Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs - Page 262 - AVS | Home Theater Discussions And Reviews
Forum Jump: 
 401Likes
Reply
 
Thread Tools
post #7831 of 14290 Old 01-09-2014, 07:16 AM
Advanced Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 794
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 234 Post(s)
Liked: 134
Quote:
Originally Posted by sdrucker View Post

My original REW question was actually about FS vs. %FS plots

From my experience it is good to look at the ETC graph in the dB scale when analysing the room and working on the acoustic treatments. And it is worth to look at the % scale of the impulse response (the pure original impulse response, not the ETC, so it will have positive and negative sides of the signal) when analysing what is made to the signal by electronic correction.
IgorZep is offline  
Sponsored Links
Advertisement
 
post #7832 of 14290 Old 01-09-2014, 08:10 AM
Advanced Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 794
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 234 Post(s)
Liked: 134
Quote:
Originally Posted by sdrucker View Post

No, I've measured my results at the sweet spot, since that's where MLP is and I'm the only one that cares.

The problem is that there is no sweet spot in a real life. We have two ears and they are not located in a single spot already. We do not listen with fixed head - it would be a nightmare if we would need to do this. Even if single chair is the only listening position we want lo listen comfortably and this means move a little bit when listening. This little bit is actually enough to totally screw up high frequency correction if it is not done right. This is why it is needed to measure an area when calibrating, even it if is quite small area, it should still be something distinctly larger than a point. And also when verifying the results we should also measure an area.
Quote:
Originally Posted by sdrucker View Post

Is that "close enough" for measurements in your book?

It depends on what you measure and when. Sometimes this is negligible, sometimes even the change in humidity, temperature and atmospheric pressure could make a big difference. When you measure and compare/sum/substract/correct late high frequency reflections (and for high frequencies such as 10kHz something larger than 1ms is a late reflection already) then all those things matter a lot.
Quote:
Originally Posted by sdrucker View Post

Igor, I certainly understand your skepticism, since I wasn't sure I believed the results myself until I repeated it, despite the claim that Trinnov makes for their technology. In fact, I've been very reluctant to do ANY posts from the R-972 I picked up because I was concerned they'd be controversial. ... but I felt obligated to post some results - even fragmentary - because I've been chattering about trying out Trinnov long enough here smile.gif.

Nothing bad at being controversial as long as you are willing to find the truth. I am interested in such comparison and I think many others in this thread would be interested also to see the conclusive data. Unfortunately everything I have seen to date is an impressive data, but not conclusive unfortunately. It brings serious doubts as it sometimes goes over what should be possible when physics is concerned. This means - you can ideally fix impulse response by EQ for a single point in space. But it will not work well for any other point. It actually will only make problem worse at every other point. It was discussed here some time ago already. So, either there are some problems with the measurements or there are "tricks" played by the Trinnov (Dirac/whatever I've seen such things elsewhere also) or I miss something. Actually comparing the acoustic measurements and trying to understand what Room EQ is doing is not an easy task - the graphs are very noisy for that. The pre-out measurements are a king to understand important bits about the Room EQ when it is going to the question - "is it careful enougs and is doing only good things or there is something that might cause bad things also that were hidden behind the acoustic measurement noises". If you can do this and share the files... It will be good thing, and it will also answer this thing about correction time window precisely:
Quote:
Originally Posted by sdrucker View Post

At any rate, their mic allegedly measures in 3D space, so if you believe what the Optimizer is doing, it's conceivable that you can get improvements at this level. I think Hoyt on the dedicated R-972 thread claimed that they correct up to 150 ms, but I've only measured to 40 ms.
Quote:
Originally Posted by sdrucker View Post

Again, repetition under more controlled conditions is the only way to see if this is really happening, hence I'm repeating with a single speaker, nobody home, and a time stamped/photographed mic position as soon as I get some free time. And also to make sure that obscure elements like whether a baby bouncy seat against the side of a wall isn't creating variations. smile.gif

What I would like to know about the conditions:
1) How is it calibrated - number of points, how large is the area. It should not be single point calibration.
2) Measurement at centre of MLP and shifted at different distances within calibrated area (with and without EQ)
3) pre-out measurement (with and without EQ, but with the same modes such as crossovers for sub etc. - for controlling correctness)
Then it will be a good comparison done and concluded from enough points of view.
pepar likes this.
IgorZep is offline  
post #7833 of 14290 Old 01-09-2014, 09:35 AM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,425
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 101 Post(s)
Liked: 65
Quote:
Originally Posted by IgorZep View Post

Quote:
Originally Posted by cfraser View Post

The final thing that made me remove the traps is Audyssey was really jacking up the THD for whatever reason (in sub-300Hz range) when the traps were there.

