Film-Tech Forum ARCHIVE: "Perfect" masking and aperture

This is topic "Perfect" masking and aperture in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=004129

Posted by Steve Kraus (Member # 476) on 10-22-2002, 04:56 PM:

If you wanted to be absolutely perfect and assuming you have a straight shot at the center of the screen (no keystoning) is it more proper to show the precise standard aperture dimension (cut right on the outline for whatever format) and then put the masking just enough into that image to hide the aperture edge fuzziness...OR...go just a bit oversize so that what ends up on the screen is the exact dimension. Which is more perfect? I vote for the latter but that means there is a chance something weird is going to project on the masking and still be a bit visible.

Posted by Tim Reed (Member # 137) on 10-22-2002, 05:46 PM:

I'd say you're right. I'd want to see the image to the precise dimensions. If you figure that the image area goes out at least to .839, which it did for scope, then that should be enough of a safe area to cover your masked shadow.

------------------
Better Projection Pays!

Posted by David Stambaugh (Member # 1102) on 10-22-2002, 05:51 PM:

Clearly this is only a "non-binding advisory vote"

but if Option #2 introduces any possibility of something weird being projected on the masking, I would vote for Option #1. I'll take nice sharp black edges with slight overshoot anytime over fuzzy edges or the possibility of something more obnoxious showing.

Posted by Tim Reed (Member # 137) on 10-22-2002, 05:58 PM:

In a theatre, with release prints, I'd agree there, David - or at least take out the difference half way. But in a screening room or production environment like Steve's, I'd go strictly for the standards.

------------------
Better Projection Pays!

Posted by John Walsh (Member # 168) on 10-22-2002, 05:58 PM:

Funny, because an installer and I were talking about just that. Apparently, when Disney does a theater, they order full open aperture plates and move in the masking, an idea I really like.

I think the answer may be somewhere in between. If the projector design places the aperture close to the film plane, the aperture will be in better focus (ie; less fuzz). If farther away, more fuzz, and probably would need an over cut aperture so it's "clean" right to the 0.825 line.

Posted by Gordon McLeod (Member # 33) on 10-22-2002, 06:47 PM:

I go for masking on spec and oversize the plate slightly

Posted by Steve Guttag (Member # 268) on 10-22-2002, 06:49 PM:

Funny this is a complaint I have the document RP-40...it never discloses "which is correct."

As to using full open apertures...for what lens? You take a precut 446 x 825 aperture and chalk up several different brands and focal lengths of lenses and you'd be amazed at the variance of what is shown. The new ISCO plus series RED lenses practically look around the aperture plate trying to get all of the light!

I say, anything beyond .825 on width is undefined...there is no obligation to protect beyond it (and likewise for height). Therefore I vote for option 1...that is...your plate just shows a slight portion of the line in question then mask it off for a crisp picture. There is a safe action area and I believe up to 10% crop on all directions is supposed to be safe.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Bob Maar (Member # 763) on 10-22-2002, 08:10 PM:

Very interesting....let us play this out....

Posted by John Hawkinson (Member # 1135) on 10-22-2002, 09:02 PM:

Well, "obviously" you have both sets of plates (and different lenses for each

), and you rehearse with the larger apertures and if they work out OK, than you use them, otherwise retreat to the smaller.

--jhawk

Posted by Matt Close (Member # 996) on 10-22-2002, 09:26 PM:

Interesting ..... I am in the midst of this very problem ..... what to do?

I am trying to get the 'right' amount of picture on the screen .. had to buy new lenses as the ones originally installed here were the next size along resulting in cropped images all the time ..... anyhoo ...

My quest for perfection has given me problems with the occasional reel, and a LOT of trailers!! It seems something from the sound track development process 'leaks' on to the picture area just a little creating a bar of purple haze on the left side of the screen ...... so I am thinking a tiny bit undersize L > R in everyday practice with masking to suit, and a tiny bit over size T > B to get the right amount of height in the image and bring in the masking.

