Official DataSat RS20i thread. (Setup Tips, Questions,General Info, etc) - Page 142 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 952Likes
Reply
 
Thread Tools
post #4231 of 5211 Old 01-02-2017, 06:22 PM
LICENSED TO THRILL
 
CINERAMAX's Avatar
 
Join Date: Sep 2000
Location: PAN-ATLANTIC
Posts: 17,636
Mentioned: 48 Post(s)
Tagged: 0 Thread(s)
Quoted: 2224 Post(s)
Liked: 1267
Send a message via Skype™ to CINERAMAX
Quote:
Originally Posted by cannga View Post
I would point out again that sound in the Wide area doesn't just disappear when you don't have Wide. This area has been reproduced for many years now by stereo imaging between eg right front and right surround. This is seen in action movie when they pan sound front to back, it's been done for many years and it doesn't disappear because we have Atmos ..
Ignorance is bliss, in this context phantom nothing works like a discrete source.

Even Xmas music sounds great on neural wides.

The width channel addresses a very specific PRIMAL NEED, and I wish Walter shared his recent research on scientific studies that he recently completed. He wassaped me for the new year and proceded to explain his findings. I told him no need to explain.

I'll admit that it is challenging to implement wides, but after him experiencing first hand THIS, it motivated to dig into the reasons the Width channel added so much verisimilitude to the sound-stage at HYPERION.




I predict that by a year from today, I will not have to do one more post expounding the need for wides, these will be embraced whole by this community in 2017. Watch.
sdrucker likes this.
CINERAMAX is offline  
Sponsored Links
Advertisement
 
post #4232 of 5211 Old 01-02-2017, 06:27 PM
AVS Forum Special Member
 
cannga's Avatar
 
Join Date: Feb 2001
Location: Palos Verdes - Lakers Land
Posts: 2,821
Mentioned: 39 Post(s)
Tagged: 0 Thread(s)
Quoted: 1507 Post(s)
Liked: 1384
Quote:
Originally Posted by Chris Austria View Post
I agree and hope 2017/18 will bring 9.1.6 to the Japs. rec. - only the masses will make changes...
I would point out again that sound in the Wide area doesn't just disappear when you don't have Wide. This area has been reproduced for many years now by stereo imaging between eg right front and right surround channel content. This is heard frequently in action movies when they pan sound front to back, it's been done for many years and it doesn't disappear because we have Atmos . Please think about it: Dolby engineers are not dumb enough to leave a sonic void here.

As we all know, Dolby Atmos is a hybrid system: channel and object. Bottom level (the important one) contains significant 7.1 channel content (probably because they've done it this way for a long time, including the critical LCR), that does NOT use Wide speakers at all, further diminishing its importance.

Many examples, easily heard: disconnect all speakers except right front and right surround, and look for sound that appear in both. That image will be in between - the wide area. One example easily heard from what I'm watching now: Blu-Ray Game of Thrones' dragon flights, last episode last season; and of course ambient/background music in all movies.

No one is arguing against principles of object audio; I just want to share my opinions about diminishing return, considered evaluation, and I'm anti hype, not wide. The reason I mentioned the Japanese receivers: Because if they have it that likely means the cost is down to much lower level for all. Until then IMHO except for complex multi speaker system no one should pay $35k (vs 10k Datasat LS10 eg - this thing is putting a lot of pressure on all competitors) for this unimportant area that is already reproduced by other speakers. It's worth at most 2 cents (kidding). BTW, don't think it's coming back (they probably got rid of it partly for the reasons above) so I wouldn't wait. All IMHO.

PS Peter sorry because of what recently happened I am afraid of any conversation with you . Kindly put me on your ignore list please.

Regards, Can
My System & Theta Casablanca Mini-Review Uncontrolled passion for music, and sound. (For CB IVa setup help, click HERE.)
Interesting Audio Diagrams :-) & High-End Speaker Reviews
JTR Subwoofer Thread I don't always listen to subwoofers, but when I do, it's JTR :-).

