Originally Posted by DanFrancis
Interesting screenshot, Amir. I'd like to see what that looks like measured with something NOT labeled "JBL Synthesis". Unfortunately, there's no indication of what that resolution is, nor is there an indication of the smoothing. It's an excellent response, to be sure, but what I'm trying to do here is utilize third-party tools to verify what the various parties involved in my experiment are "claiming".
There is nothing being "claimed" as in Audessey saying what it thinks it is going to do vs what it is doing. If the system is lying there, it is also lying to itself in how it corrects the room and something tells me that is not the case
That said, yes, I don't know what smoothing is applying to the graph so maybe there is some marketing going on there.
Then again, see how up front they are with showing the filter points and their coefficients/bandwidth. Double click on any of those points and you get a full view of the filter that you can modify at will. The filters nicely show what correction is done and at what level. It is very hard to lie there.
What was the distance from the mic to the loudspeaker? Arrival time of 35.6ms leads me to guess that the measurement distance was in the neighborhood of 20-something feet (I'm assuming roughly 10 ms latency for the DSP).
Your numbers are close. I don't have the exact figures. The JBL system uses 8 mics which we place at all seating positions and I think a couple in different places. Which one is used to arrive at the millisecond value, I don't know.
There are swings in that trace (pre-) of more than 10db in the low end, and a midbass depression of 4-5 db spanning almost the entire octave of 500-1k; 'guessing the highpass crossover is in the 600 hz range.
The response of any speaker speaker below the transition frequency is dictated by the room. So the "10 db at the low end" is to be expected. You can change the speaker and you will still get that hump or some variations thereof. The room dictates it more than the speaker.
Expanding, that is why we don't generally rely on EQ to solve low frequency response. That is actually the third step. I am writing an article on this but for a teaser, see the audio portion of this presentation on Video for Audiophiles
I am actually pleased in the way Synthesis managed to fix what it did there.
That pre-calibration trace isn't quite as rosy at it appears at first glance, and certainly the JBL hardware does a neat job of cleaning the FR. I'd really like to measure with independent hardware to see the real power/accuracy of what the ARCOS system does.
I am fine with that
. It is not a concern for me though as just like you, the results speak for themselves.
We used to have two parallel systems in our reference theater, one with planar magnetic speakers and Audyssey pro. The other as you see with JBL speakers and Synthesis EQ I showed here. Here is the picture from that presentation:
Who wouldn't fall in love with that speaker array and planar speakers?
Yet the truth of performance said otherwise.
Since Audyssey doesn't show or do post correction measurement, we did that manually and it was a disaster. It had the dips that you spoke of. And no amount of manually fixing them made a difference. The sound was just wrong, even if the response was flatter. We messed with it for a while but eventually decided to rip it all out as not one customer preferred that to our JBL system in our reference theater. Which I should say, is much cheaper.
BTW, the JBL software goes through a full calibration pass for the mics, the results of which it also shows together with the correction.
Although Synthesis hasn't appealed to me from the "sound" aspect; their technology has fascinated me for quite some time! Of course, it doesn't hurt to have the resources of the "mother ship" to play with.
Lot of people have that bias. But remarkable thing happens when you put said speakers behind screen and they can't see their ordinary/commercial look and the horns. And then they rave about their sound. Count me as one of those ultra biased people
. I was one of the people advocating the planar magnetic speakers only to eat my own words. Now I am in the process of putting the JBLs in my own personal theater. I will of course try to hide them just the same.
I guess the comment to be made is: less is more.
Definitely. As I mentioned above and in that linked presentation, Room EQ is the third step/tool in the process. The first is always, always multiple subs, optimally placed. The sins created there cannot be erased elsewhere.