Sherwood R-972 User Thread - Page 115 - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!



Forum Jump: 
 8Likes
Reply
 
Thread Tools
post #3421 of 3450 Old 04-13-2016, 02:50 PM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Sounds like the beginning of the end.

Start saving for a new receiver or preamp
1201 is online now  
Sponsored Links
Advertisement
 
post #3422 of 3450 Old 04-13-2016, 03:34 PM
Member
 
xjr15's Avatar
 
Join Date: Nov 2009
Posts: 86
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 15
Don't worry - I have a spare.

Quote:
Originally Posted by 1201 View Post
Sounds like the beginning of the end.

Start saving for a new receiver or preamp

Don't worry -- I have a spare R-972 I got on eBay a few weeks back. I NEED the Trinnov -- the remapping is such a HUGE benefit -- I have never heard anything that comes close for surround music and movies. But R-972 setup for me would also mean redoing the setup for a Velodyne SMS-1 and for a Smyth A8 (binaural emulation of multichannel over stereo headphones -- I live in an apartment) -- and setting up the R-972 alone was a nightmare that I want to avoid reliving.

When the Trinnov remapping in the R-972 works, it REALLY works, but getting it to work is not necessarily easy -- or possible.

P.S. Watched Love & Mercy BD last night -- surround effects were to die for.

Last edited by xjr15; 04-13-2016 at 03:47 PM.
xjr15 is offline  
post #3423 of 3450 Old 04-13-2016, 06:18 PM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by xjr15 View Post
Don't worry -- I have a spare R-972 I got on eBay a few weeks back. I NEED the Trinnov -- the remapping is such a HUGE benefit -- I have never heard anything that comes close for surround music and movies. But R-972 setup for me would also mean redoing the setup for a Velodyne SMS-1 and for a Smyth A8 (binaural emulation of multichannel over stereo headphones -- I live in an apartment) -- and setting up the R-972 alone was a nightmare that I want to avoid reliving.

When the Trinnov remapping in the R-972 works, it REALLY works, but getting it to work is not necessarily easy -- or possible.

P.S. Watched Love & Mercy BD last night -- surround effects were to die for.

the remapping never did anything for me, but the time correction was the cats meow. i tell the story that i bought the 972 for the kids room. i couldnt believe the sound quality after doing the measurement. i liked the sound in there better than the main room. after a while i borrowed it to the main room and it stayed there until it died.

thankfully, i learned about how wonderful time based correction is and found an affordable solution that does the same thing - and sounds the same - for peanuts.

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3424 of 3450 Old 04-13-2016, 07:30 PM
Member
 
xjr15's Avatar
 
Join Date: Nov 2009
Posts: 86
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
the remapping never did anything for me, but the time correction was the cats meow. i tell the story that i bought the 972 for the kids room. i couldnt believe the sound quality after doing the measurement. i liked the sound in there better than the main room. after a while i borrowed it to the main room and it stayed there until it died.

thankfully, i learned about how wonderful time based correction is and found an affordable solution that does the same thing - and sounds the same - for peanuts.
Time-based room correction is not unique to the R-972. I think Audyssey room correction does that and receivers with Audyssey are a dime a dozen. It's just another way of doing frequency response EQ. I don't like Audyssey at all, and Floyd Toole says he considers that most if not all such systems don't really work because they don't support enough data points. But for me -- switching Audyssey in and out -- if that's correction then I don't want it. It very audibly degraded the sound -- less detail, treble 'smear', etc., -- no contest -- the uncorrected sound was better.

The really audible benefit of the Trinnov remapping is the improvement it makes in the surround imaging -- the full 360 degree imaging you get with a good surround recording or soundtrack. I don't use the Room EQ in the R-972 but only the remapping. On many surround music recordings, the imaging is symmetric around the center line defined by the sweet spot and the center speaker but extended behind your head so you get a symmetric 360 degree 'ambisonic' soundstage. If you're not getting that, then the Trinnov setup didn't work right. I only got it to work with one of the three Trinnov 'positions' during R-972 setup, so whether it works or not is a crap shoot with the R-972.

When you experience the 3D remapping, you know it's right because of the left/right symmetry of the surround imaging on a good surround mix; you know you are finally hearing what the artist/engineer intended, because the precision and interaction of the left/right surround image and the left/right image symmetry cannot be an accident, and you will hear some of the images do clockwise and counterclockwise rotation that was not audible before the remapping.

A problem with all surround recordings, however, is that only one person (the one who gets to sit in the sweet spot) gets the intended effect and the correct imaging. In that respect it's like driving a sports car: only the driver can appreciate the handling and response of the car. For a home theater shared with family and friends, it's probably better to just stick to 3.1 audio and forget about the surround audio.

