Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs - Page 521 - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!


Forum Jump: 
 715Likes
Reply
Thread Tools
post #15601 of 15831 Old 08-07-2015, 04:57 PM
AVS Addicted Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 10,481
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 3199 Post(s)
Liked: 1750
Quote:
Originally Posted by CherylJosie View Post
Well that pretty much covers it then. Same measurement setup as me.

How do you calibrate SPL? REW speaker or subwoofer signal, externl signal, or auto detect the sensitivity factor from mic cal file?
With a downloaded sensitivity parameter, as described in the Guide.
ahblaza likes this.
AustinJerry is online now  
Sponsored Links
Advertisement
 
post #15602 of 15831 Old 08-07-2015, 06:16 PM
Advanced Member
 
CherylJosie's Avatar
 
Join Date: May 2008
Posts: 746
Mentioned: 8 Post(s)
Tagged: 1 Thread(s)
Quoted: 380 Post(s)
Liked: 140
Quote:
Originally Posted by 3ll3d00d View Post
It looks for the impulse peak and treats this as the timing ref point, you control the default behaviour via http://www.roomeqwizard.com/help/hel.../analysis.html
Quote:
Impulse Response Window Defaults

The Left Side and Right Side window selectors offer a choice of window types to be applied to the impulse response data before and after the peak. These are the defaults applied to new measurements, window types for existing measurement can be altered via the IR Windows toolbar button. By default REW will set the widths of the windows automatically to show the whole room response, to override this uncheck the Set IR Window Widths Automatically box and set the default widths you wish to be applied to new measurements
...

Impulse Response Calculation
...
The Decimate IR selection controls whether REW reduces the sample rate of the impulse response to correspond to the range of frequencies in the measurement. Selecting this option greatly reduces the impulse response size for low frequency measurements and speeds up processing of the data.
...
After REW has made a measurement it can truncate the derived impulse response to preserve the important information while minimising the storage required for the measurement file.
...
Frequency Response Calculation
...
The Allow 96 PPO Log Spacing selection controls whether REW is permitted to convert frequency responses from linearly spaced data to logarithmically spaced data at 96 points per octave.
So from this verbage I conclude the following:

  • REW measures an audio segment via microphone and then based upon the settings of the program it calculates various derived data that can be plotted and analyzed but discards the actual audio segment to save storage space
  • To change some of these parameters it is necessary to take a new measurement because relevant information has not been retained
  • It is necessary to understand the fundamentals of signal processing to make intelligent decisions about these settings -- blind guesses and rule of thumb are not going to be adequate
  • There are subtle interactions between direct and reflected sound that distort magnitude, phase, and impulse responses but these measurements must detect both the direct and reflected responses independently to adequately separate their effects
Despite being able to sort of summarize these things, they are all murky in my technical background. So now I have to brush up on my signal processing math and read the REW manual front to back to keep up with you guys.

This is becoming a little more complicated than I bargained for but I will attempt the challenge anyway.

Quote:
If you want an absolute time measurement then you need a loopback and configure REW to use it. This sends the same impulse via the selected channels and REW then compares e recorded impulse peak against the reference impulse peak to determine the delay. You can use a software loopback here BTW, it just needs the original impulse to come back in on the nominated timing input channel. The only real rule is that the impulses must have similar frequency content otherwise the shape of the impulse is v different so peak alignment will fail to give accurate data. Note this means you can use it to measure software delays as well.
So how to do this with a USB microphone and is it going to provide me with any useful information? I only want to trim my room placements and room treatments and crossovers and EQ. Not interested in doing any digital filter designs or passive crossover tweaks etc. so if my weird phase plots are really a measurement artifact combined with a nasty set of reflections that I probably will not be able to treat adequately in here anyway I may just resort to that windowing though it will make phase measurements at subwoofer crossover less accurate and I need an estimate of the derating factor to get sensible result?

Quote:
REW also uses the peak to estimate the delay and can remove this in order to remove the delay induced excess phase. You may find that manually moving the impulse to bring T=0 to the initial rise gives more accurate results. There is some degree of interpretation required here to get the "right" answer.
This sounds like a sort of calibration on its own. Is this something to do before or after taking a measurement? Or should I just let REW figure it out on its own and call it a day?

Since impulse response and frequency response are sort of duals of each other, I take it the impulse response is derived from the frequency response? My recollection is that an impulse has ideally zero duration and zero energy and that makes it basically impossible to generate it in the analog world let alone use it as a measurement signal.

