Sound Calibration questions

[Ron Higgins]

Facts:
1) Small USA based family owned theater with a handful of locations
2) Each location has a handful of auditoriums
3) It is time to check/calibrate the sound levels in all auditoriums, plus a couple of stage speakers will be replaced soon
4) Tech that used to do this, is no longer able to
5) We bought a handheld sound meter: NTI XL2 with a Class 2 microphone
6) Sound processors are JSD60's and JSD80's

Questions:
a) Is the NTI XL2 the right meter for the job?
b) Is a Class 2 microphone the correct microphone?
c) If the NTI XL2 meter is good, is there anything else needed? (ie, the Cinema Meter Option)
d) If the hardware and software is up to the task, is this something we can learn to do? We are comfortable with the JSD settings, just haven't figured out the meter
e) Are there any tutorials, or training options?

Side question, as the JSD's are getting older, what processor would you suggest as a replacement?

Thanks!

[Armand Daiguillon]

I'm sure the actual pros here will hate this response .....but......
With the JSD60 you can download the software to login pretty easily and just EQ it to "make it sound good".
With the JSD software you can manually adjust the EQ while its live playing a movie (or trailer on loop) unlike the Dolby's where it cuts sound when you enter EQ mode.
I'd just sit in the auditorium logged in via wifi and adjust the EQ and the speaker levels by ear.
I've been at SO MANY theaters where its been adjusted by pros with mics and pink noise etc .... still sounded like total crap.
At my theaters though we always get compliments on the sound though - so i'm happy with my "by ear" EQ and leveling.

[Stefan Scholz]

Microphones... You absolutely need to buy a Class 1 microphone, calibrated to a NIST standard...
No, not really. In one of my training classes on the subject, I learnt. Take a microphone A, place it in position A and connect to your analyzing tool. Replace With Mike B, C, D, ... in the same spot, adjust the input gain for the same voltage reading. The difference will be within +/- 1 dB. Move a microphone 1 foot away. You'll notice more than 6 dB deviation in frequency bands.
What is the major discrepancy between a $ 50 cheap microphone and a calibrated $ 1500 microphone? There's a few. The expensive one comes in a machined, neat to look at shell, which is sturdy and has a valuable feeling to it. The $ 50 mike feels cheap, is assembled with hot glue, and once dropped to the floor, it breaks apart. Otherwise they might have different sensitivity and little deviation in their response.
Expensive ones may have a very low sensitivity, which enables them to capture a large dynamic range, by not overloading the input amplifier.
The major advantage of an expensive one, it comes with a calibration sheet, so the manufacturer has at least checked it once for proper operation. So there's no DOA to expect, when receiving it.
What about matched octetts or quads? The answer is given above. A common production line assures quite identical characteristics. The effect of placing them in a room is far more important for deviation, that the time consuming effort to find quads or octetts. Better spend the money on a high grade preamp/ computer interface.

The NTI XL2 is a great, little tool, that is perfect for such work, once you have understood how to use it, and how to do the spatial averaging by walking through the room correctly.
As long, as your USL 80s have a network interface (there's even units which are RS 232 only, that still work flawless), you can place your pad with the control program inside the room, and adjust.
Main part after that was called "Listen!" in Dolby literature. Use familiar content play, and tweak.

[Marcel Birgelen]

First of all, there is nothing inherently wrong with either the JSD-60 and JSD-80. There is also nothing magic about them, but I suppose that's what you expect from your average cinema sound processor.

My $0.02:
a) Is the NTI XL2 the right meter for the job?
Yes

b) Is a Class 2 microphone the correct microphone?
Class 1 please.

c) If the NTI XL2 meter is good, is there anything else needed? (ie, the Cinema Meter Option)
Not really... The Cinema Meter Option is great if calibrating cinemas is your daily job...

d) If the hardware and software is up to the task, is this something we can learn to do? We are comfortable with the JSD settings, just haven't figured out the meter
Most people with a halfway functioning brain should be able to learn how to do this.

e) Are there any tutorials, or training options?
The best training is learning it from someone else IMHO.

[Jay Wyatt]

a) Is the NTI XL2 the right meter for the job?
The XL2 has some cinema calibration routines and if put in the right hands could probably yield satisfactory results. I have one but only use it for mobile SPL checks. If I had a choice I'd prefer to use a SMAART rig or a D2 with multiple mics. The thing isn't cheap and for the cost of it you're probably 1/2 - 2/3rds of the way towards a modern SMAART setup.
b) Is a Class 2 microphone the correct microphone?
Depends on the Class2 mic you have and its published frequency response. NTI recommends Class1 in its manual.
c) If the NTI XL2 meter is good, is there anything else needed? (ie, the Cinema Meter Option)
CMO might offer a nice feature set for cataloging and establishing baselines for your circuit since you mention multiple sites. You'll need a mic calibrator as well to generate a sine tone that can tell the NTI what mic gain to use. Basically, it establishes what 94dB/114dB is with a known tone. All SPL readings thereafter are based on this.
d) If the hardware and software is up to the task, is this something we can learn to do? We are comfortable with the JSD settings, just haven't figured out the meter.
Maybe, but probably best to learn from someone who knows the ropes.
e) Are there any tutorials, or training options?
For that specific measurement platform (XL2) the best bet is user guide for CMO. I don't know all that many folks who are versed in it though, most have come from the D2 or SMAART world. I'm not a huge fan of navigating the menus for the calibration routines and small display.

