-

Jump to content



Photo
- - - - -

Why do so many automatic configuration systems set centers/rears to "large"?


This topic has been archived. This means that you cannot reply to this topic.
125 replies to this topic

#1 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 05 2006 - 09:04 AM

As some of you know, I'm an advocate of using the large setting for all speakers-if they can handle that-for surround *music* systems as recommended by so many professional 5.1 mixers. But that is not exactly what I am discussing here.

What I was curious about was why the engineers that designed these automated parameter setting systems i.e. Pioneer's MCACC, Yamaha's YPAO, etc, would allow so many people's rear and center channels in their home theater systems to be set to large if that wasn't the better way to do things. But there are many people that advise ignoring this and manually re-setting them to small "because that's what they should be set to".

Not to imply that computers or engineers are infallible, but I would like to know the rationale behind that manual re-setting advice.

Because wouldn't trained audio engineers know a little bit more about this issue than the casual audio hobbyist? And as far as these systems' microphones being inaccurate as I've seen some people suggest, that can be (mostly) taken of by including correction formulas in the MCACC/YPAO/etc software. And a company doesn't want to have to deal with a multitude of lawsuits that involve their customers' speakers being fried, so I think they would design these systems very carefully.

BTW: I think the thinking behind the holy grail of crossovers (if a system even needs one at all), 80Hz, is because the huge majority of people will NOT buy a set up disc and a sound level meter, so a crossover was chosen that is a good compromise between sound quality and speaker longevity (and maybe a bit of smart marketing too: potential 5.1 customer "No WAY will I put five full-range speakers in 3 cubic foot cabinets and a subwoofer in my living room! And look at the cost!!!").

Anybody else have any thoughts on this?

#2 of 126 OFFLINE   Andrew Pratt

Andrew Pratt

    Producer

  • 3,813 posts
  • Join Date: Dec 08 1998

Posted February 05 2006 - 10:26 AM

I can't speak for those systems but with Sherwood-Newcastle's SNAP function it simply tests to see if the speakers can output a signal below the set crossover point...if it can they get set to large. Its absurd and flies against conventional wisdom but its the way the software currently works. As a result virtually all speakers other then tiny bose like cubes will most likely be set to large with this system even though they don't have the ability to play down anywhere near what a true large speaker should be capable of.

#3 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 05 2006 - 11:37 AM

Quote:
Its absurd and flies against conventional wisdom
Why?

And what is conventional wisdom?

#4 of 126 OFFLINE   Kevin C Brown

Kevin C Brown

    Producer

  • 5,713 posts
  • Join Date: Aug 03 2000

Posted February 06 2006 - 12:36 PM

To add to what Andrew said, those kinds of software test for output at low frequencies, but not necessarily the "quality" of that output. I.e., flat freq response *with* low distortion.

And even I see "conventional wisdom" as being: all small, sub on. Most of these receivers are going to be used predominantly for HT. And the low freq effects there are just too much for most speakers to *effectively* handle. Music *is* a different beast though, I agree.

And in addition, what I'm seeing these days is that most people will put too large an expenditure on the display, and not enough on the audio. (But then like some people, I grew up on the audio part, and just added video to it. Posted Image ) So even in that case, with not a lot of money spent on a *typical* speaker setup, you're just simply safer with all small.

As far as 80Hz, I think it came to be for 1 reason: THX, and that is the frequency at which point it's commonly taken that above which sounds are localizable, and below with they aren't.
If it's not worth waiting until the last minute to do, then it's not worth doing.

KevinVision 7.1 ...

#5 of 126 OFFLINE   Kevin. W

Kevin. W

    Screenwriter

  • 1,546 posts
  • Join Date: Oct 27 1999

Posted February 06 2006 - 03:35 PM

Can't you pick the speaker size and crossover then run the setup? If not why don't they allow us to do this? It makes no sense to run the auto setup and then change a couple off settings.

#6 of 126 OFFLINE   JeremyErwin

JeremyErwin

    Producer

  • 3,219 posts
  • Join Date: Feb 11 2001

Posted February 06 2006 - 05:29 PM

Somewhere I read that optimally, speakers should be flat to an octave below the crossover point. So, to crossover at 80 Hz, your speakers should be flat to 40 Hz. I haven't tested whether this advice is sound enough to follow.

