Home Theater Forum and Systems banner

41 - 60 of 110 Posts

·
Registered
Joined
·
1,784 Posts
Good morning everyone.

These are some interesting thoughts to be sure. It is my strong opinion that using a center channel in larger rooms is a must and in general that center channel should be voice matched to the mains. This is as we all know oft times not easily done as most center channel speakers are smaller and less capable than the mains which is odd considering the amount of work the center channel does. As Mr. Lumens room is somewhat smaller, there is no real reason to use a center. Having heard the Salon's, I can say that they are not as lively in the presence region as the B&W's and I would imagine if it comes down to it they may be too much speaker for this room. I know that sounds silly but they are substantial and may need a bit more room to breathe. The B&W's are well sized for the green room and leave substantive space on all sides although having said that, i think Lumen is not overly pleased with their mild tubby-ness in the lower ranges.

This is such a hard hobby made more complex by all the new digital gadgets and geegaws. I do hope our musicologist buddy does find a way to make everything work to his satisfaction.
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #42 (Edited)
A target curve suggestion: Add a tiny bump, 0.5 dB to 1 dB, between 2 kHz and 3 kHz.

My target curve file includes
1000 hz, 0 db
2000, +1
3000, +1
5000, 0

A lot of speakers have a little "liveliness" lift in that region by design. If 1 dB sounds a little extreme, try 0.5 dB. Speech intelligibility will be greatly improved.
Huge help and appreciated! Jogged my memory of a chart I came across which defined typical sounds (wooly, bright, etc.) in terms of frequency. A google search turns up many variations just as useful.

Just using this space as a notepad for the moment. I also remember dialing in a 2-3dB cut from around 2kHz-4kHz to tame perceived harshness. Should've consulted the chart instead of tuning by ear on limited source material. Or better yet, commit it to memory. Anyway, I can always develop different curves for different quality sources by taking advantage of my processor's ability to download different curves. But having to connect a laptop to run Dirac Live/LE is inconvenient. I'll have to dig out he user manual to see if I can download from a thumb drive instead.

Relationship between mains position and toe-in relative to center channel plants Salons' huge reflective side panels into listening path. Room correction s/w may compensate, but the goal is to make its job easier, so try some absorptive panels. They can be temporarily leaned in place against the speakers on the floor.

Remove wall panels at 2nd reflection points to help liven listening end of room. Would like to swap 16" dia traps with 20" ones, but too much work for one afternoon.

Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #43
Those of you still waiting for hard data will be disappointed at my lack of cooperation. The best I can do at the moment is to post the chart I've been using (see below). It's an EQ guide intended for recording engineers, but which applies to sound reproduction as well. For sound in the home, PEQ can deal with fine adjustments over a broad frequency range and vice versa. We may not have precise control at the end of the room correction chain if our AVR leaves us with only a house curve to manipulate. Individual PEQ bands which are independent of the room correction presets are needed to further refine the sound. Through a combination of manual PEQ and slight change to my house curve, I was able to coax speech intelligibility from the system. Now to perfect it (what a lofty goal)! Still would like to revamp and recal a few things in the room. Next 3-day weekend?
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #46
Lol! Insert famous excuse "It wasn't my fault" from the Blues Brothers here.

I confess! I used PEQ as a band-aid because by the time I found out I was disappointed in the auto target curve, I'd already disconnected the laptop and tidied up Ethernet cables and such.

And my room is a How Not To HT that belongs more in the critique thread than it does here! But it sounds decent enough until I can commandeer larger quarters MWUUAAHH-Hah-Hah!

Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #47
Now that I've been able to watch a few movies and concert videos, I can honestly say the sound is much improved (not just different). Voices have been lifted out of the mix so to speak. The harshness/scratchiness that originally compelled me to dial-in the 3dB dip centered at 3kHz has only slightly returned - enough to make itself noticeable, but not quite enough to be objectionable. Deep-bass vibrates the slab without creating bloat in the upper bands. "Punch" is back and improved over what I can remember (though I may have missed it for so long I forgot it wasn't there!).

Some heavy hitters have checked in here, and I still mean to take their advice regarding phantom center tests. That should clear up whether the harshness is perceived, or actually originates in the mains or the center. I did wind up taking their advice regarding low frequency masking of dialogue, and dialed-down the +3db lo-shelf to +1.5dB. The original value relied on guesswork rather than good reasoning, but that's what sounded good to me in the past, so that's what I went with. Last night I remembered Wayne's excellent House Curve article, where he states that "A house curve is perceived flat response – that is, it sounds flat, not measures flat. Thus it has to be subjective." He also mentions that a good Rule of Thumb is: "The smaller the room, the steeper the curve." I also remembered his System vs Program Compensation article, and am anxious to re-tune my curve after having reviewed both.

