AFMG Network Forum
AFMG Network Forum Index -> AFMG SysTune -> Peak to delay troubles
Post new topic  Reply to topic View previous topic :: View next topic 
Peak to delay troubles
PostPosted: Mon Aug 11, 2008 3:17 am Reply with quote
gluis
Member
 
Joined: 03 Aug 2008
Posts: 11
Location: Caracas, Venezuela




I had been experimenting with your program for some days, and there is something giving me trouble. It seems that the delay finder gives me exactly double the delay I need. It works all right, but in the offset window it says double of what is needed. I checked it with other programs (Smaart 5.4 and 6, SAT live). THe thing is that it aligns things correctly but it states double the amount needed. That would be a mess if you are relying on it to align different subsystems, not only to do TF.

One more thing; I would lie to see a function where we coud get a simple fixed points per octave TF so we could have the same resolution all over the spectrum, like some other platforms do. I think the program is terrific, but I believe this would be a welcomed addition.

Best regards,

Guillermo Sanchez
View user's profile Send private message Send e-mail Visit poster's website MSN Messenger
PostPosted: Wed Aug 13, 2008 11:27 am Reply with quote
Waldemar
Member
 
Joined: 05 Dec 2005
Posts: 112




Guillermo,
if you calculate an Impulse Response from one Mic relative to the "Out" Signal,
then you will get the Arrival Time as a sum of the "acoustical" and the "electronical"
delays.
To measure and compensate the "electronical" delay, please make a sortcut between the Output and the Input of your soundcard. The "Arrival Time" of that Impulse Response is the electronical delay. This time you have to compensate manually by typing it into the [Delay Offset] field.

For minimizing this electronical delay it is recommended to use an ASIO driver.
Some devices (like the EASERA Gateway) will show no internal delay!

On the other hand, if you make a shortcut of the second output to the second input,
you can use this 2nd input as reference channel. The delay shown then will be exactly the "acoustical" delay.

No, you are right. The magnitude of the transfer function cannot be shown as bar graph. You should select the smoothing to 1/1 oct. If you move the cursor to an octave frequency, you can read the value between the two display-windows.

_________________
Best Regards
SDA, Waldemar Richert
View user's profile Send private message
PostPosted: Wed Aug 13, 2008 3:54 pm Reply with quote
gluis
Member
 
Joined: 03 Aug 2008
Posts: 11
Location: Caracas, Venezuela




I'm shortcutting the second output to the second input and using this a the reference channel. I always do that in order to cancel the electronic delay of the sound card out of the equation, as both channels, the measured and the reference would have the same electronic delay. As I told you, I used the very exact rig with different programs and Systune always gave me exactly double the delay time of the others, and all the others gave the same exact (and correct) delay time. Besides, it was obvious as I put the microphone at 3 meters from the speaker and systune gave me 6 meters and change if you put it in the proper units (in ms it was exactly the double of the other ones, it wasn't a problem with the measuring units) it showed the peak at 0ms on the impulse response after the align to peak function and it worked ok though.

About the second issue. what I'm refering to is the ability of Smaart, SAT Live, SIM II, etc to have make a measure the tranfer function using a different FFT size per each octave, in order to have the same resolution per octave (for example the Smaart 5 uses 24 points per octave, giving you the same resolution in each one of them) It does not show as a bar graph, and I don't want it to show as a bargraph. I want to have the same resolution on each octave, without resorting to smoothing. It is common practice in most of the Dual Channel FFT on the market.

I know each designer (or team of them) have his vision about what he wants his program to do, but I think this is an important issue that you might like to look into it.

I have been making dual channel FFT measurements for a while and it is a useful tool. In fact most of the measurements I take I use Fixed point per octave method. SIM guys pay top dollars for they rigs andhave only that option btw.


Just in case it helps, I'm using a MAC Book Pro with dual 2.33Ghz processor, 4 Gb of RAM running windows XP in Boot camp, a M Audio Fast Track Pro Sound card, and B&K 4006 microphones. I was making measurements in a controlled enviroment (my living room) of a Yamaha MSP 5 loudspeaker, and comparing the results of the 4 programs I mentioned to see their ease of use, accuracy, advantages and disadvantages as I'm in the process of choosing other platforms besides the one I had been using. In fact, I did the same at a gig yesterday, but I will post about it on another thread.

Best regards,

Guillermo Sanchez
View user's profile Send private message Send e-mail Visit poster's website MSN Messenger
PostPosted: Sun Mar 08, 2009 1:11 am Reply with quote
Woodstock
Member
 
Joined: 30 Jun 2006
Posts: 4




Hi Guillermo,
you are right, I have the same problems with the delay times. Maybe the result of a delay time between the Main system and a delay line is correct, but it makes me nervous when my microphone is about 3m far from the speaker and I'm expecting about 10ms an Systune is showing me 20ms.

I agrree with you to put the FPPO on the wish list.

