Home Theater Forum and Systems banner

61 - 80 of 116 Posts

·
Registered
Joined
·
106 Posts
So, I've got a question regarding psychoacoustics in general. I'm wondering, when I read that the human hearing mechanism approaches a more anechoic state of perception as frequency increases, does this have to do with the other often referred to phenomenon that low frequencies are (to us) "omnidirectional", and that high frequencies are directional? Or are these two separate, unrelated phenomenon? Also, is there a chart or graph or something that shows the approximate time/frequency relationship of the psychoacoustic theory.

I guess I just want to further understand this phenomenon. I'm wondering if this phenomenon is at all applicable in an average car interior, or is it too small, in other words, are the reflected sounds too close in arrival time to the direct sounds to be eligible for psychoacoustic discussion?

Anybody?
 

·
Registered
Joined
·
98 Posts
Bobkatz,

Hi, my name is Nick. I very often read this and other forums but rarely do I chime in on a topic. Most of the other people on this forum are more experienced with the general types of questions I read, and I learn a lot by just reading all the Q & A. However, your question of frequency dependent windowing caught my interest as this is something I have a daily working experience with. Have you ever heard of SysTune? It's software from AFMG, a Berlin based software company that makes all kinds of awesome software that's used in the loudspeaker design and installation industry. SysTune is primarily geared toward the pro sound industry, but, I use it. And I work as an installer of custom car audio systems. Frequency dependent windowing, in SysTune this is called the TFC, which stands for time frequency constant. It works like this, when you set the window there are 3 markers that represent the right hand window. One for 8 kHz, one for 1 kHz, and one for 125Hz. Now those are just markers to kinda give you an idea of the "spread" of the TFC. In actuality it is a constantly variable time window without discrete steps. Whatever the window time you set for 8 kHz, it is twice as long for half that freq. For example, say you set your 8 kHz marker for 2ms after the peak of the IR, then your window time for for 4 kHz will be 4ms, 2 kHz-8ms, 1 kHz-16ms, and so on. This way, you can window out reflections but still keep full range response in one graph. Well, window out higher freq. reflections anyway. So as you can imagine, in a car, the reflected energy can be REALLY close to the direct arrival. So with conventional windowing, in a car, if you windowed out the reflections, you'd end up with barely useable data, it would be a time consuming process to properly perform a phase alignment of mid to tweeter in a fully active system for example. And then have to change your settings to see the next driver's interactions as you worked down the freq. scale. But with TFC, it works brilliantly. Another one that is similar to SysTune is Smaart v7. It is a program that uses what's called Multi Time Window, or MTW. however, with Smaart, the multi time window is not user adjustable, so while I do use that software at work for certain things, SysTune is my main program. Room EQ Wizard plays an integral role in my work by allowing me to create compensation files for different mics and inputs, and also for its trace arithmetic, which I use to find the cabin gain of different vehicles. Anyway, thought just in case you hadn't heard of SysTune (or Smaart) that you might find them interesting, as the very mechanism they use to perform their measurements is variable time windowing!

I have been considering the purchase of Smaart 7Di recently, but them ran across SysTune. Now I am torn as to which program might be better, because they are very expensive. I don't mind purchasing one of them, but buying both would be too cost prohibitive for me.

Smaart had the initial edge for me due to the Mac OS availability, which is my preferred platform.

But from your description, it sounds like SysTune might have an edge for the in-car environment due to the freq. dependent windowing?
 