A few passages from the latter article bear repeating:
...The notion of calling up a different house curve for everything that comes down the pike - or as I’d characterize it, deliberately un-calibrating the system - is not the proper way to do things. If certain discs or TV shows have, for example, bloated midbass – fine. I assume that’s the way the director or producer wanted it to sound, right or wrong. I’m not interested in re-tuning my whole system to “unbloat” his idea of “correct.” I have the benefit of a real time analyzer connected to my system that gives a visual display of all audio-program frequency response, so when I hear something that doesn’t sound right – bloated midbass, limited low extension, screaming treble, what have you - the RTA confirms that is indeed what the program is generating. It verifies my system is calibrated correctly and is accurately reproducing what it’s being fed – a rather satisfying feeling, actually.
I simply LOVE watching the lights on my RTA dance to the incoming signal! But I've never used that function in an official capacity. What a great way to (temporarily?) quelch Audiophilia-Nervosa! Who of us hasn't sat in the sweet spot and been underwhelmed or dissatisfied with the sound? Who hasn't wondered whether a new player or processor would be the next big step toward audio nirvana? Well, having one of these puppies in the chain is key to washing away all pyschosomatic doubt!

Certainly, if you find yourself making the same adjustments on a regular basis – e.g. always dialing up the treble a few notches – then tweaking the system a bit further is in order to finalize your settings. With a properly calibrated system, most program material should be “in the ball park” as far as sounding balanced. That’s the goal, since perfection is unobtainable. Indeed, this is the one downside to having quality electronics and good speakers all properly calibrated for the room: you will quickly find out just how bad a lot of programming really is, especially when it comes to bass.
.
<snip>
.
I typically just use the receiver's tone controls or my subwoofer remote control (for level adjustment) for times like this, but it would be easy enough to dial in a pre-set curve of some type for certain problematic TV channels or shows that you frequent, if your system has the capability. But even such pre-sets should be recognized for what they are - temporary settings.
You're just too practical and convincing, Wayne. :T
I was wondering how I'd download a multitude of curves to an AVR with no presets. Even so, I'll aim to zero-in on a satisfyingly homogenized system curve. By "homogenized," I mean sufficiently generalized such that most source material will sound good.
 

·
Premium Member
Joined
·
5,797 Posts
Now that I've been able to watch a few movies and concert videos, I can honestly say the sound is much improved (not just different). Voices have been lifted out of the mix so to speak. The harshness/scratchiness that originally compelled me to dial-in the 3dB dip centered at 3kHz has only slightly returned - enough to make itself noticeable, but not quite enough to be objectionable. Deep-bass vibrates the slab without creating bloat in the upper bands. "Punch" is back and improved over what I can remember (though I may have missed it for so long I forgot it wasn't there!).

Some heavy hitters have checked in here, and I still mean to take their advice regarding phantom center tests. That should clear up whether the harshness is perceived, or actually originates in the mains or the center. I did wind up taking their advice regarding low frequency masking of dialogue, and dialed-down the +3db lo-shelf to +1.5dB. The original value relied on guesswork rather than good reasoning, but that's what sounded good to me in the past, so that's what I went with. Last night I remembered Wayne's excellent House Curve article, where he states that "A house curve is perceived flat response – that is, it sounds flat, not measures flat. Thus it has to be subjective." He also mentions that a good Rule of Thumb is: "The smaller the room, the steeper the curve." I also remembered his System vs Program Compensation article, and am anxious to re-tune my curve after having reviewed both.

A few passages from the latter article bear repeating:

I simply LOVE watching the lights on my RTA dance to the incoming signal! But I've never used that function in an official capacity. What a great way to (temporarily?) quelch Audiophilia-Nervosa! Who of us hasn't sat in the sweet spot and been underwhelmed or dissatisfied with the sound? Who hasn't wondered whether a new player or processor would be the next big step toward audio nirvana? Well, having one of these puppies in the chain is key to washing away all pyschosomatic doubt!


You're just too practical and convincing, Wayne. :T
I was wondering how I'd download a multitude of curves to an AVR with no presets. Even so, I'll aim to zero-in on a satisfyingly homogenized system curve. By "homogenized," I mean sufficiently generalized such that most source material will sound good.
Get a MiniDSP DDRC=99A...then you can switch between tunes remotely via your remote control. You can store hundreds of calibrations if you want, but you would have to load from your collection to 1 slot before you could select remotely.
 