#7 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 06 2006 - 05:32 PM

Quote:
To add to what Andrew said, those kinds of software test for output at low frequencies, but not necessarily the "quality" of that output. I.e., flat freq response *with* low distortion.
Good point....*if* there is low bass in those channels.

Kevin: this is something I don't know much about, but I haven't noticed that much low bass out of my own fronts/center/rears (which are set to "large") with the particular movies I own. Obviously what I own is only a tiny fraction of what's available and in turn a very small sample group, so this is partly why I brought up this issue to get more data (though I am also assuming others take the grills off their speakers so they can see if their woofers jiggle Posted Image ).

Attack Of The Clones is an example of what I'm talking about. Like during the first scene, when the Princess' 40s-era bomber Posted Image passes overhead: the ship's very deep rumble is isolated in the LFE channel & the woofers* of my fronts/centers/rears only produce a light vibration (sorry, I don't own an spectrum analyzer) even at high volume settings. Disclaimer! I didn't stand by every speaker for every bass-filled scene, so this may not happen ALL the time.

Purely for the sake of argument, say that all movies don't contain low bass-say, below 80Hz-in all the satellites. If the auto system designers knew that, it would be a good idea to set the appropriate channels to large to avoid the added distortion & electrical phase problems that can occur when using filters. And since a speaker's frequency response doesn't immediately *stop* at its printed spec i.e. it rolls off gradually, maybe the designers have this in mind also?

* eight inches in the fronts & rears, one 6.5" for the center

#8 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 06 2006 - 05:46 PM

Quote:
Somewhere I read that optimally, speakers should be flat to an octave below the crossover point. So, to crossover at 80 Hz, your speakers should be flat to 40 Hz.
I've read that too.

And if this rule were ridgidly adhered to, then a LOT of people's systems are already not performing at their best. My bass-reflex Boston Acoustic CR9s with single 8" woofers get down to 42Hz (+/-3db) and they are considered to be large speakers by many people here. What about all those satellites with 5.25" and smaller drivers? But I fully realize in the real world that compromises sometimes have to be made (money/WAF/etc), so I'm just bringing this up in a devil's advocate way.

#9 of 126 OFFLINE   JeremyErwin

JeremyErwin

    Producer

  • 3,219 posts
  • Join Date: Feb 11 2001

Posted February 07 2006 - 04:02 AM

Perhaps someone could extract the rear channels from the Master and Commander soundtrack and comment. Throughout the film, the surrounds are used to reinforce the feeling of being on a ship, so there's a lot of creaking and groaning going on. I'm not sure that the ambience would still be there if the (localizable) bass were directed to the front.

Often, DVD-A dispenses with the boom track entirely, either by creating a 5 channel program, or using the sixth channel for a full range effect channel (height channels come to mind). Some of the critics I've read seem to be awfully suspicious if the "5.1" label is used too prominently. (The purpose of a DVD-A surround program is not the recreation of ambience for DSP challenged individuals...)

#10 of 126 OFFLINE   Brian L

Brian L

    Screenwriter

  • 2,882 posts
  • Join Date: Jul 08 1998

Posted February 07 2006 - 04:24 AM

When I helped set-up a friends new Denon 3805, it too selected large for the fronts and center. We manually chose small.

There are many good reasons to cross your mains over at something around 80 hertz when you have a sub; no need to re-hash that here.

As for the old chestnut about having a speaker be flat to 1 octave below the crossover, I say bunk. There are many varibales in that equation, but with the typical 12 hertz/oct crossover, you don't want to go that low.