Posted by Joe Redifer (Member # 3) on 10-23-2002, 01:07 AM:

I, without doubt or hesitation, choose option #1. For one, it is always easier to tweak an "undercut" plate than an overcut one for obvious reasons. Also, with all of the garbage on the outside of the safe image area like the flashing line at scene changes on many scope movies or boom mikes in some flat movies, I feel it is the better option even though the difference is minor.

Posted by John Pytlak (Member # 331) on 10-23-2002, 08:09 AM:

The screen masking should define the projected image area, as determined using the SMPTE 35-PA (RP40) test film. The aperture should be sized to overshoot the masking just enough to hide the "fuzzy" edge. Definitely the aperture opening should NOT be cut so large as to show the DTS timecode or negative splice lines, even if they end up on the black masking.

Caring About Composition:
here

------------------
John P. Pytlak, Senior Technical Specialist
Worldwide Technical Services, Entertainment Imaging
Research Labs, Building 69, Room 7525A
Rochester, New York, 14650-1922 USA
Tel: +1 585 477 5325 Cell: +1 585 781 4036 Fax: +1 585 722 7243
e-mail: john.pytlak@kodak.com
Web site: http://www.kodak.com/go/motion

Posted by Steve Guttag (Member # 268) on 10-23-2002, 08:27 AM:

John,

But whereas anything outside of the allowable space is undefined...say beyond .825 one should never see the .825 line once the screen is masked. Now some projectors and gates allow for a much smaller fringe area but it will have a fringe area none-the-less. The only way it can make any sense is to have a safe area (10% crop) and that space is used to hide any of the mechanics of showing the image...this includes the fringe area and any keystoning anomolies. If the film makers play by the rules then no significant information is lost by the cropping and no extranious garbage (soundtrack, splice flash, pad roller scuffing...etc) is shown.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by John Pytlak (Member # 331) on 10-23-2002, 08:42 AM:

Steve wrote: "But whereas anything outside of the allowable space is undefined..."

But the standards DO define much of what lies outside the projectable image area. For example, SMPTE 93 and SMPTE 139 define the film perforations. SMPTE 59 defines the camera aperture. SMPTE 111 defines the printed area. SMPTE 40 and SMPTE 203 define the analog track area. RP115 defines the DTS time code area.

I do agree that extraneous items may appear in the printed picture just outside the projectable image area, so filmmakers need to be diligent in keeping microphones, light stands, flags, and other items from being seen, even if only on the image that spills onto the masking.

I'm sorry that you have found it necessary to sometimes crop the image by up to 10% from the standardized projectable image area for reasons other than simple keystone correction. SMPTE 195 defines the projectable image area that should be seen by the audience. SMPTE 195 states in Note 2: "In the absence of specific instructions to the contrary, it is intended that the actual projected film image area be the largest appropriately-shaped figure that can be inscribed within the specified dimensions".

If "...extraneous garbage (soundtrack, splice flash, pad roller scuffing...etc) is shown...", someone is not meeting the published standards, and "violating" the standard projectable image area (that has actually been larger in the past). Taking another 10% crop seems the wrong way to go...we need to address the reasons the SMPTE 195 projectable area is "violated". (You oppose cyan dye tracks, which will completely eliminate soundtrack developer splashes and bleeding into the picture area).

Posted by Michael Schaffer (Member # 1204) on 10-23-2002, 08:56 AM:

I have worked mainly with Kinoton projectors where the aperture plate is quite close to the film so it has never been a big problem to get very close to the complete picture because the fuzzy area is rather narrow. Personally, I hate to see the fuzz very

much!!

Posted by Steve Guttag (Member # 268) on 10-23-2002, 10:31 AM:

Sure, if one goes out far enough, you can find the defined ones like the optical soundtrack and the perforations. It camera aperture is defined...I can't lay my hands on SMPTE 111 to see the printing dimensions. Then one has to add up the total weave/jump allowable from both camera and projector...take into account the current state of contact priniting everything...one is playing with fire to actually pull the masking until they see the actual lines and then cut. While I agree that seeing fuzz is bad, seeing all the other garbage is worse.