Last edited by cannga; 01-02-2017 at 07:07 PM.
cannga is online now  
post #4233 of 5211 Old 01-02-2017, 06:35 PM
AVS Forum Addicted Member
 
thebland's Avatar
 
Join Date: Jan 2001
Location: Detroit, Michigan USA
Posts: 25,748
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 1666 Post(s)
Liked: 1003
Peter,

The plane and crash sequences in 'Sully' last night were simply spellbinding in Atmos - there was sound everywhere - it felt like my room was the cabin of the plane. There were sounds coming fron all over - and yes, the wides, played a large role in the drama.

I have one more screen shot (clearer) and you can see all Atmos channels were firing (though the iPad mini clips off a few of my channels - so not all are seen). And yes, at a hot -21 db - not Alcons hot but pretty involving!

CINERAMAX likes this.

Goodbye to a great audio and video genius and writer... JOHN GANNON. I enjoyed your friendship, wit and a nice long run we took around Indianapolis at CEDIA years back... and for buying my Runco 980 Ultra years back... you saved my ass! Rest in peace.
thebland is online now  
 
post #4234 of 5211 Old 01-03-2017, 08:16 AM
AVS Forum Special Member
 
The Bogg's Avatar
 
Join Date: Nov 2006
Location: GTA, Ontario, Canada
Posts: 2,052
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 440 Post(s)
Liked: 179
Quote:
Originally Posted by audioguy View Post
I have a similar start but probably earlier. bought first (used) stereo in college; upgraded after I graduated from a store in St. Louis called EJ Korvettes. My first "multimedia" system was in 1983 with a Mits Front Projection TV, Laser Disc, VCR, some kind of "surround" synthesis processor and 4 speakers. And so it continues.

While height speakers are certainly a "nice to have", it is only one of a very few upgrades my wife both noticed and liked (getting a 4K projector was the other one)
Similar start here too - took me about a year working at McDonalds in high school to save up for a pair of Technics SB-M3 speakers (about 2.5k Cdn dollars at the time) and the associated gear. It was probably my first lesson in audio about room acoustics - the speakers never sounded all that good in the space I had for them. It's just as true today that the acoustics of the room are at least an equal partner to the equipment for getting good sound.

Amazing how far progress has come with room correction systems. I had an Accuphase DG-28 digital equalizer for my 2 channel system. It was decent but when I bought the Trinnov ST2 it was a real revelation and step up. I also tried Dirac on the Emotiva prepro and similarly it was a quantum leap ahead in performance. We are lucky to have not only good equipment and rooms but the current room correction systems really let us reach a level of performance that was unachievable even a few years ago.

Design by Rives...dollars by The Bogg

Click for my build thread
The Bogg is offline  
post #4235 of 5211 Old 01-03-2017, 09:50 AM
AVS Forum Special Member
 
audioguy's Avatar
 
Join Date: Jun 2001
Location: Not far from Atlanta - but far enough!
Posts: 6,484
Mentioned: 43 Post(s)
Tagged: 0 Thread(s)
Quoted: 2529 Post(s)
Liked: 1668
Quote:
Originally Posted by The Bogg View Post
Amazing how far progress has come with room correction systems. I had an Accuphase DG-28 digital equalizer for my 2 channel system. It was decent but when I bought the Trinnov ST2 it was a real revelation and step up. I also tried Dirac on the Emotiva prepro and similarly it was a quantum leap ahead in performance. We are lucky to have not only good equipment and rooms but the current room correction systems really let us reach a level of performance that was unachievable even a few years ago.
I became, sort of through a back door, the marketing director for a company called Cambridge Signal Technology, whose digital room correction product (SigTech), was the first commercially available product to the market. In some ways, it was actually superior to the best products today - but the final results were certainly not was good. I thought TacT was excellent, Audyssey could be very good (if you knew what you were doing) and Dirac blows them all away. The first time I heard the potential of what reasonable speakers could actually sound like in a normal room, with a room correction system in lace, I was simply blown away.

