Linn event at Goodwin's: digital room optimization

[Nyal Mellor]

Whilst it is always nice to see new approaches to fixing room acoustic issues I don't see how Linn's approach can really work as well as in room mic measurements.

Sure there are issues with in room measurements, not least the need to provide a mic, mic interface, computer software, dealer / user training, etc. However there is no way a software program can accurately simulate the response in a "normal" room with construction vagaries, windows, doors and other elements and an unknown set of speakers.

Even if we simplify things to speakers that radiate omnidirectionally in the low frequencies (fair enough this is most, but not all, speakers out there) then we still have the issue of construction vagaries. KYDG has purportedly shown that their CFD modeling is accurate to within a few dB when simulating response in rooms with KNOWN and PREDICTABLE construction (new build home theaters mostly).

However the Linn software, from what I can infer from the user manual (and not having a Linn player with the software here to examine) is not doing anything like KYDG's CFD modeling, nor does it account for non-rectangular rooms. FWIW we also do low frequency simulation for non-rectangular rooms using boundary element modeling but it is not as sophisticated as what KYDG are doing.

To truly do a good job with pure modeling you would need to be able to input non-rectangular spaces and locations of all windows and doors within as well as the directivity characteristics of the low frequency source and then the computer would need to solve the wave equation to predict the frequency response at the seating location. I don't see any of that in the Linn software.

Whilst I can't speak for KYDG I would suspect that when the time comes to actually calibrate and equalize the system then they take in room measurements, even after doing all the sophisticated CFD predictions. Why rely on predictions when you can collect actual data?

I'm sure when the Linn software is turned on in a rectangular, semi predictable rectangular room that there are sonic improvements. It's probably better than not bothering to correct the room issues at all. But I am not convinced that predictions will get close to real in room measurements at any point in the next few years.

 

[bonzo75]

And you heard all three in the same room and system?

No. But the meridian rooms were properly treated professional set ups while Linn was in normal homes, no carpet in one, close to the wall in both even though they do better pulled away from the wall, and no treatment. One of the rooms was really small

 

[asiufy]

Not sure where we disagree. You like Avantgarde, I like them too. I only like the trios, not the duos, while you might like the latter. I am not saying MSB is not good, just that Lampi is better

See, we don't agree in anything I'm yet to find an Avantgarde I don't like. And Lampi better than MSB? I can't agree to that...

And guys, you're all missing the point here. AFAIK, Linn doesn't claim their approach is "better than this-or-that". They just claim that, through their process, you're able to place your speakers wherever you want them, and obtain similar results as if you'd optmized their placement by hand.

I understand it's mostly marketing material, but please read the blurb on their webpage

http://www.linn.co.uk/music-systems/technology/space-optimisation

Basically, this is "DSP for the rest of us". They don't even call it DSP!

 

[Kal Rubinson]

If you can calculate peaks you can also calculate dips. It's physics.

Of course. However, RG's statement referred to the correction of nulls, not their identification.

 

[Kal Rubinson]

No. But the meridian rooms were properly treated professional set ups while Linn was in normal homes, no carpet in one, close to the wall in both even though they do better pulled away from the wall, and no treatment. One of the rooms was really small

I can see that you might have a preference based on memory of these very different situations but I would discount that as proof of anything.

 

[bonzo75]

I can see that you might have a preference based on memory of these very different situations but I would discount that as proof of anything.

Not really. I am quite sure the sound of the Linn was better. This is not like electronics where you need to A/B. But then Meridian speakers have never been to my taste - also, I am not claiming their DSP approach is better