Unless I see something to the contrary, the area beyond .825 or the appropriate height on the release prints is undefined. If it is defined then it should be stated somewhere that it's purpose is to fall on masking systems and should be protected space as such.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Joe Redifer (Member # 3) on 10-23-2002, 02:06 PM:

Well Michael Schaffer if the aperture plate is close to the film, then there is obviously no need to even consider option #2 in any scenario so

Michael Schaffer

too!

Any projector that has the aperture so far away from the film *cough* Simplex *cough* should be redesigned so that the aperture is close as possible. There is no reason to have it so far away. Any projector that does just has a bad design.

Posted by Stan Gunn (Member # 554) on 10-23-2002, 06:20 PM:

I believe that option 1 is the go, as it seems that the labs do not care much today about splices on scope, or other such things as sharp frame lines.
I cut new plates for a pair of cinemecs about 3 years ago, now they show a blue edge on the left, on NEAR all films both formats.
The plates were all cut to a tight .825 width, so what changed?
The plates,or the film. I can assure you that the film guides etc have not changed, as the guide roller at the top of the gate wears it would allow the film to move closer to the outboard possition therby reduceing any fuzz on the sound track edge.

I beleive that all plates should be cut to the correct width and height for the format being shown, my later years before forced retirement were spent in an art house twin, where we ran the film in its correct format, or as close as we could get, film done right is to show the audiance only the content of the frame and not fuzzy bits on the edge or top and bottom.

------------------
KALEE FOREVER. CARBON ARC, THE ONLY LIGHT FOR THE STARS!
ALL PARTS FOR VICTOR AND KALART VICTOR 16MM PROJECTORS.SERVICE TO 35 AND 16MM PROJECTION EQUIPMENT.
35MM sprockets made to order.

Posted by Michael Schaffer (Member # 1204) on 10-23-2002, 08:48 PM:

I hope you didn`t take offence in my including your icon in my post, Mr Redifer. I just included it because I think it is a very strong icon. Note I also included

which is very positive since dlp is digital and therefore perfect...

Posted by Steve Kraus (Member # 476) on 10-23-2002, 10:51 PM:

Hey, if you wanted perfect we'd have point sources of light, no lenses, and no fuzzy edges.

Posted by Joe Redifer (Member # 3) on 10-24-2002, 03:29 AM:

I did take a bit of offense at first, but then I looked at the assortment of your icons. But I just had to reply. And since you don't have a pic, well, I just used your name. Anyway we agree about the aperture thing, so I wasn't trying to be argumentive.

Posted by Jeffry L. Johnson (Member # 453) on 10-25-2002, 03:17 PM:

I place my masking edge at the SMPTE Standard specified dimension.

Here are the relevant SMPTE Standards.

ANSI/SMPTE 195
SMPTE Standard for Motion-Picture Film (35-mm)--Motion Picture Prints--Projectable Image Area
Dim A 0.825 in Nom (aperture width)
Dim C 0.738 in Ref (picture centerline to reference edge)
Dim D 0.324 in Min (picture edge to reference edge)
Dim E 1.151 in Max (picture edge to reference edge)

ANSI/SMPTE 59
SMPTE Standard for Motion-Picture Film (35-mm)--Camera Aperture Images and Usage
Dim A 0.864 in Nom (aperture width)
Dim C 0.738 in +/- 0.002 (picture centerline to reference edge)
Dim D 0.307 in Max (picture edge to reference edge)
Dim E 1.171 in Min (picture edge to reference edge)
Dim F 0.115 in Nom (soundtrack width)