Here is an article that I bet is near and dear to KCB's heart(!), that makes a case for what is wrong with the crossover settings in most AVR's when used with loudspeakers that go down to 40 hertz or so.

http://www.hometheat....es-6-2005.html

Brian

#11 of 126 OFFLINE   Andrew Pratt

Andrew Pratt

    Producer

  • 3,813 posts
  • Join Date: Dec 08 1998

Posted February 07 2006 - 06:43 AM

Optimally speaking for HT use a speaker needs to be able to reproduce the full audio spectrum to be considered 'large'. If it can't then it should be set to 'small' at a cross over point that's one octave higher then its F3 point (point where it's response is no longer flat)...though as the link above points out there's a number of factors at play. The problem with many of the auto EQ's is that they simply test to see if there's any output below the cross over so if you set it to the THX recomended 80 Hz and you're little 5.25" drivers can output down to say 60 Hz they well get flagged as Large. I can't speak for Pioneer or Denon's system but with Sherwood's you can't first preset the speaker sizes as it does that as part of the calibration process. You can edit this after the fact but I'm not 100 percent sure what that does to the EQ curve.

#12 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 07 2006 - 09:36 AM

Brian: that's a good article (who is "KCB"?). Some thoughts on it:

1) while most of the speakers they reviewed may get down flat to 40Hz (this was mentioned in the article), from what I've seen here and on other forums, many people don't own such speakers. Instead they own JBL E20s, Athena Micras, etc, etc. So I have to wonder what the effect of using an 80Hz xover will have on these types of speakers and the resulting sound.

>>> After reading the article, I think I am starting to realize why so many sat/sub systems (those which use small sats like the ones mentioned above) for me don't sound quite right with music.

2) I see what they mean about being able to choose different xover slopes but like they said themselves, IMO a separate "advanced" menu containing such choices definitely should be set apart from the regular set-up menu. That's getting into some very hairy audio tech territory & expecting casual HT owners (i.e. "Regular Joes") to deal with that is asking WAY too much.

>>> I think it would also help a lot if we knew how dvd movie soundtracks are mixed. Maybe Denon, Sherwood, etc know something we don't.

Maybe those auto setup systems are also detecting the cutoff slopes of the user's speakers, and if they are the right type, running them in large mode? I wish someone from Pioneer, Yamaha, etc would discuss their reasoning behind their b.m. systems and why their auto setup systems work the way they do. They must have some logical reason for designing them the way they do.....unless revealing that reason would cause unneeded worry on the customer's part? Sort of like the fact that the FDA allows a certain percentage of insect parts in food & which is perfectly safe, but if a food company started making that public some people would stop buying their brand.

#13 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 07 2006 - 10:23 AM

On a side note: I just got through reading that article mentioned in the one Brian linked, the one entitled "Miscellaneous Ramblings on Subwoofer Crossover Frequencies". As far as HT systems were concerned, I truly thought there were solid bass management hardware standards in place. But according to the authors, there aren't and b.m. systems can vary wildly.

In the articles I've read about surround music production and REproduction, there were hints about this......and now I'm starting to see why so many 5.1 mixers advise not using b.m. to play back their mixes. Not just because the playback end can disturb the music but because if THEY use some form of b.m. on their end, when combined with the user's b.m. characteristics, you could create a very messy acoustical and electrical environment that could play havoc with all those different instruments, vocals, *intentional* phase effects (which are used a lot in pop & rock music), etc, etc.

These companies should get together and have a conference to hammer out a set of basic standards that all audio manufacturers would adhere to. Yea, this is about as possible as Charlie Brown being able to kick that football but I can still dream. :wink:

#14 of 126 OFFLINE   Brian L

Brian L

    Screenwriter

  • 2,882 posts
  • Join Date: Jul 08 1998

Posted February 07 2006 - 10:34 AM

KCB = Kevin C. Brown...he posted previously in this thread, and has had a debate or two with Mr. Deering regarding the subject of the article I linked.

EDIT: Wrong Author! Sorry, it's Brian Florian that has debated Kevin...I think!!!

While I think I understand the point of the article, I would say that its small beer compared to a lot of other ways that BM can be screwed up. A few extra DB over the crossover region is not the end of the world.

While I have not researched the speakers you quoted, I would go out on a limb and say that they probably do get down to 80 hertz OK. And if they were to roll off below that at something like 12dB/Oct, then they would work perfectly with a crossover that has the same slope.

I may be over simplifying, but that's the whole THX deal; the crossover and the speakers should have complimentary slopes.