So is this derivation calculation of impulse response from frequency response why there is this ambiguity around the T=0 'right' answer? Sort of like the unknown constant of integration? How do distinct echoes and reverb come into play (as opposed to room resonances and phase shifts?) when REW calculates the timing? Or should I not ask? This seems like it might get complicated.

Quote:
I have used both Linux and Windows versions for 10 years now. I haven't noticed any limitations in the Linux version, do you have something specific that doesn't work?
Where to start...

  1. Gray box at bottom of 'preferences' window is truncated and the window cannot be resized. Scroll bar and text vanish below the bottom of the window.
  2. When the measurements sidebar has 30 unsaved measurements and throws an error on the next one, saving and deleting them becomes a nightmare. Saving and deleting the top one per the UI automated suggestion/option will preserve it but it is not removed from the window. Then the red X on the measurement will not remove it either. Must be removed from the window from the menu instead. From that point forward the measurements that actually get saved in the 'save all' may vary depending on factors I have not discerned yet. The point is, some measurements do not get saved and the UI gets all confused.
  3. The control-click method of deriving the distance from the ETC is not functional.
  4. The USB mic is not auto detected.
  5. The USB mic sensitivity factor does not satisfy the flag at beginning of measurement saying REW has not been calibrated (is it supposed to?) but running the SPL meter cal will satisfy that flag.
  6. When saving measurements, often (more than 60% of the time) the pop-up windows and forms are drawn empty of any contents and do not respond to any mouse click inside of them. Many retries required.
  7. When saving measurements, double clicking on an existing file may pop up the confirm yes/no window but the measurement data may still not be saved. Must single click on the file and then select OK from the button instead.
  8. There is no way to select the channel in the stream. No ASIO. Maybe I need to install Jack? Never successfully completed that step despite trying several times over the past few years. Output is always in stereo 2.0 format.
  9. Sometimes the 'preferences' window allows me to calibrate the SPL using SPDIF out but no sound comes out of the signal generator or measurements buttons. I have to de/re-select the sound card output.
  10. (default) and Default Device behavior changes in the sound card form. Sometimes it allows for selecting Default in the form directly below and sometimes that box is grayed out. It changes erratically. Both input and output seem to be affected.
  11. Random freezes requiring process kill
I am using the open source java that comes packaged with Ubuntu 14.04. Maybe these bugs have something to do with that but until I know more I hesitate to install any other version of Java. I broke many Linux installations by adding things.

I also am using an OS (14.04) that I installed before there was a GT730 video card driver in the Ubuntu repository. I had to install the stock Nvidia driver and even after replacing it with the driver that eventually appeared in the repository, the video card still misbehaves intermittently on flash video (weird switching between frame buffers apparently makes the video flicker in and out of its physical location and size rapidly).

Not sure if it is Linux or REW causing these issues. I try not to reboot unless I need the kernel update but I always install it so it activates on the next reboot anyway. Sometimes updates break the machine until it reboots but if I can limp along I just keep running.

UI issues with Linux are rampant. The desktop is a disaster. Linux server side is under great control but the desktop is a free for all. If you know of something that works better than Unity I would appreciate a tip. I have tried XFCE etc. and Unity is still the best for ease of use combined with features but I am considering installing Mint instead just to get away from Unity since I do not like it. Probably whatever desktop REW was ported to originally is the one to use for these measurements.
CherylJosie is online now  
post #15603 of 15831 Old 08-07-2015, 06:19 PM
Advanced Member
 
CherylJosie's Avatar
 
Join Date: May 2008
Posts: 746
Mentioned: 8 Post(s)
Tagged: 1 Thread(s)
Quoted: 380 Post(s)
Liked: 140
Quote:
Originally Posted by AustinJerry View Post
With a downloaded sensitivity parameter, as described in the Guide.
This appears not working in Linux. Maybe it is keyed off the USB auto detection that is not working either.
CherylJosie is online now  
post #15604 of 15831 Old 08-07-2015, 06:44 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
Is measuring distortion in REW purely for evaluation purposes or can it aid with stuff like speaker or sub placement and setting crossovers?

