Possible solution for DTS/Dolby/Atmos+eARC+HDCP to AES/EBU via Dante? (for Okta 8 etc)

[Littletycoon]

But those are not a good fit for Atmos setups because one, they don't support GLM, two, they are low powered (meant for museums are cafes)

Not much power required for atmos, most sound and dynamics (power requirement) is needed by lcr

 

[welwynnick]

I have – it's on by default with supporting models. I think it just adds an (imperceptible) amount of extra DSP latency. I'm on GLM 5.1.

Genelec phase linearity only adds 4ms, but a small reduction in group delay corresponds to a significant change in phase linearity and a big improvement in transient response.
This is group delay with (1) and without (2) for the 8351B for example:

If you have a flat frequency response and a flat group delay like (1) you get a step response like (2) (which is what it's supposed to look like) instead of (3),

Apart from the bass, I think the answer is to keep group delay flat to within 1ms, and phase response flat to within around 40 degrees.
Then you get time-aligned transient responses as good as the best electrostatics.
With linear phase off, the 8351 etc look like the Revel above (and all other three way speakers) but look like the Buchardt above with it on.
There are a several recent digital active speakers like Kef, Genelec, Neumann and Buchardt that use DSP to linearize phase response.
Have you tried turning the phase lineariser off? I'm curious what you'd hear?
I think there are potentially several big wins using a processor with digital outputs:

• Better amplitude equalisation
• Better phase equalisation
• Better crossovers
• Better digital to analogue conversion
• Better preamp
• Better volume control

 

[kfrancis]

Does GLM support what is sometimes called "Sound Field Management" (SFM) with multiple subwoofers, i.e. EQ'ing each subwoofer separately in order to achieve flat response at multiple listening positions at the same time? Basically the equivalent of MSO. If not, can individual subwoofers be EQ'd manually so that one can use MSO and then feed the results to GLM?

No, I don't believe so. Each sub is EQ'd independently to flat to the extent possible at the measured position(s). But, yes, you can modify the results afterwards. There are 20 notches on the subs, and it looks like GLM is only using the first 10 of them itself. You can edit all 20, though, of course.

 

[kfrancis]

Not much power required for atmos, most sound and dynamics (power requirement) is needed by lcr

Very true, but I'd prefer to use the same system for control and calibration for all the speakers in a given room, I think. I do hope to use a few of the Smart IPs for in-ceiling distributed audio elsewhere eventually. They're very clever.

I suppose the dream would be some sort of PoE++++++++ that could adequately supply the more power-hungry monitors, but I would certainly be happy with just power plus 1 other cable per speaker that I can home run to a switch instead of power plus 3 other cables that each go to different places!

 

[kandamrgam]

Not much power required for atmos, most sound and dynamics (power requirement) is needed by lcr

Atmos has no power requirement, that's not what I meant. In a typical room to set up surround sound (assuming 5m x 4m room), you would need adequate power. I doubt smart-ip speakers can meet that. They are fundamentally designed for low volume setups like cafes and art galleries.

As for power and SPL calculation, genelec provided me this excellent worksheet: https://support.genelec.com/attachm...ame=DolbyAtmosRoomDesignTool_Example.xlsb.zip You can choose your speaker and it will tell you if your speaker is powerful enough.

 

[masterhw]

Atmos has no power requirement, that's not what I meant. In a typical room to set up surround sound (assuming 5m x 4m room), you would need adequate power. I doubt smart-ip speakers can meet that. They are fundamentally designed for low volume setups like cafes and art galleries.

As for power and SPL calculation, genelec provided me this excellent worksheet: https://support.genelec.com/attachm...ame=DolbyAtmosRoomDesignTool_Example.xlsb.zip You can choose your speaker and it will tell you if your speaker is powerful enough.

I am optimistic for many home setups, which are typically limited in size and sustained output across all channels, that PoE++ would be capable of serving most use cases. Genelec's Smart IP models even have some clever DC-DC power balancing allowing them to hit brief peaks/transients beyond what the continuous power rating might suggest.

However, the subwoofers really seem to suffer the most here. LFE channels tend to be pushed the hardest in SPL, tend to be sustained (not brief peaks), and even with the Smart IP sub falling off a cliff under 35 Hz it looks considerably compromised for HT applications. I don't see a straightforward solution to this problem given Genelec's own SAM subs don't accept Dante either.

 

[Littletycoon]

Atmos has no power requirement, that's not what I meant. In a typical room to set up surround sound (assuming 5m x 4m room), you would need adequate power. I doubt smart-ip speakers can meet that. They are fundamentally designed for low volume setups like cafes and art galleries.