Last edited by xjr15; 04-15-2016 at 11:13 PM.
xjr15 is offline  
post #3425 of 3450 Old 04-14-2016, 07:05 AM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by xjr15 View Post
Time-based room correction is not unique to the R-972. I think Audyssey room correction does that and receivers with Audyssey are a dime a dozen. It's just another way of doing frequency response EQ. I don't like Audyssey at all, and Floyd Toole says he considers that most if not all such systems don't really work because they don't support enough data points. But for me -- switching Audyssey in and out -- if that's correction then I don't want it. It very audibly degraded the sound -- less detail, treble 'smear', etc., -- no contest -- the uncorrected sound was better.

The really audible benefit of the Trinnov remapping is the improvement it makes in the surround imaging -- the full 360 degree imaging you get with a good surround recording or soundtrack. I don't use the Room EQ in the R-972 but only the remapping. On many surround music recordings, the imaging is symmetric around the center line defined by the sweet spot and the center speaker but extended behind your head so you get a symmetric 360 degree 'ambisonic' soundstage. If you're not getting that, then the Trinnov setup didn't work right. I only got it to work with one of the three Trinnov 'positions' during R-972 setup, so whether it works or not is a crap shoot with the R-972.

The problem with all surround recordings, however, is that only one person (the one who gets to sit in the sweet spot) gets the intended effect and the correct imaging. In that respect it's like driving a sports car: only the driver can appreciate the handling and response of the car. For a home theater shared with family and friends, it's probably better to just stick to 2.1 audio and forget about the surround audio.
I agree with the surround being a distraction for everyone but the person in the sweet spot. that's why I have "only" a 3.1 system.


sit anywhere except the middle and you get only one side of the surround so a 5.1 system sounds like a very distracting 4.1 system.


as for comparing trinnov to audyssey- not close. audyssey does not do nearly as much time based correction as the r972 does.


the highest version, multeq xt32 only has 512 points vs 1048 for the r972. that's a pretty big difference. also the unedittable curve of the audyssey is not pleasing to many people, and finally what it does to the sub is also lacking.


the box I am using ( minidsp 2x4 hd) has 2042 points (equivalent to 1021 at 48khz)
has fully editable eq curves, actually works in the time domain, and can be used to time correct multiple subs. it also supports up to 20 frequency based parametric eq points.
imo its superior to audyssey and with all due respect the r972 as well mainly because we I get to choose what curve I want and it has way more options.


finally, the remapping correction did nothing for me. I think maybe my speakers are in a traditional location. I like how it handled the surrounds back when I handled surrounds but mostly what I noticed was that it quieted them down about 6 db compared to recommendations

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.

Last edited by 1201; 04-14-2016 at 11:08 AM.
1201 is online now  
post #3426 of 3450 Old 04-14-2016, 03:57 PM
Member
 
xjr15's Avatar
 
Join Date: Nov 2009
Posts: 86
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
I agree with the surround being a distraction for everyone but the person in the sweet spot. that's why I have "only" a 3.1 system.


sit anywhere except the middle and you get only one side of the surround so a 5.1 system sounds like a very distracting 4.1 system.


as for comparing trinnov to audyssey- not close. audyssey does not do nearly as much time based correction as the r972 does.


the highest version, multeq xt32 only has 512 points vs 1048 for the r972. that's a pretty big difference. also the unedittable curve of the audyssey is not pleasing to many people, and finally what it does to the sub is also lacking.


the box I am using ( minidsp 2x4 hd) has 2042 points (equivalent to 1021 at 48khz)
has fully editable eq curves, actually works in the time domain, and can be used to time correct multiple subs. it also supports up to 20 frequency based parametric eq points.
imo its superior to audyssey and with all due respect the r972 as well mainly because we I get to choose what curve I want and it has way more options.


finally, the remapping correction did nothing for me. I think maybe my speakers are in a traditional location. I like how it handled the surrounds back when I handled surrounds but mostly what I noticed was that it quieted them down about 6 db compared to recommendations
By "data points", I meant INPUT data points. I think Audyssey XT32 only takes 32 points, not nearly enough for a credible correction. For doing a frequency response room correction, the ideal is probably a real-time averaging input so you wouldn't be restricted to a small number of discrete measurements.

But, the real value of the R-972 is the remapping function, not the frequency response Room Correction.

The level and content of the surround channels following Trinnov remapping will, of course, vary with the source material, but I have never gotten the kind of seamless, detail-rich imaging that I get with Trinnov remapping by using only level-matching and distance/time-delay (provided by all receivers) and by very careful manual positioning of the speakers. Trinnov doesn't necessarily lower the surround levels; in many cases, just the opposite happens, and the surround channels are more prominent than before, again, depending on the source material. The positioning of the speakers and the their interaction with the room and its contents (furniture, etc.) with respect to imaging is very complex (far too complex and fine-grained to be corrected by manual speaker positioning -- which is a hit-or-miss trial-and-error process), and small errors can have big consequences, so we need a comprehensive, TRANSFER-FUNCTION-BASED solution, which, by definition, automatically includes ALL relevant variables in the correction filters.
And that's what you get with Trinnov remapping.