Last edited by PlasmaPZ80U; 08-07-2015 at 06:45 PM.
PlasmaPZ80U is offline  
post #15605 of 15831 Old 08-07-2015, 07:14 PM
AVS Addicted Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 10,481
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 3199 Post(s)
Liked: 1750
Quote:
Originally Posted by CherylJosie View Post
This appears not working in Linux. Maybe it is keyed off the USB auto detection that is not working either.
I have to admit, you Linux guys are stubborn. Just think how nice life would be if you had an inexpensive Windows laptop for running your audio tools that work so poorly in Linux. But I have been around Linux guys enough to know that you probably would rather have a root canal. All I can say is, it serves you right that you encounter compatibility issues at every turn. My sympathies.
Stereodude and ahblaza like this.
AustinJerry is online now  
post #15606 of 15831 Old 08-07-2015, 07:30 PM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,686
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 344 Post(s)
Liked: 116
Quote:
Originally Posted by PlasmaPZ80U View Post
Is measuring distortion in REW purely for evaluation purposes or can it aid with stuff like speaker or sub placement and setting crossovers?
It can aid in not blowing them up
ahblaza likes this.

My Room
My Music
Rega - Apollo, Rega - DAC, Goldpoint Passive, Classe CA-100, miniDSP
Jenzen Next

Last edited by jim19611961; 08-08-2015 at 06:49 AM.
jim19611961 is offline  
post #15607 of 15831 Old 08-07-2015, 10:10 PM
Advanced Member
 
CherylJosie's Avatar
 
Join Date: May 2008
Posts: 746
Mentioned: 8 Post(s)
Tagged: 1 Thread(s)
Quoted: 380 Post(s)
Liked: 140
Quote:
Originally Posted by AustinJerry View Post
I have to admit, you Linux guys are stubborn. Just think how nice life would be if you had an inexpensive Windows laptop for running your audio tools that work so poorly in Linux. But I have been around Linux guys enough to know that you probably would rather have a root canal. All I can say is, it serves you right that you encounter compatibility issues at every turn. My sympathies.
'sokay, whatever floats yer bloat.
CherylJosie is online now  
post #15608 of 15831 Old 08-07-2015, 10:47 PM
Advanced Member
 
CherylJosie's Avatar
 
Join Date: May 2008
Posts: 746
Mentioned: 8 Post(s)
Tagged: 1 Thread(s)
Quoted: 380 Post(s)
Liked: 140
Quote:
Originally Posted by PlasmaPZ80U View Post
Is measuring distortion in REW purely for evaluation purposes or can it aid with stuff like speaker or sub placement and setting crossovers?
Probably the only use for measuring distortion at installed venue is to detect borderline condition. Proper design and realistic expectation should handle such issues better up front.

Musicians who grew up slamming out power chords in garage bands know all about distortion and deafness. All the detriments with none of the fame for nearly all of them.

Such people have a harder time perceiving distortion due to overlap of deafness and harmonically shifted energy. Distortion measurement in real time would be good tool for such individual to ensure no one gets punished with deafening treble but then who really wants a deaf person running the sound anyway? Even clean treble can be obnoxious if done to excess. Sound person gotta have ears.
CherylJosie is online now  
post #15609 of 15831 Old 08-08-2015, 07:19 AM
AVS Special Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 1,413
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 740 Post(s)
Liked: 226
Quote:
Originally Posted by CherylJosie View Post
[LIST][*]REW measures an audio segment via microphone and then based upon the settings of the program it calculates various derived data that can be plotted and analyzed but discards the actual audio segment to save storage space[*]To change some of these parameters it is necessary to take a new measurement because relevant information has not been retained[*]It is necessary to understand the fundamentals of signal processing to make intelligent decisions about these settings -- blind guesses and rule of thumb are not going to be adequate
No this isn't correct. Why do you think this? There is an option to reduce the size of the stored impulse (see the analysis tab) but I don't believe that is the default. This means you can go back and change windows, shift the impulse around, change magnitude and so on to your hearts content.

The only exception to this is if you use the trace arithmetic to copy a measurement ( using the (a+b)/2 option), this copies the windowed impulse.
Quote:
Originally Posted by CherylJosie View Post
  • There are subtle interactions between direct and reflected sound that distort magnitude, phase, and impulse responses but these measurements must detect both the direct and reflected responses independently to adequately separate their effects
The IR is not distorted as such, it really is the sound recorded.


