Jump to content

Understanding and Optimizing Tactile Feedback


dominguez1

Recommended Posts

Around 0.5m, if you measure very close, as in 5mm, you will measure the velocity of the diaphragm, and since this is a compact horn unit, the total from the port and the woofer would not sum correctly.

 

The second blue line - velocity - is not the measured velocity, it is normalised using a transfer function created by dividing the spl on the velocity.

When using this same transfer function on the lp measurement for velocity, we get a velocity measurement that can be compared to the spl, so that we can see the difference between the near and lp situation.

How did you measure this velocity?  Do you have a write-up you can point me to?  In any case, your results do make intuitive sense to me.

Link to comment
Share on other sites

  • Replies 88
  • Created
  • Last Reply

I think one thing that's confusing a lot of people here is the interpretation of particle velocity and sound intensity.  If you measure particle velocity at a point in space where a listener's skin would be located if they were sitting there listening, you will get a different result than if you could measure the particle velocity in the same place while the person sits there.

 

The presence of your body or any other solid object necessarily alters the sound field in the air around it.  If the impedance of the object is very high relative to air at the frequency that is being considered, then almost all the sound energy that tries to impinge it will either bounce off or diffract around it, regardless of what sound intensity was measured in the air in its absence.

 

Does this make sense?

Link to comment
Share on other sites

Today I did a re-boot of the experiment that initially convinced me there is more than spl alone at play.

What started it all was when I had a configuration that never sounded right, no matter what I did - adjusting eq, delay, boosting ulf to insane levels - nothing helped, it just sounded like lots of bass and not that powerful, effortless feel.

 

Two different configurations, where frequency response and impulse response ideally is the same, but very different in sound field properties - velocity, intensity.

I listened to some music clips, and some movie scenes.

Then I measured it all - frequency response, and velocity in all 3 planes.

 

There is no contest.

 

The difference has to be experienced to be believed.

 

Examples:

 

Audiophile recordings (elevator-music; jazz with few instruments and singing lady, Michael Ruff Sheffield Labs recording):

Attack and kick from drums is not that different, from around -10dB and up you start to get the tactile feel from drums.

But the good set-up sounds more balanced in a way, like effortless, weightless bass.

The bad set-up is missing most of that addictive-bass feel that makes you want to hear just one more, and you keep turning up the volume.

 

Avratz (Infected Mushroom):

Defect set-up: Attack and kick is sharp and nice, sounds ok, but not great, like - what's so special with this recording.

Good: Attack and kick is similar, but there is this weightless, effortless feel of the bass, and when you turn it up (like +6dB) each bass-hit feels like you are pushed backwards, the difference is far from subtle.

 

Hanna The plane:

Good sounds more powerful, but most important - it sound like less bass, less rumble, but more feel.

 

Oblivion Coming in hot with bass-eq:

Defect set-up: Sound very powerful and intense, with nice weight. 

Good: Insane..

 

To sum up the impressions:

- Upper bass and attack and kick is not very different

- Music and scenes with more low frequency content - below 50-60hz - here is where the difference really stands out, sounds more powerful, more real.

- Experienced differences in bass quality may be caused by other differences like frequency response and impulse response.

 

I will post measurements.

 

And if anyone has requests for scenes I should play, I can do that now, before it gets too late.

The best thing would of course be if you could be here and listen to it yourself, but since that is not very practical, my subjective impression is what we have.

Link to comment
Share on other sites

Here are the measurements of the two different configurations.

 

A - blue lines - good configuration.

B - red lines - bad configuration with low velocity.

 

Velocity plots are 0 degrees - facing the front wall/speaker, 90 degrees, and h - horizontal.

 

We can see that the A has clearly a larger velocity across the lower part of the bass range from 20hz-60hz.

The frequency response is not a perfect match, and those differences will be audible.

Phase is fairly flat for both from 100hz down.

