AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #1)
New Posts  All Forums:Forum Nav:

"Official" Audyssey thread (FAQ in post #1) - Page 199

post #5941 of 62279
8/21/08 at 2:16pm
Quote:
Originally Posted by sterryo View Post

Looking at the bio's from the Audyssey "about us", it appears that most of the major players are no longer youngsters and they must be suffering the aging effects on their ears also (join that old fart's club--huh what did u say? sonny) and I'm sure that they want to hear the best sound from their home systems also. So let's look for some innovation for us with deteriorating hearing.


Come on Chris.
All we're asking for is a "simple" diy firmware upgrade where we old farts can dial in our age and Audyssey will provide the appropiately calibrated curve.

Maybe you could simplify the settings to "Teen", "Middle Age", and "What? Speak up!"
Reply
Reply

Gear mentioned in this thread:

post #5942 of 62279
8/21/08 at 2:24pm
Quote:
Originally Posted by Erik Garci View Post

True, it will be passed, but my receiver will make it 5.5dB lower, according to my measurements. So I'm asking if there is a good reason to make it 5.5dB lower during playback.

If a filter has a nominated cutoff, say 120Hz, then the response should be 3dB down at 120Hz, and then to continue to fall off at a predetermined curve e.g. 24 dB/octave.
I can see anything wrong with having the LPF set at 250Hz, so long as it doesn't introduce any artifacts between 120 and 250Hz.
Reply
Reply
post #5943 of 62279
8/21/08 at 2:29pm
Quote:
Originally Posted by LarryChanin View Post

Hi,

Might I remind you that home theater is not merely a solitary listener planted in the "sweet spot" in the dark. Audyssey is supposed to improve the sound quality of the audience, i.e., a group of listeners, seated in different locations.

So I guess those of us who are older enthusiasts should forget about the listening pleasure of the rest of the audience and adjust the frequency response to suit our hearing? Seems rather ironic to go to the trouble of improving the sound in most seating locations only to destroy it by selfishly catering to our own hearing deficiencies.

Perhaps Audyssey should get into the hearing aide business and tailor a EQ solution to individuals without compromising the sound quality delivered to the entire audience.

Larry

I guess your right Larry, imagine if Hugh Heffner set his system for his ears only.
Reply
Reply
post #5944 of 62279
8/21/08 at 2:42pm
Friends

I am getting ready to order the Audyssey Pro Installer End-user kit. Do any of you know positively if the soon to be out Onkyo TX-NR906 can or can NOT be Audyssey Pro calibrated? I want to get the most out of Audyssey and the 8 mic positions will not cut it.

I want to assume it IS pro capable but you know how assumptions go. The Audyssey rep I am emailing with about joining them does not know for sure since the receiver is not out yet.

Jerry
Reply
Reply
post #5945 of 62279
8/21/08 at 2:53pm
Quote:
Originally Posted by Horta View Post

Friends

I am getting ready to order the Audyssey Pro Installer End-user kit. Do any of you know positively if the soon to be out Onkyo TX-NR906 can or can NOT be Audyssey Pro calibrated? I want to get the most out of Audyssey and the 8 mic positions will not cut it.

I want to assume it IS pro capable but you know how assumptions go. The Audyssey rep I am emailing with about joining them does not know for sure since the receiver is not out yet.

Jerry

You might want to wait on the kit purchase until after CEDIA the 1st week in Sept because they will be announcing an upgraded kit.
Reply
Reply
post #5946 of 62279
8/21/08 at 2:55pm
I have an Onkyo TX-SR875 receiver. My speakers are KEF 107s as Mains (18hz-20khz), KEF 100 as Center, KEF Coda 7s as surrounds, Bose 901s VI as Rear Surrounds and dual MFW-15s powered subs. When I run Audyssey, it sets my mains to full, center to 70hz, surrounds to 90hz and rear surrounds to 60hz. When I review the speaker settings, it says that I have "Double Bass". Reading the Onkyo manual, it says that "Double Bass" shares the bass between the Mains and the Sub(s). The manual doesn't go into any detail as to what the specs are on "Double Bass", ie crossovers, etc. My first questions is: Does anyone have a good explanation as to how "Double Bass" works? If this is not the optimum setting, what would be? My mains and subs face the narrow part of my room which is 15' wide (x 25' long x 12" high). The receiver is THX Ultr-2. Thanks for your help.
Reply
Reply
post #5947 of 62279
8/21/08 at 3:27pm
Quote:
Originally Posted by Horta View Post

Friends

I am getting ready to order the Audyssey Pro Installer End-user kit. Do any of you know positively if the soon to be out Onkyo TX-NR906 can or can NOT be Audyssey Pro calibrated? I want to get the most out of Audyssey and the 8 mic positions will not cut it.