Quote:
Originally Posted by CherylJosie View Post
So how to do this with a USB microphone and is it going to provide me with any useful information? I only want to trim my room placements and room treatments and crossovers and EQ. Not interested in doing any digital filter designs or passive crossover tweaks etc. so if my weird phase plots are really a measurement artifact combined with a nasty set of reflections that I probably will not be able to treat adequately in here anyway I may just resort to that windowing though it will make phase measurements at subwoofer crossover less accurate and I need an estimate of the derating factor to get sensible result?
You can't get absolute time data without a loopback. You can get accurate relative time data from multiple measurements if each measurement contains the same reference content to align against. This is feasible if you can route the REW audio signal through a signal processor that can copy/delay channels, possibly brutefir is an option, there are probably other options too.

The utility of accurate phase data across multiple responses is to enable summing multiple measurements. If you just want to analyse a single measurement then no need to do this.




Quote:
Originally Posted by CherylJosie View Post
Since impulse response and frequency response are sort of duals of each other, I take it the impulse response is derived from the frequency response? My recollection is that an impulse has ideally zero duration and zero energy and that makes it basically impossible to generate it in the analog world let alone use it as a measurement signal.

So is this derivation calculation of impulse response from frequency response why there is this ambiguity around the T=0 'right' answer? Sort of like the unknown constant of integration? How do distinct echoes and reverb come into play (as opposed to room resonances and phase shifts?) when REW calculates the timing? Or should I not ask? This seems like it might get complicated.
The FR is computed from the IR using fft. Probably you start by reading the sections of the REW help on how it measures.


Quote:
Originally Posted by CherylJosie View Post
Where to start...

  1. Gray box at bottom of 'preferences' window is truncated and the window cannot be resized. Scroll bar and text vanish below the bottom of the window.
  2. When the measurements sidebar has 30 unsaved measurements and throws an error on the next one, saving and deleting them becomes a nightmare. Saving and deleting the top one per the UI automated suggestion/option will preserve it but it is not removed from the window. Then the red X on the measurement will not remove it either. Must be removed from the window from the menu instead. From that point forward the measurements that actually get saved in the 'save all' may vary depending on factors I have not discerned yet. The point is, some measurements do not get saved and the UI gets all confused.
  3. The control-click method of deriving the distance from the ETC is not functional.
  4. The USB mic is not auto detected.
  5. The USB mic sensitivity factor does not satisfy the flag at beginning of measurement saying REW has not been calibrated (is it supposed to?) but running the SPL meter cal will satisfy that flag.
  6. When saving measurements, often (more than 60% of the time) the pop-up windows and forms are drawn empty of any contents and do not respond to any mouse click inside of them. Many retries required.
  7. When saving measurements, double clicking on an existing file may pop up the confirm yes/no window but the measurement data may still not be saved. Must single click on the file and then select OK from the button instead.
  8. There is no way to select the channel in the stream. No ASIO. Maybe I need to install Jack? Never successfully completed that step despite trying several times over the past few years. Output is always in stereo 2.0 format.
  9. Sometimes the 'preferences' window allows me to calibrate the SPL using SPDIF out but no sound comes out of the signal generator or measurements buttons. I have to de/re-select the sound card output.
  10. (default) and Default Device behavior changes in the sound card form. Sometimes it allows for selecting Default in the form directly below and sometimes that box is grayed out. It changes erratically. Both input and output seem to be affected.
  11. Random freezes requiring process kill
I am using the open source java that comes packaged with Ubuntu 14.04. Maybe these bugs have something to do with that but until I know more I hesitate to install any other version of Java. I broke many Linux installations by adding things.

I also am using an OS (14.04) that I installed before there was a GT730 video card driver in the Ubuntu repository. I had to install the stock Nvidia driver and even after replacing it with the driver that eventually appeared in the repository, the video card still misbehaves intermittently on flash video (weird switching between frame buffers apparently makes the video flicker in and out of its physical location and size rapidly).

Not sure if it is Linux or REW causing these issues. I try not to reboot unless I need the kernel update but I always install it so it activates on the next reboot anyway. Sometimes updates break the machine until it reboots but if I can limp along I just keep running.

UI issues with Linux are rampant. The desktop is a disaster. Linux server side is under great control but the desktop is a free for all. If you know of something that works better than Unity I would appreciate a tip. I have tried XFCE etc. and Unity is still the best for ease of use combined with features but I am considering installing Mint instead just to get away from Unity since I do not like it. Probably whatever desktop REW was ported to originally is the one to use for these measurements.
One of those glitches (with deleting measurements) occurs on windows too. The others sound like Java issues. What do you get if you run java -version at the CLI? You should install the oracle JVM anyway, use the current Java 8 one. I believe there are packages available for Ubuntu.