 

post-181-0-29248400-1428698172_thumb.png

 

Link to comment
Share on other sites

And if anyone has requests for scenes I should play, I can do that now, before it gets too late.

The best thing would of course be if you could be here and listen to it yourself, but since that is not very practical, my subjective impression is what we have.

Very cool.

 

The Incredible Hulk Scene; both the University through Sonic Cannon, and the fight with Abomination

Link to comment
Share on other sites

Can you describe these to configs more? Distance from sub, etc.?

 

I may or may not come up with a better description, but it is basically a nearly square room 4.5mx4.5m approximately. the most important is that the right side wall has a large opening.

If turn this 90deg to left, so that this wall with the opening becomes the back wall, the low-frequency response would improve. A lot. 

 

Subwoofer units placed in 4 corners, radiation source close to floor, lp around 2/3-3/4 back towards back wall, no low freq absorption in room.

Link to comment
Share on other sites

Here are the measurements of the two different configurations.

 

A - blue lines - good configuration.

B - red lines - bad configuration with low velocity.

 

Velocity plots are 0 degrees - facing the front wall/speaker, 90 degrees, and h - horizontal.

 

We can see that the A has clearly a larger velocity across the lower part of the bass range from 20hz-60hz.

The frequency response is not a perfect match, and those differences will be audible.

Phase is fairly flat for both from 100hz down.

 

post-181-0-29248400-1428698172.png

Takes some examining to decipher since there are so many measurements, but I can see what you're referring to in the 20-60 band. Nice.

 

I also notice that at about 180hz, the 'good' placement and 'bad' placement velocity response start to look exactly the same. This makes sense as the frequencies shift to the far field soundfield where it is in phase with SPL. In this region, you can see that measuring SPL to understand tactile feel is enough because there is no variation in PVL between the two configurations (e.g. in the farfield, 100dbSPL 20ft away will feel the same as 100dbSPL 40ft away). However, as this chart demonstrates, PVL starts to take much different paths in the near field soundfield, necessitating an understanding of the the PVL response to fully understandg tactile feedback and Sound Intensity.

 

I am curious as to what's happening below 20hz...seems very erratic. Could it be your PVL probe is just not accurate at these frequencies?

Link to comment
Share on other sites

Velocity measurements can not be trusted below 15-20hz, and I suspect even below 30hz there is very significant noise.

Above around 1khz it should fall off, which we can also see.

 

That the good and the bad looks similar above 180hz is to be expected, as the differences in configuration only affect frequencies below 120hz.

 

What is interesting is to notice how the sound field behaves above around 200hz - it becomes more smooth, and it is directional, and overall level is higher.

This change occurs before the speakers themselves start to get directional, so this must be attributed to the room acoustics.

Now, imagine a much larger room - if it were 10x the size, we could expect this behavior to start at 20hz instead of 200hz, which gives us this desired and predictable velocity response across most of the audible frequency range.

This could be why larger rooms gives a better and more tactile bass response.

 

Not all speakers-rooms measure like this, often the response is rather erratic all the way up, suggesting that the sound field is compromised by reflections and poor directivity from the speakers.

Link to comment
Share on other sites

So the difference between A and B is simply floorplan/configuration with everything else equal (time aligned and EQ'ed subs, optimized x-overs, etc?)

 

If that is so, even more evidence that the room can be much more important than the equipment in it sometimes.

 

Great stuff.  How would an average joe with only REW measure if one setup was better than the other given the very similar FRs for A&B?

 

Also, what is happening from 100-140Hz? 1/4 wave cancellation/mode?

 

JSS

Link to comment
Share on other sites

The only difference between the configurations are delay on the individual subwoofer units.

This creates very different soundfields.

To do this properly I should have adjusted eq to make the frequency response similar - as in within +-1dB.

 

You can not see the difference on the frequency response alone, and it is not revealed by further analysis like impulse response (that means step-response for bass) and decay.

One indication of a bad situation could be to measure at different positions - the bad has a sweetspot in the lp, where freq resp is better, and it changes a lot once you move only a smaller distance to either side.

