Originally Posted by giomania
I am not sure about the intent but maybe the originator can clarify that question.
I did tag him in one of my posts in this thread when we are questioning whether different control points for this procedure are optional but he never responded.
Sorry guys, been busy with other things for a while. As the guy quoted in the FAQ let me try to answer some of these recurring questions.
1) The values for the target curve were simply an example, following a somewhat Harman-Curve type shape in the bass region. Yes, feel free to try different variations on that curve to find out what you like best. For example, if you want the sub to be flatter a little longer you can change the 55 Hz point to 65 Hz, etc. If you want to boost a bunch more (like 12 db
), it may sound better to have a more gradual transition back down to zero--instead of going to zero at 160, you could try 180, 200, etc. It's very easy to make a bunch of files with different curves and try them. I'd suggest just worrying about your L&R to make it faster to make a bunch of files and you may want to limit the correction to 500 or 1000 Hz so you can get the most apples to apples comparison with Direct mode (especially music lovers who like how the system sounds in that mode). Once you've found the curve you like, you can measure/fine-tune it and then apply it to the other speakers.
2) Yes, I do recommend applying the same curve to all the speakers except upfiring Atmos (in the bass region--the treble, limiting or not limiting, etc is an entirely separate subject and one where you may want to treat different speaker types differently). It's much easier to simply type in the same values than re-invent the wheel for each speaker. While it may seem silly to be typing in +6 db boost at 20 Hz for a speaker that rolls off at 100 Hz, for example, don't worry about it--Audyssey isn't really going to try and do that and blow your speaker. It'll get the most out of the speaker within its capability and handle the rolloff appropriately. While not a replacement for accurate measurements, the predicted "after curve" does give you a pretty accurate idea of what Audyssey will try to do
with that speaker in the bass region, given your target curve and its capabilities. If you try and boost 10 db and see the after curve for a speaker shows it only rising a couple db before rolling off again, well, now you won't be surprised when you measure it.
You're probably going to set the crossover well above the rolloff anyway. So why do it? The goal is to get the speaker close to or on your target curve as soon as it can be made to do so. If you do not and just let Audyssey do the bass "flat," Audyssey will often smooth out the rolloff, making it more gradual which results it it starting even sooner which can cause a bigger mismatch at the crossover. Even if you only get a little hump or flat-ish spot before the speaker rolls off, it'll probably match the sub at the crossover better than if it had a more gradual rolloff starting at a higher frequency.
3) Actual measurements to double check results are highly encouraged. First, regardless of any of the above, checking you get a smooth blend at the crossover point and applying the sub distance tweak as necessary is something that pretty much always has to be done with Audyssey to ensure things are working properly. After that, checking the results of the curve you made above is the only way to really know what you have. I focus these measurements on the LCR and sort of figure what happens with the surrounds/Atmos is going to happen (don't go doing the sub distance tweak to better integrate a surround speaker and mess it up for your LCR). When doing this, ensure you're only measuring a single channel so the sub isn't playing the bass for more than one speaker or it'll obviously measure too high against a single speaker. Measure the speaker set to large, then measure the speaker set to small/crossover at 250 (to get it out of the way) with the speaker unplugged so you're only measuring the curve of the sub. Those two lines should lay on top of each other for a range if you have successfully corrected both the the same curve. You may need to bump the sub up or down a db or two to put the lines on top of each other. That'll give you a good idea of the range of acceptable crossover values that won't change the tonal balance--if you put the crossover at a point they've diverged (or are about to) that will result in a mismatch and change the tonal balance one way or the other.
4) Yes, putting a curve like this on the sub will affect LFE. But you need to look at the big picture. Most typical curves like this will only have the sub down 2-3 db at 100 Hz and maybe 3-4 at 120 (depending upon details, of course). There usually isn't (or shouldn't be) much material in the LFE channel that high anyway. Also, many people lower the low-pass on the LFE anyway for some very good reasons making it pretty much a non-issue. I consider getting things right with the main channels, especially for music, much more important than having the exact perfect tonal balance of some explosion in the LFE...since nobody knows what that's supposed to sound like anyway.
I think that hits most of the most commonly asked questions lately. Let me know if you have any more.