or Connect
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #51779)
New Posts  All Forums:Forum Nav:

"Official" Audyssey thread (FAQ in post #51779) - Page 1681

post #50401 of 70906
Quote:
Originally Posted by Gooddoc View Post

Wrong. 9 dB is 9 dB, regardless of whether you boost a null or boost a flat FR

???????????? Care to think it over again? If you boost a flat FR you did nothing, just raised overall volume. Audyssey selectively boosts frequency areas where boost is needed and does not boost areas where boost is not needed, same goes for "taming" peaks at frequencies where "taming" is needed. Go figure!
post #50402 of 70906
Feri - you are confusing the measured response IN ROOM with what the amp has to do to get there.

Just because the final output is measured flat doesn't mean that you didn't need a bunch more gain in the amp to get there. As others have noted it's not Audyssey per se, the same would be true for any EQ.
post #50403 of 70906
Quote:
Originally Posted by batpig View Post

Feri - you are confusing the measured response IN ROOM with what the amp has to do to get there.

Just because the final output is measured flat doesn't mean that you didn't need a bunch more gain in the amp to get there. As others have noted it's not Audyssey per se, the same would be true for any EQ.

My dear batpig, in this case IMHO the amp being flat as hell from nearly DC to 100 kHz has noting to add or take away. It just takes an EQ'd signal, amplifies it and sends it to the speakers. But if the signal is not EQ'd the amp will still do its job, won't it? Getting back to the original question of boosting a null I don't understand why others are thinking of an overrall boost, when a boost by Audyssey is not doing anything else but compensating a null, thus making the input signal to the amp as flat as possible.

Tell me if I'm wrong. please!
post #50404 of 70906
If you set a sub to play at 80db, and Auyssey finds a -10db hole IN ROOM RESPONSE, then Audyssey would ask the amp & speaker to produce 10db more at the offending frequency to get a flat Room Response. So if your room is -10db @ 30hz, then your sub amp and speaker are going to be asked to do a lot more at 30hz.
post #50405 of 70906
Quote:
Originally Posted by D Bone View Post

If you set a sub to play at 80db, and Auyssey finds a -10db hole IN ROOM RESPONSE, then Audyssey would ask the amp & speaker to produce 10db more at the offending frequency to get a flat Room Response. So if your room is -10db @ 30hz, then your sub amp and speaker are going to be asked to do a lot more at 30hz.

No. Audyssey can't ask the amp to output more coz the amp is flat as hell and Audyssey cannot control that. What Audyssey can control is the FR of the input signal.
post #50406 of 70906
Quote:
Originally Posted by mogorf View Post

Getting back to the original question of boosting a null I don't understand why others are thinking of an overrall boost, when a boost by Audyssey is not doing anything else but compensating a null, thus making the input signal to the amp as flat as possible.

Tell me if I'm wrong. please!

yes you are wrong -- the input signal to the amp is NOT the part that is flat. The goal is to make the IN ROOM response end up flat, which will require a NON FLAT input signal to the amp, that is then amplified and becomes a (bigger) NON FLAT signal sent to the speaker. In fact, specifically, it will theoretically be an inverse of the uncorrected measured in-room response.

I have no idea what you mean by "it's not a boost, it's compensating a null". That seems a meaningless distinction. It is a boost. If it has to apply 9dB of gain to compensate for a null, that's 9dB of gain that the amp (and the the driver) have to provide, right?

Have you NOT seen people's graphs measured at the pre-outs? That's the signal that goes to the amplifier. Do they look flat to you?
post #50407 of 70906
Quote:
Originally Posted by mogorf View Post


My dear batpig, in this case IMHO the amp being flat as hell from nearly DC to 100 kHz has noting to add or take away. It just takes an EQ'd signal, amplifies it and sends it to the speakers. But if the signal is not EQ'd the amp will still do its job, won't it? Getting back to the original question of boosting a null I don't understand why others are thinking of an overrall boost, when a boost by Audyssey is not doing anything else but compensating a null, thus making the input signal to the amp as flat as possible.

Tell me if I'm wrong. please!

If you get 9dB more out of the amp, for any reason and in any circumstances, you need to 8-tuple the wattage.
post #50408 of 70906
Quote:
Originally Posted by kbarnes701 View Post

If you get 9dB more out of the amp, for any reason and in any circumstances, you need to 8-tuple the wattage.

Almost spit my Coke onto my laptop!
post #50409 of 70906
Quote:
Originally Posted by D Bone View Post


Almost spit my Coke onto my laptop!

