I spent a couple hours integrating my subwoofer (Monolith M12 v2) with my Genelec M040s using my new to me Minidsp Flex. There were 4 obvious peaks before PEQ, which required pretty big cuts, with gains of 5-9 dB, and Q 5-7. Should I allow for these big changes in adjusting my system volume downwards, and by how much? I'd appreciate any feedback on the effectiveness/thoroughness of the changes (I've attached before and after pictures of the system, and each component speaker separately. Finally, in looking at these graphs, my system looks perhaps rather too flat, is it fair to say I may have a bright room?
-
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!
Integrating subwoofer, how did I do?
- Thread starter arichman
- Start date
Not sure which is which response from the graph legend. Some simply prefer a bit of boost on the low end and a bit of a downward turn for high frequencies. Once you've got a good neutral response, experiment a bit.
- Thread Starter
- #4
Yes, the legend is not very helpful, I haven't puzzled out the graphics in REW just yet. The red line is the system response in both the "before" and the "after". The subwoofer response line drops of course after about 80 hz, again in both graphs. The remaining two lines are the left and right speaker response curves.
Keith_W
Master Contributor
- Thread Starter
- #7
There is a lot of glass on one side, but there is an insulated honeycomb blind that fills that space. Wall hangings and floor rugs, but probably a pretty reflective room on the whole, and it is a bedroom not a dedicated listening space so I don't envision anything ambitious beyond trying to shape the response curve. Regarding that, I could apply a high shelf at one end and a low shelf at the other. Or I could more simply add some gain to the subwoofer to boost that end. Is there an argument for one approach the other? At the other end, the Genelecs I have were singled out in the review on this site as not showing enough of a slope in response with increasing frequency, so the flat response curve is at least in part a property of the speakers themselves. I think they do sound bright at high volume. Is a shelf filter the right way to handle this?
- Thread Starter
- #9
A little bonus for those of you who have read this far. My more problematic system is in my living room, which is all glass down one side. I have a Flex here too (BMR Towers, another Monolith M12 v2 subwoofer), and I've never been able to get a good result out of REW, due no doubt to its complexity and my inexperience. This time I cribbed the EQ settings provided by a Wiim Pro using its RoomFit software using a calibrated microphone, and manually entered them into MiniDSP Device Console, about 5 minutes work. The before and after plot shows the system performance with and without EQ, I'll let you decide which is which. The second plot shows the individual contributions of each component, subwoofer, left and right speaker by itself, along with the system performance, all after equalization. Ignore the legend, the subwoofer is obviously the hump on the left, the system in yellow, the remaining two are the left and right speakers. Finally, I've attached all the eq settings. I didn't have much in the way of expectations given the painlessness of this process, but I at least am hugely impressed by the result. I have been listening and remain impressed...
Attachments
I would be slightly vary of implementing any of the EQ points you have set that is above the one at 201hz.
From there on up you are also affecting the direct sound you are hearing (irrespective of the reflections in the room), and also the measurements will probably not look as nice if you move the microphone around.
So an interesting experiment may be to deactive all of those EQ points that are above 201hz, and see to what extent you hear a difference, and what the difference is.
You could also try (if you didn't already) take some wider measurements to see if there are any of the dips/peaks that stay consistent when moving the microphone say 0.5-1m away from you current position in various directions, and only EQ those.
From there on up you are also affecting the direct sound you are hearing (irrespective of the reflections in the room), and also the measurements will probably not look as nice if you move the microphone around.
So an interesting experiment may be to deactive all of those EQ points that are above 201hz, and see to what extent you hear a difference, and what the difference is.
You could also try (if you didn't already) take some wider measurements to see if there are any of the dips/peaks that stay consistent when moving the microphone say 0.5-1m away from you current position in various directions, and only EQ those.
- Thread Starter
- #12
While I got what I got, I readily accept that there is less here than meets the eye, as is (perhaps) being politely suggested. This result is quite sensitive to microphone placement as I have subsequently found. Those big boosts erode essential headroom as anyone who listens to loud music may come to regret. And in any case, the result does not generalize to other systems and most importantly other rooms. While I regarded this room as somewhat challenging, I don't think so now. It is pretty big, the speakers are well away from any walls and I'm sitting pretty close. So the effect of reflections is correspondingly reduced. I have since moved this setup to a much smaller room, the corresponding results with REW are not anywhere near as perfect, but I'm enjoying it nonetheless.
Last edited:
I would argue your corresponding results are likely better. Flat graph does not necessarily mean better system result. It sounds like you understand there is nuance to correction. I always ask myself when I make an EQ change am I correcting the room, the speaker, the overlap between sub/speaker, or should I even make the change or is it a mic position artifact.
- Thread Starter
- #14
My main point in my last response was to deemphasize the overarching importance of the flat graph as you describe it, as one priority among others. But of course, that flat graph was not the product of endless tinkering but 5 minutes work. Definitely a worthwhile exercise! I was therefore also interested in understanding why this worked so well. So I have to credit whatever algorithm Wiim has worked up as effective. But the physics of the room determines what is possible, so there is clearly more to it. I speculated that the room was rather “dead” and therefore represented something of a blank slate which allowed the Wiim algorithm to succeed in its attempt to flatten things. In response, you made what I consider a stronger judgement, that the dead room which facilitated the achievement of a flat response was actually a problem. If so, at the very least it suggests an interesting irony!
Similar threads
