Slide the right curtain down to 3-400Hz, maybe try a less aggressive room curve in the bass and cut off the peaks. Have a listen, measure a sweep and MMM in REW to correlate results. Repeat until you're happy with the sound. One channel looks louder than the other 80-1000Hz, maybe move speakers for better balance -IF it's not just Dirac doing weird stuff. Always double check Dirac with REW.
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Would you recommend upgrading from the miniDSP 2x4 HD to the DDRC-24?
- Thread starter musica
- Start date
Keith_W
Master Contributor
Just checking - do you think there is a problem when you look at the graph?
- Thread Starter
- #23
The problem is that I don’t know how valid my measurements are, because my speakers are one meter apart and the listening point is one meter from the speakers, so when I take the measurements the distances are very short.
TurtlePaul
Addicted to Fun and Learning
Your target levels look all wrong. The 0dB level you are targeting below 1 khz is about 7 dB higher than the actual trend at in that range, causing a stair step down above the correction range. +8.3 dB of bass boost is probably too much bass boost. There is an 18 dB change between 50 hz and 2 khz, which is just way too much.
To me it is pretty obvious that the target should be closer to the green line below.
Last edited:
Keith_W
Master Contributor
That is a valid concern. If you have a very short listening distance, the possibility of parallax errors due to smaller movements is greater than if you were sitting from further away. Fortunately, that is easily checked by taking sweeps at several points in space - I suggest +/-20cm since that's how much your head might move during listening.
I agree with what @TurtlePaul said, there is an obvious "step" in your measurement. It is very easy to see, and I was wondering why you needed to ask about it. Would it help if we showed you some ideal FR measurements so that you can understand how to look for it?
- Thread Starter
- #26
Yes, please show me some ideal FR curves, considering that I listen in nearfield and my system is 2.1, thank you.
Keith_W
Master Contributor
Sure. Take a look at the first post in this thread for a collection of target curves. Your actual result will look quite different from that. What you need to do is train your eye to spot the deviation.
When you take your verification measurement with REW, apply 1/6 or 1/12 smoothing, then load the target curve. In the "All SPL" tab, display the left and right speaker along with the target curve. Or you could compare a mono measurement (left and right speakers measured together) with the target curve. Make sure you adjust the vertical scale to display 20-50dB (depending on how closely you want to look). You will see something like this:
This is not REW, but it should work as an illustration. Red/green = left/right speaker. Blue = target curve. You can easily spot deviations from the target curve.
When you take your verification measurement with REW, apply 1/6 or 1/12 smoothing, then load the target curve. In the "All SPL" tab, display the left and right speaker along with the target curve. Or you could compare a mono measurement (left and right speakers measured together) with the target curve. Make sure you adjust the vertical scale to display 20-50dB (depending on how closely you want to look). You will see something like this:
This is not REW, but it should work as an illustration. Red/green = left/right speaker. Blue = target curve. You can easily spot deviations from the target curve.
- Thread Starter
- #29
But is it ideal for the highs to roll off like this, isn’t it too much?
- Thread Starter
- #30
In fact, if I correct it this way it doesn’t sound good, the upper mids disappear. To hear them I have to move the right curtain up to 20 kHz, so Dirac boosts the upper mids by +8 to +10 dB. I’m not sure if I’m doing the right thing?
PeterNL
Active Member
Your Dirac measurements look very wrong and I don't know what actually happened :/
Did you use correct calibration file ? In your case it should be file without "90deg" in the filename.
Please post here again screenshot from Dirac but with automatic settings (curtains, curve...created by Dirac). Everything looks very off (1kHz point should be on the 0dB line).
I have checked for you on my Dirac Live (PC), here are my measurements and correction, also moved right curtain to around 1kHz only to see how FR is responding:
Before correction:
After correction:
Right curtain at 1kHz:
Speakers Kali Audio LP6 V2 (NO sub), distance between speakers: 85cm, distance head from speakers: 80-90cm.
Last edited:
TurtlePaul
Addicted to Fun and Learning
Can you post the graph of when you tried to fix the Dirac?
You should try the other way- reducing the DIRAC window to below 150 hz, rather than increasing it to 20 khz with 8+ dB boost across the range.
You should try the other way- reducing the DIRAC window to below 150 hz, rather than increasing it to 20 khz with 8+ dB boost across the range.
PeterNL
Active Member
So what can you choose in "select arrangement" ? How many points etc. ?
- Thread Starter
- #35
Focused sweet spot
For a single listener in a dedicated
seating position.
9 points, a parallelepiped with the head inside
I know OP has made the Dirac jump, but i wanted to point out that auto eq with REW integration is available in the minidsp 2X4hd console for a license of $30. I made the investment but my ears report no preference for the corrections i have tried. I need to revisit.
I would like to see a comparison between the two, does dirac generate a parametric equalization that is comparable to REW?
Or is the functionality or use case different?
Is dirac setup a more automated process? I found some areas in REW that had me hunting the documentation.
I would like to see a comparison between the two, does dirac generate a parametric equalization that is comparable to REW?
Or is the functionality or use case different?
Is dirac setup a more automated process? I found some areas in REW that had me hunting the documentation.
PeterNL
Active Member
Ok, should be good too. But take one thing in consideration that in Studio (my pictures) person on the chair is facing towards the speakers and your are standing behind him, but at home is mirrored RED is Right, GREEN is Left and points are virtual CUBE where front dots are bigger than rear dots but dots above head are on the same height, same for lower dots. Yes I know, sounds stupid obvious but also I made mistake once
By the way, what about mic calibration file ?
- Thread Starter
- #38
The microphone calibration file is not the 90.txt, it’s the one that points to the speakers. Could you please tell me how you took the measurements and the distances between the points in such a limited space? Thanks.Ok, should be good too. But take one thing in consideration that in Studio (my pictures) person on the chair is facing towards the speakers and your are standing behind him, but at home is mirrored RED is Right, GREEN is Left and points are virtual CUBE where front dots are bigger than rear dots but dots above head are on the same height, same for lower dots. Yes I know, sounds stupid obvious but also I made mistake once
By the way, what about mic calibration file ?
View attachment 476921
PeterNL
Active Member
It depends how far your head is moving while you listen to music. Measuring central point is most important and mic must be centered very precisely and at ears level (still Dirac recommend mic pointed upward and using stable mic boom or good tripod), after center point follow Dirac instructions and measure other points, try to make (cube, box with 20-30cm walls) or imagine cube and move mic 10cm left forward and upward from center point etc. I would try 10 or even 20cm from center point, vertical and horizontal.
I don't really understand you saying that pointing mic upward is more difficult in your limited space, could you explain ? For me it doesn't really matter if I put mic to the stand pointing speakers or pointing up, it is even easier to point it up because of easy and only one angle (just pointing ceiling) instead of pointing at the speakers (but pointing where exactly, left, right, between...).
- Thread Starter
- #40
Similar threads
