Understanding the State of the Art of Digital Room Correction

The speakers were repositioned after the biamp, so the measurements don’t compare apples to apples.  I did near field measurements before and after the biamp.  They were very similar as I recall, but I did not save them.

You can surely try using REW to optimize a simple parametric EQ.  It’s better than nothing and better than trying to do this without the REW optimization.  I did this.  It’s ok, but not fantastic.   REW can also design a convolution filter.  I also tried this and it’s a little better than the parametric.  Audiolense is a whole other level of sophistication.  The Bob Katz article is 8 years old.  
 

I have absolutely no relationship with Mitch, other than that he helped me twice to optimize my system.  His fees are less than most of the silly cables audiophiles buy.  All I can say is give it a try.

The last time I chatted with Bob he was using Acourate for his Studio A. He also gave Uli a well deserved testimonial. I demo Acourate in my video here. So, not sure what you are referring to @Ozan Bolat?

The Ozan character has been banned from AS three times now. He was famously the most abrasive person ever when discussing his whacked out ideas about room correction. Many people will recognize his cucumber username previously. 

It has been a very rainy Sunday morning in the UK, plus we have an extra hour thanks to the clocks going back. Thanks to this I was able to spend only 50 minutes of my time listening to the entire 1 hour 50 minute video without feeling guilty about a job I need to do in the garden.

Anyway, I understood much of this, but was a bit lost in other areas. This is fine, if nothing else it gives me a firm idea as to what areas I for which I need to do a bit of background reading.

I write this as someone who has recently obtained an excellent improvement to my system using Focus Fidelity Filter Designer. One takeaway I have got from the video is that both Accurate and Audiolense, neither of which I have ever actually used, seem to be both far more complicated to use than Focus Fidelity, but at the same time they appear to be rather more powerful and offer far more scope for fine tuning or optimisation. If I am completely honest, during the demonstration of Accurate Pro, I did find myself thinking this looks too hard, or at least thinking that I currently do not have enough free time to become sufficiently familiar with this to be able to obtain satisfactory results. How much ultimate performance I am missing out of by using the simpler Focus Fidelity I do not know, but at least I am getting great results now, rather than struggling to find time with software that I do not yet understand.

As I mentioned above, I did get lost in a few areas of the video, yet there were also a number of areas where I gained some excellent insight and understanding. The video well worth watching and many thanks to @mitchcofor taking the time to pull this together.

I was also fascinated by the comment at the end regarding USB microphones versus analogue with a mic preamp. To date, pretty much every recommendation I have seen regarding room measurements has started with "get a Umik-1 microphone", but if getting something else could could yield better results, this is something I could easily do. I am a little puzzled by what is the at the route of the issue though, as an analogue mic would still need A/D conversion with a different clock to the DAC, so I am struggling a little to understand the fundamental issue with a USB mic. (I'll give this some thought)

Thanks again Mitch, great stuff!

5 hours ago, Confused said:

It has been a very rainy Sunday morning in the UK, plus we have an extra hour thanks to the clocks going back. Thanks to this I was able to spend only 50 minutes of my time listening to the entire 1 hour 50 minute video without feeling guilty about a job I need to do in the garden.

 

Anyway, I understood much of this, but was a bit lost in other areas. This is fine, if nothing else it gives me a firm idea as to what areas I for which I need to do a bit of background reading.

 

I write this as someone who has recently obtained an excellent improvement to my system using Focus Fidelity Filter Designer. One takeaway I have got from the video is that both Accurate and Audiolense, neither of which I have ever actually used, seem to be both far more complicated to use than Focus Fidelity, but at the same time they appear to be rather more powerful and offer far more scope for fine tuning or optimisation. If I am completely honest, during the demonstration of Accurate Pro, I did find myself thinking this looks too hard, or at least thinking that I currently do not have enough free time to become sufficiently familiar with this to be able to obtain satisfactory results. How much ultimate performance I am missing out of by using the simpler Focus Fidelity I do not know, but at least I am getting great results now, rather than struggling to find time with software that I do not yet understand.

