NEW UPDATED Room Calibration for Film and TV Post

[abluesky]

The problem that tends to surface, when aligning a subwoofer using full-bandwidth pink noise in a "typical room", is that the in-room response of the sub isn't "flat" and may not be as extended as the "spec". So the best way to set it, is using a spectrum analyzer, or using bandwidth limited pink noise and a SPL meter, if an analyzer is not available. Otherwise, the best you can do is +/- 1 or 2 dB using full-bandwidth pink noise.

Cheers!

[Eric L]

Thanks guys for this informative last few posts. Although I have mixed in many different types of rooms, I have little experience calibrating rooms with an RTA using the different flavors of pink noise. If it is not too much trouble, maybe someone could directly answer the following for me, as I find myself in this very situation currently. I would like to mention that I have read this thread thoroughly multiple times, including all the links, and have had off thread discussions with other mixers already, yet I am still confused.

I have the Dolby -20 wideband Pink graciously offered by Dr. Sound.
I also have all the calibration files graciously offered by Pascal and his BlueSky Brethren.

Utilizing these tones and an RTA that also works in SPL mode, what numbers do I shoot for with each tone?

I am a little confused as I have already made an attempt at doing two things with these files. I first calibrated and tuned the room with Bluesky's wideband Pink. However, once I got a flat response on the RTA, the high end sounded completely wrong. Too much highend. I thought that this must not be the correct Pink Noise to use for frequency adjustments.

So I switched to the Dolby Pink and made the level and frequency adjustments and everything sounded lower than it did when I used the BlueSky pink, but the highend sounded correct and the response seems pretty flat to my ears when playing material mixed in other facilities. Although it does seem that there is less power from the speakers now.

However, now I am not sure if the LFE was done correctly, because I used the Dolby pink and measured with the RTA in SPL mode, not the band limited pink that Pascal suggests I use.

So assuming a bass managed system...

Measure the center channel first with Dolby wideband pink and set it at 78dBC
Send the center channel Bluesky band limited lowend pink (so the center channel should be silent) and measure the sub response and set it at "what"? 78dBC right?

Or do I use the dolby wideband pink and mute the center speaker while measuring the sub response and set it at "what"? Again, 78dBC right?

Now, sending Bluesky low end pink to the LFE, measure that at "what"?

Or should I use the Dolby Pink for the LFE? (I don't think so based on Pascal's last post)

When should I use the Bluesky wideband Pink? Why is it different than the Dolby Wideband Pink? I guess I do not understand the different flavors of pink, and when to use which one.

It seems that using the Dolby Pink for frequency adjustments is the correct choice? Is it? Why?

Thanks again. I hope this isn't asking too much.

[abluesky]

So assuming a bass managed system...

Measure the center channel first with Dolby wideband pink and set it at 78dBC
Send the center channel Bluesky band limited lowend pink (so the center channel should be silent) and measure the sub response and set it at "what"? 78dBC right?

Or do I use the dolby wideband pink and mute the center speaker while measuring the sub response and set it at "what"? Again, 78dBC right?

Now, sending Bluesky low end pink to the LFE, measure that at "what"?

Or should I use the Dolby Pink for the LFE? (I don't think so based on Pascal's last post)

When should I use the Bluesky wideband Pink? Why is it different than the Dolby Wideband Pink? I guess I do not understand the different flavors of pink, and when to use which one.

It seems that using the Dolby Pink for frequency adjustments is the correct choice? Is it? Why?

Thanks again. I hope this isn't asking too much.

Hello Eric:

If you are calibrating your system for -20dBFS = 78dC, the you will measure the following.

L,C,R,LS & RS = 78dBc (each)
bass-managed subwoofer (BW LF pink to any one of the main channels) = 78dBc
LFE - using the BW LF PINK (not calibrated separately in our system) = 88dBc

The full-bandwidth pink noise should only be used with a RTA. In general the bandwidth limited files are more accurate and take out some of the variables with regard to subwoofer / room response and SPL meter LF response variations. We included the pink noise file for reference and for use with a RTA. All of the test signals were created and measured on a Prism Sound dscope III audio test system and conform to the AES 17 guidelines for digital audio measurements.