[Gordon McLeod]

I would recommend the D2 it is cost effective and very good we still use our old R2 on occasion. What is handy is two windows on your laptop one the D2 display and the other JSD 80 software.
A very important thing is know the limitations of the system (amps, and particularly speakers and room acoustics so you dont make thing try to work outside their limits as that is when it gets nasty)
This is my test gear I travel with all told it weighs 445 lbs just under checked baggage limits

[Mark Gulbrandsen]

I'll second getting a D-2. I had both the R2 and the D2. I sold the D2 to another tech once I was ready to retire, but I sold the R2 here on F-T. Note that the R2 is no longer supported for repair. But I know an independent guy that can repair them.

[Steve Guttag]

The D2 remains my favorite cinema analyzer. It is, by far, the most consistent. I was just in a theatre that I hadn't tuned in quite some time (probably close to a decade)...I was changing out the screen channel amplifiers and putting in some new DSP that begins with "Q"...Since I was only working with the 3-screen channels on this go around, I referenced the remaining (Surround, subwoofer) channels. You'd think I tuned it yesterday. Everything was within SMPTE 202. There is no way my mics were in the same location as before yet the consistency, despite those differences and differences in humidity (most likely the culprit for the differences) just reinforces why I think it is best analyzer out there. It cannot do all of the tricks of SMAART but I trust its response more than SMAART...even with "calibration" files.

Since Ron already has the other analyzer, I'm sure there will be a desire to use it. However, without a multiplexer and level normalization to Mic1, it is soooo much harder to get as good a result. With the D2, you are tuning in real time and getting a VERY GOOD average for the place you are in.

As to replacement processors...your choices get limited year-by-year, it would seem.

If you don't need a multichannel analog input, the Dolby CP950 is probably the most cost-effective solution.

And, if you get a ASIO mic pre like the Octa-capture, some suitable mics, like the beyerdymanic and a mic calibrator (or, in a pinch just an SPL meter that is reasonably accurate, to set the level for each mic)...then you can let DAD tune your room for you. Most would get SMAART so you can use the mics yourself and see what things are and do manual tuning too. But learning SMAART is a chore unto itself.

There is the QSC DPM300/300H...though it presumes you have DCA or other Dataport amplifiers and is VERY QSC-centric.

Datasat still has the AP25, the last I heard and they'll be at CinemaCon this year.

You can skip the tuning entirely and go with Trinnov and their OV2 (Ovation) and let it do the tuning as that is one of its chief hallmarks. You just need to supply suitable mics and adapter for it...probably the smallest test rig you could carry.

And, if you want to jump in with both feet, there is QSYS. It is drag-n-drop DSP with not only sound, but control and video. Your imagination is the limit on that. It can run your theatre sound, your lobby displays...whatever. It doesn't auto-tune (though it wouldn't surprise me if a company, like Forward Thinking Designs is working on such a thing...they even made a SMAART interface for it). One does have to be certified to work with it but they have a step-by-step video series to teach you at your own speed. It can handle anywhere from a single up to the largest movie-plex. In case you haven't looked at the QSYS Corner thread, that is what I design most systems with...it is just incredibly flexible.

[Marcel Birgelen]

You can skip the tuning entirely and go with Trinnov and their OV2 (Ovation) and let it do the tuning as that is one of its chief hallmarks. You just need to supply suitable mics and adapter for it...probably the smallest test rig you could carry.

The Datasat AP25 (if you can get one) also comes with "auto-tuning" using Dirac Live. Dirac's software is comparable to the Audyssey Room EQ often found on AVRs.

I'm not necessarily a fan of those auto-tuning solutions. It's interesting how some of those things can come to completely different results in the same room with only a few minutes in between each measurement...

[Ron Higgins]

First of all, much thanks for all of the excellent replies. Lots of info to digest!

We have already tried to use the NTI XL2 with the Class 2 microphone and pink noise without any success (basically no clue what we were doing).

That's why we wondered if we needed something else, or if it comes down to training. We've watched the video on the NTI Cinema Meter Option page, which makes things look pretty easy.