What I still find so incredibly amazing though is how so many audio "purists", even if they have a digital only audio system, REFUSE to put such a device in their system. These, in many cases, are the same people who have no problem spending $10,000 on audio cables.

Our Theater

"I have wondered at times about what the Ten Commandments would have looked like if Moses had run them through the U.S. Congress." - Ronald Reagan

Ephesians 4:29
audioguy is offline  
post #4236 of 5211 Old 01-03-2017, 01:43 PM
AVS Forum Special Member
 
The Bogg's Avatar
 
Join Date: Nov 2006
Location: GTA, Ontario, Canada
Posts: 2,052
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 440 Post(s)
Liked: 179
Yeah, some of the guys in my "audio group" are exactly like that - great gear. I mean really great gear. Unfortunately pretty mediocre sound. But nice shiny expensive cables, lol.

Design by Rives...dollars by The Bogg

Click for my build thread
The Bogg is offline  
post #4237 of 5211 Old 01-03-2017, 01:44 PM
AVS Forum Special Member
 
Kain's Avatar
 
Join Date: Dec 2001
Location: Dubai, UAE
Posts: 3,634
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 1499 Post(s)
Liked: 437
What are your thoughts on having high-end speakers such as Meyer Sound or Alcons Audio and using them with a consumer-level processor such as the Anthem AVM 60? A waste? I've heard the Anthem's EQ (ARC) is actually quite good.
Kain is offline  
post #4238 of 5211 Old 01-03-2017, 05:28 PM
Advanced Member
 
Ganymed4's Avatar
 
Join Date: Oct 2012
Location: Germany
Posts: 913
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 190 Post(s)
Liked: 118
Quote:
Originally Posted by Kain View Post
What are your thoughts on having high-end speakers such as Meyer Sound or Alcons Audio and using them with a consumer-level processor such as the Anthem AVM 60? A waste? I've heard the Anthem's EQ (ARC) is actually quite good.
I don't think that good speakers are 'a waste' at all times. I think that they cannot unfold their full potential without good electronics. I also find that myself having inferior speakers regarding my electronics - Datasat and Audionet.de - their potential is leveraged by excellent electronics - Datasat and Dirac. But in the end, a fitting chain of electronics and speakers is best. If I look at my frequency and impulse diagrams from Dirac, I get tears in my eyes. They are really bad but the sound is still good, by all means. But I understand now, it could be better by acoustical treatment and better speakers. Just my experience.

I still don't have an answer from Datasat and will answer this in another post...

Last edited by Ganymed4; 01-03-2017 at 05:47 PM.
Ganymed4 is online now  
post #4239 of 5211 Old 01-03-2017, 05:31 PM
Senior Member
 
Billybobjimbob's Avatar
 
Join Date: Aug 2009
Posts: 383
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 42
Quote:
Originally Posted by Kain View Post
What are your thoughts on having high-end speakers such as Meyer Sound or Alcons Audio and using them with a consumer-level processor such as the Anthem AVM 60? A waste? I've heard the Anthem's EQ (ARC) is actually quite good.
Likely be better to ask that question in the Anthem thread avm 60 thread?

21inch hitachi
Billybobjimbob is offline  
post #4240 of 5211 Old 01-03-2017, 05:41 PM
Advanced Member
 
Ganymed4's Avatar
 
Join Date: Oct 2012
Location: Germany
Posts: 913
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 190 Post(s)
Liked: 118
Quote:
Originally Posted by dschulz View Post
I am curious as to owner expectations of permanent upgradability - surely there is a point at which the architecture of the RS20i will no longer support the state-of-the-art, and Datasat has to replace it with an RS30i?
I hope not, regarding the price tag. I haven't received an answer from Datasat yet about the additional 16 channels, the copying of channels to the expansion cards etc. Whatsoever, I find this really poor from Datasat, that I really have a limitation with setting up more than 4 Atmos channels with an RS20i. If the additional 8 or 16 audio channels were only intended for active x-overs, this is also poor. Regarding the 4 HDMI inputs, this is also poor related to actual consumer AVRs. This really makes paying the premium for 'the last SSP you will buy in your life' even poorer. This is really a broken promise by Datasat and if the reasons are really in the DSP architecture, this is even worse. I may say that I am really disappointed by Datasat and their lost promise, that I never need to buy another SSP in my life. I am very disappointed by this. Buy Trinnov - sorry to write this...