·
Registered
Joined
·
106 Posts
Oh boy, that is a SERIOSLY tough question to answer, they are both EXTREMELY awesome software IMO. For me, as I've used bot Smaart v5, v6, v7, and SysTune (and now SysTune Pro), SysTune is a more useful program because I can use it for things that Smaart can't do. Like for instance, one of the things that I used to use my oscilloscope for, which is determining the relative acoustic polarity between drive units in factory systems, or any system where you can't actually see the speaker. With SysTune you can choose "time signal" as one of the display options, and it's quicker than routing my interface's output into the input of my oscilloscope. Also, SysTune has delay analysis and virtual EQ, and offline investigation of stored data, the advantages of which are too numerous to go in depth on right now, but suffice it to say that they allow me to SERIOUSLY cut down on installation and tuning time of higher end, fully active systems. Smaart has none of these things. Although, as long as you know how to use the software, either one will ultimately provide the exact same results. It's just that for me, since I work with sound systems every day, yes, SysTune has the upper hand. If I were going to use recommend either one for personal, infrequent use, I'd say Smaart for sure. SysTune has a STEEP learning curve. Smaart is much more streamlined, intuitive, user interface wins hands-down. Really depends how much you're into audio test and measurement and acoustic analysis, with SysTune, I've learned exponentially more about IRs and FFT in the first few months of ownership than in all the time combined with Smaart, REW, AudioTools for iOS, etc. With either one, it's truly an eye opening experience to be able to CONTINUALLY test a loudspeaker system and make changes and in real time see how those changes affect the resulting measurements/sound. With programs like REW, although free, and totally awesome, you have to take a sweep, make a change, take a sweep, make a change...............truly two very different experiences. Oh yeah, and a lot of what I mentioned as features of SysTune is actually SysTune Pro, which is more expensive than Di, it's just about the same (+\- & $50) as Smaart v7 full version. Also, a lot of guys are using SysTune on MAC, not natively of course, but in bootcamp I think.
 

·
Registered
Joined
·
98 Posts
Sounds like SysTune is much better because of the real-time measuring tools.

I'm going to grab the demo and try it out.

How many mics are you using?


Sent from my iPhone using Tapatalk
 

·
Registered
Joined
·
261 Posts
I've generated a beta build that has a psychoacoustic smoothing option that uses a frequency-dependent window and dip limiting, there is a Windows version and a Mac version. I don't recommend using the psychoacoustic smoothing for the EQ target match, variable smoothing will produce better results.
what's the difference between ERB (equivalent rectangular bandwidth I suppose?) and Psy smoothing? ERB is, visually, noticeably nicer than Psy as the latter has odd humps. It also seems a better option than Var as Var is too smooth at high frequencies. Psy looks a lot like one of the acourate functions tbh albeit one of the ones that, on its own, is not useful and seems to be designed to be used with another function.

I would say ERB looks particularly good at dealing with HF noise, for example it really removes the cruft from a a gated speaker measurement without actually changing it's real shape at all. It's not as effective for subwoofer use though as it seems you lose a lot of (lower) frequency resolution, it seems more effective >5-600Hz to me.
 

·
Registered
Joined
·
106 Posts
Sounds like SysTune is much better because of the real-time measuring tools. I'm going to grab the demo and try it out. How many mics are you using? Sent from my iPhone using Tapatalk
So, I use a combination of a 3 mic array, and a single mic. All of the mics are Dayton EMM6, one of which was calibrated by Cross Spectrum Labs. I also have a UMM6 from CSL that I use with Smaart Tools on my iPad. But as far as SysTune goes, the 3 mic array allows me to quickly gather 3,6,9,12,15 etc. different measurement points, this way I can quickly/easily determine what is mic placement related anomalies, and what is valid data that needs to be paid attention to. Also, with SysTune you can select all inputs, or any combination of inputs for real time spatial averaging. But for time/phase alignment thru crossover, I select the mic at the middle of the array usually. I used to use a 4 mic array, but it's size and unwieldy shape made it difficult to position, and I see no real disadvantage to 3 vs. 4, but the 3 mic array has definite advantages over the 4 because it's smaller, lighter, easier to position and move. You wouldn't think one mic would make much difference but it does. Especially the added width and weight of the mounting hardware. Keep in mind, that the exact same results could ultimately be realized with a single mic, many of the choices I make in my measurement setup are directly influenced by the fact that I use it to tune systems at work, and so, since I'm being paid to do so, efficiency is key. I don't want to waste customers time or money. So I'm always trying to find ways to make the process as efficient as possible, without sacrificing accuracy or the quality of the end result.