I want to assume it IS pro capable but you know how assumptions go. The Audyssey rep I am emailing with about joining them does not know for sure since the receiver is not out yet.

Jerry

Jerry,

We can not make announcements about features on upcoming products until they are made public by the manufacturer. That would put us in violation of our NDA.

Thanks for understanding.

Chris
Reply
Reply
post #5948 of 62279
8/21/08 at 5:44pm
Quote:
Originally Posted by Nordo View Post

I guess your right Larry, imagine if Hugh Heffner set his system for his ears only.

Hi,

Very imaginative example. I have to admit it does support my personal view that ideally the home theater experience is really a social event, and as such, inherently deals with groups of people, not solitary listeners. Sure, Audyssey can improve a solitary listening experience, but what distinguishes it from simple parametric equalization is its ability to improve the listening experience of an audience.

Larry
Reply
Reply
post #5949 of 62279
8/21/08 at 7:11pm
Quote:
Originally Posted by audyssey View Post

How far below the tweeters of the front three channels will the ears be when you sit on the floor? Typically, we don't recommend taking measurements for floor sitters, but it really depends on the speaker setup.

Chris

Hi Chris,

The tweeters for the center channel will be 9 inches below ear height (it is a Definitive CLR 2000 audyssey measures it as full range, the speaker is in a cabinet) and the left and right speakers tweeter will be 6 inches above ear height while sitting on the floor (Definitive BP30 audyssey measures them at full range).
Reply
Reply
post #5950 of 62279
8/21/08 at 9:00pm
Quote:
Originally Posted by audyssey View Post

The MultEQ delay is 0 incrementally. That is because of the way DSP frameworks are set up. The audio processing through a DSP is performed in blocks rather than sample by sample. So, there is a fixed input to output delay that is directly related to the processing block size. You incur that delay the minute you put a DSP in the path, even if no algorithms are running on it. Then the architecture is such that any filtering algorithm such as MultEQ that runs on the DSP fits within the existing processing delay so it doesn't add anything more. Typical input to output delays for high end DSP chips are on the order of 10-15 ms.

Dynamic Volume is a different story. One of the innovations is that in the way it employs a look ahead so that it can assess the perceived loudness of the current and upcoming audio. There is some additional delay associated with that, but it has been optimized so as to not be perceptible given the bigger delays incurred because of the video processing.

It seems to me that the total audio delay through many receivers is borderline at the moment, and the manufacturers have to work to reduce it, not add more delay. Often the audio path goes through 2 (and sometimes 3 or 4) DSPs, and the delay is summed. I've measure over 40ms through my Yamaha, and some recent Onkyo's are over 70ms and causing a lot of complaints over noticable audio delay.

And while the audio delay is going up because of more sophisticated processing and multiple DSPs, displays are actually getting faster and the video delay is coming down (because of 120Hz refresh and the drive to reduce delays for video gamers). An adjustable video delay somewhere (source, or receiver or display) would help keep movies in sync, but it wouldn't help the gamers who want their absolute latency to be a minimum (and gamers want receivers and surround sound these days).

I think DSP processing needs to be done with significantly shorter blocks (I can certainly imagine a marketing advantage for a "low latency" receiver). Would that cause a problem for MultEQ or Dynamic Volume? What's the shortest delay you could work with?
Reply
Reply
post #5951 of 62279
8/21/08 at 10:38pm
I don't think 120 Hz refresh rate has anything to do with the video processing speed. It has to do with how the processed video signal is displayed and all of the video processing has already occurred by that stage.

