricky they are to resolve.  Hence in my case leading to the conclusion that I really should focus on getting better room treatment in place before worrying about room correction too much.

EDIT:  How off topic is all this!  One day I will post about this in the Sweet Room thread, if I ever find the time to get everything fully worked out.

[Delija]

You are right about offtopic , but I think @zoran_pavlovic is already working on speaker positioning...




With Sweet Room, there are no problems - you can play REW sweeps through Devialet and that's it.

RTA can be easily measured with Roon DSP applied, but that's not enough.

I was following this thread about external sweep in REW some 5 years ago and as far as I can see it is still not implemented "nicely" in REW:
https://www.hometheatershack.com/threads...rew.56977/

There are indeed some workarounds with manual synchronization and they should work. I really wasn't investigating further since I still use XTZ Room Analyzer. 
My bad!
https://www.avnirvana.com/threads/extern...urce.2241/
http://www.roomeqwizard.com/betahelp/hel...s.html#top
https://www.avnirvana.com/threads/check-...post-26739



Thanks, @Confused!