Last minute Summer Devialet Event at OAC with Mathieu Pernot! 12 July 2018

[Hifi_swlon]

Also, there have been a few posts about room correction.  Mathieu did make one comment about RC that has not been mentioned, it was something along the lines of if Devialet were to do room correction it would be a complete system.  So it looks to me that there is some concern about the wide variety of ways that convolutions could be  supplied, but a closed all Devialet system might work.  I am not an expert on the inner working of RC, but combined with ADH I can see some issued with how headroom is dealt with, how changes in sample rates are dealt with, how RC interacts with SAM, this kind of thing.  As stated by Ian earlier, it is clear that RC is not on the short term road map, but I came away with the impression that a complete Devialet RC system might be a possibility in the future.  But don't hold your breath, as the saying goes....

That's really interesting, and I think shows how out of touch Devialet are with RC. OK, so an 'all in' solution would be fantastic, but the reality is its far too complex a task for anyone let alone Devialet.  To do it right requires complex GUI/UX work, and lots of technology and understanding of the science.  IMO only Dirac have nailed making it user friendly, so integrating that would probably be their only hope for a full solution, and one I'd happily embrace (there may be better out there, but at the end of the day something's better than nothing, and Dirac's a pretty good something).

There's a reason Roon etc have gone for a 'simpler' approach of convolution engine only:
-Its just established maths.  There are even free codebases out there, although probably a subject for debate about SQ among the various implementations.  I'm happy with the Roon one.
-It has minimal requirements for the user GUI side.  All you need is the ability to load a convolution filter (or set of files) from disk, a headroom value to be set, and a clipping indicator. And a way to toggle between filters and enable/disable.
And that's a working system.

I believe they can, and should, implement this on the CI and, that it's far simpler than they make out.

I dont think they should even consider the bigger picture version.  It would be lovely and ideal, but it also means it will never happen.  They are the least likely company to pull off such a thing, and I don't think they need to. Anyone considering room correction and convolution is probably going to be 'happy' to get their hands dirty and do it via loaded filters. That is a completely workable stage one and would, i think, keep most of their customers that want RC very happy.

If you think about it, its probably more complex to configure the Devialet, write it to an SD card, and load it into the D, than it would be to old a convolution filter.  i.e. users with enough technical knowledge to adjust anything on their Devialet could load a convolution filter.

Mathieu made a comment that they thought the Expert range would sell better in the US if it was bigger, apparently a lot of potential US customers somehow cannot accept that the tiny slim Devialet's can be pushing out 1000w RMS, or maybe they just prefer "big".  Mathieu said something along the lines of "it might have sold better if we had just put it into a much bigger case."

They should do exactly that if they think it would sell better in the US. Make a 2000 Pro and put it in a massive chrome cube. They could still make it beautiful.