If you were building a 7.1 set up from scratch, what components would you go with?

[Harold Hallikainen]

The headroom question is interesting. Ideally a 0 dB FS peak would be properly played without distortion due to nonlinearity. I have heard it suggested that since about -20 dB FS on a soundtrack is supposed to result in 85 dB (C) at reference position from a screen speaker, we should be able to play 0 dB FS and get 105 dB (C) there. However, SMPTE pink noise has a crest factor of 12 dB, so it is only possible to go up 8 dB before it is clipped by the digital recording. So, do we really need an extra 20 dB? Is SMPTE pink noise a good representation of soundtracks? Also, there will always be some distortion due to nonlinearity. How much is acceptable? I have been impressed with total distortion measurements using multitone (also called "spectral contamination"). The tone amplitudes can be weighted to represent typical soundtrack spectral distribution. Tone frequencies are chosen to avoid both harmonic and intermodulation frequencies so both harmonic and intermodulation distortion are detected.

Harold

[Steve Guttag]

Harold, it gets more complicated than that. You have things like room gain (we are not in anechoic chambers) so there is pretty much another 3dB built in on that alone (for the LF frequencies). Most don't calculate subwoofer power right (most are WAY under specified and used "program" or "peak" power response) while others trying to use the 115dB as their reference, which is not the same as 10dB of in-band gain (hint, if you set your subwoofers up right and then compare say Center to the subwoofer using a simple SPL meter...you WON'T see a 10dB difference in SPL. It is closer to 6-7dB.

For someone to need 3-6dB of headroom in a cinema system, there would need to be something SERIOUSLY wrong and someone incompetent tuning the room (see some 10dB deficit and figure they can just boost the crap out of some frequency). And (and this is the big and)...this would be if one is playing their fader at reference.

Telling some theatre owner to have 6-8dB of "headroom" in their system, across all channels, is just nuts. Think about it, If their subwoofer system needed 2KW...what are the odds that they put in enough power and speakers to kick that up to 8KW (or put in the wall of subwoofers to get the efficiency better)? Sure, you can get a studio or screening room to go all-out...they have the money and rarely need to depend on ticket sales/popcorn to afford their systems.

[Marco Giustini]

I can agree on 6-8dB headroom. But why not 3dB? Yes, this is assuming the feature is being played at 7.0 (never happens) but assuming that, I thought it would be a good idea not to push your drivers to 100% power.

[Steve Guttag]

Worried about power compression? Are you driving them that hard for that long? For what percentage of a movie would your speakers (any of them) be at 100% power? For subwoofers, if you use the continuous power rating, they should be able to handle that for 2-hours minimum...even Tenet didn't run their subwoofers at 100% for 2-hours (and I have the data to back that up). So, to get your 3dB of headoom, do you double your subwoofers or double your amplifiers? Either way, you are doubling the cost (or possibly physical space behind the screen). For the stage/surround speakers, it is almost difficult to not have 3dB of headroom for the types of signals they play until you get into largish rooms. But no, if I calculate that I need 500-watts and the system handles 600...that's fine. And, most people (and calculators) are calculating such that the LF section has to carry the ENTIRE audio spectrum...which is not the case. For 3-way systems, where the LF section comes in the 300-400Hz range, it isn't even carrying ½ the total SPL.

Since the vast majority of the screen channels I use are double-15s...I normally have 1200-watts of continuous PN capability or closer to 2400-watts of program, normally it is just a matter of sizing the amplifier right anyway but no, I don't go for double...that's a waste.

[Marco Giustini]

2-hours minimum without breaking, no?

If I recall a JBL driver spec sheet, power compression happens pretty soon (milliseconds?) and - still if memory serves - driving a driver at 100% will basically reduce its output by 3dB because of power compression. Hence why you don't want to get even close 100%. But you know those things better than me.

It's always a matter of compromise and balance. Looking at the math, you make sure to have some headroom - if it's easy to get 2-3dB with a small investment, the why not. 600W vs 500W is already a good starting point, as long it's not vice versa. All I am saying here is that 3dB headroom is the "sweet spot", what post production uses. So why not keeping an eye on that.

[Marcel Birgelen]

even Tenet didn't run their subwoofers at 100% for 2-hours

Yeah, there were some moments of mumbling without an airplane taking off or exploding within 15 feet of the microphone.

I don't go for double...that's a waste.

Not even in Texas, where everything must be bigger according to the rule of law?

As for speaker and power amp degradation due to continuous use close to the design specifications maximums, there is very little real data available. It's a fact that stuff degrades over time, but it's uncertain to me, how much effect it really has compared to hours of usage.

[Steve Guttag]

Cinema is somewhat unique in that we know what the absolute maximum level out of any channel, any speaker is. There are no wild-cards here to consider, as one would need to consider with a live venue or even running something other than DCPs. We know what the absolute maximum is. We have a much more controlled space and can design the system around the acoustics of the specific space. While 3dB may seem like a trivial amount, it is a doubling of power or doubling of speakers. The cost of amplifiers go up geometrically while the incremental increase in power decreases. Rather then purchase unusable headroom, put the money into better equipment that will actually make the system better. One could always buy something like Meyer and then the problem is solved...amplifiers built into the speakers.

[Marcel Birgelen]

We have about 10dB of headroom in our screening room, which is obviously completely nuts, but rather a result of the size of the room. Getting more headroom in a big room will get almost exponentially expensive. So, if a room is properly measured and sized and only used for the purpose of showing movies, then the 3dB headroom should be more than sufficient IMHO.