Last edited by Ganymed4; 01-03-2017 at 08:29 PM.
Ganymed4 is online now  
post #4241 of 5211 Old 01-03-2017, 10:27 PM
Advanced Member
 
dschulz's Avatar
 
Join Date: Nov 2011
Location: Los Angeles, CA
Posts: 788
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 672 Post(s)
Liked: 439
Quote:
Originally Posted by Ganymed4 View Post
I hope not, regarding the price tag. I haven't received an answer from Datasat yet about the additional 16 channels, the copying of channels to the expansion cards etc. Whatsoever, I find this really poor from Datasat, that I really have a limitation with setting up more than 4 Atmos channels with an RS20i. If the additional 8 or 16 audio channels were only intended for active x-overs, this is also poor. Regarding the 4 HDMI inputs, this is also poor related to actual consumer AVRs. This really makes paying the premium for 'the last SSP you will buy in your life' even poorer. This is really a broken promise by Datasat and if the reasons are really in the DSP architecture, this is even worse. I may say that I am really disappointed by Datasat and their lost promise, that I never need to buy another SSP in my life. I am very disappointed by this. Buy Trinnov - sorry to write this...
The RS20i will process a maximum of 16 channels - the expansion cards will enable additional *output* channels, with individual levels and delay, but not additional processing channels for Atmos. It is currently DSP-limited to 12 Atmos channels (7.1.4 or 9.1.2 or even 5.1.6), but I think it is not unreasonable that Datasat could release a newer DSP board that will support a full 16 channels of Atmos.
dschulz is offline  
post #4242 of 5211 Old 01-03-2017, 11:30 PM
AVS Forum Special Member
 
maikeldepotter's Avatar
 
Join Date: Aug 2014
Location: Netherlands
Posts: 1,322
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1039 Post(s)
Liked: 295
Quote:
Originally Posted by dschulz View Post
The RS20i will process a maximum of 16 channels - the expansion cards will enable additional *output* channels, with individual levels and delay, but not additional processing channels for Atmos. It is currently DSP-limited to 12 Atmos channels (7.1.4 or 9.1.2 or even 5.1.6), but I think it is not unreasonable that Datasat could release a newer DSP board that will support a full 16 channels of Atmos.
Yes, it should be relatively easy for them to get to Atmos 9.1.6, even with the current 12 channel DSP-chip limitation, by parallel Atmos processing on two chips: 5.1.6 for overheads only, and 9.1.2 for base layer only.

A good idea and understanding lies at the basis of every successful project.
maikeldepotter is offline  
post #4243 of 5211 Old 01-04-2017, 01:01 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by dschulz View Post
. . . It is currently DSP-limited to 12 Atmos channels (7.1.4 or 9.1.2 or even 5.1.6), but I think it is not unreasonable that Datasat could release a newer DSP board that will support a full 16 channels of Atmos.
Dan, is this a hardware limitation, or could it potentially support more channels if Dolby were to write/make available the appropriate code to support more? (I'm assuming it is Dolby, rather than Datasat, that have written the core DSP code).
Wookii is offline  
post #4244 of 5211 Old 01-04-2017, 01:02 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by maikeldepotter View Post
Yes, it should be relatively easy for them to get to Atmos 9.1.6, even with the current 12 channel DSP-chip limitation, by parallel Atmos processing on two chips: 5.1.6 for overheads only, and 9.1.2 for base layer only.
Arguably thats what they should have done straight out of the gate on a 16 channel processor.
Wookii is offline  
post #4245 of 5211 Old 01-04-2017, 02:23 AM
Senior Member
 