I hope that helps...

Cheers!

[Eric L]

Hello Eric:

If you are calibrating your system for -20dBFS = 78dC, the you will measure the following.

L,C,R,LS & RS = 78dBc (each)
bass-managed subwoofer (BW LF pink to any one of the main channels) = 78dBc
LFE - using the BW LF PINK (not calibrated separately in our system) = 88dBc

The full-bandwidth pink noise should only be used with a RTA. In general the bandwidth limited files are more accurate and take out some of the variables with regard to subwoofer / room response and SPL meter LF response variations. We included the pink noise file for reference and for use with a RTA. All of the test signals were created and measured on a Prism Sound dscope III audio test system and conform to the AES 17 guidelines for digital audio measurements.

I hope that helps...

Cheers!

Hi Pascal. Thanks for the response. It sort of helps. I am using an RTA. So using the Blue Sky BW LF Pink, set the LFE to 88dBc? Not 82-84dBc as the Dolby Docs seem to suggest? Those must refer to using full range pink noise, correct?

When I used the Blue Sky full band pink, and then made eq adjustments to get a C weighted flat response, the top end sounded too hot when I played back audio material that was mixed in a Dolby Calibrated room.

When I re-eq'd with the Dolby Pink, it sounded flat and correct, yet it seemed to lack power (compared to what I had just had ) even though the RTA's spl measurement still said 78dBc.

So I think I understand most of what is going on, except the difference between the two pink noise files, Blue Sky and Dolby. Can you explain the differences to me? Why the difference in the high end. Is it a power difference across the octaves. I'm still confused. I'm getting different information from different sources. Neither being wrong, but my other source only uses Dolby Pink and they have suggested numbers for me to shoot for, but they are not what you are saying.

Lastly, when I listen to the two pink noises at a fixed level, the Dolby -20 pink sounds louder to my ears, has a larger waveform on the timeline in Pro Tools, and measures higher using a plug in analyzer. So this kinda answers my questions as to why the perceived lack of power after I re-calibrated. I just want to know which is correct, because they are definitely not the same.

Thanks again. As always, I appreciate your assistance.

[abluesky]

Hi Pascal. Thanks for the response. It sort of helps. I am using an RTA. So using the Blue Sky BW LF Pink, set the LFE to 88dBc? Not 82-84dBc as the Dolby Docs seem to suggest? Those must refer to using full range pink noise, correct?

When I used the Blue Sky full band pink, and then made eq adjustments to get a C weighted flat response, the top end sounded too hot when I played back audio material that was mixed in a Dolby Calibrated room.

When I re-eq'd with the Dolby Pink, it sounded flat and correct, yet it seemed to lack power (compared to what I had just had ) even though the RTA's spl measurement still said 78dBc.

So I think I understand most of what is going on, except the difference between the two pink noise files, Blue Sky and Dolby. Can you explain the differences to me? Why the difference in the high end. Is it a power difference across the octaves. I'm still confused. I'm getting different information from different sources. Neither being wrong, but my other source only uses Dolby Pink and they have suggested numbers for me to shoot for, but they are not what you are saying.

Lastly, when I listen to the two pink noises at a fixed level, the Dolby -20 pink sounds louder to my ears, has a larger waveform on the timeline in Pro Tools, and measures higher using a plug in analyzer. So this kinda answers my questions as to why the perceived lack of power after I re-calibrated. I just want to know which is correct, because they are definitely not the same.

Thanks again. As always, I appreciate your assistance.

Hello Eric:

The bandwidth limited pink noise files are level compensate for use with an SPL meter, so the numbers I listed above are correct (78 LCRLS, 78 BM SUB, 88 LFE, or 85dBC, 85dBC, 95dBc for film). They are much different from using pink noise, which is full-bandwidth. Just to be clear, in our systems you don't calibrate the LFE channel, you calibrate the bass-managed subwoofer channel. The 10dB of in-band gain is added electrically, via a 0/10dB toggle and is referenced to the subwoofer. I just want to make sure that you are not calibrating the subwoofer at +10.

