• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. 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!

Getting vocals in focus

stunta

Major Contributor
Forum Donor
Joined
Jan 1, 2018
Messages
1,155
Likes
1,399
Location
Boston, MA
The vocals in music are not in focus with my 708Ps. There seems to be this one specific position that my head needs to be in to get them to be in focus and even then they are slightly off center and only about 3 feet off the ground (the speakers are sitting on FM8 subwoofers that are on isoacoustic bases). The speakers are toed in and pointing ~1 feet either side of my ears at my MLP and the distances form an equilateral triangle.

Any suggestions on what I should be changing to get the vocals to be in focus and also not be so sensitive to me moving my head even an inch?

I have the system calibrated with Dirac Live and so I don't move my speakers fearing they will need a recalibration. If I adjust toe-in and re-calibrate, will I then be stuck in an endless loop?
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
Sounds like a symmetry issue with the room acoustics to me. Dirac may have fixed responses for one spot but if they go all over the place as you move around, that's no help.

I'm not sure I get your description, I am inferring a listening distance of 2 ft but surely that wouldn't be right with 708Ps? A plan or photos would be good to have.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
The vocals in music are not in focus with my 708Ps. There seems to be this one specific position that my head needs to be in to get them to be in focus and even then they are slightly off center and only about 3 feet off the ground (the speakers are sitting on FM8 subwoofers that are on isoacoustic bases). The speakers are toed in and pointing ~1 feet either side of my ears at my MLP and the distances form an equilateral triangle.

Any suggestions on what I should be changing to get the vocals to be in focus and also not be so sensitive to me moving my head even an inch?

I have the system calibrated with Dirac Live and so I don't move my speakers fearing they will need a recalibration. If I adjust toe-in and re-calibrate, will I then be stuck in an endless loop?

I’d move the speakers around the room to find the best spot before doing any EQ or fine calibration anyway — can save you a lot of time.

But, yeah, room acoustics and placement most likely...

Also, I know this might be obvious, there’s music out there where the vocals are just slightly off centre... can be easily confirmed with headphones. Sometimes I have found myself wondering because of certain tracks, but found the culprit is actually the record itself.
 
OP
stunta

stunta

Major Contributor
Forum Donor
Joined
Jan 1, 2018
Messages
1,155
Likes
1,399
Location
Boston, MA
Sounds like a symmetry issue with the room acoustics to me. Dirac may have fixed responses for one spot but if they go all over the place as you move around, that's no help.

I'm not sure I get your description, I am inferring a listening distance of 2 ft but surely that wouldn't be right with 708Ps? A plan or photos would be good to have.

The MLP is about 10' from the speakers and L & R are about the same distance from each other. The L speaker is flanked by walls and acoustic panels and R speaker has cork behind it with shelves and open space on its right. I am not sure if toe-in is needed for these speakers and if so how much. Note that they are sitting on top of Rythmik FM8s which I tried to toe in at the same angle - they handle 80-200Hz and treated as mains together with the 708Ps in Dirac.

Hope the photos are helpful.

IMG_20200905_005247.jpgIMG_20200905_125940.jpgIMG_20200905_125944.jpg
 
OP
stunta

stunta

Major Contributor
Forum Donor
Joined
Jan 1, 2018
Messages
1,155
Likes
1,399
Location
Boston, MA

DSJR

Major Contributor
Joined
Jan 27, 2020
Messages
3,314
Likes
4,427
Location
Suffolk Coastal, UK
Good luck here...

Our little sitting room only allows the speakers to be in one position slightly left of centre (the fireplace is in the corner). The upper mids 'pull' slightly to one side and swapping the speakers round (two pairs, one old one not) makes no difference at all. As I have a recurring issue with one ear, I can't always tell if it's the recording, the room or my damned ear infections :mad: :facepalm:
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
Changing the distance to one speaker of the pair will move the percieved center.

If the image is to the left, move the right speaker forward a little.

It doesn't take much when you're really paying attention.

---

Acoustic measurement of equal speaker distance:

If you can measure with REW:

Place the mic where you think the sweet spot is.

---

Using acoustic reference on the left speaker, measure the impulse response for the left and then the right speaker.