·
Registered
Joined
·
2,454 Posts
An easy way to check your center for clarity is to take it outside (nice to have a piece of speaker wire long enough so you don't have to pull the stereo out as well) and away from boundaries & give it a listen. I'm always floored how good speakers sound outside compared to inside. This will tell you very quickly if the room is the problem. I've come to believe that diffusion is critical to good sound.
 

·
Banned
Joined
·
4,838 Posts
Sorry if I sound like a killjoy, but my opinion is that it is easy to get caught up and lost in the "house curve follies" - so many to choose from, let me try them ALL!

It is a lot of work, and most (all?) of them will probably sound terrible to you.

My suggestion: start flat and modify from there. Keep a "flat" slot for comparison so you can A-B and compare. Be sure to listen at several different volume levels.

Another rule of thumb: The better (smoother & cleaner) the tweeters, the more highs you will enjoy, regardless of room size.

DO NOT use the target curve to try to increase bass beyond the natural rolloff at the low or high end.

But, (sigh!), if you must, you must.
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #51
Sorry if I sound like a killjoy, but my opinion is that it is easy to get caught up and lost in the "house curve follies" - so many to choose from, let me try them ALL!



It is a lot of work, and most (all?) of them will probably sound terrible to you.



My suggestion: start flat and modify from there. Keep a "flat" slot for comparison so you can A-B and compare. Be sure to listen at several different volume levels.
I value your opinion, but I think I miscommunicated. Maybe shouldn't have used the term "multitudes." I just wanted the ability to compare a few tweaks to my existing curve. I agree it's a good idea to start from flat rather than the automatically generated curve that room correction thinks you should have.



DO NOT use the target curve to try to increase bass beyond the natural rolloff at the low or high end.



But, (sigh!), if you must, you must.
But I'd rather do it right AND understand why. Not sure if I miscommunicated again. My "shelf" rolls off at the low end with a sharp knee at around 18Hz. Is that bad?



Sent from my iPhone using Tapatalk
 

·
Banned
Joined
·
4,838 Posts
...My "shelf" rolls off at the low end with a sharp knee at around 18Hz. Is that bad?
Not if the speakers' natural rolloff is at 18 Hz. IF the speakers roll off at 25 Hz and you are trying to get a little extra boost down to 18 Hz, you will probably get flabby low bass. I just spent a day figuring this out. The natural LF rolloff for most speakers is pretty steep and if you try to push that point down even just a few Hz, it can mean bass boost in the range of 6 dB or more to accomplish and the result gets harder to control and sounds icky (yes, that is an official audio term:blink:). You can use a DL target curve to cause the rolloff to match the speakers natural rolloff and get a cleaner sound.

At the HF end it can affect imaging. Use the target curve to have the system rolloff match the natural tweeter rolloff at the LP (the tweeter that rolls off first, if they do not match closely), and make sure the system rolloffs of the two channels DO match closely, and the imaging at HF will remain stable on sibilants, no smearing.

A few Hz difference at the low end (kHz at the high end) really tighten them up up nicely, each in its own way.
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #53
My room is also on the verge of to dead... I would like to make some diffusion panels like GIK has for placing over my absorption panels. I wish they would offer kits of just the wood with the pattern cutout for DIYers too.
Last weekend was fun, as I had time to experiment with room treatment and correction. I'm unsure what or how to measure for degrees of liveness, but I feel I'm successfully headed in the right direction. It's not as hard to hear quiet speech, and it's not as muffled. Notice I didn't say "not muffled at all," because there's still room for improvement. What I thought was a deficiency in center channel quality turned out to be a combination of too much high-frequency absorption coupled with a less-than-complimentary house curve. I decided to go with more of a LE/DE scheme and tame some center-channel nasties (lower-mid boxiness and upper-midrange harshness). The 2nd reflection-point panels came down, and the diffusers in each front corner were replaced with absorbers. Extra pillows and blankets were removed. DIRAC Live's house curve for the center channel was set up flat, except for a bass "shelf" of +1dB with slight dips of -2dB centered at 500Hz and at 3kHz. Room correction effects were disabled above 10kHz in all channels.

Your idea of placing a diffuser over an absorber is an interesting one. I wonder if any of the sound would actually make it through the diffuser into the panel below. I suppose it would depend on which frequencies the diffuser deflected. Do you think the panels would need to be matched so as not to interfere with each other?