Still, it would be nice to see some study of what extra headroom would bring to the table for the longevity of the equipment, although I doubt it will do very much. From what I've seen, the amount of playtime isn't as much of a factor as simply the age of the equipment, especially when it comes down to speakers, as materials tend to get porous over time and not so much the amount of usage. Blown-out drivers, while they still can happen occasionally due to manufacturing defects, shouldn't really be a big problem in any properly designed room though.

[Marco Giustini]

I don't see a reliability issue with a system specc'd with 0dB headroom - also that would imply running everything at 7.0 which is rare. It's more of a sound quality thing. Clearly doing the math for 0dB headroom is - unfortunately - a good starting point. But keeping an eye on the 3dB figure does not hurt IMHO.
We are also assuming that manufacturers don't list their specifications with a little but of enthusiasm - particularly when it comes with budget brands (And yes, at that point, let's invest in proper speakers with a 0dB headroom rather than crappy speakers with 3dB headroom!)

[Steve Guttag]

Barry Farrell at QSC did such at study and there is a video floating about (easier to find back before they eliminated "Cinema" as a division...I think some of those videos are now lost). With Q-SYS, it is possible to monitor/log power usage. The number of times a system uses a significant portion of its power capabilities during action movies is exceedingly rare (there is plenty of recovery time for speakers). I know that he has presented this information at dealer meetings and places like ICTA. In short, for most movies, power demands, even on subwoofers, are rather short duration things (look like spikes on a graph).

As to budget brands...well, that person is also less likely to invest in the proper speakers/amps for power handling too. You know...the one with the single driver subwoofer and just one of them, regardless of auditorium size. I have found that manufacturers specifications to be reasonably accurate. A potential "gotcha" is for subwoofers because sensitivity ratings were not developed for their range so one is still using 40-100Hz (or some are 50Hz to 500Hz), which isn't really the frequency response of interest. So, if you have a subwoofer that is quite sensitive from 80-100Hz but drops quickly below 60Hz, that 80-100Hz will skew the sensitivity up (accurately) yet we only really care from 80Hz on down. It is important to look at the frequency response as well as the published sensitivity to know what the number really means for the application. If the frequency response is provided, it' probably worth the exercise to re-calculate a subwoofer's sensitivity to know how it will REALLY perform.

[Elia Orselli]

Barry Farrell at QSC did such at study and there is a video floating about (easier to find back before they eliminated "Cinema" as a division...I think some of those videos are now lost). With Q-SYS, it is possible to monitor/log power usage. The number of times a system uses a significant portion of its power capabilities during action movies is exceedingly rare (there is plenty of recovery time for speakers)

Is this the video? https://www.youtube.com/watch?v=4uK8A0pk9bU

[Steve Guttag]

No, that is an older one. Some of the later ones are now "missing"...Barry did some tests with music tracks and did let it play through speakers. But he also has demonstrated via graphs the output levels of movie soundtracks. I don't recall if those made it to a video or were just in power point presentations I've seen. He charts did show up in a webinar: https://youtu.be/Q9XZYuGWhcU Look about 17 minutes in.

[Marco Giustini]

I am familiar with the video Elia posted.

Some time ago a QSC system was installed and I do remember QSC was arguing about those "new" power values advertised on their amps. There was a bit of an argument on this subject where QSC claimed that system X would give us plenty of power and headroom while someone was claiming that that was a mistake and that it would have not worked (* meaning: the system wouldn't have the expected headroom).

The system was built following QSC specifications. After the system was commissioned (no, Steve, no crazy EQ to cover up for room anomalies etc), the channels which were being argued about showed clipping on the amplifiers during loud moments (obviously when running ay 7.0).

So... I see your point, Steve, and I appreciate what QSC say. In my opinion it makes sense up to some point, but I disagree with taking those findings religiously and just downgrade the system based on them. It's my personal opinion of course.

Granted, for a budget system it's probably ok to accept some minor clipping during some very loud sequences but keep the costs down. As we all know, it's a matter of compromises.

[Steve Guttag]

I can't speak to the un-named system but I can speak to the ones I've put in. I've only ever been surprised once and it was with a specific product where the sensitivity listed, while accurate, was a VERY poor representation of the real-world sensitivity. I did my own math using a 20Hz-80Hz frequency range and after recalculating, the real-world matched the prediction. In short, even if I were to use that product again, I would be able to accurately predict the results, including with DPA-Q/CXQ amplifiers. In fact, that has already come to pass when one person asked me about using that particular product and I predicted, very accurately, what sort of response it would have (power consumption, headroom...etc.

As for just having 3dB of headroom...is there something above 0dBFS one is trying to account for? Are people playing movies that reach 0dBFS at fader levels above 7.0 (0dB)? Again, if you want to have headroom in a screening room or, in particular, a dub stage where things are being created...go for it. For typical cinemas, put the money into better equipment, not wasted capacity.

[Marco Giustini]

and I don't disagree with that. In fact, I did mention that just a few posts ago

(And yes, at that point, let's invest in proper speakers with a 0dB headroom rather than crappy speakers with 3dB headroom!)

I am NOT suggesting to cut on speaker quality to get 3dB headroom - that would be nonsense. I am suggesting to get the best one can possibly get. The target is good speakers AND some headroom. I do agree that if the budget is short, "good speakers" will have priority.