Home Theater Forum and Systems banner

1 - 20 of 60 Posts

·
Registered
Joined
·
21 Posts
Discussion Starter #1 (Edited)
Feature Request: External sweep with REW

EDIT: Modified title to feature request.

Is it possible to play a sweep externally from CD or another computer when taking a REW measurement?

Reason why I want to do this is that I want to confirm the digital EQ parameters I have applied in my audio playback software have flattened out the freq response as REW has predicted.

Thanks
 

·
Banned
Joined
·
516 Posts
Simply make a full range sweep of the system with REW!

The filters determined in that section of the program have nothing to do with the measurement portion of the program!
 

·
Registered
Joined
·
21 Posts
Discussion Starter #3
Sorry I was not clear enough in my first post.

I have measured my system with REW, found some peaks in the 20-100hz region, and have then created filters to flatten them.

On my playback system (macmini running PureMusic) I then apply these filters using a VST EQ plugin, and all music played back on the macmini through the PureMusic playback software has the EQ applied and the 20-100hz region sounds alot better. The modes seem to have been tamed.

But I now want to confirm - to see the frequency response of my system, with the digital EQ applied. Running a sweep in REW would not include the digital EQ so I would just get the same result. What I would like to do is play a recorded sweep (WAV file) through PureMusic (with the EQ VST active) and then have REW simply record and show my the data. I.E. REW would 'wait' for the sweep to come and simply process and display the measurement data. In theory this should now look the same as the 'predicted' result from within REW when making the filters.

Hope that makes sense!
 

·
Registered
Joined
·
21 Posts
Discussion Starter #5
One way you could do it would be to get a broadband pink noise signal into PureMusic, then use REW’s RTA feature.

Regards,
Wayne
Thanks Wayne, I have done this but really I would like to take sweeps so I can do detailed analysis.

I guess it's not possible? Maybe I should submit a new feature request....
 

·
Registered
Joined
·
4 Posts
I'm curious about this as well. My problem is measuring my system in 5.1 configuration. The fronts and sub is ok, but I'm having trouble measuring the center and surround channels, one by one.

When using the multichannel inputs on the AVR (Yamaha RXV767) all eq and other DSP-curciuts are bypassed. With an external sweep source, one could use a multichannel calibration disc.

EDIT: seems JohnM managed to write a reply before me. Looking forward to the new functionality :)
 

·
Premium Member
Joined
·
14 Posts
This is on my list of features to implement, near the top.
Fabulous!!!

John, are you able to advise on the below?

In the meantime, Im doing tests to determine if running live REW sweeps through JRiver MC loopback mode will yield valid results. My purposed workflow is:

1) Take measurements in REW in standard fashion to obtain "Before" results
2) Create adjustment filters in REW or otherwise
3) Load adjustments into JRiver MC 64 bit convolution engine

http://yabb.jriver.com/interact/index.php?topic=68828.0

4) Run new live sweeps in REW patched through JRiver loopback feature (with convolution applied) to measure "After" results

http://yabb.jriver.com/interact/index.php?topic=70242.0

5) Compare "before and "after" results

My question is, whether JRiver loopback will add time domain issues that will skew "after" results. If so can this be mitigated by doing either or both of:

A) Doing "soundcard" calibration running through JRiver MC loopback to offset any influence, and/or:
B) Doing BOTH the "before" AND "after" measurements through JRiver MC loopback, so no new variable is introduced between before/after?


Also John, could you advise what measurement/graph in REW I'd use to determine if loopback is altering the signal time domain or otherwise?, In other words, if I were to compare two identical measurements, one straight and one in loopback, what REW test would I do to determine if they differ, and by how much??

Thanks!!
 

·
Premium Member
Joined
·
14 Posts
What do you mean by "time domain issues", delay? Any delay difference would not affect the SPL response.
By "time domain" (vs frequency domain) issues I mean delay, phase change, etc. All the things that show up in the waterfall, excess group delay, etc. I am not an expert in these things, but I understand the world of audio measurements generally includes two components, frequency domain and time domain, possibly you use different labels (or Im just very confused).

My concern about using JRMC loopback mode when measuring convolution-corrected sweeps in REW (reflecting the corrections), is that the measurements may be out of phase and or show delay as compared to the pre-corrected REW measurements (which do not go through loopback, but instead go to the measurement soundcard).

Am I making myself clear, or not? Does any of this matter? Does the REW mic loopback solve all of these time-related issues?

The context of all of this, is trying to create and compare X/Os and then room correction in REW/Audiolense/Acourate. I want to use REW for baseline measurements of "before and "after" corrections. Because REW requires "live sweeps" vs external (as of today), I intend to play your measurement sweeps through JRMC convolution by means of its loopback feature. This way I can compare XO's/DRC convolution filters produced in REW, Audiolense, Acourate, etc, with each other and against the uncorrected signal. I will be able to see more than FR and can also look at excess group delay, waterfall, phase, etc.

Should I be thinking about this some other way?

Thanks
 

·
REW Author
Joined
·
6,711 Posts
Does any of this matter?
No. There are lots of time delays when making measurements, from buffering and processing delays in various pieces of equipment to the time-of-flight delay from speaker to mic. Use the "Estimate IR Delay" feature in REW to remove the delays from each measurement.
 

·
Registered
Joined
·
21 Posts
Discussion Starter #18
Perhaps a quick and easy way to implement this would be to allow a set of parametric EQ filters to be loaded (or manually entered) in the 'Measurement' dialogue window which would then be applied on the sweep? If the resulting measurement is then flat it would confirm proper setup of EQ filters...
 

·
Registered
Joined
·
70 Posts
Another quick way of getting this to work, at least for me, would be if I could save the sweep as a wav to play back through JRiver. REW could then be "fooled" into thinking it is playing the sweep...
 

·
Super Moderator
Joined
·
9,204 Posts
1 - 20 of 60 Posts
Top