I certainly agree that the Revel center would be a better match for the Salons, and I would head that way too if I were you. But depending on the source of the problem, your intelligibility may not improve just by moving to the matching Revel center. That's all I'm sayin' :)
Good call! After a little work I was able to transform the center's tonal character. It's now much better, not just different. That'll learn me not to make changes without looking at multiple causes. Your phantom-center suggestion is still in the running, I just need to find more time to run the tests. The motivation will come when I'm acclimated to my new-found sound!
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #54
...IF the speakers roll off at 25 Hz and you are trying to get a little extra boost down to 18 Hz, you will probably get flabby low bass.
Can't get anything past you! That's exactly what I had, only I attributed it to the ported nature of the sub. :duh:

I just spent a day figuring this out. The natural LF rolloff for most speakers is pretty steep and if you try to push that point down even just a few Hz, it can mean bass boost in the range of 6 dB or more to accomplish and the result gets harder to control and sounds icky (yes, that is an official audio term:blink:). You can use a DL target curve to cause the rolloff to match the speakers natural rolloff and get a cleaner sound.
Appreciate you sharing your findings! In trying to push the response down using the house curve, I think I was confusing room gain with natural rolloff. Embarrassingly enough, I didn't think things through. My reasoning neglected to account for boosting outside of the driver's capability. Sort of like trying to boost a null. It seems to me as if both eat up headroom, but boosting natural cutoff has a much more audible effect. Incidentally, DIRAC's suggested house curve matched the low-frequency rolloff very closely. This time I left it alone. I'm not sure why the sub was easier to integrate through the xover region. Could it have to do with distortion products (harmonics)? I mean, wouldn't a boosted driver generate harmonics related to the overdriven fundamental?

At the HF end it can affect imaging. Use the target curve to have the system rolloff match the natural tweeter rolloff at the LP (the tweeter that rolls off first, if they do not match closely), and make sure the system rolloffs of the two channels DO match closely, and the imaging at HF will remain stable on sibilants, no smearing.
Arrrghhh! I missed this during my re-tune. I believe my L/R mains currently do not track in the upper region. Easy enough to remedy given your tips. Thanks again! :T
 

·
Premium Member
Joined
·
5,797 Posts
Last weekend was fun, as I had time to experiment with room treatment and correction. I'm unsure what or how to measure for degrees of liveness, but I feel I'm successfully headed in the right direction. It's not as hard to hear quiet speech, and it's not as muffled. Notice I didn't say "not muffled at all," because there's still room for improvement. What I thought was a deficiency in center channel quality turned out to be a combination of too much high-frequency absorption coupled with a less-than-complimentary house curve. I decided to go with more of a LE/DE scheme and tame some center-channel nasties (lower-mid boxiness and upper-midrange harshness). The 2nd reflection-point panels came down, and the diffusers in each front corner were replaced with absorbers. Extra pillows and blankets were removed. DIRAC Live's house curve for the center channel was set up flat, except for a bass "shelf" of +1dB with slight dips of -2dB centered at 500Hz and at 3kHz. Room correction effects were disabled above 10kHz in all channels.

Your idea of placing a diffuser over an absorber is an interesting one. I wonder if any of the sound would actually make it through the diffuser into the panel below. I suppose it would depend on which frequencies the diffuser deflected. Do you think the panels would need to be matched so as not to interfere with each other?


Good call! After a little work I was able to transform the center's tonal character. It's now much better, not just different. That'll learn me not to make changes without looking at multiple causes. Your phantom-center suggestion is still in the running, I just need to find more time to run the tests. The motivation will come when I'm acclimated to my new-found sound!
From what i understand it is more of a decoration than a true diffuser, but by having a solid wood surface it should reduce absorption of the higher frequencies.
 

·
Banned
Joined
·
4,838 Posts
LE/DE scheme
You don't hear DE/LE or LE/DE terms used that much these days. I personally prefer DE/LE because it tends to keep you on guard against early reflections that are more disruptive to SS&I. The later LE reflections are usually late enough to add to spaciousness without being disruptive to SS&I.

