DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   Alternative Imaging Methods (https://www.dvinfo.net/forum/alternative-imaging-methods/)
-   -   Focal length conversions (https://www.dvinfo.net/forum/alternative-imaging-methods/15881-focal-length-conversions.html)

Charles Papert October 17th, 2003 02:27 AM

Focal length conversions
OK, so we have this figure of 7.2x as a conversion between the field of view and focal lengths of 1/3" DV and 35mm lenses. But this is apparently 35mm still photography, not 35mm motion picture (Academy) format. Does anyone know the exact calculation of the magnifaction factor for Academy? I'm coming up with 4.4x, since 2/3" video to 35mm has a 2.2x conversion. Is this correct?

The other thing to consider with this is that given the specs of the stock Canon lenses for the XL1, the 35mm equivalents as stated may be, again, still photography equivalents rather than cine equivalents. Thus the wide angle of the standard lens as listed as 39mm equivalent may actually be closer to 23mm...? But yet it doesn't seem to be.

Confusing yes--I could probably figure this out by dragging my XL1 to set and comparing it to a 35mm lens equivalent, but I haven't the energy, especially since you guys are "wicked smart".

Jeff Donald October 17th, 2003 04:44 AM

I have the size for Academy film format as .825 X .602. Can anyone confirm that? If that is correct then the conversion factor is 4.5X. Conversion factors can be varied. The conversion factor can be based on height, width or diagonal. Each will give a slightly different calculation. I've always based mine on height. If I'd used width the factor would be 4.2X, diagonal 4.3X, so take your pick.

Mizell Wilson October 17th, 2003 08:26 AM

If it doesn't get answered by Monday I'll bring my notes in but Jeff brings up the most important point to consider in this process: when we talk about film e.g. 35mm we are talking about it's length across the horizontal whereas when we say 1/3" CCD we're talking about length across the diagonal.

The next thing to consider is the actual taking frame sizes including the fact that the CCDs don't always use the full theoretical diagonal length of it's chip and in film we make way for perfs and such.

I spent a long midnight into the morning session reliving my high shcool trig days to transcribe all the length, width, and heights of the frames. One nice thing...with our much maligned SD aspect ratio of 4:3, the diagonal is automatically 5.

I'll leave ya with that for the weekend :)


Rob Belics October 17th, 2003 11:37 AM

The camera aperture used for calculating dof is .868 x .631

Mizell Wilson October 17th, 2003 11:55 AM

.868" gigs up with what I've got in a chart for Academy taking.

If we figure .33" for the diagonal of our 4:3 aspect ratio CCD, then the resulting horizontal length is .264" ( (.33" / 5) *4)

Which then technically means that from 1/3" CCD to Academy is a 3.29x jump and 2/3" CCD to Academy is a 1.64x jump.

Anybody want to check my math?

The next thing to consider is that the 35mm still frame is much larger, so the 7.2x might be a throw back to that. My still customers that have then used the Mini35 have mentioned significant differences from what they see in their SLR rigs to what they see in the Mini35 whereas customers used to shooting 35mm motion film have not noticed as much of a difference.

thoughts, suggestions, koans?


ps...i don't happen to have any still photography textbooks or tech guides...does anyone happen to know the measurements for that frame?

Jeff Donald October 17th, 2003 12:20 PM

I normally use the American Cinematographers Manual for this stuff, but I haven't seen mine since I moved two years ago. I got the figure from two web sources that claimed .825 x .602 was the largest printable size for Academy Film Format. The new size doesn't change things much. The diagonal factor on Rob's size is 4.51X, height 4.57X and width 4.40X. If the chip goes smaller (I used 4.82mm X 3.64mm/ dia. 6.04mm) you could expect another tenth or two of change.

Mizell Wilson October 17th, 2003 12:29 PM

.825" x .602" is the Academy projection frame, but for depth of field and angle of view purposes the Academy taking frame is .868" x .631" I'm getting this from Jon Fauer's "shooting digital video" but I imagine it matches what's in the ASC manual.

Jeff Donald October 17th, 2003 12:44 PM

The math is the Pythagorean theory. The 7.2X is the factor used for the Canon EF lenses when used on the XL1. The measurements are the same for the 1/3 inch chip and 35mm film is actually 36mm x 24mm. Printable dimensions are 1mm less.

Mizell Wilson October 17th, 2003 12:44 PM

actually, i think the first decision we need to make is whether we're going to talk mm or " here :)

Jeff, where did you happen to come up with that frame size for the video CCD? My calculations were based on a diagonal of 8.382mm or .33" i.e. 1/3", leaving us with a .198" heigth by .264" width or 5.03mm x 6.71mm.

Minutiae behind, I think we've already shown that the 7.2x must come from a comparison based on the 35mm still frame and since it was a still lens adapter for the XL1s that hit the market first, that number has obviously propagated itself wrong as the use of PL lenses has been introduced.

Anyone want to take a shot as to why I'm coming up with smaller conversion factors?


Jeff Donald October 17th, 2003 12:52 PM

The 1/3 inch is a hold over from the old tube (Saticon, Vitacon, etc.) days of video. It has no relationship to the actual dimensions of the chip. The size of the chip was supplied by Canon. There are several threads here that confirm this. So does the math. The diagonal of 35mm film is 43.266mm. Divide the diagonal of the CCD into it and you get 7.16X

Mizell Wilson October 17th, 2003 01:08 PM


thanks for clarifying those measurements, I had never actually seen them myself.

Has anybody else had the realization of what that diagonal acutally represents in inches and to how we've been misled in chip size :)

I really don't like marketing sometimes. How many of you out there knew that Titanic was 2 hours and 72 minutes long, at least according to the marketing dept.


Charles Papert October 17th, 2003 06:36 PM

I think the only way I'm going to be truly happy is to do a side-by-side comparison of cameras shooting a framing chart to figure out a real-life scenario vs the math, which is apparently fraught with interpretation and obfuscation! If I get around to this, I'll let you guys know what I see...

Jeff Donald October 17th, 2003 07:27 PM

You were right Charles. I think you'll find it is about a factor of about 4.4x I would be curious to hear how it visually measure up to your eye. If nothing else Mizell got a little education from a couple of us old timers. Surely you remember Saticon, Vidicon, Newvicon and Plumbicon tubes, Charles.

Charles Papert October 18th, 2003 01:18 AM

I think the first cameras I ever used were Vidicon (do you think the Sony EIAJ reel to reel format cameras, the little red cigar box ones, were Vidicon? I would think so)...then I remember the 1/2" VHS two piece home video systems were initially Newvicon, then Saticon--right? And the creme de la creme in the 80's, the Ikegami HL-79, had those oh-so-desirable Plumbicon tubes.

I also recall going to a trade show in NYC around '84 and seeing a demo of one of the first 3-chip broadcast cameras. A shot of the sun setting drew audible gasps from the audience. Shooting the sun on video without worrying about burning the tube! How could this be?

How times change.

Jeff Donald October 18th, 2003 05:46 AM

Ooooooh, the HL-79, that was a camera. I remember shooting some projects with it in the '80's. Those were the days.

All times are GMT -6. The time now is 02:52 PM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2020 The Digital Video Information Network