Have you pre-out measurements confirming this?
Quote:
Originally Posted by cfraser View Post

Don't want up to 5% THD from Audyssey when the speakers are like 0.1% by themselves in the same range.

0.1% THD speakers? Where have you found those? According to published measurements I've found the best speaker drivers are usually showing 0.3% and rarely go down to 0.1% and only in some very limited frequency range. And a lot more THD for the bass range...

No pre-out measurements, don't need them to "demonstrate" the effect. Essentially "prove" as far as I'm concerned, in this room/circumstance, which is the only one I care about. As mentioned, it's not there with everything identical except Audyssey is off. As also mentioned, same thing with different AVR with different Audyssey version. That's good enough. This is not a court of law, and even if it was... smile.gif Audyssey On = problem, Audyssey Off = no problem. Pretty simple. The take-away for me is to be wary of what you make Audyssey try to do. Nobody claimed it would be perfect, and who ever heard of a filter of any type, never mind a very complicated one, that doesn't introduce some distortion? For myself, I have learned to help it with PEQ (pre-Audyssey cal) if possible.

Speakers I was talking about are Maggies. And I specifically did say the issues re distortion were only in certain ranges. Planars are noted for low distortion, but not in the bass. I am perfectly aware that 3% across the full range would generally be considered a good speaker.

Are you the DBX police by any chance? You sound like it. They never help, only question. Just asking...
cfraser is offline  
post #7834 of 14290 Old 01-09-2014, 10:22 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,563
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 97


My THD is running between 0.15% and 0.3% at the LP depending on the frequency (down to 50hz). 0.1% might be achievable, but it would have to be exceptional.

My Room
My Music
Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)
Jenzen Next
jim19611961 is online now  
post #7835 of 14290 Old 01-09-2014, 10:57 AM
AVS Club Gold
 
sdrucker's Avatar
 
Join Date: Feb 2006
Location: Chicago, IL
Posts: 2,376
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 181
Quote:
Originally Posted by IgorZep View Post

The problem is that there is no sweet spot in a real life. We have two ears and they are not located in a single spot already. We do not listen with fixed head - it would be a nightmare if we would need to do this. Even if single chair is the only listening position we want lo listen comfortably and this means move a little bit when listening. This little bit is actually enough to totally screw up high frequency correction if it is not done right. This is why it is needed to measure an area when calibrating, even it if is quite small area, it should still be something distinctly larger than a point. And also when verifying the results we should also measure an area. .

This is going to be of philosophical rather than of practical interest to most of you, but until I move this to a separate thread, I'll answer Igor here.

Conceptually I agree with you, which is one of the advantages to Audyssey's measurement philosophy over several other REQ methods, including use of PEQ. However, there's a couple of practical problems:
1) The "consumer" version of Trinnov only captures one measurement for a given position to be measured and calibrated. I can't do spatial averaging, which the ADA and Trinnov Pro units can (and you can directly manipulate the weights used in the averaging to emphasize, say, the MLP related measurements more than others if desired). Nor can we do spatial averaging with REW, at least directly. We can average but not spatially average.
2) In the Trinnov world. the calibration works on both the horizontal and vertical plane to capture an accurate measurement. That is, I aim my mic specifically to achieve a 0 degree horizontal angle for my center channel, measuring at ear height, at the center of my plasma TV display. This is critical so that the 2D/3D remapping works as part of the algorithm. I don't know how measuring "off axis" from MLP to capture a larger physical calibration space would impact the basic EQ (delay and level calibration assumed) without remapping. The unknown is over how large an area in space - the radius - the Trinnov four-element mic is seeking to correct.

YMMV, but assuming that we believe that the advantages of using Trinnov has positive utility, even with disadvantages from the way we ideally would measure, the more relevant question to me would be how stable the post-calibration REW measurements are over a given space. I can do measurements with and without Trinnov at, say, two other positions to the left and right of MLP by a fixed amount, and measure in front of MLP, and we can compare the results on FR and the %FS plot (after your comment)
Quote:
It depends on what you measure and when. Sometimes this is negligible, sometimes even the change in humidity, temperature and atmospheric pressure could make a big difference. When you measure and compare/sum/substract/correct late high frequency reflections (and for high frequencies such as 10kHz something larger than 1ms is a late reflection already) then all those things matter a lot.

Meaning they matter more than for a frequency response plot or a waterfall? In that case, we want to do more than simply make sure that the physical mic position - the UMM-6 I use with REW - is in the same place. I would literally need to do REQ vs. no-REQ at the same point of time with NO variation whatsoever with the mic, room conditions, stable weather etc. I guess I better read the weather report and watch the humidity on my iPad's weather app as long as I'm at it. smile.gif. OCD but not insane if you're a methodologist LOL.