Last edited by xjr15; 04-14-2016 at 05:28 PM.
xjr15 is offline  
post #3427 of 3450 Old 04-14-2016, 09:00 PM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
i liked the trinnov surround "remapping" because it basically lowered the surround speaker level to where you could barely hear them. in my kids room i used to sit about 2 feet away from one speaker and could barely hear it.

as far as surround level,i measured it with a rew(room eq wizard) sweep to be 6 dbs down.

imo, the big thing trinnov did was the 1048 point time based correction. nothing more or less.how do i know this? i turned off the remapping feature and it still sounded just as sweet. when i tried the dirac live demo i didnt think it could or would come close to trinnov but lo and behold it sounded even better than the r972. and it was stereo only so no remapping involved. now granted this was a software demo with no hardware limitations so they probably used 64000 taps to really market the product but it sounded amazing.

then finally the 972 died and i found the 2x4hd after a lot of searching,and guess what it produces the same rich, transparent sound as the r972 and the dirac live demo.

you know whats common among all three? time based correction.

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3428 of 3450 Old 04-15-2016, 12:17 AM
Member
 
xjr15's Avatar
 
Join Date: Nov 2009
Posts: 86
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
i liked the trinnov surround "remapping" because it basically lowered the surround speaker level to where you could barely hear them. in my kids room i used to sit about 2 feet away from one speaker and could barely hear it.

as far as surround level,i measured it with a rew(room eq wizard) sweep to be 6 dbs down.

imo, the big thing trinnov did was the 1048 point time based correction. nothing more or less.how do i know this? i turned off the remapping feature and it still sounded just as sweet. when i tried the dirac live demo i didnt think it could or would come close to trinnov but lo and behold it sounded even better than the r972. and it was stereo only so no remapping involved. now granted this was a software demo with no hardware limitations so they probably used 64000 taps to really market the product but it sounded amazing.

then finally the 972 died and i found the 2x4hd after a lot of searching,and guess what it produces the same rich, transparent sound as the r972 and the dirac live demo.

you know whats common among all three? time based correction.

I don't know why I am bothering to answer a post like this, as it contains, and is based on, so much misunderstanding and confusion, ...Where to begin? I'm NOT bothering to reply to the next one, if you post an 'answer ' to this reply.

Without understanding what you are measuring, measurement is worthless and meaningless. Trinnov remapping is not "lowering the surround speaker level to where you could barely hear them". You don't specify what the INPUT SIGNAL is. With much material, including most conventional classical music recordings, there will not be any imaging in the rear or to the sides and this is expected: the surrounds will only convey hall ambience and low-level audience noise if it is a live concert. Again, the level depends on the signal. If the remapping is setup correctly, the surround channels will often be higher in level than the front channels. Watch almost any recently released action movie BD and learn (unless the Trinnov remapping function is not working properly or set up properly in your unit). Again, a "room EQ sweep" will not tell you anything about Trinnov remapping because it's not the right kind of signal with which to test the remapping function.

Again, your obsession with what you call "time-based correction" is naïve in the extreme. "Correction" of what, exactly? It seems you are talking about frequency response correction filters based on impulse response tests, and that will work, but you have to have many data points (microphone positions) to do that properly, and the fact that something "sounds sweet/amazing/better" to you is irrelevant and says nothing.

Again, Room EQ is one thing, speaker location remapping quite another. Totally different, and incomparable, functions (But you don't seem to understand that). Before using electronic frequency response EQ, you should probably try, if you haven't already, some room treatments (Sonex foam, RPG diffusers, etc.)

The purpose of the remapping function is to correct imaging problems. Room EQ, on the other hand primarily corrects tonal imbalances. They address different problems and are not mutually exclusive or incompatible, but they are not comparable. But you seem to be confusing the two somehow.

Anyway, I hope you continue to experience your "rich, transparent sound" even if you don't understand the issues.
xjr15 is offline  
post #3429 of 3450 Old 04-15-2016, 04:43 AM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by xjr15 View Post
Again, your obsession with what you call "time-based correction" is naïve in the extreme. "Correction" of what, exactly? It seems you are talking about frequency response correction filters based on impulse response tests, and that will work, but you have to have many data points (microphone positions) to do that properly, and the fact that something "sounds sweet/amazing/better" to you is irrelevant and says nothing.

.
this may be the biggest case of pot calling kettle black that ive dealt with here at avsforum. LMAO.

so,in a few words ill show you the error of your ways.
"Trinnov Room Optimizer" does 2 things.

1. time based correction
2. remapping

ive said that it sounded just as good without remapping activated
you are saying it will sound just as good with remapping on but no time based correction
and to that I say bul###*@


good luck

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.