Put the two measurements into the Overlays window.

Observe the time alignment.

Move speakers if misaligned, or move your idea of where the sweet spot is.

20us misaligned - 1/4" difference over 10 feet - the green speaker is closer to the mic

1599332681041.png


When I first did this, I found my sweet spot wasn't where I thought it was (by a few inches), and made some adjustments.
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
Also, for a trick I picked up from Earl Geddes, you can try crossing the speaker axes in front of the MLP. This tends to give directional speakers a larger sweet spot. It would also be reducing adjacent wall reflections. You can use a mirror to find out where the reflection points are (place on wall until you can see a speaker from the MLP), so then you can contemplate your possible treatment options.

10' is not a small distance for an 8". A side wall next to one speaker with open space next to the other is an absolute classic scenario for symmetry issues. Happy reflection point hunting, I'd say.
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
It doesn't take much time difference to change image location.

Experiment:

Headphones, and delay one channel at miniDSP.

Delay resolution is 0.02ms, equalling 6.8mm or 0.271" difference in distance.

I can hear the image step toward one side with each additional 0.02ms delay.

At 0.4ms / 137.6mm / 5.42" time imbalance the perception is that all the sound is coming from one earpiece, though both are emitting sound at the same SPL (mono signal).

1599339367949.png




Ok, so headphones aren't speakers, but the same rules apply, its's just more obvious in the phones as there are no room reflections, and no sound sneaking around from one ear to the other.
 

RPG

Member
Forum Donor
Joined
Feb 12, 2020
Messages
81
Likes
86
Location
Seattle
Changing the distance to one speaker of the pair will move the percieved center.

If the image is to the left, move the right speaker forward a little.

It doesn't take much when you're really paying attention.

---

Acoustic measurement of equal speaker distance:

If you can measure with REW:

Place the mic where you think the sweet spot is.

---

Using acoustic reference on the left speaker, measure the impulse response for the left and then the right speaker.

Put the two measurements into the Overlays window.

Observe the time alignment.

Move speakers if misaligned, or move your idea of where the sweet spot is.

20us misaligned - 1/4" difference over 10 feet - the green speaker is closer to the mic

View attachment 81567

When I first did this, I found my sweet spot wasn't where I thought it was (by a few inches), and made some adjustments.

Ray......just absorbing the knowledge, here. Does a balance control work the same way, or is it preferable to change the speaker's distance in order to adjust the center/sweet spot???
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
Does a balance control work the same way,

No.

Balance changes the relative volume of the speakers, not the timing.

The effect can be similar, but finding the acoustic center for "sweet spot" listening is better.

---

Experiment:

Headphones, again.

Reduce the volume on one channel:

-9dB gives a similar directional offset as the 0.4ms time adjustment.

Change in spatial perception is not as clear, though, not as natural sounding.

Hint: -9dB is a lot.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
I have the system calibrated with Dirac Live and so I don't move my speakers

Mark your speaker positions and move the "weaker" one forward (toward the sweet spot) a little at a time.

See if the balance problem you hear clears up.

Mono (correlated) pink noise could be a good test tone for that in addition to whatever music exhibits the problem.
 

RPG

Member
Forum Donor
Joined
Feb 12, 2020
Messages
81
Likes
86
Location
Seattle
No.

Balance changes the relative volume of the speakers, not the timing.

The effect can be similar, but finding the acoustic center for "sweet spot" listening is better.

---

Experiment:

Headphones, again.

Reduce the volume on one channel:

-9dB gives a similar directional offset as the 0.4ms time adjustment.

Change in spatial perception is not as clear, though, not as natural sounding.

Hint: -9dB is a lot.

So interesting to learn about this, thank you. I have a similar room---wall about a meter to the right of the right speaker, but wide open to the left of the left speaker. I don't have any DRC software that can effect a delay, but I do have an OmniMic room measurement mic and software, so I could measure the impulse response and, if need be, physically move a speaker a bit closer. One question on that.....what did you use to generate the REW impulse measurement....a single tone or a sweep, or....????
 