Strange, but the bad one actually sounds better in other seats, and that is also where this cancellation of velocity will not occur.

 

The 100-200hz range is a mess due to boundary reflections, mostly the ceiling, moving the mic changes the response very much in this range.

This can be fixed by acoustic damping and also by increasing the acoustic size of the subwoofer system - easily done by stacking 2 horns in each front corner.

Link to comment
Share on other sites

Interesting.  I wish I could see the SPL plots from 200 Hz down with no smoothing, but it looks like the velocity plots would be too messy

to look at without.  Do realize that changing delays will definitely alter the impulse response of the combined sound.  It does look like the "A" SPL curve is a bit smoother in the mid-bass than the "B" curve but with only 1/6 octave, it's hard to tell what's going on there.

Link to comment
Share on other sites

Some more plots from the good and bad - test. 

 

Impulse step response, blue = good, the bad one actually looks better:

post-181-0-79576300-1428765596.png

 

Zoomed fr no smoothing, below 200hz there is not much difference to smoothing at 1/6 for spl, but the velocity is very different, and as I will soon show you, this is due to reflections causing the velocity to look like this:

post-181-0-90130000-1428765632.png

post-181-0-79576300-1428765596_thumb.png

post-181-0-90130000-1428765632_thumb.png

Link to comment
Share on other sites

Continuing the A and B configuration test, the waterfall shows some differences I believe correlates with the listening impression.

 

The bad has a resonance at 80hz, I think this is what causes a thickness in the attack on drums and in the upper bass.

The bad also has a dip at 50hz, and this may be what causes the total lack of punch on 50hz drums.

 

All waterfall plots have 50hz noise, that is what makes it look like the 50hz lives on forever.

 

My original bad configuration had great 50hz punch and a freq resp more equal to A.

My best good configs have had much better velocity below 30hz.

I did this test very quickly, using what is in the room right now, and I think both A and B configurations could be improved for this purpose - to investigate differences due to sound field properties.

The A - good - I think could have better velocity below 30hz.

The B could have a smoother freq resp more equal to A, with less ringing at 50hz and 80hz.

 

post-181-0-23111200-1428767600_thumb.png

post-181-0-90617900-1428767615_thumb.png

 

post-181-0-80219900-1428767632_thumb.png

post-181-0-55076100-1428767647_thumb.png

 

Link to comment
Share on other sites

 

One factor that I believe is being missed is that your SPL measurements for matching the 2 systems output are taken at a single point in space at the microphone element and in no way take into account the tactile vibrations which are the actual function being looked at. The phone was being set on the couch for the vibration measurement...The couch is a very large piece of furniture occupying a significant volume of space in the room relative to a microphone. In effect the entire couch is the microphone at that point. When you sit on a couch you are now in contact with a very large surface area of the couch. The subs were in very different parts of the room and affected by the room reflections and boundaries much differently. The total amount of energy from each sub system physically imparted into the whole of the couch as vibration will be much much different. This explains the difference in tactile sensation. The near-field sub would have a much more focused and directed energy on the couch than a sub 15 to 20ft away. The 2 subs may have had the same SPL at the spot measured but what about back behind the couch in front of the near sub? Near the floor on the left front corner of the couch, what about the back right corner? Under the couch in the middle? That's what I'm saying you would need hundreds of SPL measurements and a few accelerometer placements to get anywhere close to understanding the total energy impacting the couch from each sub system and calibrating them to each other.

 

A single point SPL measurement where the phone will be placed on the couch simply does not have the resolving power to look into the vibration issue.  In a complex acoustic environment with direct physical contact with other objects you will have the audible component of sound and also the physical sensations. The two are separate issues to a large extent on that we agree. However I'm not yet convinced that it has anything to do with particle velocity or sound intensity.

 

 

@Ricci, I took some additional readings around the couch for both the FV15HPs and the FTW21s.