Last edited by 1201; 04-15-2016 at 04:50 AM.
1201 is online now  
post #3430 of 3450 Old 04-15-2016, 06:38 AM
Advanced Member
 
StevenLansing's Avatar
 
Join Date: Mar 2007
Posts: 650
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 15
Quote:
Originally Posted by xjr15 View Post
Time-based room correction is not unique to the R-972. I think Audyssey room correction does that and receivers with Audyssey are a dime a dozen. It's just another way of doing frequency response EQ. I don't like Audyssey at all, and Floyd Toole says he considers that most if not all such systems don't really work because they don't support enough data points.
Audyssey MultEQ only corrects some 'minor' issues in the time domain as a 'direct' result of it's frequency (magnitude) response correction, because it doesn't use a process to 'seperately' correct issues in the time domain independent of frequency, like Trinnov is capable of.

Quote:
Originally Posted by 1201 View Post
as for comparing trinnov to audyssey- not close. audyssey does not do nearly as much time based correction as the r972 does.
That's because it doesn't have the ability to independently correct time domain issues separately from frequency (magnitude) response issues. That's how minimum phase filters work. Correction in one domain influences the other.


Quote:
Originally Posted by 1201 View Post
the highest version, multeq xt32 only has 512 points vs 1048 for the r972. that's a pretty big difference.s
XT32 has 16,000 FIR filter 'taps' available for the full frequency response range for each main channel and the subwoofer output vs. 1024 FIR filter taps (from 300 Hz up to 20Khz and below 300 Hz, 3 or 4 IIR filters) available for each main channel and 7 IIR filters for the subwoofer output with the R972 version of Trinnov. As Curt Hoyt has pointed out before in this thread, it's not just about the amount of filter taps,but the 'quality' of the filters as well.
StevenLansing is offline  
post #3431 of 3450 Old 04-15-2016, 10:35 AM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by StevenLansing View Post


XT32 has 16,000 FIR filter 'taps' available for the full frequency response range for each main channel and the subwoofer output vs. 1024 FIR filter taps (from 300 Hz up to 20Khz and below 300 Hz, 3 or 4 IIR filters) available for each main channel and 7 IIR filters for the subwoofer output with the R972 version of Trinnov. As Curt Hoyt has pointed out before in this thread, it's not just about the amount of filter taps,but the 'quality' of the filters as well.

good post above. one thing I find hard to believe is that they have 16000 filters available. it would take an immense amount of processing power to do that in real time and I doubt the sharc processor can handle it, and the delay would be immense - 166 ms delay


do you perchance have anything from audyssey that says they have 16000 actual filters?


im thinking they start with 16000 filters and then boil it down to 512.
with DRC for example, it goes through all the calculations with a default of 64k filters then reduces it down to final desired filter length


64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3432 of 3450 Old 04-15-2016, 11:15 AM
Member
 
xjr15's Avatar
 
Join Date: Nov 2009
Posts: 86
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
this may be the biggest case of pot calling kettle black that ive dealt with here at avsforum. LMAO.

so,in a few words ill show you the error of your ways.
"Trinnov Room Optimizer" does 2 things.

1. time based correction
2. remapping

ive said that it sounded just as good without remapping activated
you are saying it will sound just as good with remapping on but no time based correction
and to that I say bul###*@


good luck

Infantile vulgarity, as in the above post, has no place in an adult discussion, and is generally evidence of
weakness and stupidity.

And "time based" correction is a meaningless term. "Time-base correction" refers to synchronization methods for analog recording media and is not relevant here. The correct term in the present context is "time-domain correction", not "time based correction".

All I am saying is that the R-972 room EQ did not work for me, and that the Audyssey room correction that I had tried also did not work for me, but that the R-972 speaker remapping was revelatory. If you're not noticing much of a change with the remapping, it may be that your R-972 was/is not working correctly, or it may be that the source material you listen to (for instance, classical music) will not benefit very much from the remapping. Whether (and to what extent) a particular room correction product is frequency-domain and/or time-domain correction is not an issue for me; if it sounds bad, I don't want it.

As I said above, when you experience the 3D remapping, you will know it's right because of the left/right symmetry of the surround imaging on a good surround mix; you know you are finally hearing what the artist/engineer intended, because the precision and interaction of the left/right sides of the surround image and the left/right image symmetry cannot be an accident, and you will hear some of the images do clockwise and counterclockwise rotation that was not audible before the remapping. This is remarkable and not at all subtle.

Conventional frequency-domain and time-domain room correction will, in comparison, be subtle and fairly minor. And as David Rich remarked in his review of the original Trinnov multichannel professional optimizer in the $ensible Sound, corrections to tonal balance are pointless if the imaging is wrong.