Feri has his wires crossed somewhere, which puts him out of phase with the rest of us, for a moment at least
post #50410 of 70906
Quote:
Originally Posted by batpig View Post

Feri - you are confusing the measured response IN ROOM with what the amp has to do to get there.

Just because the final output is measured flat doesn't mean that you didn't need a bunch more gain in the amp to get there. As others have noted it's not Audyssey per se, the same would be true for any EQ.

Gotcha!!
post #50411 of 70906
Quote:
Originally Posted by mogorf View Post


Gotcha!!

Phew! . I thought you'd been on the Palinka
post #50412 of 70906
So, i never meant to say my subs couldnt take the +9db hump to correct the dip or null or depression, or what ever you want to call it. My thought was, that if i had a sub with a small driver, Audyssey could easily over drive in order to satisfy it calculations.
As it stands, i had about 5% @ 55hz, hardly noticable. Its all in the graphs. Only ones that are missing are the subs, close mic to establish we were flat from the start up close. I remember measuring it but never saved it.
At this point i have a bunch of work left, between getting a MiniDSP and build another sub. Then i can do a real soup to nutz measurements and it will follow a nice beginning, middle and end......more to come
post #50413 of 70906
Seems to be some confusion here so maybe this might help:
First off, technically, the term null is generally used to describe a modal null or power response dip resulting from destructive interference.

By that terminology, a null at a certain frequency can't really be tamed/corrected by boosting the EQ because boosting the signal boosts the direct and reflected waves causing the destructive interference, i.e., the waves are still cancelling each other.

That graph doesn't show much because there's so much smoothing applied. To see if it is a null or a frequency response dip in the bass region, the graph should have no smoothing applied. The fact that simple EQ can correct it though tends to point to something other than an actual null.

Generally, nulls at a specific frequency and location can really only be tamed by either using bass traps (absorb the reflected wave, reduce its amplitude and thereby reduce the intensity of the cancellation), reposition the sub (alter the geometry to the reflecting surfaces) or use multiple subs.

Now when it comes to the EQ vs power demand, yes 9db is 9db. It will create a power draw approximately 8 times greater than normal around that frequency region. There's a lot of material/audio encoded at 55Hz in drums, explosions etc. That could mean the difference between drawing 150 watts and 1200 watts.


Max
post #50414 of 70906
Quote:
Originally Posted by mogorf View Post


???????????? Care to think it over again?

Nope
post #50415 of 70906
The required excursion for a driver to reach a certain SPL is frequency dependent. The principle is that a driver will require little excursion at most frequencies but, from some cut-off point, will need to use dramatically increasing excursion to go to lower frequencies. Nothing particularly surprising here but, as far as I can tell, a lot of drivers have an additional (relatively narrow: <1 octave) excursion hump somewhere above the lower point of no return for the driver, so to speak. This hump, again as far I can tell, is dramatic, easily going to and above the excursion limit, even if you feed the driver less than its thermal-limit power.

If you happen to have eq'ing boost the hump frequencies, your driver gets into trouble very quickly, indeed.
post #50416 of 70906
Quote:
Originally Posted by mjf_uk View Post

http://www.amazon.com/Samson-BL3-Ult...9169266&sr=1-1

It's the same stand that Audyssey supply with their Pro Kit.

Thanks. I ordered it! This should be in the FAQ (pardon if its there already)
post #50417 of 70906
Quote:
Originally Posted by ReneV View Post

The required excursion for a driver to reach a certain SPL is frequency dependent. The principle is that a driver will require little excursion at most frequencies but, from some cut-off point, will need to use dramatically increasing excursion to go to lower frequencies. Nothing particularly surprising here but, as far as I can tell, a lot of drivers have an additional (relatively narrow: <1 octave) excursion hump somewhere above the lower point of no return for the driver, so to speak. This hump, again as far I can tell, is dramatic, easily going to and above the excursion limit, even if you feed the driver less than its thermal-limit power.

If you happen to have eq'ing boost the hump frequencies, your driver gets into trouble very quickly, indeed.

That would be below the tuning frequency of the sub, and thus uncorrected by Audyssey.
There have been instances where the normalization have pushed gain below F3, though.
post #50418 of 70906
Quote:
Originally Posted by atledreier View Post

That would be below the tuning frequency of the sub, and thus uncorrected by Audyssey.
There have been instances where the normalization have pushed gain below F3, though.