ANSI/SMPTE 111
SMPTE Standard for Motion-Picture Film (35-mm)--Prints Made on Continuous Contact Printers--Exposed Areas for Picture and Audio
Dim A 1.184 in +/- 0.002 (picture edge to reference edge)
Dim B 0.304 in +/- 0.002 (picture edge to reference edge)
Dim C 0.192 in +/- 0.001 (sound edge to reference edge)
Dim D 0.308 in +/- 0.002 (sound edge to reference edge)
Dim F 0.738 Nom (picture centerline to reference edge)
Note 1: overlap of picture and audio areas of 0.000 to 0.008 inches

Comparing SMPTE 195 Dim A with SMPTE 59 Dim A, there is an extra 0.039 inches of image available outside the projectable image area. Granted, this is not intended for projection but should be acceptable to hide fuzzy aperture edges.

Comparing SMPTE 195 Dim D with SMPTE 111 Dim D shows 0.016 inches between the projectable image edge and the edge of the exposed photographic audio.

ANSI/SMPTE 40
SMPTE Standard for Motion-Picture Film (35-mm)--Photographic Audio Records--Release Prints
Dim A 0.308 in Nom (printed area from SMPTE 111)
Dim B 0.192 in Nom (printed area from SMPTE 111)
Dim C 0.205 in +/- 0.001 (audio record edge to reference edge)
Dim D 0.281 in +/- 0.001 (audio record edge to reference edge)

RP 115
SMPTE Recommended Practice
Dimensions of Photographic Control and Data Record on 35-mm Motion-Picture Release Prints
Dim A 0.292 in +/- 0.001 (data record edge to reference edge)
Dim B 0.302 in +/- 0.001 (data record edge to reference edge)

I believe DTS timecode uses the RP 115 photographic control and data record area that is part of the opaque area not used for the audio record of SMPTE 40.

I feel this shows that the masking can be safely placed at the SMPTE 195 dimensions. If you follow this and can see soundtrack or DTS timecode then either the reel is faulty or the aperture plate was filed out too far.

Posted by Steve Guttag (Member # 268) on 10-25-2002, 05:40 PM:

And just how do you make THAT conclusion? Have you measured the fringe area on modern lenses? Take what you measure, then add in an allowance for image movement in the camera and printer. Then add in allowable movement in the projector. Finally, add in allowable variances in film dimensions, slitting and perforation plus shrinkage and what do you have left?

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Adam Martin (Member # 641) on 10-25-2002, 06:19 PM:

Super 8!

Posted by Michael Schaffer (Member # 1204) on 10-25-2002, 07:03 PM:

Joe, wait until I submit a picture and then you will be truly offended. All of my friends agree that I have the most stupid grin imaginable!

Posted by Jeffry L. Johnson (Member # 453) on 10-26-2002, 07:30 AM:

In over 20 years I've only very rarely had a problem with the masking and aperture plates that I have set up at 0.825 inches. I have had problems with anamorphic at 0.839 or 0.838 inches but we don't have to do that anymore.

The only times I have noticed problems were with bad reels or shrunken original elements. The vast majority of release prints have absolutely no problems with masking set at 0.825 inches.

Posted by Steve Guttag (Member # 268) on 10-26-2002, 09:08 AM:

The fact that you have had ANY problems shows the flaw in the plan. Furthermore, when the scope specification was .839 it should have been safe too...but it wasn't. Your theory also doesn't take into account that with some focal lengths, the fringe area will be rather huge...particularly nowadays with the short focal lengths.

So lets throw another type of screen into the mix...the "wrap around screen" which is essentially unmasked...that is the fringe area will show. Certainly you shouldn't cut beyond the .825 line on those.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Pat Moore (Member # 380) on 10-26-2002, 11:41 AM:

Guess I'll chime in with my two pennies worth..

Given Steve K's proposition of the ideal setup, being dead-on screen center with no keystone issues, I'd still set masking width to .825. So-called studio viewing rooms are probably as close to this ideal as I've ever seen in real life and I don't recall any that go beyond .825. If, for whatever reason, they want to see past that edge they can pull the aperture or even file a different plate, but .825 is certainly the norm.