Billybobjimbob's Avatar
 
Join Date: Aug 2009
Posts: 383
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 42
Quote:
Originally Posted by Wookii View Post
Dan, is this a hardware limitation, or could it potentially support more channels if Dolby were to write/make available the appropriate code to support more? (I'm assuming it is Dolby, rather than Datasat, that have written the core DSP code).
It's not a Dolby issue. I emailed Dolby when all this came to light - they make all the code available, it's down to the manufacturer how they implement it via 3rd party dsp chips.

21inch hitachi
Billybobjimbob is offline  
post #4246 of 5211 Old 01-04-2017, 02:45 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by Billybobjimbob View Post
It's not a Dolby issue. I emailed Dolby when all this came to light - they make all the code available, it's down to the manufacturer how they implement it via 3rd party dsp chips.
Yeah, that wasn't what I was referring to Asif. The DSP chip that Datasat have used is, I believe, a standardised unit, the same one as used by Denon/Marantz/Onkyo etc - the core code (decoding/channel output) of the DSP is, I am assuming, written by someone other than the manufacturer (either Dolby or the DSP chip manufacturer) so all the manufacturer is doing is implementing the chip, as is, into their own architecture.

So whilst Dolby make the entire code set available to OEM's to allow 32 channels of Atmos, those manufacturers using this standard DSP chip will not change the processing ability of the chip until that third party (either Dolby or the DSP chip manufacturer) writes new code for it (they may not even have the ability/rights to change the DSP code themselves) - if, in fact, they ever do.

So I was just trying to establish if Dan knew whether the DSP chip had a hard ceiling of 12 output channels or if it was physically possible for more to be added if new core code was written for it at some future point by Dolby/DSP manufacturer - i.e. when the likes of Denon/Marantz demand more output channels to sell the next gen of AVR. If this does happen, and D&M et al utilise the same chip with a new higher channel count, then there may be the possibility of a future Datatsat firmware upgrade to add it rather than a costly hardware upgrade.
Wookii is offline  
post #4247 of 5211 Old 01-04-2017, 03:16 AM
Senior Member
 
Billybobjimbob's Avatar
 
Join Date: Aug 2009
Posts: 383
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 42
Quote:
Originally Posted by Wookii View Post
Yeah, that wasn't what I was referring to Asif. The DSP chip that Datasat have used is, I believe, a standardised unit, the same one as used by Denon/Marantz/Onkyo etc - the core code (decoding/channel output) of the DSP is, I am assuming, written by someone other than the manufacturer (either Dolby or the DSP chip manufacturer) so all the manufacturer is doing is implementing the chip, as is, into their own architecture.

So whilst Dolby make the entire code set available to OEM's to allow 32 channels of Atmos, those manufacturers using this standard DSP chip will not change the processing ability of the chip until that third party (either Dolby or the DSP chip manufacturer) writes new code for it (they may not even have the ability/rights to change the DSP code themselves) - if, in fact, they ever do.

So I was just trying to establish if Dan knew whether the DSP chip had a hard ceiling of 12 output channels or if it was physically possible for more to be added if new core code was written for it at some future point by Dolby/DSP manufacturer - i.e. when the likes of Denon/Marantz demand more output channels to sell the next gen of AVR. If this does happen, and D&M et al utilise the same chip with a new higher channel count, then there may be the possibility of a future Datatsat firmware upgrade to add it rather than a costly hardware upgrade.
I see what you're saying now.

My initial thoughts were that Datasat may utilise an alternative approach so they could utilise and harness the ability of all 16 channels. I think it's been documented that Accurus have done their programming in house to get past and around the limit. But it did appear Datasat went with the 'off the shelf' chip option.

