Feature request

[TomSleebus]

Good day,
Hope this is a good place for a feature request, if not, apologies, and please direct me to the correct channel.

I know some people creating FIR filters out of REW measurements, including myself . Could it be made an option to load e.g. a convolver FIR file containing the 'emulated room' (or anti-room if you want), so one can apply correction and see the result in a sweep?

I understand the alternatives are: for windows: equalize APO e.g., this is also a vst host where one can easily load any convolver plugin, selecting the FIR. I have not yet found the alternative on mac, except for 'virtual routing' software, which makes things rather complicated.

Appreciate your time,
Greetings from Belgium
Tom

 

[John Mulcahy]

Not sure I follow what you are after, but you can do convolution on responses in REW using the trace arithmetic A*B operation. That is the way a correction filter is most commonly tested against a response measurement.

 

[TomSleebus]

Thank you John.

My approach would be 1) measure room with REW 2) create a convolver filter based on eq corrections and phase alignment, importing/exporting with rephase, indeed using the A*B aerythmic etc and then 3) test the filter created in 2). There are a few options to do this testing:

1) you select in REW the normal output, activate Equalizer APO (windows only), which acts as VST host, selecting a convolver plugin, loading the FIR filter created with REW (and rephase). The measured response will include the 'correction' from the filter
2) Use sonarworks or equivalent, in that case the (eq) filter is created by sonarworks, selecting the virtual device of sonarworks in REW measures the room including the correction
3) use a virtual routing solution to point REW's output to e.g. a DAW (or other VST host), applying a convolver VST plugin with the filter created above, having rew recording via the microphone the 'corrected' output

OR

and this is my feature request, would it be possible with REW, when doing a sweep, to load a convolver VST plugin, loading the 'room correction filter' natively in REW so you don't need an external solution for this

Hope this is more clear
Have a nice evening
Tom

 

[John Mulcahy]

Have you found circumstances under which a measurement with a filter in place differs from doing A*B on an unfiltered measurement and the filter response?

 

[TomSleebus]

No, I didn't find what you describe yet, however I think the unfiltered measurement with an A*B filter response as result is 'just' (if I can minimize it that way) a visual representation. My goal would be to have the 'corrected' sweep coming out of the speakers so it can be measured with a mic. I'm not sure if I'm missing some trivial things (apologies if I do).

 

[John Mulcahy]

'Just' a visual representation, but nonetheless exactly the result that should be measured with the filters in place (because physics, as younger folk might say )

An example of when that would not be the case is if the filtering were only in the path to some of the drivers receiving the signal but not others, however filtering in REW would then not be appropriate either as there would be no signal to use for the unfiltered path.

The question does come up from time to time, but I have yet to see a compelling reason to add such a feature.

 

[TomSleebus]

Ok i see, it would be a 'native' replacement for CamillaDSP, JRiver, EqualizerAPO etc, the thread you mentioned should have been a clue to me, I didn't do a thorough search before posting my request, apologies for that. Thanx for the consideration.

 

[EarlK]

Ok i see, it would be a 'native' replacement for CamillaDSP, JRiver, EqualizerAPO etc, the thread you mentioned should have been a clue to me, I didn't do a thorough search before posting my request, apologies for that. Thanx for the consideration.

Not to be too inflammatory but;

What you're asking John to implement is simply another form of visually-based navel-gazing.

The whole point for creating an EQ filter is to see ( hear ) if it improves one's own listening experience.

One has to actually listen to any newly created filter ( not simply look at it ) to arrive at a reasonable conclusion ( of whether or not to permanently use the filter ).

That means playing music through the filter and auditioning the results.

EarlK

 

[TomSleebus]

Hi EarIK,
I do agree on the point where one has to hear if a filter fits. When happy, set it, forget it and start listening/making music again. 100% with you.
To my context, which you may refer to as navel-gazing (no offense taken); I help producers and studio builders with the acoustic treatment, usually providing a filter as 'the last step in the correction chain'. The main interest here would be to understand if 1) the EQ correction is providing the desired results and 2) the phase corrections are giving an aligned result, both also in the time domain. Albeit this can be predicted in REW (and please, take my compliments and congrats for the amazing product - once I generate an income of what I do, the author can expect a donation for sure), the actual result will depend on room modes and SBIR and any possible combination of those. The customers are usually reluctant to add additional software to or reconfigure their computers, not to mess up existing routing, existing filters etc. 'REWers', if I may, on the other hand are reluctant adding the multitude of drivers of sound cards for measuring other people environments. The feature could be categorized as 'nice to have' or 'convenient'.
You might be correct that such a feature is only to seek confirmation of the personal obsessive compulsive behavior about being sure everything is perfect (which is not of course), but I also see it as a deliverable of provided service, not in theory (predicted) but in practice (actual).
I did not take your comment as inflammatory or offensive, my reply being an elaboration to the question context, not as a "yes but I..." one.
Wish you a fine remainder of the day,
Sincerely,
Tom