But buying a R-972 for "the kid's room" makes no sense, since the R-972 is all about optimizing the audio at the sweet spot; if you're using it for background music or to entertain the kids, that's like buying a Ferrari to run errands to the local supermarket.

Last edited by xjr15; 04-22-2016 at 10:44 PM.
xjr15 is offline  
post #3433 of 3450 Old 04-16-2016, 02:22 PM
Advanced Member
 
Sam1000's Avatar
 
Join Date: Nov 2005
Posts: 532
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 15
Need help...

I ran a calibration for a stereo pair (disconnected all other speakers per Curt's notes). The calibration process completed successfully and at the end of it, it powered down the unit as expected.
The problem is, it powers up only for couple of seconds and goes to standby mode. I did a factory reset, but that hasn't help. Is the unit bricked? or can I save it somehow?

Speakers: Pioneer/TAD S-1EX, S-7EX, Ascend HTM200
Pre/Pro: Bel Canto Pre-6, Sherwood 972
Amp: D-Sonic M3-4500, M3-1200-A
Source: SonyBDP-780, Hugo Chord DAC, Squeezelite on RPi
Sam1000 is offline  
post #3434 of 3450 Old 04-16-2016, 03:43 PM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by Sam1000 View Post
Need help...

I ran a calibration for a stereo pair (disconnected all other speakers per Curt's notes). The calibration process completed successfully and at the end of it, it powered down the unit as expected.
The problem is, it powers up only for couple of seconds and goes to standby mode. I did a factory reset, but that hasn't help. Is the unit bricked? or can I save it somehow?
the beginning of the end,or maybe the end.

start saving for a new receiver.

you could unscrew the top and make sure all the boards are fully seated.

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3435 of 3450 Old 04-16-2016, 05:18 PM
Advanced Member
 
Sam1000's Avatar
 
Join Date: Nov 2005
Posts: 532
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 50 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
the beginning of the end,or maybe the end.

start saving for a new receiver.

you could unscrew the top and make sure all the boards are fully seated.
The boards seem to be seated. Not sure if I trying to load a firmware would help..Need to find the latest firmware files..

Speakers: Pioneer/TAD S-1EX, S-7EX, Ascend HTM200
Pre/Pro: Bel Canto Pre-6, Sherwood 972
Amp: D-Sonic M3-4500, M3-1200-A
Source: SonyBDP-780, Hugo Chord DAC, Squeezelite on RPi
Sam1000 is offline  
post #3436 of 3450 Old 04-20-2016, 10:23 PM
Advanced Member
 
StevenLansing's Avatar
 
Join Date: Mar 2007
Posts: 650
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
good post above. one thing I find hard to believe is that they have 16000 filters available. it would take an immense amount of processing power to do that in real time and I doubt the sharc processor can handle it, and the delay would be immense - 166 ms delay


do you perchance have anything from audyssey that says they have 16000 actual filters?


im thinking they start with 16000 filters and then boil it down to 512.
with DRC for example, it goes through all the calculations with a default of 64k filters then reduces it down to final desired filter length



Sorry for the late response,

That Audyssey chart refers to how many 'times' more correction each flavor of MultEQ has over 2EQ.Those aren't filter tap numbers. Audyssey won't release how many filter taps each version actually uses,however the CTO has mentioned in the past that the MultEQ XT version in AVR's and certain processors used 512 filter taps for the main channels and 4x that for the subwoofer. It was also stated that MultEQ XT32 has 32x more resolution than XT for the main channel correction and uses the same amount of correction resolution as the mains for the subwoofer channel.

The math and measurements taken showing filter impulse at the pre-amp outputs,both indicate fairly accurately how many filter taps are capable of being utilized.

There are methods, such as frequency 'warping','multirate' processing and others, that allow you to generate higher resolution correction more efficiently while using less MIPS than brute force approaches.

Trinnov by the way,has been rumored to be able to generate as many as 64,000 FIR filter taps in it's latest software versions for the High-End/Professional units.

Last edited by StevenLansing; 04-21-2016 at 01:36 AM.
StevenLansing is offline  
post #3437 of 3450 Old 04-20-2016, 11:04 PM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by StevenLansing View Post
Sorry for the late response,

That Audyssey chart refers to how many 'times' more correction each flavor of MultEQ has over 2EQ.Those aren't filter tap numbers. Audyssey won't release how many filter taps each version actually uses,however the CTO has mentioned in the past that the MultEQ XT version in AVR's and certain processors used 512 filter taps for the main channels and 4x that for the subwoofer. It was also stated that MultEQ XT32 has 32x more resolution than XT for the main channel correction and uses the same amount of correction resolution as the mains for the subwoofer channel.

The math and acoustic measurements taken showing were the bass correction with MultEQ and XT start tapering off as frequency gets lower, both indicate fairly accurately how many filter taps are capable of being utilized.