[Marco Giustini]

I'd say training. As Stefan said, a class 1 mike is recommended but you won't get "no success" from a class 2 one

Sound can be tricky. The basic step is to "dial the curve". Depending on the sound system there are quite a few places where to tweak things. An RTA with many options does not help if it's your first time. The final SPL is just a very very small part of the calibration.

I think I've now watched the video you mention. I'm afraid a 6m23s video won't be enough to train anybody on how to calibrate a sound system! Also, I feel the scenario they're giving (with those readings so tightly following the X-Curve) is Science Fiction

So I would say: before you evaluate/purchase more equipment, have yourself trained somehow. Then, you will have a better background to make better decisions.

[Harold Hallikainen]

The JSD-60 and JSD-80 both have RTA firmware in them. Usually a microphone multiplexer and four microphones are used with them since room acoustics will affect the frequency response at a particular location. The JSD-60 also has automatic equalization which some people like and others do not. It adjusts the graphic equalizer to get the measured response within the SMPTE 202 limits. It attempts to provide smooth equalization without big jumps band to band. The JSD-60 can also be set to use parametric equalizers instead of graphic. The graphic equalizer is made with a bunch of biquad filters. When in parametric mode, the filter type (high pass, low pass, band pass, shelving) become configurable along with the frequency and Q. Use of the parametric equalizers allows you to use fewer filters and adjust for the frequency response trend.

An alternative to using multiple microphones and a multiplexer is to use a moving microphone. I've read a paper on that, but have not tried it myself.

I participated in the tests and writing of the SMPTE report on auditorium frequency response ( https://f.hubspotusercontent00.net/h...r0994-2014.pdf ). I believe that the report suggests that we are not trying to "tune the room," but, instead adjust the frequency response of the direct sound from the loudspeaker through the screen and through the air to the reference listening position WITHOUT the effects of room reflections. The analysis method in the report is to determine the frequency response from the impulse response without sound that arrives after the direct sound (reflections). It is my believe (though there is not universal agreement) that the X-curve in SMPTE ST 202 is due to high frequency attenuation of the perforated screen. In the report above, one of the dub stages had a woven screen and had to apply electronic high frequency roll off to get the X curve. Theaters and dub states with perforated screens had to apply no equalization or slight equalization to (in my opinion) adjust the HF rolloff for that particular screen to match the "standard" screen. I further see this as a pre-emphasis / de-emphasis system where high frequencies are boosted on the record side and attenuated (by the perforated screen) on the playback side. This is similar to pre emphasis / de emphasis used in FM broadcasting, magnetic tape recording, etc. An "executive director of digital audio mastering" for a major studio says, however, that he does not boost the highs to compensate for the perforated screen. I replied that if he is doing the mix on a dub stage with a perforated screen and does not boost the highs to compensate for the HF attenuation of the screen, he is creating a mix that sounds bad.

For further reading, see SMPTE RP 2096-1:2017 Cinema Sound System Baseline Setup and Calibration.

Harold

[Stefan Scholz]

What you're looking for is the direct field of the loudspeaker system without reflections and room influence. This is, what our hearing mechanism is used to. Early reflections up to 30 ms are surpressed, and that's why no reflected travel should any position in a room being longer than 50 ft/17 m.
An X curve or whatever is theory, and does not exist when it comes to hearing. The curve is the result of direct sound, integrated over time with reflections on a certain listening position. High frequency attenuation in air is higher, than at lower frequencies, lower frequencies are harder to absorb on walls and ceilings, therefore reflections raise the measured level. Plu, the effect of a screen, especially micro perforated screens have a substantial attenuation.
Adjusting to a curve "X" or formerly "U" in a backwards position was found to give the integrated over time response of a flat response in the near field, measured in a very large room compared to today's theatres.
With perforated plastic screens the amount of hf boost depends on the type of loudspeaker used, some incorporate the screen effects into the horn designs, others are not that good in doing so. A standard perforated screen with the speaker front approx 1 ft behind requires a boost of roughly 15 dB at 16 kHz, substantial attenuation happening.

This is too much theory here. I do my work using my D2, and on rare occasion on my old R2 on the R2, as the 1993 laptop DELL computer that belongs to it has a crackling and brittle plastic shell. Over years I learnt how to use it, how to adjust, and get very consistent results in many rooms. For some others I use a Smaart 8 microphone setup
But this is over the top for the beginner here. The NTI XL 2 is a very precise measuring tool, rather expensive. With the cinema extension (afaik Meyer cinema extension) the results are as good as the R2/D2 method. You just have to walk through the room to get the spatial integration. There are vast improvements over R"/ D2. Built in small LCD screen, whole kit is very handy and battery powered, so easy to carry around, even compared to D2 and Toughbook computer.

I still do not agree on class 1 microphones. These are mandatory if you are into the area of justiciable results upfront a gvmt institution or court trial. Ant in that case have to be frequently checked against an official (NIST etc) standard. None of that is the intention in room tuning. The NTI class 1 microphone adds another grand to the price, for what? A certificate?