I guess I have sidetracked the premise of the thread....sorry about that. Maybe the Denon's and Pioneer's of the world do know something that we don't, but I will stick to setting ALL of my loudspeakers to "small" for all the reasons so often discussed:

1. Flexibility in placement of my main channels for best imaging, since that location does not normally occur at the same spot that produces optimum bass response. Take the low bass away form the mains, and put it in a sub, and you can position both where they can do what they do best.

2. Giving my AVR or main channel amp a breather by not having to deal with low frequencies.

Another quibble I have with some of the auto-setup programs is that they don't EQ down in to the bass, where it is arguably the most important. I think some of the higher priced stuff does (Audessy, is it?), but the more affordable stuff stops at 100 hertz or so.

A side note: I was over at the Outlaw site, and there is a post from Outlaw indicating that the ICBM is now going out of production. I would have never guessed when I bought mine in early 2003 that three years later, they are still worth their weight in gold if you have one of the many Uni Players with sub-standard BM.

I retired mine to a 2CH system when I bought my Denon 3910, and it is doing yoeman's work there with my little baby NHT Absolute Zero's and my old war horse Klipsch sub.

Brian

#15 of 126 OFFLINE   Brian L

Brian L

    Screenwriter

  • 2,882 posts
  • Join Date: Jul 08 1998

Posted February 07 2006 - 10:37 AM

Quote:
These companies should get together and have a conference to hammer out a set of basic standards that all audio manufacturers would adhere to. Yea, this is about as possible as Charlie Brown being able to kick that football but I can still dream.


Lance, I think we were posting at the same time...I missed this before I posted.

Actually, Gary Reber at WSR has been pushing for a set of standards for MC mixing that takes BM in to account. I am not sure where it stands now, but he has gotten a lot of resistance for a lot of reasons.

Brian

#16 of 126 OFFLINE   JeremyErwin

JeremyErwin

    Producer

  • 3,219 posts
  • Join Date: Feb 11 2001

Posted February 07 2006 - 10:39 AM

I don't even know the crossover type in my receiver, much less the slope of my speakers. If one had the relevant data, one could write a computer program to optimize this out. But most receivers don't allow this level of customization so it's all for nought.

#17 of 126 OFFLINE   Kevin C Brown

Kevin C Brown

    Producer

  • 5,713 posts
  • Join Date: Aug 03 2000

Posted February 07 2006 - 12:26 PM

Hee, hee. Posted Image I firmly uphold that, for example, if you're going to use 80 hz as a crossover for a (set of) speaker(s) to a sub, then those speakers should be flat to at least 1/2 an octave lower than 80 Hz, or to about 60 Hz. A full octave lower, 40 Hz, is even better.

The reason why is easy: crossovers have slope, and a speaker's natural rolloff has slope, and you do not want them to add. One example: a speaker is -3 dB at 80 Hz. (Which, unfortunately, a lot of satellites specifically designed for HT aren't even that good...) But for the typical 12/24 crossover, it puts the speakers and sub both at - 3dB at 80 Hz, (crossover math: -3 + -3 = 0...), but now you add in the speaker's - 3dB, and all of the sudden you have a dip at the crossover. The idea being, the roll off of the speaker has to be sufficiently far enough away from the crossover itself so the two slopes do not add, to avoid a dip in response there.

THX certified speakers are different, because the slope of the crossover is designed to be used in conjunction with the slopes of the speakers.

Lance- This is extreme heresay, but I came across a dude on AVS once who somehow was attached to actual movie soundtrack mastering. And what he said, that surprised me a little bit, is that in his experience, low freq material was mixed in equally, monophonically, to the L, C, R, and LFE channels. I think the idea is to try and minimize phase issues between all of those speakers, as well as get the highest gain of the signal. (Although if everything is going to crossover to the sub, you could also likely be overloading the sub's signal.) I don't know if this is true for all soundtracks though. I suppose an interesting experiment would be to listen to some demo low freq sections of few DVDs, all large, sub on vs all large sub on (but disconnected) to see how much low freq material is steered to the sub just by mastering alone. Of course, this assumes the the person's mains would be able to handle enough bass that it's a valid experiment to be able to tell the difference. Posted Image

I've come across a few arguments each way for BM, mastering, etc. My conclusion is that there should have never been a separate track allowed for LFE. Having a 5.1, 6.1, or 7.1 physical setup is fine. But there shouldn't be a separate LFE channel allowed for the source. Just let the receivers and pre/pros do all the BM necessary. Eliminates the common phase problem of crossed over mains info being added to non-crossed over LFE info and the difference in phase that results. Or, using the same low pass for mains and the LFE, but then throwing away the upper part of the LFE signal entirely.
If it's not worth waiting until the last minute to do, then it's not worth doing.