As a quick side note: If you are using full-bandwidth pink noise, with an RTA, then you will need to first concentrate on getting the relative level between SAT & SUB correct. What I would do is first is set the levels of the system using the bandwidth limited pink noise files, with an SPL meter (using our directions / levels). After this point, I would use the full-bandwidth pink noise to confirm that the level between SAT and bass-managed sub is correct (relatively flat through the crossover point). If the SUB level is a little high or low, you can then tweak it to match the response of the SAT.

With regard to EQ: You want to be very careful using a RTA to EQ a nearfield system. RTA's only measure level Vs frequency, not level / frequency Vs time (which the way we hear). I would not do much, if any EQ above 500 to 2kHz at the absolute highest point and would be careful using any EQ in a small room in general. This is because when you sit nearfield dominant, your ear will ignore many of the reflections of the room, which a RTA will not and the perceived response will almost always be very different from what you measure using a RTA. Also, when you say "made adjustments to get C-weighted flat response", this is a little confusing. Some RTA's will allow you to do 'display weighting', however the display should not have any weighting when you are adjusting frequency - the display should be flat - only the SPL measurement should be weighted.

With regard to the difference in perceived and measured level with Dolby pink: As I mentioned above, the test signals were created and measured on the Prism Sound dscope III audio test system and conform to the AES 17 guidelines for digital audio measurements. Because pink noise has a high crest factor, the levels of the pink noise signals were made using a time averaging technique, where multiple time samples are captured and averaged over a ten second interval to calculate the RMS level. The discrepancy you are seeing may be due differences in the crest factor of the Dolby pink noise and the averaging used to determine the RMS value.

Feel free to e-mail me directly if you need any any additional questions. [email protected]

Cheers!

[Eric L]

Hi Pascal,

Thanks again. Your last post and an off thread e-mail has helped clear things up for me. I am sorry if some of my comments are confusing. Considering my recent state of mind, I hope that is understandable.

I'm sorry I was not clear that this is not my personal room I am working on. Unfortunately for me, this facility is not using one of Blue Sky's systems. It also is not technically a near field system. It is a mid-field speaker set up, behind a perf screen, in a smallish room, (22' x 15' x 9'). The speakers are made by one of those companies with three initials in their name, starting with J and ending with L, and the facility had not even been using the built in bass management crossover network. It was only after I listened to the room that I felt that two things were wrong.

First, the low-end was off and not being controlled very well, which is why I have re-set up the system (temporarily) using the built in bass management. They are hesitant to spend any money, but I think I have convinced them to buy a BMC MkII. I do not think that I will get it before the end of this current mix, however.

Second, because of the perf screen, the upper mid range and high end were being seriously attenuated. They already had EQs in place, which had been pre-set for I different room! ugh! Hence my need to re-Eq the room.

My reference to "flat at C weight" was mentioned only because with the RTA I am using, I have to use an in line XLR filter that can be set at C or A weight an inserted between the mic and the RTA. The RTA itself has no "weighting" display that can be toggled on or off.

Thanks for all your help. If I have any other issues I will contact you via e-mail. Oh, and no, of course I was not calibrating the sub at +10, only the LFE. My only real confusion was brought on by the different responses of the two pink files. I erroneously assumed that wide band pink noise would all be the same, but now I know that this is not the case and I feel very confident following the steps you outline in your last post.

Again, I thank you for your terrific customer support. This has gone above and beyond what was expected from a company I buy speakers from.

[subwoof]

Congratulations Marti, on having over 36,000 views on this original post! I think I might have been number 36,001 to see it.

Mike

[Floydio]

36 002nd viewer and I think I've finally got it!

Pascal, Richard, Marti, thanks for your help.
If you'd like to see my "how-to" I've typed up, please send me a message.

Floydio,
now living in the zen garden of LFE pink noise.

[Arno Peeters]

If you'd like to see my "how-to" I've typed up, please send me a message.

Yep, always good to have around: can you send it to infoATtapetvDOTnl?

[abluesky]

Pascal, Richard, Marti, thanks for your help.
If you'd like to see my "how-to" I've typed up, please send me a message.

Sure - [email protected]

Cheers!