Your idea of placing a diffuser over an absorber is an interesting one. I wonder if any of the sound would actually make it through the diffuser into the panel below. I suppose it would depend on which frequencies the diffuser deflected. Do you think the panels would need to be matched so as not to interfere with each other?
Matching only in overall character, like ratio of open to closed space of diffuser. Think in terms of wavelength. At 1 kHz, wavelength is about 1 foot. A 1-inch diameter hole is invisible to frequencies below 10 kHz or so (a very rough approximation, but gives you a ballpark idea). But a diffusor with 50% open area is another matter, quite open at LF and not much at HF.
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #57
You don't hear DE/LE or LE/DE terms used that much these days. I personally prefer DE/LE because it tends to keep you on guard against early reflections that are more disruptive to SS&I. The later LE reflections are usually late enough to add to spaciousness without being disruptive to SS&I.
Sometimes I forget to check the date of the articles I read. Many of the concepts are as timeless as the laws of physics, but others catch me off guard. Looks like I misused the term and meant to write DE/LE because:
  • The back wall (listening end) is fairly dead (two 20" traps and one panel behind LP; one panel at each 1st reflection point).
  • The front wall (speaker end) is supposed to be live (despite having activated the mains' rear tweeters and despite the TV's large reflective display).
It's easy to identify my acoustic treatments' contribution by listening to how my voice changes when I first walk into the LP end, and then how it keeps changing as I move through it to the speaker end. The acoustics still clearly need improvement. Removing the 2nd reflection panels did not entirely solve the issue with muffled sound at the LP. I think there's still too much hi-frequency absorption as provided by the dual-action traps; so if I remove them to liven the sound at the high end, then SQ will suffer at the low end. WIN-LOSE. Improvement may be possible but like anything else worthwhile, takes effort and time. I'm tempted to swap the dual-acting traps for more conventional bass-only types, but it's not in the budget right now. So what are we left to work with? Assuming the speakers have already been positioned as best as possible for SS&I considerations, here are the variables:
  • Rear tweets on/off
  • Bass trap #x of y in location z
  • Bass trap diffuser section aimed in 45deg increments
That's a lot of combinations and experimentation! Listener fatigue is bound to set in. Wouldn't you question the validity of a listening comparison separated by hours if not days? Well, that's exactly what will happen when trying to setup each and every combination listed. Time (and sanity) might be preserved by performing only the most promising trials. For instance, changes could be made in L/R pairs if the room is acoustically symmetrical - there would be no need to control left and right tweeters individually. If we recognize the orientation of diffuser sections to be the major contributor of muffled sound, then we can concentrate on which setup to perform first. Is this the order in which you'd approach a room-tune:
  1. SS&I
  2. Acoustic treatment for reflections
  3. Acoustic treatment for bass
  4. Room correction
 

·
Registered
Joined
·
2,261 Posts
Discussion Starter #58
More thoughts (sorry, I can't help it!):

Room correction could be trying to undo my well-intentioned acoustic tuning. Dirac Live yields room calibrations rich in SS&I, so if my speakers are correctly set up, less acoustic treatment should be needed, right?

I believe it's time to get back to basics, take some measurements without the traps, then introduce them individually and measure again. Then guess what? Move them around and measure some more! The vibrating trap method can help a lot here.

Or... My basic complaint of muffled sound could just be high-frequency hearing loss!




Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
1,784 Posts
I would tend to think your first two presumptions to be correct. One must always begin with a zero sum game in that the room should be an empty vessel into which sound should be introduced in its raw format first. Change can them be made to the room a tiny bit at a time to obtain the results you are looking for. We who are not professionals tend to add things to a room without having full knowledge of whether those things are good or bad. You have a good bit of sound attenuation in your room that apparently might be causing your time of discontent. Us regular guys then might expect a bit of coding to do or undue something that might be beyond its control. I think, and it is just that, Jack-think, that when you empty the room of equipment, maybe empty the room of traps and what have you and treat the room like a newborn, en empty vessel that needs to be eased into conformity.

Just random thoughts to be taken or not taken irrespective of your hearing acuity 0:)
 

·
Registered
Joined
·
4,008 Posts
More thoughts (sorry, I can't help it!):

Room correction could be trying to undo my well-intentioned acoustic tuning. Dirac Live yields room calibrations rich in SS&I, so if my speakers are correctly set up, less acoustic treatment should be needed, right?

I believe it's time to get back to basics, take some measurements without the traps, then introduce them individually and measure again. Then guess what? Move them around and measure some more! The vibrating trap method can help a lot here.

Or... My basic complaint of muffled sound could just be high-frequency hearing loss!




Sent from my iPhone using Tapatalk


Hmm. I wonder too. My thinking is if you've selected and placed your acoustic bolsters, Dirac should have less to do. Kinda like tuning your subs with minidsp before you run audyssey therefore lightening its processing workload. And yes, I also believe if you've placed your speakers correctly (?), Dirac will have less to do there as well. Am I all wet? What I mean to say is taking care of the physical aspects first, like placement and treatment, should be done before relying on electronics. Like subs. Placement first, treatment second, and EQ last. Whatchathinkaboutdat?
 
41 - 60 of 110 Posts
Top