I don't like unity personally. I have been a KDE user (on Debian unstable) since kde4 came out, haven't upgraded to kde5 yet. I find Linux less error prone and more stable than windows tbh.

Re sound devices, are you using alsa or pulse? No need to use jack for REW, you just need to be able to decipher how Linux names audio devices and know how to configure your audio subsystem. This is one area of Linux that remains rather obtuse.

The mic sensitivity issue.. You should report it over at HTS and see if there is some known issue or a bug to be fixed.
ahblaza likes this.
3ll3d00d is offline  
post #15610 of 15831 Old 08-08-2015, 08:21 AM
AVS Special Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 1,189
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 274 Post(s)
Liked: 108
Quote:
Originally Posted by CherylJosie View Post
Probably the only use for measuring distortion at installed venue is to detect borderline condition. Proper design and realistic expectation should handle such issues better up front.
Maybe but we had a lengthy discussion about distortion on this list. Using REW to measure distortion and finding ways to reduce it to single digit levels made my music sound much better.

I think you're right in that the distortion levels are not always palpable IYKWIM but when you reduce them that is palpable.
ahblaza likes this.
artur9 is online now  
post #15611 of 15831 Old 08-08-2015, 08:23 AM
AVS Special Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 1,189
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 274 Post(s)
Liked: 108
Re: Loopback

My laptop does not have HDMI so I use its analog output. Sometimes I use its Toslink output.

Could/should I use the headphone output from the Laptop and plug it into the Mic jack to get loopback? Is that an insanely bad idea?
artur9 is online now  
post #15612 of 15831 Old 08-08-2015, 03:11 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
Quote:
Originally Posted by artur9 View Post
Quote:
Originally Posted by CherylJosie View Post
Probably the only use for measuring distortion at installed venue is to detect borderline condition. Proper design and realistic expectation should handle such issues better up front.
Maybe but we had a lengthy discussion about distortion on this list. Using REW to measure distortion and finding ways to reduce it to single digit levels made my music sound much better.

I think you're right in that the distortion levels are not always palpable IYKWIM but when you reduce them that is palpable.
How does one reduce distortion?
PlasmaPZ80U is offline  
post #15613 of 15831 Old 08-08-2015, 03:20 PM
AVS Addicted Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 10,481
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 3199 Post(s)
Liked: 1750
Quote:
Originally Posted by PlasmaPZ80U View Post
How does one reduce distortion?
Easy answer is to restrict the output level for the driver so that it is operating in its "low distortion" range. This can vary from speaker to speaker, of course. One of the values of running distortion measurements at increasingly higher output levels is to gain an understanding of how far you can drive the speaker with distorting it.
ahblaza likes this.
AustinJerry is online now  
post #15614 of 15831 Old 08-08-2015, 03:51 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
Quote:
Originally Posted by AustinJerry View Post
Quote:
Originally Posted by PlasmaPZ80U View Post
How does one reduce distortion?
Easy answer is to restrict the output level for the driver so that it is operating in its "low distortion" range. This can vary from speaker to speaker, of course. One of the values of running distortion measurements at increasingly higher output levels is to gain an understanding of how far you can drive the speaker with distorting it.
So, the point of measuring distortion is to know how loud you can listen to without harming the speakers?
PlasmaPZ80U is offline  
post #15615 of 15831 Old 08-08-2015, 04:00 PM
AVS Addicted Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 10,481
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 3199 Post(s)
Liked: 1750
Quote:
Originally Posted by PlasmaPZ80U View Post
So, the point of measuring distortion is to know how loud you can listen to without harming the speakers?
No, one reason to measure is to ensure that there isn't excessive distortion at the levels you are accustomed to listening at.

For example, I rarely listen to movies or music at levels above MV= -10. At those levels distortion is very low. Do I really care how loud I can go without significant distortion? No.
ahblaza likes this.
AustinJerry is online now  
post #15616 of 15831 Old 08-08-2015, 04:07 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
Quote:
Originally Posted by AustinJerry View Post
Quote:
Originally Posted by PlasmaPZ80U View Post
So, the point of measuring distortion is to know how loud you can listen to without harming the speakers?
No, one reason to measure is to ensure that there isn't excessive distortion at the levels you are accustomed to listening at.