Most modern receivers do have an option for video delay in order to achieve lipsync. I know my Denon 2807 has and it's now 2 years old and had been around for quite a while when I bought it so it's not a recent receiver and the feature was common, or becoming common, when the 2807 was released. People just have to set things up properly. It's not particularly hard but you do have to know about these features if you're ever going to think of using them. The old RTFM acronym applies but unfortunately many people don't do that while the components are becoming more feature ridden and sophisticated so many people miss things in their setup which they really shouldn't.
Reply
Reply
post #5952 of 62279
8/22/08 at 2:24am
Quote:
Originally Posted by LarryChanin View Post

Hi,

Very imaginative example. I have to admit it does support my personal view that ideally the home theater experience is really a social event, and as such, inherently deals with groups of people, not solitary listeners. Sure, Audyssey can improve a solitary listening experience, but what distinguishes it from simple parametric equalization is its ability to improve the listening experience of an audience.

Larry

Your right. After all, the whole purpose of Audyssey is to produce the best equalising possible for the maximum number of people in a typical home environment.
Reply
Reply
post #5953 of 62279
8/22/08 at 2:50am
Quote:
Originally Posted by Charles R View Post

My fronts have built-in powered subwoofers and Audyssey has set them from Full to 150Hz (perhaps even higher once) all based on the microphone's placement.

Interesting speakers, what are they ?

Just a thought, but perhaps the active subwoofer component is slightly out-of-phase with the passive bass/mid/treble ? This could cause all sorts of cancellation effects at the point of crossover that would appear to be location/listering position related - but I assume, unless they are home made - the manufacturer would have done something to remedy this in the design ?
Reply
Reply
post #5954 of 62279
8/22/08 at 5:07am
Quote:
Originally Posted by David Aiken View Post

Most modern receivers do have an option for video delay in order to achieve lipsync.

There is also the Lip Sync option in HDMI 1.3. All devices must include the option though because any delay is calculated in negotiation between devices. Not many displays include it it yet. I wonder if Audyssey's delay works outside of this.
Reply
Reply
post #5955 of 62279
8/22/08 at 5:44am
Quote:
Originally Posted by Gary J View Post

There is also the Lip Sync option in HDMI 1.3. All devices must include the option though because any delay is calculated in negotiation between devices. Not many displays include it it yet. I wonder if Audyssey's delay works outside of this.

HDMI auto lip sync is pretty useless. For one thing it assumes the video delay is more than the audio delay, when in fact that's no longer true in all cases. It can only add more audio delay, which is not always what you want (e.g. when playing video games with surround processing on some receivers, with the TV in a low latency "game mode", the sound will be lagging noticably behind the video). And for gamers, delaying the video to match the audio is not an option, the audio delay really needs to come back down to where it was a couple of generations of receiver ago (before they started to use multiple DSPs chained together).

But while much of the basic DSP processing could easily work with less delay, some algorithms need a certain minimum delay (e.g. FIR filters with a lot of taps, as used by MultEQ).
Reply
Reply
post #5956 of 62279
8/22/08 at 6:02am
Quote:
Originally Posted by David Aiken View Post

I don't think 120 Hz refresh rate has anything to do with the video processing speed. It has to do with how the processed video signal is displayed and all of the video processing has already occurred by that stage.

A lot of video processing is frame based, and the total delay ends up being a certain number of frames. The delay in milliseconds will be less on displays that process and display at a faster frame rate. On top of that we're now seeing special "game modes" on TVs that skip some of the processing in order to reduce the delay.
Quote:


Most modern receivers do have an option for video delay in order to achieve lipsync. I know my Denon 2807 has and it's now 2 years old and had been around for quite a while when I bought it so it's not a recent receiver and the feature was common, or becoming common, when the 2807 was released. People just have to set things up properly. It's not particularly hard but you do have to know about these features if you're ever going to think of using them. The old RTFM acronym applies but unfortunately many people don't do that while the components are becoming more feature ridden and sophisticated so many people miss things in their setup which they really shouldn't.

I think you're mistaken about the video delay on the 2807. Almost all receivers only allow for an additional audio delay to improve lip sync. Some people have genuine complaints about the audio delay, and there's currently no way to correct it in all cases.