Even if the receiver manufacturers utilise new chips that allow 9.1.6, it doesn't necessarily mean that these would be compaitable with the Rs20i as a lot was made about the internal architecture of the Rs20i and that also being a limiting factor..unless the Datasat engineers have identified a workaround that would allow the next gen of chips to be utilised.

If Datasat had wanted to utilise their own custom/bespoke method of implementing the 16 channels - they could possibly have done this the first time around- at a likely expense both fiscally and as far as time was concerned- I suspect the more convenient route to keep things as straight forward as possible was to go the off the shelf route.

The bottom line is, any Dsp based processor would be in the same boat, so it's not a matter of only Datasat falling short here.

I emailed Theta as the Datasat Atmos board was coming to market and they were already looking at different solutions to up the 7.1.4 limit.

21inch hitachi
Billybobjimbob is offline  
post #4248 of 5211 Old 01-04-2017, 03:48 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by Billybobjimbob View Post
Even if the receiver manufacturers utilise new chips that allow 9.1.6, it doesn't necessarily mean that these would be compaitable with the Rs20i as a lot was made about the internal architecture of the Rs20i and that also being a limiting factor..unless the Datasat engineers have identified a workaround that would allow the next gen of chips to be utilised.
The RS20i has 16 internal 'processing' channels, so in theory 9.1.6 (16 channels) should be possible if the DSP also supported it (along with 7.1.8, 5.1.10, 13.1.2 etc etc). Of course Datasat would probably need to re-write/update their own internal processing code and interface.

I believe the way 'other' manufacturers have worked around the channel count issue is running two of these standard chips in parallel and then routing the appropriate user selected channels as required as Maikel highlighted above.
Wookii is offline  
post #4249 of 5211 Old 01-04-2017, 03:59 AM
Senior Member
 
Billybobjimbob's Avatar
 
Join Date: Aug 2009
Posts: 383
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 42
Quote:
Originally Posted by Wookii View Post
The RS20i has 16 internal 'processing' channels, so in theory 9.1.6 (16 channels) should be possible if the DSP also supported it (along with 7.1.8, 5.1.10, 13.1.2 etc etc). Of course Datasat would probably need to re-write/update their own internal processing code and interface.

I believe the way 'other' manufacturers have worked around the channel count issue is running two of these standard chips in parallel and then routing the appropriate user selected channels as required as Maikel highlighted above.
Well, I was hoping for 9.1.6 the first time around. If that configuration can become a reality, then it's great news as the sound the Rs20i can produce is awesome - it'd be a fitting move to get the most out of all the channels available.

21inch hitachi
Billybobjimbob is offline  
post #4250 of 5211 Old 01-04-2017, 04:09 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by Billybobjimbob View Post
Well, I was hoping for 9.1.6 the first time around. If that configuration can become a reality, then it's great news as the sound the Rs20i can produce is awesome - it'd be a fitting move to get the most out of all the channels available.
Absolutely agreed on both counts - but given the speed of the DTS:X roll-out, I'm not going to be holding my breath for seeing it in this decade!
Wookii is offline  
post #4251 of 5211 Old 01-04-2017, 04:28 AM
Advanced Member
 
Lasalle's Avatar
 
Join Date: Aug 2005
Location: Sarasota, FL
Posts: 593
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 332 Post(s)
Liked: 210
Quote:
Originally Posted by maikeldepotter View Post
Yes, it should be relatively easy for them to get to Atmos 9.1.6, even with the current 12 channel DSP-chip limitation, by parallel Atmos processing on two chips: 5.1.6 for overheads only, and 9.1.2 for base layer only.
I had this discussion with the engineers from Trinnov. My understanding is this is not viable because the rendering process requires shared memory. The DSP's do not have a shared memory buss so they can't parallel process the rendering. A 9.1.6 render will be different than discretely adding a 5.1.6 and 9.1.2 as the objects rendered in the lower channel counts be fully deployed in the lower speaker count. The 9 and 6 layers of the added renders will have redundancies.
Lasalle is offline  
post #4252 of 5211 Old 01-04-2017, 04:35 AM
Senior Member
 