That leaves aside the question of "what is truth", and "when is it true"? And one of the reasons I published my one plot series was to get a discussion rolling about how to measure impulse response with a complex EQ method. Even counting the Pro (fully featured) Trinnov solutions, there's only been a handful of graphs I've seen on AVS, and only the one I linked to beside mine has looked at impulse response AFAIK. As that's a primary reason to use Trinnov along with remapping, it's certainly relevant.
Quote:
This means - you can ideally fix impulse response by EQ for a single point in space. But it will not work well for any other point. It actually will only make problem worse at every other point. It was discussed here some time ago already. So, either there are some problems with the measurements or there are "tricks" played by the Trinnov (Dirac/whatever I've seen such things elsewhere also) or I miss something. Actually comparing the acoustic measurements and trying to understand what Room EQ is doing is not an easy task - the graphs are very noisy for that.

Defnitely a labor intensive one! And your comment about whether impulse response is only relevant to single point of space suggests that what Dirac and Trinnov are doing is a game of "whack a mole", as we'd say in the States. The question is over how large an area is the correction stable to the degree that differences are imperceptible? If it's a foot from the measurement position, and you're interested in MLP, that's one thing. If it's +/- an inch before differences are noted by the listener, it's pointless unless we sit with our head in a vice..
Quote:
The pre-out measurements are a king to understand important bits about the Room EQ when it is going to the question - "is it careful enougs and is doing only good things or there is something that might cause bad things also that were hidden behind the acoustic measurement noises". If you can do this and share the files... It will be good thing, and it will also answer this thing about correction time window precisely:

Since we're both science-minded, I'm willing to do that, even if the R-972 version I'm using is of mostly academic interest for most on the REW thread.

I also have the practical problem of testing pre-outs, which I've never done. I'll have to read Jerry's guide on doing it, since I have the REW/HDMI version of REW and I've only used ASIO for my drivers, not Java.

One more practical problem, at least for me: for strictly Trinnov vs. no Trinnov testing, I've used the R-972's ability to turn on/off Trinnov. However, matching crossovers with and without calibration will be problematic. That's because Trinnov sets them precisely (e.g. 76 Hz for mains, believe it or not, and they can't be changed), while the non-Trinnov bass management setting is rounded to the nearest 10 Hz (e.g. 80 Hz). Less of an issue with this level of difference, but more of one for some other crossovers.
Quote:
What I would like to know about the conditions:
1) How is it calibrated - number of points, how large is the area. It should not be single point calibration.
2) Measurement at centre of MLP and shifted at different distances within calibrated area (with and without EQ)
3) pre-out measurement (with and without EQ, but with the same modes such as crossovers for sub etc. - for controlling correctness)
Then it will be a good comparison done and concluded from enough points of view.

Just to be clear - should I literally be measuring a speaker with the crossover engaged with NO subs on, both with and without calibration? And if I do an % FS plot to get "pure" impulse response, should I be using the same graph axes for the X axis (ms) as I would for an ETC plot? I can do, say, +/- 40% or 50% for the Y axis, but that depends on what we consider a goal for impulse response on this scale. In db, we've discussed reducing the IR to - 20 db or better. What would a reasonable standard for an % FS plot be (assumedly an absolute deviation from 0, but you tell me)..

Audio Gear: Sherwood R-972 (Trinnov Altitude - coming soon!), NAD M27 amp
Display: Panasonic VT50, Lumagen Radiance Mini
Misc.: Oppo 103, Apple TV, JRiver ID NUC
PSB Imagine T2, Center, and Surrounds; HSU ULS-15 subs (2)
sdrucker is online now  
post #7836 of 14290 Old 01-09-2014, 11:32 AM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 9,195
Mentioned: 14 Post(s)
Tagged: 0 Thread(s)
Quoted: 2154 Post(s)
Liked: 1162
Quote:
Originally Posted by cfraser View Post


No pre-out measurements, don't need them to "demonstrate" the effect. Essentially "prove" as far as I'm concerned, in this room/circumstance, which is the only one I care about. As mentioned, it's not there with everything identical except Audyssey is off. As also mentioned, same thing with different AVR with different Audyssey version. That's good enough. This is not a court of law, and even if it was... smile.gif Audyssey On = problem, Audyssey Off = no problem. Pretty simple. The take-away for me is to be wary of what you make Audyssey try to do. Nobody claimed it would be perfect, and who ever heard of a filter of any type, never mind a very complicated one, that doesn't introduce some distortion? For myself, I have learned to help it with PEQ (pre-Audyssey cal) if possible.

Speakers I was talking about are Maggies. And I specifically did say the issues re distortion were only in certain ranges. Planars are noted for low distortion, but not in the bass. I am perfectly aware that 3% across the full range would generally be considered a good speaker.