Last edited:
OP
stunta

stunta

Major Contributor
Forum Donor
Joined
Jan 1, 2018
Messages
1,155
Likes
1,399
Location
Boston, MA

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
what did you use to generate the REW impulse measurement....a single tone or a sweep, or....????

Sweep.

Long ago I was suspicious about sweep generating impulse and step responses. I mean, the sweep tone has no edges, right?

That meant it was time for an experiment. So I generated a step (single full scale bit) and a step (10Hz signal to the speakers is a repetition of steps) and recorded the result with a UMIK-1.

The waveform from the microphone of the actual impulse and step excitation was all but exactly the same as the results from calculations from the sweep.

https://www.audiosciencereview.com/forum/index.php?threads/impulse-response.1765/#post-44352

https://www.audiosciencereview.com/forum/index.php?threads/impulse-response.1765/#post-44440
 

RPG

Member
Forum Donor
Joined
Feb 12, 2020
Messages
81
Likes
86
Location
Seattle
Sweep.

Long ago I was suspicious about sweep generating impulse and step responses. I mean, the sweep tone has no edges, right?

That meant it was time for an experiment. So I generated a step (single full scale bit) and a step (10Hz signal to the speakers is a repetition of steps) and recorded the result with a UMIK-1.

The waveform from the microphone of the actual impulse and step excitation was all but exactly the same as the results from calculations from the sweep.

https://www.audiosciencereview.com/forum/index.php?threads/impulse-response.1765/#post-44352

https://www.audiosciencereview.com/forum/index.php?threads/impulse-response.1765/#post-44440

Excellent, thank you.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
Instead of testing left and right for impulse, you can take the impulse with both speakers playing.

If they aren't aligned well, there will be two impulses.

This is a "both speaker" sweep, with the mic about a foot off center:

1599345186786.png
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
With the crossover at 200Hz, I figured at 10' it would give me enough SPL and it does.
I meant more in terms of dispersion than SPL, which should be more than ample if crossed over as high as 200 Hz (you could try more like 100-150, the JBLs should easily have enough oomph). If we look at Genelec's monitor selection guide, the similarly-sized 8350 easily manages enough SPL way out to >10 m, but direct to reflected sound takes a nosedive much past 2 m even in a well-treated room. Now JBLs tend to favor a more narrow dispersion than Genelecs, so you'd probably be pushing that limit out to around the 3 m mark or so (your 10'), but that still assumes a room with around a 0.3 s RT60.
 
OP
stunta

stunta

Major Contributor
Forum Donor
Joined
Jan 1, 2018
Messages
1,155
Likes
1,399
Location
Boston, MA
Man, if ASR members could come together and write a wiki or something as a practical guide to home audio, how amazing would that be?

Thank you for the responses. Long weekend is here so time to get that Umik-1 out and nudge those speakers around a bit.
 

garbulky

Major Contributor
Joined
Feb 14, 2018
Messages
1,510
Likes
827
The vocals in music are not in focus with my 708Ps. There seems to be this one specific position that my head needs to be in to get them to be in focus and even then they are slightly off center and only about 3 feet off the ground (the speakers are sitting on FM8 subwoofers that are on isoacoustic bases). The speakers are toed in and pointing ~1 feet either side of my ears at my MLP and the distances form an equilateral triangle.

Any suggestions on what I should be changing to get the vocals to be in focus and also not be so sensitive to me moving my head even an inch?

I have the system calibrated with Dirac Live and so I don't move my speakers fearing they will need a recalibration. If I adjust toe-in and re-calibrate, will I then be stuck in an endless loop?
Sounds like you're going to have to recalibrate. As has been suggested reset to normal levels. Then spend lots and lots of time experimenting with these things:
Increasing and decreasing:
the space between the speakers.
the space between the speakers and the rear walls
The distance from the speakers to you.
The toe in.
If possible center the speakers or get them away from corner walls
Bring the speakers in front of the plane of the tv or cabinets. So it should be closer to you than the tv or cabinets.
This will take a long long time as you gotta experiment with all these different combinations. The biggest differentiator is toe in and space between the speakers. Also if the speakers are too low or too high that will affect it too.
In my setup the speakers are farther apart than they are close to me.
AFTER all of that. Then work on DIRAC. But...you may not need to.
 
Top Bottom