I guess receivers could have their "game mode" too, which skips most of the processing. A similar mode exists now on many receivers (e.g. "Pure Direct" on Yamaha), but that skips DSP processing entirely and so you lose bass management as well as EQ. Giving up EQ for low latency might be a trade off gamers would accept, but they wouldn't want to lose bass management. And most normal users wouldn't want to give up any features just to get perfect lip sync. If all the processing could fit in the 10-15ms Chris mentioned that would be great, but it's not happening with the latest generation of receivers.
Reply
Reply
post #5957 of 62279
8/22/08 at 6:20am
Quote:
Originally Posted by kriktsemaj99 View Post

HDMI auto lip sync is pretty useless. For one thing it assumes the video delay is more than the audio delay, when in fact that's no longer true in all cases.

No just the vast majority of cases. You seem to be gaming oriented. Also in a AVR I have HDMI Lip Sync is an option. It does not have to be turned on.
Reply
Reply
post #5958 of 62279
8/22/08 at 6:56am
Quote:
Originally Posted by audyssey View Post

EXT. IN multichannel input. Audio coming through there does not go through the DSP and therefore does not see bass management, MultEQ, or any other digital process.
Chris

Are there any receivers where this would not be the case? I don't see the usefulness of the whole EXT. IN, if there is no bass management! Or is this an Audyssey policy not to allow Audyssey processing on the multi-channel analog inputs?
Reply
Reply
post #5959 of 62279
8/22/08 at 7:08am
Quote:
Originally Posted by Jarno View Post

Are there any receivers where this would not be the case? I don't see the usefulness of the whole EXT. IN, if there is no bass management! Or is this an Audyssey policy not to allow Audyssey processing on the multi-channel analog inputs?

The issue is simply whether or not the AVR or pre/pro has ADCs to digitize the multichannel analog input. Some AVRs do (some Denons and, perhaps, others) and so do some pre/pros (I have an Anthem D2 which does it very well and a Meridian 861 where it is the only way in).
Reply
Reply
post #5960 of 62279
8/22/08 at 7:19am
Quote:
Originally Posted by Jarno View Post

Are there any receivers where this would not be the case? I don't see the usefulness of the whole EXT. IN, if there is no bass management! Or is this an Audyssey policy not to allow Audyssey processing on the multi-channel analog inputs?

The multichannel external analog input arose as a way of playing DVD-A and SACD before AVR's and pre/pro's had the abilty to decode the formats. And with the newer Dolby and DTS codecs and yesterday's AVR's and pre/pro's (again) not having onboard decoding, the input again has a raison d'être.

The first time around, the input was a simple bypass of everything but the volume control. At that time, most processing including bass management was digital. Processing the analog bypass would require digitization which was something that most early gear did not offer. Soon, rudimentary bass management was included in DVD/SACD players.

Now most - all? - AVR's and pre/pro's can accept multichannel inputs decoded in external players via HDMI . . in the digital domain. IN this case, bass management *is* done in the AVR or pre/pro.

Not processing an analog mc input has nothing to do with Audyssey policies.
Reply
Reply
post #5961 of 62279
8/22/08 at 7:48am
Quote:
Originally Posted by kriktsemaj99 View Post

HDMI auto lip sync is pretty useless. For one thing it assumes the video delay is more than the audio delay, when in fact that's no longer true in all cases.

Quote:
Originally Posted by Gary J View Post

No just the vast majority of cases.

The danger is the audio delay will be a problem for everyone soon if we don't see change in the trend of diminishing video delay and increasing audio delay. The receiver manufacturers (and Audyssey etc.) need to be aware that end-to-end audio delay can't be increased any further, and in fact it really needs to be reduced.

Many people are still using older generation receivers that don't have this problem. Right now I've only heard of the new Onkyos (805 etc) actually being returned for their unacceptable audio delay, but it's borderline in many others (my Yamaha RX-V1800 is OK for movies when set to 0ms delay, but I sometimes wish it could go lower). And I'm not only talking about gaming, I was just saying that this (large) group of people are the most demanding on latency, and they increasingly are using surround receivers instead of just TV speakers.

Unfortunately what often happens is that the industry doesn't see the problem coming and only reacts afterwards, leaving a whole bunch of people with a problem they can't fix without buying new equipment.
Reply
Reply
post #5962 of 62279
8/22/08 at 7:59am
Quote:
Originally Posted by kriktsemaj99 View Post

The danger is the audio delay will be a problem for everyone soon if we don't see change in the trend of diminishing video delay and increasing audio delay. The receiver manufacturers (and Audyssey etc.) need to be aware that end-to-end audio delay can't be increased any further, and in fact it really needs to be reduced.