Are you the DBX police by any chance? You sound like it. They never help, only question. Just asking...

 

I think what concerns us is that you are the first person (that I know of) that is making a claim that Audyssey is increasing distortion. To eliminate room effects, looking at the pre-out measurements would allow us to see what is actually going on. No one doubts that you prefer Audyssey off.  It is just that we are inquisitive types who would like to understand what is going on in your particular case.

kbarnes701 and IgorZep like this.
AustinJerry is online now  
post #7837 of 14290 Old 01-09-2014, 11:57 AM
AVS Special Member
 
Selden Ball's Avatar
 
Join Date: Jan 2001
Location: upstate NY
Posts: 8,570
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1253 Post(s)
Liked: 671
Quote:
Originally Posted by AustinJerry View Post

Quote:
Originally Posted by cfraser View Post

No pre-out measurements, don't need them to "demonstrate" the effect. Essentially "prove" as far as I'm concerned, in this room/circumstance, which is the only one I care about. As mentioned, it's not there with everything identical except Audyssey is off. As also mentioned, same thing with different AVR with different Audyssey version. That's good enough. This is not a court of law, and even if it was... smile.gif Audyssey On = problem, Audyssey Off = no problem. Pretty simple. The take-away for me is to be wary of what you make Audyssey try to do. Nobody claimed it would be perfect, and who ever heard of a filter of any type, never mind a very complicated one, that doesn't introduce some distortion? For myself, I have learned to help it with PEQ (pre-Audyssey cal) if possible.


Speakers I was talking about are Maggies. And I specifically did say the issues re distortion were only in certain ranges. Planars are noted for low distortion, but not in the bass. I am perfectly aware that 3% across the full range would generally be considered a good speaker.


Are you the DBX police by any chance? You sound like it. They never help, only question. Just asking...

I think what concerns us is that you are the first person (that I know of) that is making a claim that Audyssey is increasing distortion. To eliminate room effects, looking at the pre-out measurements would allow us to see what is actually going on. No one doubts that you prefer Audyssey off.  It is just that we are inquisitive types who would like to understand what is going on in your particular case.

I seem to recall that Igor posted a plot not too long ago (in the Audyssey thread, perhaps) showing distortions introduced by Audyssey XT, and likening it to listening from "behind leaves".

Selden

Marantz SR7009+MM9000/Atmos 7.1.4 (FH+TM:DefTech PM1000)/LCR+TM amped
Selden Ball is offline  
post #7838 of 14290 Old 01-09-2014, 12:11 PM
Advanced Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 794
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 234 Post(s)
Liked: 134
Quote:
Originally Posted by AustinJerry View Post

It is just that we are inquisitive types who would like to understand what is going on in your particular case.

Exactly smile.gif
IgorZep is offline  
post #7839 of 14290 Old 01-09-2014, 12:15 PM
Advanced Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 794
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 234 Post(s)
Liked: 134
Quote:
Originally Posted by Selden Ball View Post

I seem to recall that Igor posted a plot not too long ago (in the Audyssey thread, perhaps) showing distortions introduced by Audyssey XT, and likening it to listening from "behind leaves".

Yes, but this was a kind of linear distortion, not the nonlinear/harmonic (THD) one. Nonlinear distortion something that is not expected from the digital filter, this doesn't mean that there could not be such distortion at all, but the level of non-linear unwanted digital noise in well designed digital filter must be close to unmeasurable.
IgorZep is offline  
post #7840 of 14290 Old 01-09-2014, 01:16 PM
FMW
AVS Special Member
 
FMW's Avatar
 
Join Date: Jun 2007
Posts: 6,118
Mentioned: 5 Post(s)
Tagged: 1 Thread(s)
Quoted: 886 Post(s)
Liked: 1082
I'm about to get involved with REW. Would someone please look over my equipment list to make sure everything will work and that I'm not missing anything? Some of the stuff is on order. Thanks for any feedback.

Computer - old HP notebook
AV Receiver - Pioneer Elite VSX 92
Connection to AV reciever - Turtle Beach USB notebook soundcard and mini optical to toslink cable
Microphone - MXL condensor recording mic with omni cartridge. Mic is pretty flat for frequency response
Mic Preamp - Behringer 302 USB preamp/mixer.

I already own everything except the "connection to AV receiver" items which are on their way. Will this get the job done?

Read more: http://www.hometheatershack.com/forums/rew-forum/72347-equipment-check.html#ixzz2pwE5xGrb
FMW is offline  
post #7841 of 14290 Old 01-09-2014, 01:23 PM
AVS Club Gold
 
sdrucker's Avatar
 
Join Date: Feb 2006
Location: Chicago, IL
Posts: 2,376
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 181
Igor - I pinged Curt Hoyt from Trinnov about the issue on off-axis measurement and IR calibration decaying outside of a single point in space, and it turns out they researched the problem back in 2005. He emailed me a excerpt of the research they did on it. However, it's copyright protected, so I can't post it on AVS without the author's permission..

