Perfect Speaker Placement - Put next to the back wall as much as possible.

[ecwl]

I personally think that all speaker placement advice online is for the 20th century when microphones and RTA’s are not readily accessible for most people. Now that USB microphones are cheap and ubiquitous (e.g. miniDSP UMIK-1) and RoomEQ Wizard is free, I think all optimal speaker placements should be assisted by the microphone and REW.

 

That said, I still felt tremendously grateful to Jim Smith’s book Get Better Sound because it explained so many things to me about what I should be listening for when placing speakers. However, because Jim Smith has golden ears and lots of experience, he can set things up without a microphone. I fully acknowledge I can’t so I think now, I’ve developed a modified method of setting up systems.

 

1) Assuming you have the option to move your listening seat forward and backward, measure the bass frequency response of pink noise at the listening seat and see how smooth the bass is

2) Move the microphone 6” at a time forward or backwards and see if you can get a better and flatter bass frequency response. The flattest response you can get will be your new seating position.

3) If you have the option to move your speakers forward or backward (and the patience), put the microphone at the new listening seat and move the speakers forward or backward 6” at a time and see if you can even get flatter frequency response. You probably can’t or even if you can, the difference is not big. The flattest bass response is your approximate new speaker position.

4) Now sit in the new seat, and everything is going to sound weird because the midrange and treble are going to be off. Start listening to vocal music and see if the speakers are too far apart or too close together. If they are too far apart, the voice is going to sound thin. So Just keep moving the speakers 2” at a time left and right to get them further apart until the sound is too thin and then move them back together until you get a warmth in the voice

5) Now is the sound too bright or too soft in the trebles? Toe your speakers towards your ears if you think the speaker is too soft or toe them out if you think the trebles or too bright. This may affect your soundstage width again so you may have to move your speakers left and right to get the vocals to be warm but not excessively so.

 

At least that’s my take on all of this. The reality is that most of us have limitations on where we can put our speakers and listening seat. So we can really just toe in and out our speakers which is unfortunate. So at the end of the day, we end up needing DSP (parametric EQ/convolution filter) to fix the bass issues. I personally think it’s very difficult to use bass traps to properly address bass issues (even though it is sometimes possible but not always). But first reflection points should have something to diffuse/absorb the sound so that images don’t smear. If you have a TV in-between the speakers, you should have a blanket to cover the TV.

 

But at the end of the day, I still think everything starts with a USB microphone. Unless people think they live in the 20th century

[ecwl]

@semente This is both fascinating and expected in many ways.

First, you pulled your speakers further away from the wall but you actually get more low bass in the 20-35Hz region. Of course, this is because you actually moved your listening seat first so you can take advantage of the natural room peaks and troughs in the bass in your room. That’s why the 3m listening position has the smoothest bass response and deeper bass.

Second, moving the speakers didn’t change the bass frequency response as dramatically as moving the seat which is always the case. However, I do find soundstage depth is often better when speakers are further from walls even though that effect is not always measurable or translates clearly to me in measurements.

Third is the more complicated part. For some reason, by moving the seat to 3m and the speakers at 0.75m, your left speaker now has a very odd frequency response from 200-800Hz. My guess is some sort of interaction between that left front window space and the left speaker. My suspicion is that while the bass is better, the midrange would start sounding funny. And my guess is that your solution would either be to move the speakers back to 0.6m (or even 0.65m) or you have to draw the curtains (potentially thicker curtains) on that left front window. In fact, you may have to do both. Hard to know. Moreover, because you would in the end have increased the distance between your speakers and your listening seat (compared to where you had it at the beginning), you will probably have to toe out your speakers slightly to compensate.

I suspect you have already found your optimal seating position and you are extremely close to the optimal speaker position that is achievable. To get even more out of your system would probably require parametric EQ or more likely a sophisticated convolution filter.

[ecwl]

8 minutes ago, MarkusBarkus said:

Choosing not to measure is a bit of a red pill/blue pill thing. It sounds good, but what if I measure and the results say it sounds bad? 😉 Funny. 

On one hand, definitely, if you think your system sounds good, don’t measure. Because you may simply not want to know what issues remain as some of these issues may not even be fixable.

 

On the other hand, I’ve seen so many people place bass traps all over their room that’s actually doing nothing or having a detrimental effect on the sound without realizing it. Because they’re treating the room blindly. And I would argue in many instances, bass traps are less effective than we think.

 

I didn’t start out wanting to measure. I just want to make sure my subwoofers properly integrates with my speakers and I found that it’s impossible to do it by ear precisely since the subs are coming in at 30-40Hz. But yeah, I noticed a lot of unpleasant bass peaks and troughs that I didn’t even realize were there. It took me a long time to acknowledge that’s totally normal in almost all rooms. I’ve learnt to accept that I like the sound that I’ve tuned and continued to enjoy my music. But later, I realized I could push the system further by introducing judicious DSPs. 

 

Another thing I found with measurements is that when you first start out, you don’t actually know what’s best. Like in @semente’s case, the seat was moved from 2.6m to 2.75m and it was much later that the seat was moved to 3m. Since the bass frequency response is always going to be uneven, it’s sometimes hard to recognize 3m is more optimal than 2.6/2.75m. This is where posting the measurements on a forum can help as others who have done it before can say, what’s better or even that it’s the best that you can do. 

 

Another reason not to measure is if you can’t change anything. If you cannot move your seat and you cannot move your speakers and you cannot DSP, there is virtually no point to measure unless you’re planning to buy a new pair of speakers and you want to know how much more you’ll get out of the system. If you can’t change the frequency response, why find out what it looks like. You can’t change it. 