As for power and SPL calculation, genelec provided me this excellent worksheet: https://support.genelec.com/attachm...ame=DolbyAtmosRoomDesignTool_Example.xlsb.zip You can choose your speaker and it will tell you if your speaker is powerful enough.

I'm not 100% sure but genelec smart ip line 4410, 4420,4430 seem to be the same as 8010, 8020/8320, 8030/8330. If that is true they seem to do absolutely fine in terms of SPL in a 20m2 room, https://www.genelec.com/correct-monitors

For subwoofers you just use dante converters to aes/ebu or analog. Maybe also ravenna converters are available.

 

[kandamrgam]

I am optimistic for many home setups, which are typically limited in size and sustained output across all channels, that PoE++ would be capable of serving most use cases. Genelec's Smart IP models even have some clever DC-DC power balancing allowing them to hit brief peaks/transients beyond what the continuous power rating might suggest.

However, the subwoofers really seem to suffer the most here. LFE channels tend to be pushed the hardest in SPL, tend to be sustained (not brief peaks), and even with the Smart IP sub falling off a cliff under 35 Hz it looks considerably compromised for HT applications. I don't see a straightforward solution to this problem given Genelec's own SAM subs don't accept Dante either.

I have multiple Genelec sources telling me that they don't recommend smart-ip speakers for Atmos. Are you OK with not using GLM ?

 

[S G]

Me too.
These are the processors that I'm aware of that decode and output multi-channel digital audio.
The 8 channel units just extract digital audio from HDMI.
I believe there will be a new option in the near future - the Audio Control Hyperion DPR-16.

Processor	AES/EBU	AES/EBU	AES67	AES67	Dante	Dante	SINAD	DNR
Model	Chan	Rate	Chan	Rate	Chan	Rate	dB	dB
Arvus H2-4D	16	192	16	96	16	48	90	102
Arvus H1-D	-	-	16	96	16	48	-	-
Arvus HDMI-2A	8	192	-	-	-	-	-	-
Audiopraise VanityPRO	8	192	-	-	-	-	-	-
Datasat RS20i	16	192	-	-	-	-	101	107
Datasat AP25	16	192	16	96	-	-	101	107
JBL Synthesis SDP-75	16	96	-	-	-	-	100	108
JBL Synthesis SDP-55	-	-	-	-	16	48	95	102
JBL Synthesis SDP-58	-	-	-	-	16	48	95	102
JBL Synthesis SDR-35	-	-	-	-	16	48	100	112
JBL Synthesis SDR-38	-	-	-	-	16	48	100	112
Lyngdorf MP-60	16	96	16	48	-	-	103	113
Lyngdorf MP-60 2.1	16	96	16	48	-	-	103	113
SMYTH A16 Pro Dante	-	-	-	-	16	96	-	-
SMYTH A16 Pro AES3	16	96	-	-	-	-	-	-
StormAudio ISP 16 Mk3	32	48	32	48	-	-	99	107
StormAudio ISP Mk1	32	48	-	-	AVB	Opt	99	107
StormAudio ISP Mk2	32	48	32	48	-	-	99	107
StormAudio ISP Mk3	32	48	32	48	-	-	99	107
StormAudio ISP Evo	32	48	32	48	32	48	-	-
Trinnov Altitude 32	16	96	-	-	-	-	100	108
Trinnov Altitude 16	-	-	Future	Future	Future	Future	100	108

Do you know if they limit resolution to 16-bit over AES/EBU to comply with HDCP obligations over sending protected content over unprotected outputs?

I am particularly curious about StormAudio products.

 

[Kal Rubinson]

I am particularly curious about StormAudio products.

What are you curious about (aside from the specs)?
I reviewed the ISP Evo a while back: stereophile/stormaudio-isp-evo

 

[S G]

What are you curious about (aside from the specs)?
I reviewed the ISP Evo a while back: stereophile/stormaudio-isp-evo

I just recently learned about HDCP requirement that devices that receive HDCP encoded input must limit the unprotected outputs to 48kHz/16-bit, and AES/EBU is unprotected, as far as I am aware.

That kinda defeats the purpose of the whole endeavor to go all digital on processing/decoding and adding your own high-SINAD DAC, such as dac8 PRO, if you are going to be limited by 96dB maximum of 16-bit audio.

chelgrian mentioned that maybe HDCP requirement of limiting resolution was relaxed to apply only to video, and as long as audio is decoded from original bitstream and output as multichannel audio, the resolution does not need to be limited anymore. But I don't know if it was just a speculation, and I would really like to get some confirmation before dropping big $.