Here's a quick summary:
He did a test where he looked 15 degrees off-axis for a JBL loudspeaker in front of a perforated screen, and 30 degrees off-axis for an Apogee three-way horn speaker used at Paramount Studios (Trinnov at the time was strictly used for studio purposes, not for home audio). On his charts, there's a clear difference between the 0 and 15 degrees off-axis responses for both IR and frequency response without the Trinnov Optimizer, both on IR and frequency response. Likewise for 30 degrees vs. on-axis.

Inspecting the post-Trinnov Optimizer results reveal that the IR responses between the on axis (0) and off-axis (15/30) positions resemble one another considerably more than not, and as measured show changes in a positive direction from the pre-TO results. From my quick read, the impulse response comparison is more consistent than that of the frequency response, where there's more variation. I have no idea if the plots were filtered by frequency, or on what scale the IR is measured (db, % FS, or something else). The only thing is that he's look at LT 10 ms, which is VERY early. Do you think that might translate into improvement beyond 10 ms, or is that for us (me) to test LOL?

BTW he did say that the effectiveness was room and speaker dependent, particularly driven by the idiosyncratic characteristics of a speaker off-axis above Schroder. And I don't know how this works for a center channel, which may be prone to different room modes than L/R mains by nature of the room.

He also noted that this test was done on an early version of Trinnov Optimizer (v1.0 vs. the v2.0 I have in my Sherwood), and that he expected decay at 30 degrees off axis. The consistency was a surprise for him too.

Edit: I took a quick look at the 165-page paper you referenced in the Audyssey thread by Putzey. I'm not an EE, so I'm not competent to comment on what he's saying, but what I think you said on the Audyssey thread about the impact of "trivial" IR off-axis on frequency response in an XT context may fit what I cited from Hoyt. Meaning that even if the IR control is relatively stable for very early reflections moderately (LT 30 degrees) off axis, the resulting frequency response correction may be worse. OTOH, if I'm listening within 15 degrees of my calibration's on-axis individual speaker presentation, why should I care? Is it any worse than what we'd see from an XT32 on vs. off-axis calibration, or possibly better?

That brings up another question: if you could have better impulse response control or better frequency response management, but not both, what's a better tradeoff?

I probably just gave 90% of the readers of this thread a headache by this answer LOL.

Audio Gear: Sherwood R-972 (Trinnov Altitude - coming soon!), NAD M27 amp
Display: Panasonic VT50, Lumagen Radiance Mini
Misc.: Oppo 103, Apple TV, JRiver ID NUC
PSB Imagine T2, Center, and Surrounds; HSU ULS-15 subs (2)
sdrucker is online now  
post #7842 of 14290 Old 01-09-2014, 01:39 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 9,195
Mentioned: 14 Post(s)
Tagged: 0 Thread(s)
Quoted: 2154 Post(s)
Liked: 1162
Quote:
Originally Posted by FMW View Post

I'm about to get involved with REW. Would someone please look over my equipment list to make sure everything will work and that I'm not missing anything? Some of the stuff is on order. Thanks for any feedback.

Computer - old HP notebook
AV Receiver - Pioneer Elite VSX 92
Connection to AV reciever - Turtle Beach USB notebook soundcard and mini optical to toslink cable
Microphone - MXL condensor recording mic with omni cartridge. Mic is pretty flat for frequency response
Mic Preamp - Behringer 302 USB preamp/mixer.

I already own everything except the "connection to AV receiver" items which are on their way. Will this get the job done?

Read more: http://www.hometheatershack.com/forums/rew-forum/72347-equipment-check.html#ixzz2pwE5xGrb

You are using what we call a "legacy" REW kit. This thread was established to promote use of the plug-and-play USB mics with current generation laptops with HDMI connections. Our focus is to quickly come up to speed with measurement techniques and then using those measurements to analyze and improve our listening room acoustics.

My concern is that you will expect assistance with your legacy kit, and we are not prepared to do that. The appropriate forum for legacy support would be the REW forum over on Home Theater Shack. Having said that, if you don't need REW support and simply want to post measurements for feedback, then that would be fine.
AustinJerry is online now  
post #7843 of 14290 Old 01-09-2014, 02:30 PM
Advanced Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 936
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked: 60
Quote:
Originally Posted by sdrucker View Post


I probably just gave 90% of the readers of this thread a headache by this answer LOL.