Billybobjimbob's Avatar
 
Join Date: Aug 2009
Posts: 383
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 42
Quote:
Originally Posted by Wookii View Post
Absolutely agreed on both counts - but given the speed of the DTS:X roll-out, I'm not going to be holding my breath for seeing it in this decade!
We'll if it does eventually turn up in the next decade, don't forget to give that loose change in your pocket a jingle

21inch hitachi
Billybobjimbob is offline  
post #4253 of 5211 Old 01-04-2017, 05:04 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by Billybobjimbob View Post
We'll if it does eventually turn up in the next decade, don't forget to give that loose change in your pocket a jingle
I've got three kids - all my loose change is spoken for, for the next two decades at least lol
Billybobjimbob likes this.
Wookii is offline  
post #4254 of 5211 Old 01-04-2017, 05:29 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by Lasalle View Post
I had this discussion with the engineers from Trinnov. My understanding is this is not viable because the rendering process requires shared memory. The DSP's do not have a shared memory buss so they can't parallel process the rendering. A 9.1.6 render will be different than discretely adding a 5.1.6 and 9.1.2 as the objects rendered in the lower channel counts be fully deployed in the lower speaker count. The 9 and 6 layers of the added renders will have redundancies.
What I'm surprised about is that if Trinnov can get the max channel count working on a what is essentially a PC, why haven't one of the other manufacturers managed to get it working on a more powerful DSP, or on some sort of FPGA style chip.
Wookii is offline  
post #4255 of 5211 Old 01-04-2017, 05:39 AM
AVS Forum Special Member
 
maikeldepotter's Avatar
 
Join Date: Aug 2014
Location: Netherlands
Posts: 1,322
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1039 Post(s)
Liked: 295
Quote:
Originally Posted by Lasalle View Post
A 9.1.6 render will be different than discretely adding a 5.1.6 and 9.1.2 as the objects rendered in the lower channel counts be fully deployed in the lower speaker count. The 9 and 6 layers of the added renders will have redundancies.
That is what I initially thought as well. But apparently, adding one pair of overheads will make the Atmos renderer send ALL overhead labeled sound to that one pair. There is nothing left in the base layer. That means there is no redundancy of overhead information when putting the 6 top speakers of a 5.1.6 rendition, over the 9 base level speakers of 9.1.2 rendition. You could in fact do this at home with two 11.1 Atmos enabled receivers if you don't mind operating two volume controls.

A good idea and understanding lies at the basis of every successful project.
maikeldepotter is offline  
post #4256 of 5211 Old 01-04-2017, 06:19 AM
Advanced Member
 
Lasalle's Avatar
 
Join Date: Aug 2005
Location: Sarasota, FL
Posts: 593
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 332 Post(s)
Liked: 210
Quote:
Originally Posted by maikeldepotter View Post
That is what I initially thought as well. But apparently, adding one pair of overheads will make the Atmos renderer send ALL overhead labeled sound to that one pair. There is nothing left in the base layer. That means there is no redundancy of overhead information when putting the 6 top speakers of a 5.1.6 rendition, over the 9 base level speakers of 9.1.2 rendition. You could in fact do this at home with two 11.1 Atmos enabled receivers if you don't mind operating two volume controls.
Interesting, where did you get this information from? That approach to 9.1.2 would eliminate all front to back panning for objects in the height layer. I'm not sure why they would do that, but could be the case. Also I thought the current chipsets could only render 4 top layer positions. They may change with the new 9.1.4 products, but I haven't seen any 7.1.6 announced (not sure on Emotiva and Acurus on the .6). It should be noted that no manufacturer has adopted a 2 chip DSP approach even though it has been discussed for over a year.
Official DataSat RS20i thread. (Setup Tips, Questions,General Info, etc)
Lasalle is offline  
post #4257 of 5211 Old 01-04-2017, 06:30 AM
Advanced Member
 