Do you think that this is a temporary situation? Video has orders of magnitude more data to process than audio, but video signal processors have been where the action is for a while now and they are f-a-s-t. Perhaps the next generation of audio signal processors will restore balance?
Reply
Reply
post #5963 of 62279
8/22/08 at 8:35am
Quote:
Originally Posted by kriktsemaj99 View Post

the trend of diminishing video delay

Again this is only in gaming. In recent history we have HDTV with 2 million pixels to process. We have video processors being put in all kinds of devices. The DVDO processors are putting 5 fields in a buffer, analyzing them, applying picture controls, de-interlacing them and scaling them. That is a lot of processing.

I think Chris said it best - "There is some additional delay associated with that, but it has been optimized so as to not be perceptible given the bigger delays incurred because of the video processing."
Reply
Reply
post #5964 of 62279
8/22/08 at 8:51am
Thank you everyone again for the replies. For those that have followed my saga.....

Come to find out from the head engineer of my speakers that my speakers had no dacron stuffing. No wonder they sounded like the rapp'n granny
The manufacturer is going to take care of me, so no gripes with them.


This is my first home theater system and I knew something sounded wrong yet I persisted to tell myself that these are quality speakers and receiver..... I must be doing something wrong.

Score:
Audyssey 1
Speakers 0


Quote:
Originally Posted by audyssey View Post

There are two possible explanations for this:

1) Your front speakers are placed in a location that gives a huge bump at 200 Hz or huge dip a little lower (150 Hz or so). This can happen because of a cabinet or sometimes from the interaction of the speaker with the three walls in the corner. The algorithm sees that and believes that the speaker roll-off is where the downward slope of the bump is.

2) Your front woofers are disconnected or damaged and they are not producing bass. This is easy to check. Sometimes on biwired speakers the connectors on the back come loose, or there may be a disconnected wire somewhere.

Chris
Reply
Reply
post #5965 of 62279
8/22/08 at 9:55am
Quote:
Originally Posted by Gary J View Post

Again this is only in gaming. In recent history we have HDTV with 2 million pixels to process. We have video processors being put in all kinds of devices. The DVDO processors are putting 5 fields in a buffer, analyzing them, applying picture controls, de-interlacing them and scaling them. That is a lot of processing.

I think Chris said it best - "There is some additional delay associated with that, but it has been optimized so as to not be perceptible given the bigger delays incurred because of the video processing."

I don't want to derail the thread, but I do think the discussion is relevant because Audyssey algorithms introduce added latency.

The days of putting 5 fields in a buffer for complex de-interlacing and scaling will soon be behind us. We'll mostly be feeding pristine progressive sources straight to the display with no need for scaling or noise reduction etc. Display refresh will be 120Hz or higher, and the video delay will be very small.

The last few years we got used to a much bigger video delay, and noone worried about audio delay because everybody assumed the video processing took longer. Receivers have been designed around that assumption, and multiple DSPs were just cascaded one after another because it was the easiest way to add more processing power. MultEQ or Dynamic Volume might have an acceptable latency of their own, but that might not be good enough in a badly designed receiver that adds a ton of extra delay. And what if the whole industry decides that "low-latency" is the new buzzword? Do they have to compromise on the EQ algorithms to make them run in 5 or 10ms?
Reply
Reply
post #5966 of 62279
8/22/08 at 10:14am
"I am getting ready to order the Audyssey Pro Installer End-user kit."

Chris, is one available now to consumers?

If so, what are the differences, if any, from the Pro kit?
Reply
Reply
post #5967 of 62279
8/22/08 at 10:18am
Quote:
Originally Posted by noah katz View Post

"I am getting ready to order the Audyssey Pro Installer End-user kit."

Chris, is one available now to consumers?

If so, what are the differences, if any, from the Pro kit?

I don't recall that "End User" is in the kit name. Is it?
Reply
Reply
post #5968 of 62279
8/22/08 at 10:36am
Quote:
Originally Posted by kriktsemaj99 View Post

I don't want to derail the thread, but I do think the discussion is relevant because Audyssey algorithms introduce added latency.