I'm loving it but it does seem somewhat off-topic. I suppose once you get the REW questions straightened out then it will be time to create a new thread.
artur9 is online now  
post #7844 of 14290 Old 01-09-2014, 02:33 PM
AVS Club Gold
 
sdrucker's Avatar
 
Join Date: Feb 2006
Location: Chicago, IL
Posts: 2,376
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 181
Quote:
Originally Posted by artur9 View Post

I'm loving it but it does seem somewhat off-topic. I suppose once you get the REW questions straightened out then it will be time to create a new thread.

Yup, considering that nobody else here that's active probably knows what I'm talking about other than Igor, Sanjay or Markus. You do but you went in a different direction.

Audio Gear: Sherwood R-972 (Trinnov Altitude - coming soon!), NAD M27 amp
Display: Panasonic VT50, Lumagen Radiance Mini
Misc.: Oppo 103, Apple TV, JRiver ID NUC
PSB Imagine T2, Center, and Surrounds; HSU ULS-15 subs (2)
sdrucker is online now  
post #7845 of 14290 Old 01-09-2014, 02:53 PM
Advanced Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 794
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 234 Post(s)
Liked: 134
Quote:
Originally Posted by sdrucker View Post

That brings up another question: if you could have better impulse response control or better frequency response management, but not both, what's a better tradeoff?

There is no tradeoff. Frequency response is a product of impulse response - they are the same thing. Once you get ideal impulse response you will get perfectly flat frequency response also.
IgorZep is offline  
post #7846 of 14290 Old 01-09-2014, 03:41 PM
AVS Club Gold
 
sdrucker's Avatar
 
Join Date: Feb 2006
Location: Chicago, IL
Posts: 2,376
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 181
I understand that they're mathematically related by a Fourier transformation, so ideal IR will yield flat FR. But what about "moderately" off-axis cases, like a speaker 15 degrees off-axis? I think what I'm asking is whether the falloff from ideal (or as close to it as you get) on-axis is 1:1 or the decay impacts one more than the other, everything else constant, with respect to early reflections. Assuming that the mathematics (matrix operations) make that a logical question to ask.

BTW I'm probably not going to be able to do a round of measurements until tomorrow night, more probably Saturday at the earliest (I have the nerve to want to actually watch a BluRay tonight). So if anyone has any further ideas about how I should look at IR with the % FS plot as per Igor, let me know. Specifically if there's any standards about how to determine what "good" is, like there are with a db based ETC plot.

Audio Gear: Sherwood R-972 (Trinnov Altitude - coming soon!), NAD M27 amp
Display: Panasonic VT50, Lumagen Radiance Mini
Misc.: Oppo 103, Apple TV, JRiver ID NUC
PSB Imagine T2, Center, and Surrounds; HSU ULS-15 subs (2)
sdrucker is online now  
post #7847 of 14290 Old 01-09-2014, 03:45 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 25,316
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 287 Post(s)
Liked: 186
Quote:
Originally Posted by artur9 View Post

This Powerpoint by Geddes may help.

Quote:
Originally Posted by andyc56 View Post

The accompanying video is also well worth watching. The beginning of the PowerPoint presentation covers the missing beginning of the video.

Thanks for the Gedlee links! Bookmarked for later reference. I can't seem to take the time to do any more than skimming the thread to keep up where I am now.

Jeff
pepar is online now  
post #7848 of 14290 Old 01-09-2014, 08:38 PM
Advanced Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 936
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked: 60
How do I use the ETC (or other REW feature) to decide between the following materials to treat my ceiling?

I want to basically eliminate the ceiling as a source of reflections but I do not want the room to be too dead. On the other hand, the family does not want any sound to come through the ceiling into the rest of the house rolleyes.gif

.....................................................125HZ...250HZ...500HZ...1000HZ...2000HZ...4000HZ...NRC
703, plain, on wall....1" (25mm) ........0.11......0.28.........0.68........0.90.....0.93......0.96....0.70
703, plain 16" air gap 1" (25mm)........0.65......0.94.........0.76........0.98.....1.00.....1.14.....0.90
FINE FISSURED................................0.35......0.34.........0.55........0.72......0.65......0.62...0.55

The first 2 are from Bob Gold's excellent site.
The 3rd is from Armstrong's product literature.

I included the 2nd number as these will be drop ceiling tiles and there will be an air gap behind them. Up until I fill the joist area with SafeNSound.

In case you're curious, the first two are driven by these black ceiling tiles on eBay

The 3rd is Armstrong's Fine Fissured Homestyle tiles available at Lowes.
artur9 is online now  
post #7849 of 14290 Old 01-09-2014, 09:24 PM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 432
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 136 Post(s)
Liked: 56
Been a busy week here.

Marcus, my first guess would be that it's a driver misalignment. The speaker is considerably below the MLP and isn't angled far enough up. Band limited ETC shows the problem occurs below tweeter frequencies. I haven't had a chance to do any measurements.