What you would need to own though, is a class 1 sound calibrator to check that your measured levels are displayed correctly before the work is started. Once this is done, you're even fine with the $ 50 plastic shell variant.
You don't need to measure at ear damaging levels, even 85dBc is harmful long term, just a level that is high enough not to be influenced by ambient noise.

And then, make speaker position as close as possible to the screen, best not more than 1" for hf, maybe a few. This helps to protrude the screen material.
Start in linear setting and try to achieve a flat display curve with minimum EQ setting. Remember, neighboring frequencies can never be + maximum to - maximun.on the nest.
Then, Listen, Listen, Listen, readjust, tweak if you feel it is required.
THe final step is setting the playback levels, where at fader 7.0 the linear 1/3 octave band section should read around 70 dB, and the LFE section 10 dB above that level.
SPL is for information only, an 85 dBc reading is valid for full range speakers only, which employ a capable bass section.
For LFE it is important to bandwdth limit these electrically at 80 Hz, and to have the usable linear section reading around 80 dB in the 1/3 octave bands. This might sum up on a SPL indicatorto the magic 91 dBc published.

[Harold Hallikainen]

Following up a bit, https://f.hubspotusercontent00.net/h...14.pdf#page=87 says that from 800 Hz up, room reflections have a minimal effect (results were the same with a long window time). This would indicate that the use of continuous pink noise and an RTA gives similar results to an impulse response measurement for frequencies 800 Hz and up. Also, measurements with microphones close to the screen ("close field") showed essentially the same high frequency response as measurements in the auditorium. Measurements farther back showed a little less HF due to air attenuation, but otherwise showed the X curve (due to screen HF attenuation).

https://f.hubspotusercontent00.net/h...14.pdf#page=88 further explains that the screen is a low pass filter.

For low frequencies (below 800 Hz), reflections become more of an issue. I've heard some discussion that many auditoriums do not have sufficient LFE even though they look good on an RTA. This is due to the reflections adding to the direct sound and giving a higher RTA reading while it is desired to have the frequency response be determined based on direct sound, without the reflections.

As Stefan point out, above, our brains "tune out" reflections. I've also heard that an orchestra does not adjust its frequency spectrum based on the room. We hear it correctly, or as we would expect it to be. If we are farther from the orchestra, we get less high frequency due to air attenuation, but that's the real world!

Finally, it's complicated. SMPTE did come up with the above linked procedure for initial calibration and a simpler procedure for "maintenance calibration" that compares current measurements with previous ones. SMPTE currently has several groups working on cinema sound, especially finding a way to measure the quality of the sound such that a number corresponds with the perceived quality (people and the measurements agree on which one sounds better). A danger in the use of perceived quality is that the listener may prefer a sound that is different from what the content producer preferred. Ideally, we hear what the producer produced.

Again, it's complicated. But the measurement instruments and techniques suggested above should do the job, especially on the high frequencies.

Harold

[Randy Stankey]

I agree with all the advice above but I'd also like to add that there is a lot of personal preference involved in tuning a sound system. Some people want that "razor sharp" sound while others might want a "fuller" sound but others want to hear booming bass. Always tune to specifications, first. Then listen to the system and decide whether it's right for you. You can make adjustments for preference, afterward. I've always stayed toward the middle of the road, so to speak. I tune to spec and then make minor adjustments, concentrating on making the dialogue sound clear. I don't go for razor sharp or booming bass.

When serving the public, it's best to do what most people will think sounds best. You're always going to have a few outliers. You're always going to have people who want to hear trumpets cut like a knife or people who want to feel the bass hitting them in the stomach. While it might be fun, most people don't care for those things.

Use the tools you have and follow the advice that the pros have given. Be objective and do it by the book. Once you get things where they are "supposed" to be, put the analyzer down, run a few movies and just listen. Get things to sound "good" but don't go picking minor details until you've had some time to decide. Maybe a day or two. Maybe a week. Get used to how things sound.

In this time, ask some people what they think. Ask other people who work there. Ask some customers that you trust. Watch and listen to the people walking out, after the movie is over and see if you can hear comments. Buy a few customers some free popcorn to tell you their opinions. Oh! Don't forget to ask the boss! They're the one who is ultimately responsible. Take all that advice, put it in the proverbial pot and stir. Then, go back to make your final adjustments based on what you learned.

Yes, there is an objective, technical way to tune a sound system but the bottom line is that you are in business to make money. It's often best to be conservative but, if you've got a customer base that wants to hear sound so sharp you can shave with it then you should consider it.

BTW: If you can find a copy of the old Twentieth Century Fox intro, play that! If you can play that, at volume, and it can send chills up your spine but, when the movie plays, you can still hear clear dialogue, I'd say you've got it just about right.