One more thing, maybe I do something wrong.
When I capture a trace and the coherence incl. masking is activated, the trace is stored without the coherence information.
So the whole trace is visible again and the coherence trace is lost. I would like to see this information by recalling the stored traces also!
Woodstock
View user's profile Send private message
PostPosted: Sun Mar 08, 2009 2:16 am Reply with quote
gluis
Member
 
Joined: 03 Aug 2008
Posts: 11
Location: Caracas, Venezuela




Hello Woodstock,

If you are doing it wrong, I am too, because the same thing happens to me. For example, I wanted to store a coherence masked curve and compare it to a non masked one but it is not possible. The mask is applied/unapplied equally to stored and live curves, so you can't compare both. That's a bummer. In fact, I'm quite surprised at the difference coherence masking introduce to the curve. Which one is the closest to reality? how can we know? I did the tests fairly close to a studio speaker place on the edge of a table to avoid as much reflections as possible, with a B&K 4006, so it surprised me to see that much change, especially in the high frequencies. For now, I play with Systune, but make my eq decisions on other measuring platforms that had been proven accurate and consistent time after time. I hope someone clarifies/rectifies the situation but until that time, I put my trust in other products, even having purchased Systune.
View user's profile Send private message Send e-mail Visit poster's website MSN Messenger
Re: Peak to delay troubles
PostPosted: Tue Oct 06, 2009 6:16 am Reply with quote
stevenliddle
Member
 
Joined: 24 May 2008
Posts: 27
Location: Sweden




Quote:
One more thing; I would lie to see a function where we coud get a simple fixed points per octave TF so we could have the same resolution all over the spectrum, like some other platforms do. I think the program is terrific, but I believe this would be a welcomed addition.

Best regards,

Guillermo Sanchez


A lot of SMAART users ask for this as they are accustomed to it. One of the main benefits of Systune is the time windowing options that are available. Take look at the manual for details about this, especially the Time Frequency Constant window. It is a more advanced function than FPPO and is not limited to 24 PPO. The resolution of the window changes constantly with frequency. Once you start using this with the correct window length (use the impulse response or even better, the ETC to determine where the first reflection is and place the window cursor before it in time) you will start to see transfer functions with less ripple as the reflections are windowed out rather effectively.
View user's profile Send private message Visit poster's website
PostPosted: Wed Nov 04, 2009 2:30 pm Reply with quote
stofferl2003
Member
 
Joined: 12 Feb 2008
Posts: 5
Location: Vienna, Austria




Hi all,

i recently upgraded from Systune V 1.0.4 to V 1.1.6 and I got once more some trouble with the delay finder.
As long, as I use the Output as reference, everthing is fine and the results are correct. But in the normal Dual FFT situation, where for example Ch1 of the interface provides the reference signal (coming from console) and Ch2 provides the measurment signal (mic), the result are something like the expected plus another 36ms. Even if i just feed the reference signal via a Y-split cable to both channels, it shows me 36ms offset (where it should be 0ms). I can follow this in the IR-window as well.
I rechecked this with different interfaces (Motu828, internal soundcard of laptop, Alesis Multimix).
If I just watch the relativ results between two IR-measurement with different distances of the mic to the speaker, this is fine. But the absolut time value ist just wrong.
Setup:
Dell Precision M65 (Dual Core T2400@1,83GHz / 2GB Ram)

Is there anybody with a similar problem out there? Or am I just doing something completly wrong?

btw: smaart 5.4 is showing correct results within the same setups

Cheers Christoph
View user's profile Send private message
PostPosted: Wed Nov 04, 2009 5:58 pm Reply with quote
Bruce
Member
 
Joined: 19 Apr 2005
Posts: 459
Location: Minneapolis, MN, USA




Christoph,

Go to the Start menu and select EASERA SysTune (use default settings) under AFMG | EASERA SysTune to reset the SysTune configuration file to the defaults. Then check to see what the latency shows.

_________________
Best Regards,
Bruce C. Olson
View user's profile Send private message Send e-mail Visit poster's website
PostPosted: Wed Nov 04, 2009 8:03 pm Reply with quote
stofferl2003
Member
 
Joined: 12 Feb 2008
Posts: 5
Location: Vienna, Austria




Hi Bruce,
thanks for your reply. As you probably supposed, problems are gone when i start systune with the default settings.
But does your crystal ball Wink give you any information, what´s happening in my "not default settings"?

Christoph
View user's profile Send private message
PostPosted: Wed Nov 04, 2009 10:39 pm Reply with quote
Bruce
Member
 
Joined: 19 Apr 2005
Posts: 459
Location: Minneapolis, MN, USA




Just the change from 1.0 to 1.1. We try to guess correct modifications to the config file, but with all the different sound cards out there it can be a bit difficult. What I am saying is this should be a one-time occurrence going from v1.0 to v1.1. You should now not see the problem any more.

_________________
Best Regards,
Bruce C. Olson
View user's profile Send private message Send e-mail Visit poster's website
PostPosted: Thu Nov 05, 2009 8:42 am Reply with quote
stofferl2003
Member
 
Joined: 12 Feb 2008
Posts: 5
Location: Vienna, Austria




...it´s good to know. that there is a quick support, when you´re in trouble.
thank you Bruce
View user's profile Send private message
AFMG Network Forum Index -> AFMG SysTune
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
All times are GMT  
Page 1 of 1  

  
  
 Post new topic  Reply to topic  


Powered by phpBB © 2001-2003 phpBB Group
Theme created by Vjacheslav Trushkin
Variation by CodeWeavers and AFMG