Keith, wrapping myself in pink fluffy is probably just a little overboard tongue.gif

Did someone mention Double Bass Array!
Audionut11 is offline  
post #7850 of 14290 Old 01-09-2014, 09:34 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 25,316
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 287 Post(s)
Liked: 186
Quote:
Originally Posted by artur9 View Post


I want to basically eliminate the ceiling as a source of reflections but I do not want the room to be too dead. On the other hand, the family does not want any sound to come through the ceiling into the rest of the house.

I included the 2nd number as these will be drop ceiling tiles and there will be an air gap behind them. Up until I fill the joist area with SafeNSound.
.

Acoustical treatment to achieve good sound in a room is something different from acoustical isolation, AKA reducing/eliminating sound from passing from the room into other rooms. I don't think REW can help with the latter other than measure the sound in the "unwanted" room to determine how much and what frequencies are being transmitted into it which might give you some hints on how big a problem you have. Likely, you will need to find a theater construction thread for isolation techniques.

That said, filling the cavity above the suspended ceiling will be necessary to reduce sound from going into rooms above.

Jeff
pepar is online now  
post #7851 of 14290 Old 01-10-2014, 02:59 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 21,603
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 3747 Post(s)
Liked: 2910
Quote:
Originally Posted by FMW View Post

I'm about to get involved with REW. Would someone please look over my equipment list to make sure everything will work and that I'm not missing anything? Some of the stuff is on order. Thanks for any feedback.

Computer - old HP notebook
AV Receiver - Pioneer Elite VSX 92
Connection to AV reciever - Turtle Beach USB notebook soundcard and mini optical to toslink cable
Microphone - MXL condensor recording mic with omni cartridge. Mic is pretty flat for frequency response
Mic Preamp - Behringer 302 USB preamp/mixer.

I already own everything except the "connection to AV receiver" items which are on their way. Will this get the job done?

Read more: http://www.hometheatershack.com/forums/rew-forum/72347-equipment-check.html#ixzz2pwE5xGrb

 

Wrong thread. That is 'legacy REW'. This thread, as the title suggests, is for people using an HDMI connection to the laptop/AVR and a USB mic.

kbarnes701 is offline  
post #7852 of 14290 Old 01-10-2014, 03:02 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 21,603
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 3747 Post(s)
Liked: 2910
Quote:
Originally Posted by Audionut11 View Post

Been a busy week here.

Marcus, my first guess would be that it's a driver misalignment. The speaker is considerably below the MLP and isn't angled far enough up. Band limited ETC shows the problem occurs below tweeter frequencies. I haven't had a chance to do any measurements.

Keith, wrapping myself in pink fluffy is probably just a little overboard tongue.gif

Did someone mention Double Bass Array!

 

Think of the money you'd save on heating bills. You could spend it on more treatments :)

kbarnes701 is offline  
post #7853 of 14290 Old 01-10-2014, 04:33 AM
Advanced Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 936
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked: 60
Quote:
Originally Posted by pepar View Post

Acoustical treatment to achieve good sound in a room is something different from acoustical isolation,
Jeff

I shouldn't have mentioned that as it's a distraction from my real question. Can ETC or something else in REW tell me which frequencies are the predominant reflections? If so, how would I go about using it to determine that?

For example, the band limited IR? Do I do a full range sweep then filter in REW or do I need to do something else?
artur9 is online now  
post #7854 of 14290 Old 01-10-2014, 04:41 AM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 25,316
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 287 Post(s)
Liked: 186
Quote:
Originally Posted by kbarnes701 View Post

Think of the money you'd save on heating bills. You could spend it on more treatments smile.gif

Oh yes, there would be "treatments" indicated if anyone saw us do that, perhaps involuntary ones.

Jeff
pepar is online now  
post #7855 of 14290 Old 01-10-2014, 05:11 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 21,603
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 3747 Post(s)
Liked: 2910
Quote:
Originally Posted by pepar View Post
 
Quote:
Originally Posted by kbarnes701 View Post

Think of the money you'd save on heating bills. You could spend it on more treatments smile.gif

Oh yes, there would be "treatments" indicated if anyone saw us do that, perhaps involuntary ones.

Jeff

 

LOL. (Hums: "they're coming to take me away, ha ha, they're coming to take me away.....)

kbarnes701 is offline  
post #7856 of 14290 Old 01-10-2014, 05:19 AM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 25,316
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 287 Post(s)
Liked: 186
Quote:
Originally Posted by artur9 View Post

I shouldn't have mentioned that as it's a distraction from my real question. Can ETC or something else in REW tell me which frequencies are the predominant reflections? If so, how would I go about using it to determine that?