There are methods, such as frequency 'warping','multirate' processing and others, that allow you to generate higher resolution correction more efficiently while using less MIPS than brute force approaches.

Trinnov by the way,has been rumored to be able to generate as many as 64,000 FIR filter taps in it's latest software versions for the High-End/Professional units.

im just going to say that I dont believe for one split second that xt32 has 16000 taps or anywhere close.

I do not believe that they have a processor capable of that many taps on 5.1 channels.

btw, Im still researching raspberry pi based solutions that will allow one to use as many as 132k taps.

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3438 of 3450 Old 04-21-2016, 01:56 AM
Advanced Member
 
StevenLansing's Avatar
 
Join Date: Mar 2007
Posts: 650
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
im just going to say that I dont believe for one split second that xt32 has 16000 taps or anywhere close.

I do not believe that they have a processor capable of that many taps on 5.1 channels.

btw, Im still researching raspberry pi based solutions that will allow one to use as many as 132k taps.


Sheer number of filter taps alone won't necessarily indicate how 'good' the resulting filter correction is.

If you do more research on Audyssey you 'will' find that they do indeed have the capability with XT32 to create 16K+ filter taps using the DSP resources available in AVR's. It's been done for many years now.

The various techniques they and others use are what allow them to create those filters with tens of thousands of taps for each channel while using the processing power available to them in the DSP's. That's why many of those techniques are patented and part of those companies various (secret sauce) IPs.

Brute force approaches to creating FIR filters are the more processor intensive way of creating filters.
StevenLansing is offline  
post #3439 of 3450 Old 04-21-2016, 06:26 AM
RUR
Innocent Bystander
 
RUR's Avatar
 
Join Date: Feb 2007
Location: California Republic
Posts: 2,711
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 331 Post(s)
Liked: 186
Quote:
Originally Posted by StevenLansing View Post
Trinnov by the way,has been rumored to be able to generate as many as 64,000 FIR filter taps in it's latest software versions for the High-End/Professional units.
It's not just rumor.

Quote:
Originally Posted by StevenLansing View Post
Sheer number of filter taps alone won't necessarily indicate how 'good' the resulting filter correction is.
Yup.
RUR is offline  
post #3440 of 3450 Old 04-21-2016, 07:59 AM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
Quote:
Originally Posted by StevenLansing View Post

If you do more research on Audyssey you 'will' find that they do indeed have the capability with XT32 to create 16K+ filter taps using the DSP resources available in AVR's. It's been done for many years now.

The various techniques they and others use are what allow them to create those filters with tens of thousands of taps for each channel while using the processing power available to them in the DSP's. That's why many of those techniques are patented and part of those companies various (secret sauce) IPs.

Brute force approaches to creating FIR filters are the more processor intensive way of creating filters.
would you mind linking me to an avr that can do real time convolving of 16000 samples on 5.1 channels. I haven't seen this and I don't believe it exists. and even if it does exist its certainly not going to be the sharc processor used for xt32


also, isn't it convenient to hide behind " secret sauce" technology when they don't want to disclose number of taps?


I think the bottom line is no one knows how many taps xt32 uses, but I'll bet money its not close to 16k per channel

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3441 of 3450 Old 04-22-2016, 04:08 PM
Advanced Member
 
StevenLansing's Avatar
 
Join Date: Mar 2007
Posts: 650
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 15
Quote:
Originally Posted by RUR View Post
It's not just rumor.
Is it just a matter of selecting how many filter taps you want to use or it application specific?
StevenLansing is offline  
post #3442 of 3450 Old 04-22-2016, 04:46 PM
Advanced Member
 
StevenLansing's Avatar
 
Join Date: Mar 2007
Posts: 650
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 15
Quote:
Originally Posted by 1201 View Post
would you mind linking me to an avr that can do real time convolving of 16000 samples on 5.1 channels. I haven't seen this and I don't believe it exists. and even if it does exist its certainly not going to be the sharc processor used for xt32


also, isn't it convenient to hide behind " secret sauce" technology when they don't want to disclose number of taps?


I think the bottom line is no one knows how many taps xt32 uses, but I'll bet money its not close to 16k per channel


Here are just a few snippets from a Q&A with the CTO of Audyssey many years ago when MultEQ XT was released.This is all I'm going to post further on the subject, since it's getting off topic. Keep in mind that XT32 'does' actually use multirate filtering,which XT didn't. It allows XT32 to have the correction resolution equivalent to 16K taps which is more computationally efficient.I was just pointing out that your earlier post about XT32 vs Trinnov and filter taps was incorrect for both.



"The largest filter length we currently use is 512 taps. There is no partitioning. There is, however, something that we call "dynamic frequency allocation" that allows us to have a higher resolution at lower frequencies"


"512 taps is not a lot really. This gives you something like 90Hz resolution (48Khz / 512). So I guess that you are using some kind of multirate filtering?"

