New White Paper posting

[gedlee]

Originally Posted by Speedskater

Dr. Geddes, I just noticed while driving up Route 23 to Midland, MI (to of course go speedskating) that your family name is all over the Ann Arbor area. Why you even have a road.

There is a single road that goes into Ann Arbor, but there is no established connection between the origin of the road and my family. It apears that the road existed before my family came to the US from Ireland, although it could have been some previous relatives.

My paternal grandfather was the black sheep of the family (he was a drunk) and so I have no connection to his relatives at all. Although I know that he had many. They just never talked to him.

[Lazarus500]

Thanks for the white paper. It is nice to have all of that info in one place. Did you know that you taught all of that to people who were willing to pay attention and assimilate the info from the 'Geddes on Waveguides' thread? It is amazing how that works when one listens and assimilates.

This sentence below comes from the 3rd full paragraph of page 5 of 14 on the pdf.

"It is well understood that the ideal frequency
response is basically a flat curve, although a lot of
evidence suggest a small drop in the respond at
HFs (mostly room averaged power response) in
smaller rooms."

Is that to read response not respond and end in something like "is advantageous."

Thanks again for all you have taught me.

rick

[pjpoes]

Hey Dr. Geddes, I have read a few times that people find the most desirable frequency response to vary with level, but that, in terms of small rooms, a HF rolloff is desirable only at mid to high volume levels, with somewhat of a boost in both the bass and HF's at low to mid volume levels. While I'm somewhat referring to the loudness curves out there, I'm also referencing things like the new dynamic loudness compensation systems developed by Dolby, DTS, THX, and Audyssey. The only one I've heard is THX Loudness Plus which I honestly would say is among the most impressive new effects I've yet heard. I hope that my next surround processor possesses some form of this as I think its very effective at improving the perceived integrity of the soundtrack. It's my understanding that this technology combines dynamic range control, a modern digital interpretation of a loudness control which varies with level relative to reference, and changes the level of the rear surrounds.

From that standpoint, it would seem like the goal then of the speaker design is to have a speaker with a relatively flat response, constant directivity as you of course do, and then allow this post processing to impact the timberal balance desired for the room. Admittedly, adding a small amount of HF roll off isn't all that hard to do in your crossover design, doesn't involve extra components, and could be made flat if the builder so desires.

[markus76]

pjpoes

Measured Dolby Volume in the Emotiva UMC-1 lately:



Now compare the lower 4 curves from above to what they should look like:



Either the UMC-1 is flawed or Dolby Volume.

[pjpoes]

I can't speak to what you have presented, I don't know enough about any of it. I have read on Audyssey's site the argument that Thom gives, which is that the original and subsequent loudness curves were crudely measured and are incorrect. That new studies were done for Audyssey's undertaking, and these results were utilized to develop their curves. I don't know if it's true or not that the original, and especially subsequent loudness curves are incorrect, but I know they are still references a lot in current research.

I haven't heard Dolby Volume or Audyssey's, only the THX system, and I really liked the effect. I did read on Dolby's website that Dolby was focused on intelligibility and bass perception. It isn't clear to me if changing the high frequency response was a part of what they did, and from those measurements, would imply to me that they are simply changing the relative bass level, which actually seems consistent with their verbal claims. Conversely, the other two systems (it appears DTS doesnt have a version of this?) do discuss changes in accordance with loudness curves, and imply that the response changes at both ends, and not just the relative bass level. This may imply that the Dolby system is inferior to the other approaches.

If you buy into Audyssey's initial claims that a room's response can be corrected via FIR filter correction for both time and amplitude in a beneficial way, their approach then looks the most interesting as it seems to function in conjunction with these filters. However, I get the impression that Dr. Geddes feels that the negatives outweigh the positives on this approach to room acoustics. I am still very interested in trying this system, and have admittedly never really heard a proper Audyssey MultiEQ XT setup professionally installed, and would really like to. I've considered buying a processor with it many times, but have always wondered if the negatives would outweigh the positives.

[markus76]

Originally Posted by pjpoes

I can't speak to what you have presented, I don't know enough about any of it. I have read on Audyssey's site the argument that Thom gives, which is that the original and subsequent loudness curves were crudely measured and are incorrect.

Here's how the ISO 226:2003 loudness curves were obtained:

http://www.nedo.go.jp/itd/grant-e/re...pdf/is-01e.pdf

[gedlee]

I agree with Markus, the ISO curves are accurate and the implimentation is flawed. Of course the manufacturer is going to claim they are right. This is audio, its done all the time.

[taj]

To be of most benefit, the loudness curve would need to be applied from end to end at an established baseline. Being an SPL issue, this would not only require that the output system (the Home Theatre system) but also the input system (the re-recording mixer at the film post studio) be calibrated to a predetermined Flat@Some_SPL level.

I suspect the film industry may be doing this with set standards, but the music industry (my background) sure doesn't.

In a music recording studio, the producer (not the engineer as most non-music industry folks believe) decides what volume the mix is made at based on his or her own whim. And for that reason I've always advocated that music lovers should try adjusting the playback volume to find a level that sounds most spectrally balanced. That will probably be close to the level it was mixed at. It's a crap-shoot. Even on a single CD from a single artist, different songs may have been mixed at completely different SPL levels.

And as a side note, music is mixed MUCH louder than most people would expect. And IMO, typically sounds better when played back at somewhat higher levels than are typical. Critical listeners take note.

..Todd

[pjpoes]

Thats no shock that music is mixed loud, just read the hearing damage (tinitus) studies done with recording engineers which show pretty high average sound levels exposure in their job.

Movies are better than music with levels, but they do vary a lot. If a movie is certified THX, then we know the standard. Really, the THX standard was a great idea, matching the playback equipment to the sound engineering levels meant that everything was properly controlled. The problem was that there is no requirement that studios follow it unless they seek the expensive THX certification on the movie. Many engineers follow that or related standards, but still, plenty of movies vary from this standard, which makes some volume hunting necessary.

The loudness curve argument you give is the same as the one Audyssey gives. They claim this is why the Audyssey correction is necessary first for it to work properly. While I'm sure this statement is true, I don't see how the loudness curve isn't still just as effective without. It's still going to change the response in accordance with the curves relative to the response of the speaker uncorrected. That means that the speaker response will change, in room, in the same relative way that the curve suggests, while maintaining the character of the speakers themselves.

[taj]

And an afterthought on that same topic...

I once toured with a keyboard player who worked on the music score for a dramatic TV show. He worked on that all day and FedEx'ed his work to the director daily. He monitored using headphones since he worked in his hotel rooms. What level did monitor at? Probably whatever level drowned out the air conditioning unit in the room.

Then it was mixed by others into the TV show's audio tracks. So it's unlikely there will ever be a real 'sweet spot' even in a standardized industry, unless the renegade music industry plays along -- and that's also unlikely, knowing musicians as I do. Art is whim.

..Todd