Audio Processor in the IMB Server

[Marcel Birgelen]

I'm a bit flabbergasted... your honkeytonk mix panel from the local electronics store is more sophisticated than this AIB-2000 box. I guess this thing is only a viable solution for small rooms that will definitely not play anything other than DCI content. If you need to buy another box to do audio processing for external analog sources, why not simply buy a CP750 if you need a rather cheap solution?

[Stefan Scholz]

In the first place, that thing is intended for small (eg office sized cinema rooms) in Asia, or for screening movies in a lobby space or a private environment. How many of these do actually require a 2 ch Analog input? I doubt, that non sync music is a big deal today. If you need pre show entertainment, it can play sound only DCPs made specificly for the venue.
In my understanding the intention was to use active powered speakers, driven by AES 3, or with the breakout box, by analog signals. I would also love AES67/ Dante capability, that would make the system even simpler to set up. It seems to decode and deembed HDMI audio, so even most conventional non DCI source players will perform with a "full" sound processing. Actually, how many of the up to date home units offers analog outputs. All my players for instance, an old Sony BluRay, a Toshiba PinkRay, the FireTV, my Surface pro, all these offer HDMI connectivity. The analog in ports were the ones not used in years, I have to admit. Even the Sony DAT at least offers AES3 connectivity. Sorry, but I see no use for any analog in these days. Yes, my gramophone record players, oh I forgot. But gramo record or casette tape? Neverd heard it's live outside hipsters.

Made for the specific mini cinema market, that solution could do quite well, and with AES 3 compatible active speaker systems, and HDMI connected players, most, or nearly all of what you need in real life today can be handled. Let's see it, it's not targeted for multi complexes and large screen venues. You can do 7.1, and you can use CH 7/8 for that, no need for HI VI in most of these, and if uses 17/18, and still biamp is possible.
It will not beat a QSys or Yamaha (in my case) networked system, and it's not required.

[Steve Guttag]

Stefan, I have no doubt that there is a market for it, but I think it is a tiny one outside of probably the reason they put it in (as mentioned, I suspect that there was a specific market that asked for it).

As for analog, the vast majority of amplifier options have analog inputs (particularly the "value priced" ones. The grown digital inputs I see are AES67/Dante but that also comes with a cost overhead. Powered speakers really cuts down on suitable speaker choices and if they have to also have AES3 inputs, you've cut the pie even slimmer.

Where it would fit best is in an existing system that has something like the QSC DCM series of monitor/crossovers. The DCM-10D and DCM-30D even have digital inputs. If you have amplifiers that have some form of crossover or option (e.g. a Crown DSi series or a QSC DCA, or ISA amp with XC-3 crossovers, you won't have address having any sort of booth monitoring, if there is a booth.

As for HDMI, the SR-1000 will de-embed but I don't think it will decode (Dolby or DTS). One has to be careful about channel routing then as HDMI channels use 7/8 for back surrounds...most DCPs put the disability channels there so some channel routing will be needed.

Currently Tascam has two players with full analog outputs (7.1 plus 2-channel mixdown in consumer or professional format). BD-MP1 and BD-MP4K. As for 2-channel stuff...in our installs, if the site has music or some such source. They also seem to get used, in a pinch, for some form of laptop/power point shows.

I just think that the appeal of an audio processor within the server will have a limited, specific purpose, appeal rather than a general purpose. That is true about Dolby's IMS3000...I don't want to sound like I'm picking on GDC. As a concept, I think it probably isn't the best fit for most installations. That said, Dolby has done it, GDC has now done it and QSC was attempting it (CMS-5000), sort of. It had QLAN spigots so it would have been hard to claim it was an audio processor on-board (other than an MDA type) so certainly somebody thinks putting the fader and EQ in the server could be a thing.

[Harold Hallikainen]

I think the CMS-5000 just did object-based audio rendering in the server. I always thought that was better than external rendering with all the content transfer, security, and synchronization problems to be dealt with by an external box. QLAN provided a way of getting all the rendered audio channels "out of the box." QLAN handles lots of audio channels with much less hardware than AES3. The CMS-5000 also watermarked all those audio outputs before they left the box. It was then up to QSYS to handle the rest of the audio requirements (routing, EQ, delays, volume control, etc.).