Edit: Smaart is also fully real time.
 

·
Registered
Joined
·
98 Posts
Oh. I took you post to say Smaart had to be swept, analyzed, reswept? While SysTune could be viewed and adjusted in real time?

I like the multiple mic array idea because special averaging in a pain in the ___. I hate doing it in REW because of the time it takes to sweep, move, sweep, repeat. Especially this time of year when I'm working in a hot garage.

What's the mic array look like?
Where do you buy one?

I have that same Dayton mic from Cross Specrum Labs. Might get a few more.

I have a miniDSP USB mic, too. But I can't use it for IR with REW because it's USB.


Sent from my iPhone using Tapatalk
 

·
Premium Member
Joined
·
6,711 Posts
what's the difference between ERB (equivalent rectangular bandwidth I suppose?) and Psy smoothing? ERB is, visually, noticeably nicer than Psy as the latter has odd humps. It also seems a better option than Var as Var is too smooth at high frequencies. Psy looks a lot like one of the acourate functions tbh albeit one of the ones that, on its own, is not useful and seems to be designed to be used with another function.

I would say ERB looks particularly good at dealing with HF noise, for example it really removes the cruft from a a gated speaker measurement without actually changing it's real shape at all. It's not as effective for subwoofer use though as it seems you lose a lot of (lower) frequency resolution, it seems more effective >5-600Hz to me.
ERB follows the profile of the ear's Equivalent Rectangular Bandwidth, which is pretty much 1/6th octave above 1 kHz and levels out to 25 Hz or so at low frequencies. I'm not sure it has much practical application though. The variable smoothing option is deliberately quite smoothed at HF (1/3rd octave above 10 kHz) to provide a good input for the automatic EQ. The psychoacoustic smoothing in the beta version is to emulate the response Bob provided, a 15 cycle frequency-dependent window combined with what looks like a form of dip limiting consisting of taking the greater of the response or the 1/3rd octave smoothed response. That won't make it into a release version, I have other ideas about what psychoacoustic smoothing should consist of and frequency-dependent windowing will be provided as part of the IR window options.
 

·
Registered
Joined
·
106 Posts
Oh. I took you post to say Smaart had to be swept, analyzed, reswept? While SysTune could be viewed and adjusted in real time? I like the multiple mic array idea because special averaging in a pain in the ___. I hate doing it in REW because of the time it takes to sweep, move, sweep, repeat. Especially this time of year when I'm working in a hot garage. What's the mic array look like? Where do you buy one? I have that same Dayton mic from Cross Specrum Labs. Might get a few more. I have a miniDSP USB mic, too. But I can't use it for IR with REW because it's USB. Sent from my iPhone using Tapatalk
Oh, ok gotcha. Yeah, I guess I should have been more clear, Smaart is just like SysTune in that respect. Totally real time.
 

·
Registered
Joined
·
106 Posts
. That won't make it into a release version, I have other ideas about what psychoacoustic smoothing should consist of and frequency-dependent windowing will be provided as part of the IR window options.
Right on John! That is absolutely awesome. I'm a big fan of frequency dependent windowing, not necessarily for psychoacoustic reasons, but still, I can't wait!
 

·
Premium Member
Joined
·
6,711 Posts
New beta builds with frequency dependent window options in the IR windows dialog (defaults in Analysis preferences) and revised psychoacoustic smoothing are now available for Windows and Mac.
 

·
Premium Member
Joined
·
1,562 Posts
JohnM,
Thanks for this feature! While I don't have particular use for it in SPL analysis, It sure works great for phase analysis. See below.