 

As I mentioned above, I did get lost in a few areas of the video, yet there were also a number of areas where I gained some excellent insight and understanding. The video well worth watching and many thanks to @mitchcofor taking the time to pull this together.

 

I was also fascinated by the comment at the end regarding USB microphones versus analogue with a mic preamp. To date, pretty much every recommendation I have seen regarding room measurements has started with "get a Umik-1 microphone", but if getting something else could could yield better results, this is something I could easily do. I am a little puzzled by what is the at the route of the issue though, as an analogue mic would still need A/D conversion with a different clock to the DAC, so I am struggling a little to understand the fundamental issue with a USB mic. (I'll give this some thought)

 

Thanks again Mitch, great stuff!

I've used an expensive pro measuring mic and a UMIK; don't think in this application there's much of a difference in actuality. In Audiolense the program tells you if your measurement is a good one. I think some of the other programs do this, too. @mitchco can correct me if this is incorrect.

You do need to have a calibration file for the mic to get a good measurement,  and for some of the higher end mics you need a mic pre-amp. 

Hi Juergen,

Thank you so much for your comment! I really appreciate it.

Kind regards,

Mitch

A question to @mitchco

There were a number of references to "further reading" in the video, with links in the actual presentation.

Would it be possible to "cut and paste" these links into this thread? 

If so, this would be very both very useful and very much appreciated.

I'd share my Audiolense experience but it's part of the video.  😉

6 hours ago, mitchco said:

Hi @ConfusedI have uploaded the presentation with links in the YT video description.

Ah, I had missed that. Perfect, many thanks.

Hi @Confused you did not miss it, I forgot to put it up which I did yesterday. Thanks for the reminder :-)

Kind regards,

Mitch

Mitch, That is certainly a very helpful presentation and I made some real progress in my understanding. I am curious to what degree are fir filters developed in one system effectively functional in the DSP processing engines used in various playback programs. In my case, I use Moode as my digital music player which now has incorporated CamillaDSP, and I would love to give it a try. I have a lot of studying to do, probably starting with your book, but does it make sense to use one package to analyze impulse responses and develop filters and then use those filters with processing engines from different developers.

Thanks,  Skip

Cheers @Skip Pack Convolution is a mathematical operation. So if the math is implemented correctly in software, then the filters should measure and sound identical regardless of convolver and OS platform.

Kind regards,

Mitch

I got the electronic version of your book and screamed through it (I'm sure a second pass is necessary). On the heels of the video presentation, the book was a nice boost. I guess the heart of my question regards the ease and generality of getting a filter generated in one system to work in systematically unrelated convolvers. Are formats fairly standard, or do I need enough understanding to bridge mismatches in filter file expressions. I'm looking to do stereo DRC only, at least for now, not crossover work.

Thanks once more!

@Skip PackCheers! Yes, there is a standard convolver .cfg file format described here: http://convolver.sourceforge.net/config.html For most stereo convolutions, just loading the impulse file is enough as described here: http://convolver.sourceforge.net/usage.html#an_impulse_response_(IR)_file

Virtually every convolver I have worked with follows the .cfg format or can load the IR.wav file if it is simple stereo. One or two convolvers prefer a mono IR.wav file (e.g. HQPlayer) but for the most part, just load the IR file. Quite a few convolvers will accept the stereo IR.wav files at different sample rates in a .zip file. The idea is that the convolver will automatically load the correct sample rate filter based on the source sample rate. This can also be done with the .cfg file. The zip of IR stereo.wav files is a convenience.

Virtually all DRC FIR filter designer software will output the correct IR.wav file with .cfg if required whether stereo or MCH. Some will also output a proprietary format, but it is usually for their companion convolver. Both Acourate and Audiolense can output the formats mentioned above, plus their own proprietary format made to be consumed by their respective convolvers.

Dirac is the only proprietary solution that is a closed source software solution. Meaning the Dirac filter designer only works with their convolver and does not accept other convolution files.

Good luck!