KevinVision 7.1 ...

#18 of 126 OFFLINE   Andrew Pratt

Andrew Pratt

    Producer

  • 3,813 posts
  • Join Date: Dec 08 1998

Posted February 07 2006 - 03:53 PM

Sherwood's SNAP EQ does work on the subwoofers and does a pretty decent job at that. Its not nearly as good as what I can dial in with my BFD but then I wouldn't expect it to either.

#19 of 126 OFFLINE   LanceJ

LanceJ

    Producer

  • 3,168 posts
  • Join Date: Oct 26 2002

Posted February 07 2006 - 04:50 PM

Quote:
Maybe the Denon's and Pioneer's of the world do know something that we don't, but I will stick to setting ALL of my loudspeakers to "small" for all the reasons so often discussed:
I don't blame you and until the 5.1 music formats came along, that's what I told my customers waaaay back in the early to mid 90s (by that time all our receivers had separate sub outputs) and how I had my own system set for two years. With this configuration I never noticed anything out of the ordinary (though switching out my little Radio Shack Minimus 7 speakers w/5" woofers for some Pioneers with 8" woofers made a HUGE difference).

The only way to figure out what is happening would be to take one system and listen to it with no b.m., and then with b.m. This would be quite time consuming, even assuming the speakers could handle it, and the results would really only be applicable to THAT system.

Quote:
But most receivers don't allow this level of customization so it's all for nought.
After reading that Secrets article, I agree. Right now I'm just trying to figure out what is *possible* so when it comes time to buy different speakers or buy a new receiver or just move my system to a different room, I'll know better what to do.

Quote:
is that in his experience, low freq material was mixed in equally, monophonically, to the L, C, R, and LFE channels.
Wow. That's....confusing. Because the LFE channel was designed specifically for low bass effects so the other channels wouldn't get overloaded and vice-versa.

Quote:
The reason why is easy: crossovers have slope, and a speaker's natural rolloff has slope, and you do not want them to add.
I thought it was interesting that in the Secrets article regarding this issue they said "Ported speakers will never fit". I'm assuming this is because ported speakers have rather sharp cutoffs below their rated low end limit, whereas sealed designs slooooowly cutoff (I'm not trying to start a sealed vs. ported debate, because every one of my system's speakers are ported). Where's Advent and Acoustic Research-the REAL versions-when you need them?! Posted Image

For more illumination on this issue, check out this discussion over at stevehoffman.tv:

"Hooking up TWO subs"

In it I asked (post #16) some questions related to this. IIRC "Serious Fun" is a professional 5.1 mixer & I've also seen his posts on a surround pro forum. I asked more questions on the 2nd page too (which I just read the answers to tonight! [I don't visit there much anymore]). And make sure to read post #26 on time aligning the sub with your sats. Unfortunately my 4.5 year old receiver doesn't have this capability.

One thing I haven't mentioned but will bring up just to make sure y'all know I'm aware of it: even if running five full-range satellites on the "large" setting was the best thing to do, the fact that each speaker is in a different physical location *could* seriously effect their bass response. I just don't know what is worse sonically-speaking, that, or the filtering/phase/xover problems.

#20 of 126 OFFLINE   Kevin C Brown

Kevin C Brown

    Producer

  • 5,713 posts
  • Join Date: Aug 03 2000

Posted February 07 2006 - 06:22 PM

It *is* interesting to think about and discuss all this, without a lot of sniping back and forth... (Once again, just appreciating the civility here at HTF vs at least 2 other places I can think of... Posted Image )
If it's not worth waiting until the last minute to do, then it's not worth doing.

KevinVision 7.1 ...


Back to Receivers/Separates/Amps



Forum Nav Content I Follow