For example, I rarely listen to movies or music at levels above MV= -10. At those levels distortion is very low. Do I really care how loud I can go without significant distortion? No.
Well, I rarely listen past -35 MV and since YPAO doesn't calibrate to reference level that happens to be around -25 to -22.5 MV after factoring in the difference.
PlasmaPZ80U is offline  
post #15617 of 15831 Old 08-08-2015, 06:12 PM
AVS Special Member
 
Frank D's Avatar
 
Join Date: Sep 2001
Location: Ontario, Canada
Posts: 1,546
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked: 45
Quote:
Originally Posted by AustinJerry View Post
No, one reason to measure is to ensure that there isn't excessive distortion at the levels you are accustomed to listening at.

For example, I rarely listen to movies or music at levels above MV= -10. At those levels distortion is very low. Do I really care how loud I can go without significant distortion? No.
At about 10 db below reference (which is what I assume you mean by MV=-10??) distortion is not bad at all and many decent speakers should handle that with ease. However I find that when I go above that 10 db below reference, distortion starts to set in. The higher I go with volume the more I can hear it. Still looking for a speaker to play cleanly at reference level.

Would be nice if REW showed a scale on the left side to show distortion. Also when moving my cursor over the distortion measurements it seems to change drastically from say .05 % THD to say 4% THD but I do not really see the distortion line change by that much. Anyone know why the distortion line show like that?
Frank D is online now  
post #15618 of 15831 Old 08-09-2015, 05:38 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,686
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 344 Post(s)
Liked: 116
Quote:
Originally Posted by Frank D View Post
At about 10 db below reference (which is what I assume you mean by MV=-10??) distortion is not bad at all and many decent speakers should handle that with ease. However I find that when I go above that 10 db below reference, distortion starts to set in. The higher I go with volume the more I can hear it. Still looking for a speaker to play cleanly at reference level.

Would be nice if REW showed a scale on the left side to show distortion. Also when moving my cursor over the distortion measurements it seems to change drastically from say .05 % THD to say 4% THD but I do not really see the distortion line change by that much. Anyone know why the distortion line show like that?
The distortion percentage is about two things. Distortion level vs output. A level distortion line will only yield a constant distortion percentage if the measured output (FR) is also level or flat.

What SPLdb are you calling reference?

My Room
My Music
Rega - Apollo, Rega - DAC, Goldpoint Passive, Classe CA-100, miniDSP
Jenzen Next
jim19611961 is offline  
post #15619 of 15831 Old 08-09-2015, 06:16 AM
AVS Addicted Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 10,481
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 3199 Post(s)
Liked: 1750
Quote:
Originally Posted by jim19611961 View Post

What SPLdb are you calling reference?
Not Frank, but reference AFAIK is 105dB measured at the MLP. A typical AVR's test tones run at MV=0, -30dBFS, which measures 75dB at the MLP.
AustinJerry is online now  
post #15620 of 15831 Old 08-09-2015, 07:14 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,686
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 344 Post(s)
Liked: 116
My take is this then.

The 105SPLdb reference is probably easy to achieve in the bass department given most people have subs and you can just keep adding woofage and watts to get to whatever SPL you want. Its starting 1/2 to 1 octave above the sub(s) XO frequency, where the sub(s) are no longer helping, where getting to 105SPLdb might be difficult. Here, your relying on the mains predominantly.

Lets take a speaker that has a 87SPL at 1w/m efficiency. At 2.5m, that same speaker will only be about 79SPLdb. So:

1W = 79SPLdb
10W = 89SPLdb
100W = 99SPLdb
300W = 105SPLdb

Well, not everyone has 300 watts, nor have speakers that can handle that. Some do. But you gain 6db by driving both channels. So you could reach 105SPLdb at 100 watts, 2 channels driven, with the theoretical example above.