Harold

[Marcel Birgelen]

I consider the CMS-5000 design to be superior of that of the SR1000. But I'm pretty sure QSC will follow-up with a product that does have proper "networked audio" output in the form of AES67. The lack of expandability options on an IMB-design will simply not allow them to retrofit a dedicated network port for anything like AES67 on the SR1000.

The more processing power we can put into the IMB, the more functionality we can put in there. And while I still somehow like to have separate boxes for separated functionality, it's clear that the future will consolidate those functions even more.

[Tony Bandiera Jr]

Tony, How does one bypass in Q-SYS if an amp or stage speaker fails? Like this...in seconds:

Bypass.JPG

Or, if the client is so inclined (on top of the above), they can have a spare amplifier, already in the rack, already being fed the signal (but in standby).

AmpBypass.JPG

The speaker terminal block on a DPA-Q (CX-Q) are large Phoenix type connectors (can take up to #8 wire) and plug in for up to 4 channels.

It depends on several factors as to amps behind the screen. In the case of above, the amp rack is in its own room, on ground level and the load-center is about 3-feet from it. If the theatre has a baffle wall (and this one does), it is more viable as a speaker room can be created to keep the amps more accessible and keep the gummy bears out. It isn't for EVERY space but one could also have an amplifier "closet" to just keep the amps accessible, and out of the public-eye or casual observance. Remember too, these amps can be up to 8-channels so two amps normally does a typical 7.1 theatre for behind the screen and at 2U each, it isn't a lot of space you are trying to find.

That said, what if they put amplifier modules in the speaker? Installation would be as simple as an Ethernet cable and AC. Design it so the amp module(s) come out easily for service/replacement. But again, it won't be for every space but it could be a good fit if the theatre is designed with access in mind.

Thanks for posting this..that's even better than the "old school" way of course. (I got out of active work in the biz before Q-SYS took off and none of the private rooms I was working on around 2014 were using it yet.)

Now my question would be, how many of the sites you have done Q-SYS with have any staff competent enough to get into the system to do this emergency bypass? (I am excluding the obvious option to do it remotely by NOC for this question.) I have a hunch that it is about the same as the former cinemas back in the day who had at least one person on staff who could do a hot repatch as I 'd mentioned.)

[Steve Guttag]

Tony,

I think that quite a few...I put that bypass button pretty much front and center (well, off-center but hardly hidden):

Screen Shot 2021-05-23 at 2.34.06 PM.png

I put quite a bit of thought into "what ifs" despite not needing them once...yet. The day will come but I think we're ready. That is particularly true where we have a spare amp. I thought about putting the spare amp on relays and such but then I've introduced a point of failure for a rare occurrence item. I suppose, I could also automate if something goes amiss with one of the amps to automatically switch into Bypass too (if that "Ok" section changes to a fault on any of the amps but again, there is but so much effort I'm going to put into it, at this stage, to protect for the rare occurrence thing. However, that is certainly possible...even to automate it to the point that a spare amp could be switched into replace a failed one. I just never want the "backup" system to be the source of failure.

-Steve

[Harold Hallikainen]

Do most amplifiers have a relay on the output that delays the speaker connection on power on (to avoid a speaker blowing pop as the output swings wildly on power up)? If so, it MIGHT be interesting to just put two amplifiers in parallel and ensure that only one is powered at a time.

[Marcel Birgelen]

Do most amplifiers have a relay on the output that delays the speaker connection on power on (to avoid a speaker blowing pop as the output swings wildly on power up)?

Any amp that takes itself somewhat serious should have such a basic protection. The way they do this varies between models, but a relay is a pretty common method. Other method involve slowly opening a transistor or valve. Some actually monitor the output and wait until it's stable, others are simply time-delayed.

[Steve Guttag]

I've definitely seen it both ways...with and without a relay. Regardless, there is some form of protection. Kintek's old KT-100 and KT-102 amplifiers were notorious for the big-bang when they turned off.

In any event, I'd be none too eager to depend on amp relays to only have one on at a time.

QSC makes a snazzy thing for their Dataport based amps for Q-SYS, for those that want that sort of backup system. Again, in a Q-SYS system, it wouldn't be too hard to have relays on the outputs of the amps and create a buss so one could switch the spare amp to backup any of the channels...but then you've introduced several points of failure too.

[Harold Hallikainen]

I was also concerned about adding points of failure by adding external relays. That's why I thought relying on internal relays would not make things any worse in terms of reliability. But, it would be really good to make sure both amps are not powered at the same time if the outputs are connected in parallel...