I actually damaged a sub driver because of this, and spent a bit of time (as in `not a lot') looking into people's experiences, and I came away wih the impression that what I am looking at is not unique to me. So:

I have 2 identical ported subs with a 14.5Hz port tuning that use MFW-15/RSS390HF-4 drivers. The modelled excursion profiles of the two drivers is almost identical, with a hump through the 20ies Hz.

The subs are +/-3dB flat to below 12Hz in-room, and are MultEQ corrected there, too.


EDIT [correction]: the driver was damaged already; I just pushed it over the edge by eq'ing in the 20'ies Hz.

EDIT: the plot includes a HPF to prevent the excursion going through the roof around 10Hz.
LL
post #50419 of 70906
That looks like a design flaw if anything? Such excursion above tuning?
post #50420 of 70906
Quote:
Originally Posted by atledreier View Post

That looks like a design flaw if anything? Such excursion above tuning?

The plotted excursion is for 300W, which I never feed them and, besides, is beyond the thermal power limit of the MFW-15. The design is deliberate, but feel free to call it a flaw

EDIT: THD is <5% throughout all frequencies.
post #50421 of 70906
Just seems a bit weird, is all. I'd never tune a driver like that. But if it works...
post #50422 of 70906
Quote:
Originally Posted by atledreier View Post

Just seems a bit weird, is all. I'd never tune a driver like that. But if it works...

Seriously OT but how would you avoid a hump above the port tuning frequency? The only issue, as far as I can tell, is how big the design allows the hump to be.
post #50423 of 70906
Yeah, let's get back on topic.

Audyssey assume you have enough headroom to take advantage of their technology. Some don't and get burned.
post #50424 of 70906
Quote:
Originally Posted by atledreier View Post

Yeah, let's get back on topic.

Audyssey assume you have enough headroom to take advantage of their technology. Some don't and get burned.

Hmmmm interesting, that works.
post #50425 of 70906
Hello all, My apologies if this has already been covered.

I want to get some information on how to correctly measure SPL levels using a R/S Meter because the measurements I am getting are very different from the auto calibration.

Background Info
My motivation behind using a SPL meter is to verify proper trims. I always liked the way the Auddyssey would calibrate my room, but at times felt like the surrounds were not as enveloping and that the speaker pans were not as seamless as they should be.

Equipment
Analog R/S SPL Meter (Not sure of the Model number)
I am using Auddysey XT on a Denon 3808 (8 point Cal)
7.1 Setup using Klipsch RF-3 II, RC-62, RS-52 for front center and surrounds.
Emotiva XPA-5 on Front, Center, SL & SR
22'L x 10.5'W x 7.5'H room with a 1.5 long 8' drop ceiling that runs the width of the room in the center. (MLP is about 1-2 ft behind this drop ceiling.)

Testing
I mount the SPL meter on a microphone stand and place it at the MLP. Using internal test tones I adjust the trims so that the meter reads 75dm (C-Weighted). Should I aim the SPL meter directly at all speakers, or should I aim it upwards ala the auddyssey mic? Specifically, should I aim the SPL meter at the bi-pole / Dipole speakers?

Side questions
Is it better to use internal test tones, or tones off a Disc? I have the AIX Calibration Blu-Ray

Thank you for your time.
post #50426 of 70906
Quote:
Originally Posted by kgveteran View Post

here is the HD graph that depicts Audyssey trying to boost the null..


check out the 5% distortion at 55hz..bingo.

How does one read distortion from this plot? I see no data labeled distortion.
post #50427 of 70906
Quote:
Originally Posted by Rumble Devo View Post


Testing
I mount the SPL meter on a microphone stand and place it at the MLP. Using internal test tones I adjust the trims so that the meter reads 75dm (C-Weighted). Should I aim the SPL meter directly at all speakers, or should I aim it upwards ala the auddyssey mic? Specifically, should I aim the SPL meter at the bi-pole / Dipole speakers?

Side questions
Is it better to use internal test tones, or tones off a Disc? I have the AIX Calibration Blu-Ray

Thank you for your time.

The SPL should be placed at the MLP, with height at ear-level, with the SPL pointed at the ceiling.

It is not uncommon for the SPL level readings to differ from 75 dB, which is what the Audyssey calibration should be setting your levels at. This is because the SPL is likely to be less accurate than the Audyssey mic. The important thing to check with the SPL is the relative levels across all your speakers. The SPL should show the same level for all speakers, within approx 1 dB. If this is the case, I recommend trusting the Audyssey levels and not changing anything.
post #50428 of 70906
Quote:
Originally Posted by Rumble Devo View Post

I want to get some information on how to correctly measure SPL levels using a R/S Meter because the measurements I am getting are very different from the auto calibration.