 

Truth is, we all listen in rooms with imperfect bass response. I think if you look at the Stereophile reviewers rooms, they all have weird bass peaks and troughs. The only room I’ve seen with extremely good measured frequency response is the Music Vault the Ultra Audio reviewer used to have.

 

To me, measuring means finding out something you don’t want to know about your system and accepting it. But it also means that there may be an opportunity to improve upon it in a more scientific manner, rather than just randomly moving things around or adding new components or bass traps. Hence, measuring is the most effective way to upgrade a stereo system in my mind (assuming you can move things around to improve things further).

[ecwl]

12 minutes ago, MarkusBarkus said:

How does treating a room and tweaking positions through a careful listening process stack up when you actually measure the results? 

 

BTW: I think measuring is probably the only way to really nail sub settings. You are quite right @ecwl

 

From my experience, measurement is generally not helpful with adjusting midrange and treble as it’s fairly easy to tune by ear. I guess theoretically if your tweeter is peaky and firing right at you, you can see that in a measurement but you can hear it so easily, you can just toe in and out the speakers or move them apart. 

 

But I think this so called careful listening process is over-rated as I’ve seen it at my primary audio dealer store so often. People often have say 5 favorite tracks they like to use to test systems. And inevitably, two of them would not be great recordings to begin with. And frequently, they would have tuned their current home system to play those 2 tracks well through an odd frequency response. Now, as a result, they’ve been listening to all their music to this uneven frequency response and over time, they just assumed that’s how their music should sound. Their ears have adapted to the odd bass peaks and troughs here and there.  So when they demo a system that is more optimally setup, they’ll say, yeah, I’m hearing so much more for the other 3 tracks but I think those 2 favorite track of mine are sounding worse. Moreover, they’ll try to convince my dealer that in those 3 tracks that sound better, it’s missing the bass peaks that they’re used to hearing. Happens ALL THE TIME.

 

In fact, it happens to me all the time too. Once I’ve setup my system, over time, I just start ignoring all its flaws and just enjoy the music. And then when I’m ready to upgrade, I start listening to other systems, I can only hear how my system is better than other systems but I struggle to hear how my system is inferior to other systems. Like right now, my bookshelf speakers clearly is not as transparent and dynamic as it can be because it’s not as anchored and decoupled from the carpet as it possibly can but I’ve been ignoring the issue for a long time.

 

But I find that inevitably, once you’ve achieved a more even bass response and adjusted your speaker toe-in/separation to re-balance the midrange and treble, over time, you’ll always get more out of all your music. It just takes a little bit of time to appreciate all the improvements.

 

As a side note, there are still really two ways to determine subwoofer integration but I still think both require a USB microphone and REW. You can do it the way I currently do which is to have the subwoofer come in where the speaker drops off. Obviously, you need the microphone to help you set the volume and crossover frequency. Or you can set the crossover frequency much higher and just determine how loud the subwoofer should be. You would still need to set the subwoofer phase where the microphone would be helpful. But more importantly, I find it’s difficult even if you do the latter to set the subwoofer volume accurately without a microphone. Pre-microphone, I think my left subwoofer was 2dB louder than my right.

 

So even with a USB microphone, there is still a lot of room for personal preferences and taste. Listening is always key in the long run.

[ecwl]

4 minutes ago, ASRMichael said:

Speaker position / room measurements calculator 

 

http://noaudiophile.com/speakercalc/

I have seen so many of these calculators and people on forums talking about them.

 

What I want to see is somebody actually using the calculators on their room and then writing down the predictions. And then using their USB microphones and posting their actual frequency response.

 

I can tell you I think in most cases, the calculator’s prediction and the measured frequency response would deviate quite significantly. But feel free to link me to posts where these responses correlate extremely well. When they do, they’re usually dedicated home theatre/stereo rooms. But even then...

[ecwl]

2 hours ago, ASRMichael said:

Off topic but I actually turn of DSP for certain types of music, like Rock. I feel there needs to be resonance in certain types of music. Rock for example needs resonance to sound great! 

 

 

I don’t know if you really like the resonance with the convolution filter off.

My gut feeling just looking at the frequency response is that your midrange is 4-6dB lower in volume than your bass without correction. Whereas the convolution filter Mitch Barrett built for you actually lowered your bass to around the same level (maybe 1-2dB higher) to your midrange. 

 

The ideal way to get better sound for your bass for you, would be to ask Mitch Barrett to build a new convolution filter with say may be a 4dB-6dB rise in the low bass and see if you like the sound better. And then you can switch between the Rock convolution filter setting and the non-Rock convolution filter setting.

 

Alternatively, if you can try to add a high shelf filter to raise the bass from 20-100Hz by 4-6dB on top of your convolution filter. The problem is that it’s slightly unpredictable how the high shelf filter would interact with the convolution filter which is why it’s always better to build the target frequency response into the convolution filter itself.

[ecwl]

@ASRMichael Haha... I dropped by my audio dealer to make me two new grounding cables for my NAS and network switch. He just setup some used Revel subwoofers he’s trying to demo and sell. He always prefers to setup the subs at 6-10dB higher to demo the sound.

 

So when I got home, I actually just added a low-shelf filter at 100Hz, +6dB and Q=1 in Roon in addition to running the convolution filter. So now my system sounds much more like how he tuned it. Not my preference. But It has its charms. And I can see why it’s important for selling subwoofers.

 

Bottomline is, you can clearly do this on top of your convolution filter without major sonic penalties. So I would try that first at home. And once you know what low-shelf filter you like the most, just let Mitch Barrett know and see if he can just incorporate that frequency response into the convolution filter for your rock music. If he can’t or charges too much extra, at least you can still just add the low-shelf filter.