 

I first calibrated to ~96.6db at 15hz at the LP for each set of subs.

 

I then moved the mic starting at the bottom of the couch, and moving up to the top of the couch. I did this for the width of the back of the couch and front of the couch for both sets of subs. As you'll see, the 15hz sine wave does not last long, so you'll see a drop in SPL when it stops. It took 3 sets of playing the sine wave to measure the length of the couch for each test.

 

The next two videos are measurements from the FV15HPs.

 

https://youtu.be/5V051nAiLJA

 

Focus on the SPL meter reading at the top right.

 

This is measuring the back of the couch with just the FV15HPs. I start at the bottom and then move toward the top. It looks as though as I approach the top of the couch, it has the most SPL.

 

In this example, the FV15HPs are nearfield to the couch. You can see that the SPL reading varies between 90 and 100 most of the time. There were occasional peaks of 102, and the max spl for this test was recorded at 106. Again, ~95% of the readings were between 90db and 100db.

 

https://youtu.be/w1YFirvjk4Y

 

This is measuring the front of the couch with just the FV15HPs.

 

Further away from the subs, the SPL is much more constant. It hovers right around 100db, with a max of 100.8db.

 

The next two videos are of the FTW21s by themselves:

 

https://youtu.be/ymyCT1JpObY

 

FTW21 back of couch.

 

You can see that the SPL is pretty constant at around 96-97db. Its max is 97.6db.

 

https://youtu.be/3fePEopgsEE

 

FTW21 front of couch.

 

SPL is pretty constant at 95db, while its max is at 96.1db.

 

Overall, the FV15HPs have a higher average of SPL. If I had to guess it would be an average of 98-99db for the FV15HPs.

 

The FTW21 had pretty much had an average of 96db or so.

 

I'd estimate the average SPL difference between the two sets of subs would be approximately 2-3db. Although there is a slight difference in SPL, the vibration associated with each are magnitudes different. As I mentioned before, I had to raise the volume of the FTW21s by 20db to get closer to the vibration readings of the FV15HPs nearfield.

 

This as well as okv's recent tests is pretty compelling evidence that SPL alone does not tell the whole story when it comes to tactile feedback and Sound Intensity.

Link to comment
Share on other sites

Now I am going to show something interesting.

 

Here is an outdoor measurement of a speaker, conditions are semi-free-field.

Observe how the velocity now is even smoother that the spl.

Blue = 1m, red = 2m.

Dark = spl, lighter = velocity 0 degrees.

 

post-181-0-50439600-1428767880.png

 

Now if we include the 90 degrees velocity, we see that it is not smooth at all, comb filtering and cancellations:

post-181-0-03311900-1428768078.png

 

 

If we compare these plots to the A configuration from the test, we can make some interesting observations.

 

First, we see that the free field has much larger velocity at lower frequencies, and it is smooth.

Even the best peaking range around 40hz in A only reaches up to the free-field level, it is never much better.

 

Since there are indeed reflections, yet still the measured velocity resembles a completely developed free-field scenario with no reflections, it is reasonable to believe that it is not necessary to create a completely dead room to achieve the same soundfield properties in a small room.

 

We see that the velocity tends to increase at lower frequencies closer to the source.

This is expected.

We also see that the velocity falls off slightly, this is not expected, and indicates that my frequency correction curve is not exactly correct, there is also a small peak around 1khz that does not make sense.

 

Recreating this same soundfield as seen in the free-field measurements can be done, even in a small room.

If this is sufficient, that is a viable solution for getting the level 3 on bass quality completed.

post-181-0-50439600-1428767880_thumb.png

post-181-0-03311900-1428768078_thumb.png

Link to comment
Share on other sites

Depends on the configuration - placement of subwoofer units, and room.

 

If you don't sit close to the rear wall, and you don't take out the rear wall acoustically (absorb everything), multisub is required to fill the holes.

And that works, you fill the holes, get a nice flat frequency response, and if you make some effort a nice impulse response can also be achieved.