​

 

[Kal Rubinson]

I just recently learned about HDCP requirement that devices that receive HDCP encoded input must limit the unprotected outputs to 48kHz/16-bit, and AES/EBU is unprotected, as far as I am aware.

That kinda defeats the purpose of the whole endeavor to go all digital on processing/decoding and adding your own high-SINAD DAC, such as dac8 PRO, if you are going to be limited by 96dB maximum of 16-bit audio.

chelgrian mentioned that maybe HDCP requirement of limiting resolution was relaxed to apply only to video, and as long as audio is decoded from original bitstream and output as multichannel audio, the resolution does not need to be limited anymore. But I don't know if it was just a speculation, and I would really like to get some confirmation before dropping big $.

​

Frankly, I am less concerned by that than by the issue that all sources piped through the Storm are limited to 48kHz. It means that I do not want to pipe my library of hi-rez MCH through it.

 

[S G]

Frankly, I am less concerned by that than by the issue that all sources piped through the Storm are limited to 48kHz. It means that I do not want to pipe my library of hi-rez MCH through it.

I just did a search for 4K movies on www.blu-ray.com similar to what Welwynnick did here and I see:

4219 4K movies total.

3257 have 48kHz audio.

9 have 96kHz.

352 have 16-bit

3137 have 24-bit.

So it seems it is more important to have support for 24-bit than 96kHz.

For me personally, I feel like I can make a tradeoff to get Dirac ART in StormAudio and give up 96kHz. But giving up 24-bit seems a bit too much. And, like I said, defeats the purpose of having a standalone DAC. Might as well just get StormAudio model with analog output that has 100dB SINAD according to ASR review, still edging out 96dB theoretical maximum of 16-bit.

 

[Kal Rubinson]

I just did a search for 4K movies on www.blu-ray.com similar to what Welwynnick did here and I see:

I have no interest in movies. I am interested only in music.

 

[welwynnick]

Do you know if they limit resolution to 16-bit over AES/EBU to comply with HDCP obligations over sending protected content over unprotected outputs?
I am particularly curious about StormAudio products.

I don't think its because of HDCP, because all the other processors output 24/96 or 24/192 over AES/EBU.
The 16 refers to the number of channels, by the way.
StormAudio output 48kHz audio over AES/EBU because their internal Dirac processing is limited to 48kHz.
I think that will frustrate them intensely.

 

[S G]

I don't think its because of HDCP, because all the other processors output 24/96 or 24/192 over AES/EBU.
The 16 refers to the number of channels, by the way.
StormAudio output 48kHz audio over AES/EBU because their internal Dirac processing is limited to 48kHz.
I think that will frustrate them intensely.

Dirac ART uses MIMO filters. The processing power needed for those has to come from somewhere. Limiting the frequency is one of those places.

 

[welwynnick]

Then they should have more processing power.
I've heard so many excuses.

 

[Kal Rubinson]

Dirac ART uses MIMO filters. The processing power needed for those has to come from somewhere. Limiting the frequency is one of those places.

Storm says it can do 96khz but they have not implemented it.

 

[welwynnick]

Storm know perfectly well that they should be processing 96kHz audio. They will be using a large portion of their available budgets and resources towards achieving this, but in vain. Internally they will be ripping each other to shreds trying to achieve that, while conveying a decorous and professional message externally. They're using the best spin in the world to explain that this is intentional, that this is how they wanted it to be, that this is all that is necessary because of the material available, that this what achieves the optimum result. I know, because I have been that person on the inside of a high technology organisation trying to deliver something that was technically just a bit too far out of our grasp. It's transparently clear to those that have seen both sides, what we hear from Storm Audio is the external message that corresponds to the mayhem inside. It happens a lot.

 

[chelgrian]

Storm says it can do 96khz but they have not implemented it.

The thing is these things are aimed at video and the vast majority of film is at 48k partly because that has been broadcast standard sampling rate for decades and partially because almost all Atmos sound tracks are 48k because using 96k halves the number of objects it can handle from 128 to 64. For various reasons to do with the way films are mixed 64 can turn out to simply not be enough.

I think Theta make a point of putting enough DSP in the Casablanca to run Dirac at 96k but I’mnot sure anyone else does.

I believe Lyngdorf RoomPerfect runs at whatever the sample rate of the input is.