Ganymed4's Avatar
 
Join Date: Oct 2012
Location: Germany
Posts: 913
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 190 Post(s)
Liked: 118
Quote:
Originally Posted by Wookii View Post
What I'm surprised about is that if Trinnov can get the max channel count working on a what is essentially a PC, why haven't one of the other manufacturers managed to get it working on a more powerful DSP, or on some sort of FPGA style chip.
My guess is that FPGAs and PC are completely different worlds in the sense of SSP manufacturers. Datasat comes from the AP20 whereas Trinnov was building new from ground up. I also guess that their REQ was always based on PC technology and code as Dirac is also. DSPs have known limitations but they can be put together, however the argument with the shared memory is a good one, because even you can couple two or more DSPs, it is - as far as I know - not possible for them to access a common memory area aka shared memory. This could be important to access the meta information contained in an Atmos data stream. That is all guessing and I have only parts or the information necessary. Yeah it seems that only new hardware could change this situation but this doesn't work with the RS20i, then it would be a new machine.
Well, then I have to be OK with a 7.x.4 setup for the next years, I guess.
Ganymed4 is online now  
post #4258 of 5211 Old 01-04-2017, 07:25 AM
Advanced Member
 
Lasalle's Avatar
 
Join Date: Aug 2005
Location: Sarasota, FL
Posts: 593
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 332 Post(s)
Liked: 210
Quote:
Originally Posted by Wookii View Post
What I'm surprised about is that if Trinnov can get the max channel count working on a what is essentially a PC, why haven't one of the other manufacturers managed to get it working on a more powerful DSP, or on some sort of FPGA style chip.
Trinnov is a Linux computer, very scaleable in MIPS and Memory. The architecture can accommodate multiple CPU boards (although I don't think they need them) and Terabytes of shared memory.
Lasalle is offline  
post #4259 of 5211 Old 01-04-2017, 07:32 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 1,808
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 1291 Post(s)
Liked: 465
Quote:
Originally Posted by Lasalle View Post
Trinnov is a Linux computer, very scaleable in MIPS and Memory. The architecture can accommodate multiple CPU boards (although I don't think they need them) and Terabytes of shared memory.
Yeah it uses an Intel Core i7 I believe - standard desktop style hardware platform. I would image that Atmos channel decoding uses a mere fraction of that processing power though. Surely if Lumagen can get there immense video processing on an FPGA, Atmos processing could easily be accomodated by one also (though I agree with Ganymed4's comment that such a platform may well be alien to most SSP manufacturers).
Wookii is offline  
post #4260 of 5211 Old 01-04-2017, 12:38 PM
AVS Forum Special Member
 
maikeldepotter's Avatar
 
Join Date: Aug 2014
Location: Netherlands
Posts: 1,322
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1039 Post(s)
Liked: 295
Quote:
Originally Posted by Lasalle View Post
Interesting, where did you get this information from?
IIRC some posts from @sdurani , who seems to be well documented.

Quote:
That approach to 9.1.2 would eliminate all front to back panning for objects in the height layer. I'm not sure why they would do that, but could be the case.
Indeed.

Quote:
Also I thought the current chipsets could only render 4 top layer positions. They may change with the new 9.1.4 products, but I haven't seen any 7.1.6 announced (not sure on Emotiva and Acurus on the .6).
You are right. The more accessible option with two AVRs would be 9.1.2 + 5.1.4= 9.1.4 with the former D/M models. Acurus does 7.1.6 and 9.1.2, but for some obscure reason no 9.1.4....

Quote:
It should be noted that no manufacturer has adopted a 2 chip DSP approach even though it has been discussed for over a year.
I have no clue what's the reason for that. Maybe there is another solution which will be appearing soon (whatever 'soon' means)...

A good idea and understanding lies at the basis of every successful project.
maikeldepotter is offline  
Sponsored Links
Advertisement
 
Reply Ultra Hi-End HT Gear ($20,000+)

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


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