We certainly have to deal with keystone in the theatre world though, especially with stadium seating. You generally can't "fix" it with perspective control, so we end up compromising by necessity. Picture width at the top is at the widest point -- still .825 for me -- and the width at the bottom is cropped by the masking hanging straight vertical.

Given the quality of prints we sometimes see, I think going beyond .825 asks for trouble. I'll give up 2% of the picture width to know I have a clean presentation on screen.

Steve G -- I'm not a floating screen fan myself but I'd agree that I would not go past .825 width-wise.

Pat

Posted by John Pytlak (Member # 331) on 10-26-2002, 09:20 PM:

If you want a margin for error, 2% of screen width seems reasonable. AFAIC, 10% is excessive --- if unwanted junk shows up on screen with more than a few percent cropping, the standards are being totally violated.

Posted by Steve Guttag (Member # 268) on 10-27-2002, 12:02 AM:

John,

Is there ANYWHERE in the standards that defines the area beyond .825 to some other dimension (and likewise on the heights) as for use to hide aperture plate shadow on masking material?

If not how can you or anyone claim it is safe? What if the area beyond .825 is clear on the print? Is it violating any of the standards (I know some are)?

If that is a safe area, then let it be defined as such and written into the standards so people can be accountable to it.

Where did your 2% number come from? Just personal experience or actual field tests? Is it in the standards anywhere?

I'll admit I'll have to go through them myself to dig up the 10% number. As I recall the allowable amount of cropping is not to exceed 10% with less than 5% desired.

Now, I'll admit that 10% is a bit excessive and I can't imagine cropping that much...even in extreme situations.

Based on experience...I'd say that about 2.5% of the image is lost or in the aperture shadow...it is a matter of if it is safe to spill that onto the masking or not...but to declare it safe, then site the standard that say so. Where does it say there WILL be safe image on a RELEASE print that is greater than 2.5% beyond the projectable image dimensions? It has to be greater to allow for the absolute worst case cumlative from camera, through printing and projection (all working within standards yet at the worst of the extremes). That is what a "safe area" is!

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Brad Miller (Member # 2) on 10-27-2002, 01:05 AM:

There is also the clear edgemarks on the inboard side of the film on Deluxe prints to take into account. Frequently this is on the black lead in of the studio logos where the start of the audio is. So is it better to chop that off and miss the opening bit of the studio fanfare, let it run on screen as a distraction, or crop enough to block it out?

I wish someone would get word to them to stop that nonsense.

Posted by Jeffry L. Johnson (Member # 453) on 10-27-2002, 10:59 AM:

here

Film Notes for the Reel People
H-50-7 (c)1978
The Case for Test Films
Define Lines
The horizontal or vertical lines at the top, bottom, or sides of the test pattern indicate the size of the film image being projected. The numbers, expressed as decimal parts of one inch (825 = 0.825), give the actual dimensions of that image. While there has never been an established limit for allowable film image cutoff, a value of 10 percent is generally regarded as the outer limit, particularly if it includes both the width and the height. For instance, if you are projecting a wide-screen release with a 2.35:1 ratio and are using the proper aperture plate and backup lens, you should see the entire test pattern on the screen. The lower screen edges should show the figures "839" and should include the arrows that include the projector aperture width. Likewise, the top and bottom of the screen should just reveal the lines marked "700." If the vertical lines marked "745," or the horizontal lines marked "640," cannot be seen, you are losing more than 10 percent of the film image width, or height.

Posted by Frank Angel (Member # 248) on 10-27-2002, 02:05 PM:

Forget about all those numbers -- the real problem is, most theatres use standard velour as masking material. All the fuss about how much you need to crop because of the fuzzy fringe that will play on the mask and thus seen by the audience becomes moot if the mask is made of the correct material. We found standard velour is not absorbent enough to effectively perform its intended function, i.e., MASKING the shadow edge. If you have more that an inch or two of overshoot, it is still seen by the audience.