My Room
My Music
Rega - Apollo, Rega - DAC, Goldpoint Passive, Classe CA-100, miniDSP
Jenzen Next
jim19611961 is offline  
post #15621 of 15831 Old 08-09-2015, 07:35 AM
AVS Special Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 1,413
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 740 Post(s)
Liked: 226
FWIW sub reference is 115dB, add 5dB (give or take a few) if using bass management and then add more if you have the usual upwards tilt at the low end. If you want to do this and dig below 20Hz then it quickly gets to be quite a big ask.
3ll3d00d is offline  
post #15622 of 15831 Old 08-09-2015, 08:19 AM
AVS Special Member
 
Frank D's Avatar
 
Join Date: Sep 2001
Location: Ontario, Canada
Posts: 1,546
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked: 45
Quote:
Originally Posted by jim19611961 View Post
The distortion percentage is about two things. Distortion level vs output. A level distortion line will only yield a constant distortion percentage if the measured output (FR) is also level or flat.

What SPLdb are you calling reference?
I call SPL reference 85 db with 20 db peaks at MLP.
Frank D is online now  
post #15623 of 15831 Old 08-09-2015, 08:54 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,686
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 344 Post(s)
Liked: 116
Quote:
Originally Posted by Frank D View Post
I call SPL reference 85 db with 20 db peaks at MLP.
I wouldn't think you would have problems reaching that SPLdb unless:

1) Inefficient speakers (<85SPLdb 1w/m)
2) Not much power (<100 watts)
3) LP > 3m from speakers
4) Or some combination of the above

Most music isn't dynamic enough to need 20db headroom. Movies, yup.

My Room
My Music
Rega - Apollo, Rega - DAC, Goldpoint Passive, Classe CA-100, miniDSP
Jenzen Next

Last edited by jim19611961; 08-09-2015 at 08:58 AM.
jim19611961 is offline  
post #15624 of 15831 Old 08-09-2015, 09:37 AM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
I'm confused, do movies have 75dB average with 105dB peaks (30dB difference) or 85dB average with 105dB peaks (20dB difference)?

And is LFE 85dB average with 115dB peaks or 95dB average with 115dB peaks?

And to clarify an older topic I'm still confused about, when watching a typical BD movie, does the LFE channel actually play 10dB higher than the other channels straight out of the subwoofer or does an internal offset from disk to player to AVR make it play at the same level as the other channels?

In short, when playing at reference level (0MV on a AVR that is calibrated to 75dB with -30.0dBFS signal), what is the loudest spl the sub can output and what is the loudest the other speakers can? Assuming bass management at 80Hz and not running the sub trim hot.

Last edited by PlasmaPZ80U; 08-09-2015 at 09:39 AM.
PlasmaPZ80U is offline  
post #15625 of 15831 Old 08-09-2015, 09:52 AM
AVS Special Member
 
Soulburner's Avatar
 
Join Date: Oct 2007
Posts: 1,690
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 194 Post(s)
Liked: 254
Quote:
Originally Posted by Frank D View Post
At about 10 db below reference (which is what I assume you mean by MV=-10??) distortion is not bad at all and many decent speakers should handle that with ease. However I find that when I go above that 10 db below reference, distortion starts to set in. The higher I go with volume the more I can hear it. Still looking for a speaker to play cleanly at reference level.
http://reaction-audio.myshopify.com/...tions/speakers
Soulburner is online now  
post #15626 of 15831 Old 08-09-2015, 10:32 AM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
Quote:
Originally Posted by AustinJerry View Post
Quote:
Originally Posted by PlasmaPZ80U View Post
I'm confused, do movies have 75dB average with 105dB peaks (30dB difference) or 85dB average with 105dB peaks (20dB difference)?

And is LFE 85dB average with 115dB peaks or 95dB average with 115dB peaks?

And to clarify an older topic I'm still confused about, when watching a typical BD movie, does the LFE channel actually play 10dB higher than the other channels straight out of the subwoofer or does an internal offset from disk to player to AVR make it play at the same level as the other channels?

In short, when playing at reference level (0MV on a AVR that is calibrated to 75dB with -30.0dBFS signal), what is the loudest spl the sub can output and what is the loudest the other speakers can? Assuming bass management at 80Hz and not running the sub trim hot.
These are peculiar questions. There is no such thing as an "average" level with movies. An action movie will have a much louder average level that a reflective movie filled with dialog. The 105dB reference (115dB for LFE) is simply a standard.

This has been answered a number of times. The LFE channel is ENCODED on the disk 10dB lower than the other channels. It is expected that the AVR add back the 10dB during playback so that the LFE and other channels have the appropriate relative levels.