That can happen, as the methods are different.

Quote:
Background Info
My motivation behind using a SPL meter is to verify proper trims. I always liked the way the Auddyssey would calibrate my room, but at times felt like the surrounds were not as enveloping and that the speaker pans were not as seamless as they should be.

I find it is hard to get seamless pans with dipoles because they sound so different spectrally and spatially than the L/C/R direct radiators. But they should be good on envelopment, so maybe we are talking about a level problem.

Quote:
Equipment
Analog R/S SPL Meter (Not sure of the Model number)

We have used R/S analog meters for decades at Dolby, and I still use them in my theater, and find them totally reliable for balancing speakers. If all the levels are equal, but differ from Audyssey, no need to adjust, as it's only overall loudness at issue.


Quote:
Testing
I mount the SPL meter on a microphone stand and place it at the MLP. Using internal test tones I adjust the trims so that the meter reads 75dm (C-Weighted). Should I aim the SPL meter directly at all speakers, or should I aim it upwards ala the auddyssey mic? Specifically, should I aim the SPL meter at the bi-pole / Dipole speakers?

If it makes very little or no difference, then it doesn't matter how you aim. I find it makes a little bit of difference, not much, but I am not using dipole/bipoles.

If it does make a few dB difference, then aim the mic toward each speaker.

Quote:
Side questions
Is it better to use internal test tones, or tones off a Disc? I have the AIX Calibration Blu-Ray

Usually, internal tones are best, and some discs are equally good. But not AIX. It uses wideband noise, which is pretty useless for channel balance cals because response errors at freq extremes will influence the levels, which are best matched in the midrange. That's why test tones use filtered pink noise. As an aside, the AIX tones are "85 dB" while all others use 75 dB. I guess Mr. Waldrep didn't get the memo.
post #50429 of 70906
Quote:
Originally Posted by Rumble Devo View Post

Hello all, My apologies if this has already been covered.

I want to get some information on how to correctly measure SPL levels using a R/S Meter because the measurements I am getting are very different from the auto calibration.

Background Info
My motivation behind using a SPL meter is to verify proper trims. I always liked the way the Auddyssey would calibrate my room, but at times felt like the surrounds were not as enveloping and that the speaker pans were not as seamless as they should be.

Equipment
Analog R/S SPL Meter (Not sure of the Model number)
I am using Auddysey XT on a Denon 3808 (8 point Cal)
7.1 Setup using Klipsch RF-3 II, RC-62, RS-52 for front center and surrounds.
Emotiva XPA-5 on Front, Center, SL & SR
22'L x 10.5'W x 7.5'H room with a 1.5 long 8' drop ceiling that runs the width of the room in the center. (MLP is about 1-2 ft behind this drop ceiling.)

Testing
I mount the SPL meter on a microphone stand and place it at the MLP. Using internal test tones I adjust the trims so that the meter reads 75dm (C-Weighted). Should I aim the SPL meter directly at all speakers, or should I aim it upwards ala the auddyssey mic? Specifically, should I aim the SPL meter at the bi-pole / Dipole speakers?

Side questions
Is it better to use internal test tones, or tones off a Disc? I have the AIX Calibration Blu-Ray

Thank you for your time.

It's very unlikely that an inexpensive consumer SPL meter will give a reading as accurate as Audyssey does so I would tend to rely more on the Audyssey settings for the trims than a Rat Shack meter. What trim levels is Audyssey setting for you?

It is also very difficult to put the SPL meter in the exact same spot as you had the Audyssey mic in.

The Rat Shack meter is good for looking at relative levels (this speaker is +2dB compared with that speaker) but not so good at setting absolute levels (this speaker is reading 72dB). The meter is also more or less entirely useless for setting subwoofer levels.

Of course, personal preference comes into it and if you feel the need to raise the trim levels of your surrounds, then by all means do so - you don't need a Rat Shack meter for that - just go into settings and bump them up a couple of dB to suit your taste.

The Rat Shack meter should be placed in the exact spot that the Audyssey mic was in for the first test position and pointed towards the ceiling. Any other position or orientation will give a skewed, and therefore useless, result.
post #50430 of 70906
Quote:
Originally Posted by atledreier View Post

Audyssey assume you have enough headroom to take advantage of their technology. Some don't and get burned.

IIRC that's not the case, according to what Chris has said, i.e., Audyssey doesn't correct below the low freq rolloff point so as to not overdrive the woofers.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #51779)