We discovered a velour that we call AB velour -- Absolute Black velour. Light projecting on it at normal SMPTE spec brightness levels will simply not be seen, no matter how much overshoot you wind up with. Thus, if you need to mask into the image enough for the fringe to be totally eliminated and to do that means you have 2 or 3 inches of picture spill on the mask, not to worry -- it will not be seen on this AB velour. That solves a host of problems in the practical world.

Anyone who wants a swatch of this stuff to check out, just email me. It is truly amazing. It is also very lightweight compared to standard velour. In one installation, all I did was get up on a ladder and simply glue a 12 in strip of it to the existing mask with cloth glue....never even had to deal with taking down the old mask.

Posted by Steve Guttag (Member # 268) on 10-27-2002, 05:39 PM:

See now here is another case where 70mm come in to save the day...the fringe area around the 70mm plate is almost non-existant. And with most projectors...you can pull the aperture plate and not show too much beyond what is projectable!

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Michael Schaffer (Member # 1204) on 10-27-2002, 07:48 PM:

Frank, is AB velour acoustically transparent? When you play in 1:1,66 or 1:1,85, the left and right speakers are behind the masking.

Posted by John Walsh (Member # 168) on 10-27-2002, 07:52 PM:

On the "instruction sheet" that comes in the can with the 35PA, there's something written about how; "... 10% crop should not be tolerated, with less that 5% perferred."

You are always forced to crop a little in at least one direction, unless you have both adjustable top and side masking, and/or use a variable length lens. Most places don't have these.

Assuming little or no image distortion (keystoning, pin cushioning, etc.) and using regular fixed top/bottom masking with adjustable sides, you should be able to keep top/bottom cropping to less than 2%, and no crop at the sides. I usually favor 1.85 (no cropping) and crop 2.39 the 2%.

Posted by Joe Redifer (Member # 3) on 10-27-2002, 10:06 PM:

Frank, perhaps we could eliminate aperture plates altogether by carpeting the front of the auditorium floor and ceiling with this masking as well!

Posted by Frank Angel (Member # 248) on 10-27-2002, 11:26 PM:

Mike, actually in our theatre the speakers are in a space just above the screen so acoustic transparency isn't an issue for us. But that problem of left and right speaker clusters being covered by the mask would be less of a problem with my AB velour because it is about half the weight of standard velour that typically covers the L/R speaker clusters. It would be interesting do some tests with and without the AB velour to see how much attenuation it causes, but my gut says it wouldn't be very significant. In a home theatre setup we used only about 6 inches of velour on the mask edge but then used black china silk in front of the speakers. China silk is acoustically transparant but it isn't very good at light absortion so you don't want to use it where there is any image spill.

How do commercial theatres deal with the problem of the masking velour covering the left and right theatre clusters when playing 1.66 or even 1.85 on smaller screens? Is the room equalized with or without the mask in place for the smaller width ratios? If it is eq-ed with the mask at scope, would that make any sense give that there are more flat films released than scope films?

As for Joe's idea....I think it should be submitted as an Academy Award nominee for Technical Achievement....mention that it would be oh so sheek.

Frank

Posted by John Pytlak (Member # 331) on 10-28-2002, 07:43 AM:

Steve wrote: "Where does it say there WILL be safe image on a RELEASE print that is greater than 2.5% beyond the projectable image dimensions? It has to be greater to allow for the absolute worst case cumlative from camera, through printing and projection (all working within standards yet at the worst of the extremes). That is what a "safe area" is!"

The standards don't specify that the area beyond the projectable image areas of SMPTE 195 will be free of distracting elements, but if everyone is doing their job correctly, there shouldn't be anything there that the masking can't absorb. The masking defines the projectable image area on the screen per SMPTE 195, the aperture is cut to place the fuzzy edge onto the masking.

The (IMHO) excessive 10% crop figure came from you: "The only way it can make any sense is to have a safe area (10% crop) and that space is used to hide any of the mechanics of showing the image". The more reasonable 2% crop suggestion for a margin of error was Pat Moore's.