The loudest level that a speaker can produce depends on the efficiency of the speaker, the power driving it, and the ability of the AVR to go above MV=0. On AVR's with Audyssey room correction, if Audyssey is adding significant boos to the speakers, the AVR may be restricted WRT how high the MV can go above zero.

When you ask questions like this, is it to satisfy your curiosity, or do you actually want to drive your system significantly above Reference levels?
Curiousity; I never go near reference level.
PlasmaPZ80U is offline  
post #15627 of 15831 Old 08-09-2015, 12:00 PM
AVS Addicted Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 10,481
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 3199 Post(s)
Liked: 1750
Quote:
Originally Posted by PlasmaPZ80U View Post
Curiousity; I never go near reference level.
Have you ever given consideration to the fact that it takes time to answer your questions, time better spent helping someone who has a real issue?
muffinmcfluffin likes this.
AustinJerry is online now  
post #15628 of 15831 Old 08-09-2015, 01:06 PM
AVS Special Member
 
Frank D's Avatar
 
Join Date: Sep 2001
Location: Ontario, Canada
Posts: 1,546
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked: 45
Quote:
Originally Posted by jim19611961 View Post
I wouldn't think you would have problems reaching that SPLdb unless:

1) Inefficient speakers (<85SPLdb 1w/m)
2) Not much power (<100 watts)
3) LP > 3m from speakers
4) Or some combination of the above

Most music isn't dynamic enough to need 20db headroom. Movies, yup.
Sure all good points but to address each.
1) Speaker efficiency is 94 db and rated 8 ohms using Coaxial Speaker
2) Using Sunfire Cinema Grand that is rated 200 watts RMS into 8 ohm (and 400 watts into 4 ohm)
3) MLP is only 8 feet so less then 3 m. Treated, dedicated, small HT room (11 ft w x 17.5 ft long x 7.5 ft high).

I measured distortion using REW, UMIK-1 mic pointed at speaker using 0 degree cal file and got for my centre channel:

A. Vol on XMC-1 set to -10 (so spl of 75 db using pink noise just to give you an idea at MLP ie not at 1 foot where it would be louder), no DIRAC, measured from 1 foot away:

Some of the higher distortion spots:
850 hz THD = 5.5%, 3rd harmonic= 5.5%
1400 hz THD = 2%, 3rd = 1.3%
1250 hz THD = 7.5%, 3rd = 5.8%
Most other areas are below 1% so I did not list them.

B. Same volume etc as above but at MLP that is 8 feet away:
200 hz THD = 3.5%, 3rd = 3.5%
300 hz THD= 6.2%, 3rd = 6.1%
500 hz THD = 3.6%, 3rd = 3.6%
675 hz THD = 2.5%, 3rd = 2.5%
800 hz THD = 6.4%, 3rd = 6.4%
1000 hz THD and 3rd = 3.3%
1250 hz THD = 7.5%, 3rd = 6%
1350 hz THD and 3rd = 5%
Most other areas are below 1% so I did not list them.


So what do you guys think of those REW distortion numbers?

Last edited by Frank D; 08-09-2015 at 01:12 PM.
Frank D is online now  
post #15629 of 15831 Old 08-09-2015, 01:07 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 8,442
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 658 Post(s)
Liked: 274
Quote:
Originally Posted by AustinJerry View Post
Quote:
Originally Posted by PlasmaPZ80U View Post
Curiousity; I never go near reference level.
Have you ever given consideration to the fact that it takes time to answer your questions, time better spent helping someone who has a real issue?
I learn plenty from asking questions, both on this forum and elsewhere in life. It was always encouraged in the schools and colleges I went to.

However, I don't believe my asking a question obligates anyone here to answer it if they don't prefer to. Don't think there's anything else for me to say on the matter.
Alan P and Soulburner like this.
PlasmaPZ80U is offline  
post #15630 of 15831 Old 08-09-2015, 01:15 PM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,686
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 344 Post(s)
Liked: 116
Quote:
Originally Posted by Frank D View Post

So what do you guys think of those REW distortion numbers?
Their too high. Much too high. Something isnt right. All of mine are below 1% from 25hz up.
ahblaza likes this.

My Room
My Music
Rega - Apollo, Rega - DAC, Goldpoint Passive, Classe CA-100, miniDSP
Jenzen Next
jim19611961 is offline  
Sponsored Links
Advertisement
 
Reply Audio Theory, Setup, and Chat

Tags
Dayton , Dayton Audio , Room Equilizer Wizard Rew



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