What happens to the soundfield is undetermined, and it is likely you end up with something close to B - the bad.

 

None of the multisub set-up approaches I have seen address the issue of soundfield, actually the seem to ignore the impulse response as well.

 

This is not just about getting a tactile response that shakes and rattles and kicks you in the chest, it is about getting good bass sound quality.

From these experiments and measurements it looks like it is necessary to get rid of low frequency reflections inside the room.

The A configuration is better, but it still leaves a lot compared to free-field.

Link to comment
Share on other sites

A note on eq, and why a house-curve with boost at the lowest frequencies often sounds better.

 

In a small room you get spl room gain at very low f.

If you eq for flat, the acoustic energy delivered from the speaker system falls at lower freqs.

A house curve will then get you a flatter total energy response, and that may be why we prefer some boost.

 

Also if you eq by taking down peaks, the energy delivered at those frequencies will be reduced.

With less eq the spl response may not be as flat, but the total energy will be flatter.

Link to comment
Share on other sites

Continuing the A and B configuration test, the waterfall shows some differences I believe correlates with the listening impression.

 

...

It's a tough call to me whether A or B has the better step response.  I prefer to look at the impulse responses instead of step responses.  IMO, your frequency response looks a bit better for A than for B.  I also think the waterfall for A looks better than it does for B.  I know from experience that even subtle improvements to the waterfall can be audible, but of course, it may not be the only factor at work here.

 

A note on eq, and why a house-curve with boost at the lowest frequencies often sounds better.

 

In a small room you get spl room gain at very low f.

If you eq for flat, the acoustic energy delivered from the speaker system falls at lower freqs.

A house curve will then get you a flatter total energy response, and that may be why we prefer some boost.

 

Also if you eq by taking down peaks, the energy delivered at those frequencies will be reduced.

With less eq the spl response may not be as flat, but the total energy will be flatter.

I think this is a bit oversimplified.  As the sub adds energy to the room, the room surfaces, room contents, and the air itself remove energy from the room.  The frequency response measurement indicates the SPL at each frequency when these two processes are in balance.  How much power the sub has at a particular frequency depends on its anechoic/ground-plane response, which depends a lot on the woofer and cabinet design.  How much power the room dissipates at a particular frequency is a lot more complicated.  In regions of resonance and low-end gain, the room doesn't dissipate energy as well, so the SPL rises a lot until the room is dissipating energy at the same rate that the sub is adding.

 

By EQing down a resonance you not only change the rate at which the sub adds energy at steady state for that frequency but also the transient profile.  With a PEQ dip, the sub will initially add energy at the same rate that it did without EQ, and then it will gradually reduce its energy contribution at that frequency.  Ideally, the system reaches steady state sooner, giving quicker attack and decay as well as a more balanced SPL.  Of course, this won't give you higher particle velocity unless you are using EQ applied independently to multiple subs to shape the sound field.

 

My own believe for why house curves sound better is that smaller rooms typically present transient sounds at higher sound levels than the larger rooms that content is typically mixed in.  This necessitates a lower overall playback level and a house curve to restore the tonal balance lost due to the lower playback level.

 

Unfortunately, I can't contribute my own experiments on this topic because I have a suspended wood floor that transmits vibrations quite well.  I also have a wide open room and may have bass conditions more similar to far field.  As it is, I could probably do with a bit less vibration in the upper bass range, where it seem my sofa and/or floor take a bit longer to decay than the sound does.

 

By the way, don't underestimate bass traps.  They made a huge difference in the enjoyment of bass in my space.  Even though the traps primarily work at > 80 Hz, they improve the response of the harmonics that give the lower bass hits and notes speed and definition.  A lot of tactile feel is also involved at these higher frequencies.  They also tend to make those lower bass notes sound louder.  Just looking at your responses (-10 dB from 95-140 Hz or so), I have no doubt that regardless of what your sound field looks like, you are missing out on a lot of bass that is very tactile.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...