I am more optimistic than your view of "all working within standards yet at the worst of the extremes". Severely cropping the image to deal with an absolute worst case situation seems like overkill.

Posted by Steve Guttag (Member # 268) on 10-28-2002, 08:22 AM:

If the area beyond the "projectable image area" is undefined, your recommendation of hiding it safely on the masking is reckless. Masking, as others have already discussed, varies from installation to installation. Some absorb more than others, some less. Some much less with time. If you can't guarantee that there is safe material project on the masking, then you can't recommend that someone project it...it is that simple.

As to cropping the picture...if it is in the safe action area...there is nothing wrong with it...nothing of interest is supposed to be there.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Michael Schaffer (Member # 1204) on 10-28-2002, 08:38 AM:

If your speakers are above the screen, where are they aimed? Are they visible to the audience or hidden? I know an engineer who installs all his cinemas with the speakers above the screen aimed at the area in front of the audience so that the sound is supposed to be reflected into the auditorium. I have never been in any of cinemas, so I have no idea if that concept works out.
For equalising a room, the masking should be set to Scope, and the masking should be acoustically transparent. I have no idea what the definition for transparent is, how much attenuation is allowed, but I am sure some of the gentlemen on this forum will know. We always used a very simply procedure to test the transparency, by blowing through the cloth.
It would seem a good idea to use heavy dark just for the edges to blacken out the fuzz and light silk for the rest. The outer speakers would not be affected then. It is equally important to check the cable tension regularly, as slackened cables will result in imprecise positioning of the masking.

Posted by Frank Angel (Member # 248) on 10-29-2002, 12:34 AM:

The speaker are in a loft that is just above the top of the proscenium and jutts out about 15 feet in front of it, but not further out than the first row of seats. Even still, we have a difficult time with speech intelligibility in the first five rows, which are off-axis. There are four constant Q horns covering all quadrants of the theatre for the center channel with the sides consisting of similar clusters but with two additional horns at a down angle to help cover the front rows. Except for folks who insist in sitting in the very front rows, we get surprisingly even coverage for such a large hall. All speaker components are JBL Pro except the bins for the subwoofers -- the speakers themselves are 18in JBLs but the bins are a fiberglass proprietary folded horn structure designed by Community Sound to produce a very solid, tight bottom end below 30 hz, plus or minus five or six of your fillings.

As for the "engineer who installs all his cinemas with the speakers above the screen aimed at the area in front of the audience so that the sound is supposed to be reflected into the auditorium. " I am not an acoustician, but that sounds highly dubious. How does he know what his reflective surface is doing to the sound; no reflecting surface reflect the entire audio spectrum evenly. It will absorb some frequencies and create sharp spike in others -- tying to fix that with 1/3rd or even 1/6th octave equalizers can be a bitch. Plus, if he is reflecting off the front of the audience seating, how does he get even coverage, to say nothing of the affect people and their clothing will have on the reflected sound hitting them? Sounds risky. But then, I never argue with something that someone found to work....if indeed it does work.

You are right, the heavy velour should only be used as a border at the edge of the mask -- china silk, which is acoustically transparent, should be used for the rest of the mask material. Unfortunately because china silk is pretty expensive, a lot of installations don't spring for it, much to the detriment of their eq. One of the places I work has full velour for the mask -- 26lb velour at that. It completely covers Left and Right speakers. Nice goin', designer guys --same theatre with the full glass front wall and a ceiling that dips down in front of the change-over corner of the screen.

I always though you gauged "acoustically transparent" by simply playing your pink noise. looking at your RTA, put the material in front of the speaker and what you see is what you hear. You shouldn't see any difference, or at least nothing significent. Define significent, well, anything that you can hear, say +/- 3db. With china silk I would be surprised if it is any more than +/- 1 db and only in the very high end -- probably not even as much coloration as standard speaker grillcloth.

Frank