"True if you follow a brute-force FIR method as listed in all the textbooks. We don't use multirate, but another method that gives us a few Hz of resolution where it is needed the most."

"... but it seems that you are not aware of other methods that have nothing to do with downsampling or multi-rate filtering. There are other principles in addition to the ones you mention and I do agree with them. In fact we use them. We just would prefer not to reveal them in a public forum."

"Regarding your latency question, that is very valid. Typically, the latency of an FIR filter increases with the number of filter coefficients. We are fortunate that this issue has been addressed by the DSP manufacturers in the following way. There is a framework (think of it as an operating system) that they provide. Every process in that framework works like a plug-in. There is a fixed input to output delay associated with each DSP platform and the architecture is such that any plug in (such as MultEQ) within that framework does not add any additional delay. This DSP delay is uniform across frequency so that there are no group delay problems."

Last edited by StevenLansing; 04-22-2016 at 05:43 PM.
StevenLansing is offline  
post #3443 of 3450 Old 04-22-2016, 08:28 PM
Advanced Member
 
Join Date: Jun 2015
Posts: 704
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 122
thanks for posting.

in essence the above tells me nothing.

i will continue to be highly skeptical of any far out claims of 16k taps

64 base level speakers $10k
28 height speakers $10k,
14 subwoofers $10k
90% of the movie sound coming from the center channel - priceless.
1201 is online now  
post #3444 of 3450 Old 04-23-2016, 09:04 AM
RUR
Innocent Bystander
 
RUR's Avatar
 
Join Date: Feb 2007
Location: California Republic
Posts: 2,711
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 331 Post(s)
Liked: 186
Quote:
Originally Posted by StevenLansing View Post
Is it just a matter of selecting how many filter taps you want to use or it application specific?
FIR filter length is selectable by the user and the resultant correction may be used for any I/O.
RUR is offline  
post #3445 of 3450 Old 05-11-2016, 04:55 AM
Member
 
xjr15's Avatar
 
Join Date: Nov 2009
Posts: 86
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 15
Quote:
Originally Posted by xjr15 View Post
Infantile vulgarity, as in the above post, has no place in an adult discussion, and is generally evidence of
weakness and stupidity.

And "time based" correction is a meaningless term. "Time-base correction" refers to synchronization methods for analog recording media and is not relevant here. The correct term in the present context is "time-domain correction", not "time based correction".

All I am saying is that the R-972 room EQ did not work for me, and that the Audyssey room correction that I had tried also did not work for me, but that the R-972 speaker remapping was revelatory. If you're not noticing much of a change with the remapping, it may be that your R-972 was/is not working correctly, or it may be that the source material you listen to (for instance, classical music) will not benefit very much from the remapping. Whether (and to what extent) a particular room correction product is frequency-domain and/or time-domain correction is not an issue for me; if it sounds bad, I don't want it.

As I said above, when you experience the 3D remapping, you will know it's right because of the left/right symmetry of the surround imaging on a good surround mix; you know you are finally hearing what the artist/engineer intended, because the precision and interaction of the left/right sides of the surround image and the left/right image symmetry cannot be an accident, and you will hear some of the images do clockwise and counterclockwise rotation that was not audible before the remapping. This is remarkable and not at all subtle.

Conventional frequency-domain and time-domain room correction will, in comparison, be subtle and fairly minor. And as David Rich remarked in his review of the original Trinnov multichannel professional optimizer in the $ensible Sound, corrections to tonal balance are pointless if the imaging is wrong.

But buying a R-972 for "the kid's room" makes no sense, since the R-972 is all about optimizing the audio at the sweet spot; if you're using it for background music or to entertain the kids, that's like buying a Ferrari to run errands to the local supermarket.
My point being that even if you have been very careful (manually) setting up the speakers to adhere to ITU or SMPTE standards, it is unlikely that the positions and directions will be correct, and errors of only a few degrees or a few inches can cause the surround image to collapse. That's where the Trinnov electronic remapping comes to the rescue, and that's why it is needed.

Last edited by xjr15; 05-11-2016 at 03:49 PM.
xjr15 is offline  
post #3446 of 3450 Old 06-15-2016, 05:49 AM
Senior Member
 
JayS99's Avatar
 
Join Date: Nov 2006
Posts: 421
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 11
Can I switch gears and ask for some help. I happen to love my 972 (or did). I now have two, neither of which work to some degree (one I actually wound up getting for free because it was partially DOA on arrival so I got a refund and owner did not want it back). Here's my dilemma .. My first 972 lost all video (all HDMI channels) and the ability to do setup but does process digital and analog audio in. The second one I picked up can switch video fine, goes into setup, has ZERO audio (both digital and analog) .. won't even generate test tones... SOMEWHERE between both of them I'm thinking I've got one good 972 (or hoping).