For example, the band limited IR? Do I do a full range sweep then filter in REW or do I need to do something else?

REW is a tool that can tell you all that, and this thread can help with the details on how to do it.

First reflection point treatments are typically as broadband as possible, i.e. reach as deep as possible. So the 1" of 703 on the wall wouldn't be ideal. I would be interested in knowing from the experts here if REW can tell us that.
Jeff
pepar is online now  
post #7857 of 14290 Old 01-10-2014, 05:30 AM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 25,316
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 287 Post(s)
Liked: 186
Quote:
Originally Posted by kbarnes701 View Post

LOL. (Hums: "they're coming to take me away, ha ha, they're coming to take me away.....)

Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs While In Solitary Confinement

Jeff
pepar is online now  
post #7858 of 14290 Old 01-10-2014, 05:32 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 21,603
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 3747 Post(s)
Liked: 2910
Quote:
Originally Posted by pepar View Post
 
Quote:
Originally Posted by kbarnes701 View Post

LOL. (Hums: "they're coming to take me away, ha ha, they're coming to take me away.....)

Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs While In Solitary Confinement

Jeff

 

All that rubber lining the walls can make for an overly dead environment :)

kbarnes701 is offline  
post #7859 of 14290 Old 01-10-2014, 07:26 AM
Advanced Member
 
StevenLansing's Avatar
 
Join Date: Mar 2007
Posts: 641
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 14
Quote:
Originally Posted by sdrucker View Post



One more practical problem, at least for me: for strictly Trinnov vs. no Trinnov testing, I've used the R-972's ability to turn on/off Trinnov. However, matching crossovers with and without calibration will be problematic. That's because Trinnov sets them precisely (e.g. 76 Hz for mains, believe it or not, and they can't be changed), while the non-Trinnov bass management setting is rounded to the nearest 10 Hz (e.g. 80 Hz). Less of an issue with this level of difference, but more of one for some other crossovers.
.

Did you see this in the frequency response measurements or was the 76Hz point what was reported by the Trinnov result screen? I was under the impression that Trinnov's crossover choice was overlaid over the R-972's crossover slopes creating some mismatch. I believe that was the main reason why Dr. Rich found the crossover splice between sub and mains to be less than ideal when he tested the R-972 with Trinnov engaged.
StevenLansing is offline  
post #7860 of 14290 Old 01-10-2014, 08:12 AM
AVS Club Gold
 
sdrucker's Avatar
 
Join Date: Feb 2006
Location: Chicago, IL
Posts: 2,376
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 181
Quote:
Originally Posted by StevenLansing View Post

Did you see this in the frequency response measurements or was the 76Hz point what was reported by the Trinnov result screen? I was under the impression that Trinnov's crossover choice was overlaid over the R-972's crossover slopes creating some mismatch. I believe that was the main reason why Dr. Rich found the crossover splice between sub and mains to be less than ideal when he tested the R-972 with Trinnov engaged.

The 76 Hz was what was reported on the Trinnov screen, and what is saved for that position after I rebooted the receiver. I actually set the R-972's manual crossover screen to the same settings as Trinnov (as close as I could get, meaning rounding upward 80 Hz instead of 76 Hz in this case) for my speakers anyway, to do the pre/post comparison (wanting the crossovers "pre" to be identical to those "post"), but as far as I can tell the Trinnov calculations don't get copied into the "manual" settings in any form. The opposite might be a different case.

At any rate the original crossovers I had before I did any Trinnov run were fairly close (using my last Audyssey run as a "point estimate', I picked 80 Hz for my mains and center anyway, and for my surrounds, I think I used something about 20 Hz lower than the crossover Trinnov came up with), so at least theoretically it shouldn't matter too much as long as any changes with manual crossovers are moving them upward - which rounding upward is doing. I also did a distance tweak, in the Audyssey fashion, with the subs connected to my MiniDSP to try to optimize the L/R+subs crossover splice/region, which is something outside of Trinnov or the R-972 as such.

Regardless, this isn't the place to discuss the R-972's functionality, as this is WAY off topic on a thread discussing REW measuring techniques and room improvement with acoustic treatments for the vast majority of active posters. Let's take this up on the R-972 thread. The rest of you: ignore this post and the one I'm answering smile.gif.

Audio Gear: Sherwood R-972 (Trinnov Altitude - coming soon!), NAD M27 amp
Display: Panasonic VT50, Lumagen Radiance Mini
Misc.: Oppo 103, Apple TV, JRiver ID NUC
PSB Imagine T2, Center, and Surrounds; HSU ULS-15 subs (2)
sdrucker is online now  
Sponsored Links
Advertisement
 
Reply Audio Theory, Setup, and Chat

Tags
Dayton , Dayton Audio , Room Equilizer Wizard Rew

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off