Harold

[Marcel Birgelen]

What I've learned over the years is that completely redundant systems, especially those that include some kind of auto-failover, are extremely hard to do right. Often those systems end up being overly complex, expensive and it's the redundancy that's causing more problems than it's actually solving. Even semi-automated systems are often far more complex than you'd initially think. This is especially true if you want to protect yourself against events that don't happen all that often.

One of the primary issues of any system doing any kind of automatic fail-over is to reliably detect all the possible failure modes of such a system. Even something apparently simple like monitoring power input can be more complex than you think, especially if you're dealing with stuff like 3-phase power.

Then, the next hard thing to do is to get the thing you want to fail-over to into the same state as the thing you're coming from. In case of appliances that are MOSTLY state-less, like amplifiers, this may be relatively easy, but in case of more complex stuff like playout servers this requires a whole lot more infrastructure to keep the state synced between the primary and the backup system. Then there is the risk that the state the failed component is in, is causing the problem in the first place. A bug in System A, will likely trigger the same bug in Standby System A'.

In this particular case, where we're considering a semi-automated fail-over, that can be triggered at the push of a button, we're already seeing the possible complications:

- Having two amplifiers feeding into the same speaker may have disastrous results for that speaker.
- Putting in some kind of active switch between the amps and the speaker may solve this problem, but introduces extra complexity into the system and another SPOF.
- Putting in two remote-controlled relays between the amps and the speakers removes the SPOF, but introduces only more complexity and doesn't entirely remove the risk of two amps feeding into the same speaker, as one of the relay-boxes may have a hanging relay, for example.

So, over the years I've become more of a fan of having backups in the way of having some easily accessible spare equipment available. If possible, already installed in the rack/location you're likely going to need it. This combined with some clear instructions and people that are trained to execute those instructions.

In this case, swapping some cabling from Amp A to Amp A' is both the cheapest and most likely one of the most reliable solutions. The only disadvantage is that you can't do it remotely, but then again, any professional cinema operation should at least always have one person on site with some decent knowledge of the systems in that same building.

[Carsten Kurz]

How often do amplifiers actually fail? Sure, in ATMOS installations, there is a higher statistical rate. At the same time, in ATMOS installations, there are easy means to mask it.

I am thinking for a good while now about easy ways to detect individual speaker failures in classic parallel surround setups. Easy in the sense that it allows them to be checked often (enough).

[Harold Hallikainen]

Back when I was building radio stations, we'd use the older equipment to build a parallel backup station. We did a system in Los Angeles where an FM station had two transmitters and two antennas with no antenna switch. On changeover to the standby transmitter, the standby would be brought up at 50% power, then the main dropped. The standby would then be brought up to 100% power. To switch back, the reverse was done. There were separate audio processors and studio to transmitter links for each transmitter. The only thing in common was the tower. Some stations have auxiliary sites so not even the tower is in common.

I agree that the failure rate of amplifiers may be low enough that efforts to quickly get around failures could actually make the problem worse. The relatively common method of getting around a failed center channel by temporarily feeding center to left and right is good. To parallel the broadcast analogy, you'd have duplicate speakers and amplifiers, then choose which one to drive. Or maybe drive them all for more headroom normally.

On amplifier output relays, I'm thinking now that class D amplifiers that rely on pulse duration modulation would not require such a relay. It would be very easy to not enable the output transistors until the PDM circuitry is up and running.

Speaking of pulse duration modulation, I heard of a very clever idea a few years ago. A typical pulse duration modulator generates a single pulse of varying width at a certain frequency, either through analog techniques (triangle or sawtooth wave driving an analog comparator) or digital techniques (a digital counter driving a digital comparator). The interesting technique is to use a random number generator instead of a counter. The random number generator drives a digital comparator that provides an output if the random number is below a threshold corresponding to the instantaneous audio sample. As the sample rises, the random number is below the sample more often. The result is a high frequency series of pulses whose DC average corresponds to the audio sample. But the ripple frequency is much higher than with the other method, so lower cost filtering can be used. I guess, though, that as the sample voltage increases, the output pulse frequency decreases as more of the random pulses combine to longer pulses.

I have never implemented the random number method but thought it was very clever.

Harold

[Carsten Kurz]

Isn't that the idea behind certain DAC technologies?