Have tried all the resets in the world I can find online.. none work.. have left them unplugged and retried, didn't work. I want to swap boards (first thing would be to try the HDMI boards).. I've been inside and built a lot of PCs in my day, but there are just some connectors inside the 972 that I'm not familiar with. Sounds like some folks have had theirs apart. Anyone out there willing to spend a few minutes going over disassembly with me? Specifically, there seems to be a bridge type connector on top of the HDMI boards that I'm not familiar with. Also, folks have said that HDMI 1 is on a different board than the rest, but I swear (unless I'm missing something very obvious) that all the HDMI inputs look like they're on the same board.

Any and all thoughts would be great appreciated..

Jay S.

Last edited by JayS99; 06-15-2016 at 06:24 AM.
JayS99 is offline  
post #3447 of 3450 Old 06-15-2016, 12:35 PM
AVS Addicted Member
 
Charles R's Avatar
 
Join Date: Nov 2000
Location: Indianapolis
Posts: 11,808
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 421
Quote:
Originally Posted by JayS99 View Post
Any and all thoughts would be great appreciated.
Can't really help outside of if you haven't already tried insert/reinsert the boards and see if that helps. With one unit I had (and others have reported the same) it magically came to life. During shipping one of the boards became unseated.

| Mitsubishi HC7900DW Projector | 16x9 120" Screen | 7.2.4 Boston Acoustics & Power Sound Audio | Yamaha RX-A3060 (on order) & AudioSource AMP110 | NVIDIA Shield | 20TB Media Server | TiVo
Charles R is offline  
post #3448 of 3450 Old 06-15-2016, 02:03 PM
Senior Member
 
JayS99's Avatar
 
Join Date: Nov 2006
Posts: 421
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 11
Quote:
Originally Posted by Charles R View Post
Can't really help outside of if you haven't already tried insert/reinsert the boards and see if that helps. With one unit I had (and others have reported the same) it magically came to life. During shipping one of the boards became unseated.
Charles.. Part of the issue. The HDMI board appears to be paired with another board with a top connector.. Haven't had much luck figuring out how to loosen the boards to the point where I can do that. Have unscrewed from outside, but that isn't helping much. :-(

Anyone have any experience with actually removing the boards??

Thanks again..
Jay S.
JayS99 is offline  
post #3449 of 3450 Old 07-20-2016, 09:03 PM
Member
 
KK20's Avatar
 
Join Date: May 2003
Posts: 132
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 10
Side surround speaker placement

Due to my room configuration, my side surrounds (7.1 configuration) are located at ceiling height (10 feet high) directed (angled) to the central listening position.

In researching Atmos systems, I was wondering if there would be any advantage to replacing the high, angled side surrounds to a ceiling mounted speaker pointing down to the seating area? The thought is that this might trick the R972's Trinnov to add a height surround similar to the Atmos concept.

Note that I'd test this theory, however, it will involve ceiling surgery (holes and the like), so I'd prefer to get opinions before I cut holes in my drywall!

KK20

Last edited by KK20; 07-21-2016 at 08:44 AM.
KK20 is offline  
post #3450 of 3450 Old Today, 08:12 AM
Senior Member
 
jasonmichaelh's Avatar
 
Join Date: May 2006
Location: AZ
Posts: 360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 33
Quote:
Originally Posted by KK20 View Post
Due to my room configuration, my side surrounds (7.1 configuration) are located at ceiling height (10 feet high) directed (angled) to the central listening position.

In researching Atmos systems, I was wondering if there would be any advantage to replacing the high, angled side surrounds to a ceiling mounted speaker pointing down to the seating area? The thought is that this might trick the R972's Trinnov to add a height surround similar to the Atmos concept.

Note that I'd test this theory, however, it will involve ceiling surgery (holes and the like), so I'd prefer to get opinions before I cut holes in my drywall!
If you are using 3D remap, I would guess that the Trinnov correction is using your fronts and rears that are closer to ear level to create phantom side surrounds at ear level (maybe side surrounds are in use too). It doesn't seem like you would get much improvement moving the side surround from 10ft high to ceiling mounted, unless maybe your side walls are really far away. In fact, if the 10ft high side surrounds are currently used by the Trinnov remap as part of the phantom side surround, it would likely use them less for that, if you moved them to the ceiling.

Sony XBR-65X900A | Oppo BDP-103D | UBD-K8500
Mirage OMD-28 | OMD-C2 | OMD-5 | Energy Veritas 2.0R x4
Yamaha CX-A5100 | NAD T973 x 2
SVS PB12Plus/2 | Dayton UM18-22 Sealed x2 w/ iNuke 6000DSP
Monster Power HTPS 2000
jasonmichaelh is online now  
Sponsored Links
Advertisement
 
Reply Receivers, Amps, and Processors

Tags
Sherwood Newcastle R 972



Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off