Right next to the title, there is a link for more about this video (you will have to open the video in the YT app or on their own site). There is a zip file there, but this doesn’t contain a pdf file.
1 Like
Thank you Serkan.
Dear Serkan! Yesterday I made a room correction in accordance with your guide. There was a loss of dynamics and punch, especially the sub-bass was missing. I did one measurement with the umik 1 microphone in the center in the LP. What am I doing wrong? Unfortunately, I can only take measurements in the evening when I’m at home.
Most likely you didn’t level match signal levels for your comparison to yield a valid conclusion, no?
1 Like
Of course, I compared the signal level, I understand that it is necessary to compare at the same volume.
1 Like
Maybe you are just accustomed to too much bass in the sound and when you apply room correction, the bass level sounds too low suddenly. Can you show us the measurements before and after (or at least REW prediction of after)?
2 Likes
Yes, of course, I will upload files with measurements and corrections in the evening.
1 Like
Dear Serkan, colleagues, good evening!
I send files of my measurements. Please tell me what am I doing wrong. I will be very grateful for your help.
@OCA Dear Serkan and community members. Can someone advise me? Above, I posted my measurements, but unfortunately I did not receive any feedback from you. I really hope for your help, thanks in advance to all who responded.
Your speakers has opposite polarities with each other cancelling each other out especially between 60-100Hz, you need to reverse one of them. In case these are combined responses with the subs active then it might be one or both of the subs. You have to time align each sub with the speaker first.
2 Likes
Dear Serkan, thanks for the recommendation. It looks like my measurements are not correct. I will take measurements again at the listening point. Perhaps I set something wrong in the REW parameters when measuring my acoustic systems. When the time comes, I’ll post my new measurements here.
@OCA Dear Serkan! I took new measurements in REW, can you see now they are correctly made? And if you have the opportunity, please comment on my RT60. And what should I do with room mode at 42 hertz? How to defeat her?
And here are my room settings:
https://drive.google.com/file/d/1_8Mbc_8i3jfnB1LuSMDBa37lAABJTzzN/view?usp=sharing
Convolution1 is a clean, fast and precise stereo filter for Roon. Your left and right speakers have opposite phase responses between 70-90Hz and cancel each other’s bass response when played together although individually they don’t have dips at that region. This is probably the lack of bass you were referring to. This should be corrected by better placement of speakers, reflections from 2-2.5m away of each speaker are cancelling out each other at the LP. Do you have an odd ceiling shape or something?
Anyway, I tried to correct it a little bit with phase manipulation. This is Convolution2. This is a slower filter. It corrects the dip by about 2.5dB and this was the limit before pre-ringing. It might even be causing some ringing already.
I added all REW steps as mdat files although I doubt that you’ll be able to understand them at this stage 
FYI your speakers crossover phase filter is a 24dB/octave, polarity inverted linearization filter @ 1630Hz.
1 Like
@OCA Dear Serkan! Thank you for your help! I will study the files and definitely try them on my system. Thank you very much!
Expect to dial in around 5-6dB headroom adjustment to avoid clipping. The crossover phase correction filter alone will cause that much. You can adjust clipping with this song (guaranteed nothing else will ever clip if this headroom is adjusted for this):
1 Like
So I have previously used various methods of Room Correction - DIRAC, Magnus’s RTA MMM REW method, simple PEQ, etc. I started on this new method using @OCA 's great video describing all 4 filters.
Unfortunately, my room/measurements seem to be so inconsistent that I can find no identifiable room mode/standing wave patterns in my measurements to create a VBA filter!
So I decided to just skip the VBA filter and move on to the 2nd inverse IR filter.
But during this process, my A over B - using L0 / Target - was getting generated at around -40Db instead of centered around 0Db. And then the 1/A filter has a similarly low target. Double checked everything and couldn’t figure this out.
Then I realized that I think it’s because my initial measurements have negative values in the upper frequencies above 20k … I am really stumped how this can occur when doing Sweep measurements … any thoughts anyone???
There’s nothing wrong with your procedure. You must be using a very recent early access version. John (the author of REW) recently revised (Feb 17) trace artihmetic basics in REW. A over B would show up 110dB below the measurement instead of around 0dB but 1/A conculdes the operation as before (around 0dB) so your filter results shouldn’t be changed. Check this link:
1 Like
Ah yes, I downloaded EA version a couple of days ago. And so yes, my 1/A appears to be centered around 0dB BUT it has a target value of -47.6dB! How do I do the SPL offset to reach standardized 3dB target?? If I add 50.6 then the Filter of course jumps up to the 50dB range.
[EDIT: I see this in the change info:
- Changed: The handling of SPL offsets and IR scaling for the results of a division have been changed to allow the resulting IR to be exported and used directly without normalisation or scaling, but note that float format should be used if exporting to WAV without normalisation
- Changed: Do not use rectangular windows for inversion results due to artefacts in subsequent processing, use Tukey0.01 instead
]
1 Like
As per VBA, I can’t see any bass peaks in your response above either. Are you using heavy room treatment or is there some sort of DSP in your setup?It seems like there’s nothing to correct in the bass response.
1 Like