Measurement process seemed similar to me probably because I have spent way too much time with REW over the years and am very familiar with it. It’s true that I had already taken successful measurements with Impala this morning within 10 minutes of installing it whereas with REW there was a long learning curve.
I checked the REW manual you have shared. You are aligning impulse peaks rather than IR starts which is exactly how I do it in REW. But I was more interested in the transfer function applied final average measurement produced in FF to which you base your correction
I agree that a working demo would be hackable, would have already produced the wav files needed and eventually would ruin the business and I also believe the price is quite reasonable for something that really works. May be, you should consider convolving a demo song with FF filters for users based on their measurements and send them to hear the potential results.
Btw, @Markus_Hubner has posted this info under this guide discussion which has been replaced with my new method some time ago which is here:
In this method, I am inverting the response over a Harman curve and generating minimum phase FIR filters all in REW. It’s quite efficient and people are consistently getting good results with it so far but there’s space for improvement in group delay and phase alignment in the bass area. My attempts to further develop it by time reversing excess phase below 500Hz have consistently failed with massive pre-echo effects. So in fact, just to learn how you guys do it, I would be willing to cough $250 and probably will one of these days
Great to see you @Focus_Fidelity dropping in and adding some insights.
@OCA You are definitely a REW expert and when I look back at my first approaches with REW I felt lost and it was far too time consuming.
As you mentioned, 10 minutes with Impala and Focus Fidelity Designer and I already achieved a performance I never had with any other digital room correction software. I have far too little knowledge to know in detail why and what FFD is doing to achieve this that easily but at the end it’s just what performs for users anyway.
Very much looking forward to an exciting exchange here. I for example also learned about the „clock drift correction“ problem and its impact on measurements as well as derived filters.
So eager to see things evolving.
There are limits on what can be done with excess phase inversion. For example, excess phase zeros with low damping or position dependence cannot be safely inverted.
Please note that the license for the filter designer (Focus Fidelity - Digital Room Correction) does not allow for “reverse engineering” the software.
Thanks for the info on excess phase inversion. I wasn’t intending to reverse engineer the software, I am afraid I wouldn’t be able to even if it were possible I was hoping to be able to compare the convolved response with the original to see what happens in the phase domain though.
If you take measurements with REW at multiple positions in your room (across your sofa, for example) and then compare the excess group delay plots. You’ll get some idea of the variation and the difficulty with arriving at group delay correction that is robust across a listening area.
You should see group delay correction that provides good improvements to the shape of the step response. Step response tends to reveal more than impulse response plots. A couple of examples in images here Focus Fidelity - Digital Room Correction
Group delay correction is an area where Focus Fidelity will be doing further work, the filter generation algorithm is subject to improvements from time to time.
Hi all, quite new to this matter, I’m trying to follow this great guide to do some room correction.
My room it’s acoustically a real mess, anyway I want to experiment and learn a bit
Can someone help me undertanding what, at point 3 of the quick guide,
Decrease modal peaks
means? Should I try to reduce the dips other than those I see as excess group delay (point 2 of the guide “not to equalize”) or I should rather act on those with modal filter?
This is the only point I didn’t understand as it’s not reported in the complete guide and - again - this matter is quite new for me.
Thank you for your kind help and reply.
@OCA I want to thank you for the detailed guide, with the written and the video parts I was able to follow the entire procedure and the result was excellent!
I have more than 12 dB of volume disalignement on measure that actually are not there so I had to correct less in roon speakers setup but in terms of sound I got a great improvement.
Thanks again for sharing.
Quick question for REW users - do your last settings retain when you close and reopen the program or do you have to go through the same “setup” process each time?