The days of putting 5 fields in a buffer for complex de-interlacing and scaling will soon be behind us. We'll mostly be feeding pristine progressive sources straight to the display with no need for scaling or noise reduction etc. Display refresh will be 120Hz or higher, and the video delay will be very small.

The last few years we got used to a much bigger video delay, and noone worried about audio delay because everybody assumed the video processing took longer. Receivers have been designed around that assumption, and multiple DSPs were just cascaded one after another because it was the easiest way to add more processing power. MultEQ or Dynamic Volume might have an acceptable latency of their own, but that might not be good enough in a badly designed receiver that adds a ton of extra delay. And what if the whole industry decides that "low-latency" is the new buzzword? Do they have to compromise on the EQ algorithms to make them run in 5 or 10ms?

I'm not sure how much clearer I can make this. The DSP architecture introduces 15-20 ms of latency depending on the chip model. This has NOTHING to do with Audyssey. You can turn off every digital process including PLIIx, bass management, and everything else and you will still get 10-20 ms if your signal goes through a DSP chip. Always. Every time. Even when Audyssey is off. The only way to avoid the latency is to not send your signal through the DSP chip.

The issue of delay in DSP chips also has nothing to do with a badly designed receiver. It is a matter of resources. Typical high end DSP chips have about 300-400 MIPS. When you add up all the digital algorithms required to make a complex AVR product, you will find that they require 600-800 MIPS, particularly if you want to support the HD audio codecs. So, the only solution is to cascade two chips. That doubles the latency as kriktsemaj99 correctly points out. Some of the very high end models even have three DSP chips cascaded.

So, unless we see a sudden move to analog circuitry and give up on all advanced audio signal processing I'm afraid that the DSP chip delay is here to stay. Faster DSP chips come out every year, but that doesn't affect the throughput delay. It only allows them to run more algorithms on one chip so that will bring some benefit as the total number of chips is reduced. But don't hold your breath. As the available MIPS go up so will the available algorithms coming out from the many companies competing for real estate on these chips.

Chris
Reply
Reply
post #5969 of 62279
8/22/08 at 10:41am
Quote:
Originally Posted by noah katz View Post

"I am getting ready to order the Audyssey Pro Installer End-user kit."

Chris, is one available now to consumers?

If so, what are the differences, if any, from the Pro kit?

The Audyssey policy is to first put you in touch with a local installer that is qualified to perform MultEQ Pro calibrations. In some cases this has not been possible and we have sold the kit directly to consumers. The kit is identical, but as a consumer you can only use it for the product that you own. You can't perform calibrations for your friends and family. Also, the price for consumers is significantly higher as we are not really in the business of stocking and selling kits. These are normally made available to professional installers through our distribution network.

Chris
Reply
Reply
post #5970 of 62279
8/22/08 at 10:44am
Quote:
Originally Posted by audyssey View Post

I'm not sure how much clearer I can make this. The DSP architecture introduces 15-20 ms of latency depending on the chip model. This has NOTHING to do with Audyssey. You can turn off every digital process including PLIIx, bass management, and everything else and you will still get 10-20 ms if your signal goes through a DSP chip. Always. Every time. Even when Audyssey is off. The only way to avoid the latency is to not send your signal through the DSP chip.

The issue of delay in DSP chips also has nothing to do with a badly designed receiver. It is a matter of resources. Typical high end DSP chips have about 300-400 MIPS. When you add up all the digital algorithms required to make a complex AVR product, you will find that they require 600-800 MIPS, particularly if you want to support the HD audio codecs. So, the only solution is to cascade two chips. That doubles the latency as kriktsemaj99 correctly points out. Some of the very high end models even have three DSP chips cascaded.

So, unless we see a sudden move to analog circuitry and give up on all advanced audio signal processing I'm afraid that the DSP chip delay is here to stay. Faster DSP chips come out every year, but that doesn't affect the throughput delay. It only allows them to run more algorithms on one chip so that will bring some benefit as the total number of chips is reduced. But don't hold your breath. As the available MIPS go up so will the available algorithms coming out from the many companies competing for real estate on these chips.

Chris

OK. So what is the latency of the SEQ?
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors

Gear mentioned in this thread:

AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #1)