Phase without FDW.PNG

FDW.PNG
 

·
Registered
Joined
·
106 Posts
From the SysTune manual, referring to their TFC (frequency dependent) window

"For each frequency data point there is a different window length, namely for each doubling of the frequency the window length is cut in half. For example, the window length at 1 kHz is twice as long as the length at 2 kHz but only half as long as at 500 Hz. The actual window for each frequency is a Tukey 50% window. "

Can you give us a description of the time/frequency relationship used in the beta version you mentioned. I haven't been able to test it out yet, but I will real soon.
 

·
Premium Member
Joined
·
6,711 Posts
The window width varies continuously with frequency, width is per the user selection (e.g. 15 cycles is 15 ms at 1 kHz, 1.5 ms at 10 kHz, 150 ms at 100 Hz etc). The window shape is Gaussian.

Edit: Here is the relevant part of the help file:

In addition to the left and right windows a frequency-dependent Gaussian window can be applied. This is a window whose width varies inversely with frequency, getting progressively narrower as frequency increases. The width of the window can be specified as a number of cycles or an octave fraction. If the width is in cycles then the width (between the half amplitude points of the window) at any frequency will be the number of cycles times the period of that frequency, for example a 15 cycle window will have a width at 1 kHz of 15 * (1/1000) = 0.015 s or 15 ms. The corresponding octave fraction has an effect similar to applying a smoothing of the same octave fraction, except the variable window excludes progressively more of the late arriving sound as frequency increases rather than just averaging it out - this has similarities with the way the ear increasingly picks out the direct sound from the speaker at higher frequencies.

If Add frequency dependent window is selected the window is applied after first applying the selected left and right windows. The FDW is centred on the window reference time - for best results this should be at the peak of the impulse.
 

·
Banned
Joined
·
4,838 Posts
Wow, John, great work! Thanks so much.

I see that the Psychoacoustical smoothing must be applied to a curve that has full original data resolution, not one with 1/96th oct data (or averaged, or any amount of previous smoothing, etc.), as with the previous beta. Could we get a brief explanation of how that smoothing is done, please?
 

·
Premium Member
Joined
·
6,711 Posts
I see that the Psychoacoustical smoothing must be applied to a curve that has full original data resolution, not one with 1/96th oct data (or averaged, or any amount of previous smoothing, etc.), as with the previous beta. Could we get a brief explanation of how that smoothing is done, please?
Could you elaborate on that, please? Previous smoothing settings shouldn't make any difference. Log spaced data will give slightly different results as some averaging has already happened at HF to convert the data to log spaced. The PSY smoothing is 1/3 octave below 100 Hz, 1/6 octave above 1 kHz. It calculates a cubic mean (cube root of average of the cubed values) to place more emphasis on peaks.
 

·
Banned
Joined
·
4,838 Posts
Could you elaborate on that, please? Previous smoothing settings shouldn't make any difference. Log spaced data will give slightly different results as some averaging has already happened at HF to convert the data to log spaced. The PSY smoothing is 1/3 octave below 100 Hz, 1/6 octave above 1 kHz. It calculates a cubic mean (cube root of average of the cubed values) to place more emphasis on peaks.

Red_Psy 1 from raw_ unsmoothed measurement_ Purple_Psy 2 from copy of Psy 1.jpg

Red plot is a regular measurement with Psychoacoustic smoothing added.

Purple plot is first plot with no smoothing, then click the Average the Responses button to make a "copy," (do it all the time, super convenient, realizing it contains 1/96th oct data and no phase), then Psychoacoustic smoothing added. Difference is small, as you say, but 1 dB above 10 kHz (still pretty small).
 

·
Premium Member
Joined
·
6,711 Posts
That will be due to the lack of phase data I think, causing the measurement to be treated a little differently. I will take a look at it tonight. If you convert to log by Apply Windows with convert to log allowed you will find the results are much closer.
 
61 - 80 of 116 Posts
Top