100 Year Old Miles Brothers Film on SF's Market Street

[Tim Tyler]

"Morley Safer reports on a mystery that was solved about a 100-year-old film

that we now know was made on San Francisco's Market Street

just days before the 1906 earthquake."

 

http://www.cbsnews.com/video/watch/?id=6966797n

[John Sprung]

Very interesting. The camera they show in the 60 minutes piece is an Ensign Cinematograph. (I have one myself.) It couldn't have been used to make the Market Street film, because it only holds 200 feet.

 

 

 

 

-- J.S.

[Tim Tyler]

The full length clip at http://www.archive.org/details/TripDown1905 contains some breaks in the first few minutes that could have been camera reloads, but the last 7 minutes or so seems to be a continuous take.

[Leo Anthony Vale]

Very interesting. The camera they show in the 60 minutes piece is an Ensign Cinematograph. (I have one myself.) It couldn't have been used to make the Market Street film, because it only holds 200 feet.

 

 

It's only stated that the Ensign was one of the Miles' cameras, not that it was the camera used for the Market St.film.

[Dom Jaeger]

What a fascinating piece of footage. I was initially amazed that there were so many automobiles in 1906, until I read the descriptor that mentioned the scene was probably staged to give the appearance of prosperity - a careful examination had shown it was the same few autos circling back and forth!

 

I wonder if they might have used more than one camera? The cataloguer's remarks mention a section that is repeated. It also says that the repeated section of 2060 frames amounted to 70 seconds, which gives a frame rate of just under 30 fps. I've read that silent films were cranked at quite a variety of speeds, from Edison's early recommendation of 46 fps (quickly abandoned) to Bitzer cranking parts of Birth of a Nation at 12, so maybe 30 is not unreasonable? Edison films from around 1900 project well at about 24 fps according to this interesting article: http://www.cinemaweb.com/silentfilm/bookshelf/18_kb_2.htm

A 1000 ft roll (rare but not unheard of at the time) would give about 9 minutes at that speed.

[Carl Looper]

I though I'd seen that before.

 

That footage is the footage that the Voodoo Camera Tracker uses to demonstrate camera tracking

 

http://www.digilab.uni-hannover.de/docs/manual.html#figure14

 

Carl

[John Sprung]

It also says that the repeated section of 2060 frames amounted to 70 seconds, which gives a frame rate of just under 30 fps. I've read that silent films were cranked at quite a variety of speeds, from Edison's early recommendation of 46 fps (quickly abandoned) to Bitzer cranking parts of Birth of a Nation at 12, so maybe 30 is not unreasonable?

 

One thing that was a de facto standard very early on was that you got 8 frames per turn of the crank. That worked well for the widely used speed of 16 fps, but it makes 24 quite difficult to do for a long take. Even at 16 fps, whoever cranked this did a lot of work that day. Is that 2060 a count of original film frames, or of video frames? With 16 fps film, it's quite common to transfer each film frame to two video frames in NTSC/ATSC.

 

 

 

-- J.S.

[Tim Tyler]

I analyzed the MPEG2 available at http://www.archive.org/download/TripDown1905/TripDown1905.mpeg

 

The file is 29.97 and the film was transferred at 12 fps.

 

The motion of some of the actions seem a little slow to me though, especially people jogging across the street. If I had to guess I'd say the camera was cranked at 16 fps.

[K Borowski]

I agree, it was probably shot at 16 fps. At that rate (were speed governors common on cameras then?) It'd need a little over 400 feet to accomplish a continuous 7-minute take. Of course, maybe with Edison perfs being a different size than current ones, that number is off.

 

I really wish I could see the clip in HD. My computer is too slow to play the 60 minutes version back. Too bad I missed the original airing on television. Tell me they didn't crop it in HD. . .

[John Sprung]

(were speed governors common on cameras then?)

maybe with Edison perfs being a different size than current ones, that number is off.

 

No, centrifugal governors weren't used until they started using big clock springs to drive the camera. They didn't even bother with flywheels. Cranking technique was good enough.

 

Edison's perf pitch wasn't all that different, he had 4 perfs per frame, and the perf to perf distance was 3/16". His aperture was 1" wide by 3/4" high.

 

 

 

 

-- J.S.

[K Borowski]

No, centrifugal governors weren't used until they started using big clock springs to drive the camera. They didn't even bother with flywheels. Cranking technique was good enough.

 

Edison's perf pitch wasn't all that different, he had 4 perfs per frame, and the perf to perf distance was 3/16". His aperture was 1" wide by 3/4" high.

 

That looks pretty damned smooth for a hand-processed print of a hand-cranked, hand-processed nitrate negative from 104 years ago. At that time, film was, I think still being hand-COATED, maybe by George Eastman himself. It's incredible that the dimensions from that time are almost identical to S35 4-perf. today.

 

I'm unfamiliar with how that pitch compares to current day (0.1866 inches?), but even that seems similar, with 3/16 being 0.1875" What I did notice on one of the links, believe it was the 60 Minutes segment (that my computer was too slow to play back in HD, unfortunately) was that the perfs seemed by eye to be narrower, more of a sqquare than a modern neg. or print perf. And they had trouble with the lower-def. transfer I could play back, with the telecine losing its loop several times.

 

John Pytlak would be all over this.

[K Borowski]

Edison films from around 1900 project well at about 24 fps according to this interesting article: http://www.cinemaweb.com/silentfilm/bookshelf/18_kb_2.htm

A 1000 ft roll (rare but not unheard of at the time) would give about 9 minutes at that speed.

 

Ahh, but if you look closely, it has more to do with dramatic films being shot at a faster rate than advertised to compensate for projection houses cranking at faster speeds (mostly to save money).

 

I think that a documentary would look most natural projected at the exact framerate it was photographed at.

 

 

I'm wondering if 1,000 foot rolls of film were really that available in 1906 though. The fun little Kodak encyclopedia I printed out over seven years ago is still up. . .

 

http://motion.kodak.com/US/en/motion/Products/Chronology_Of_Film/index.htm

 

 

The captioned photo at the bottom of the page would seem to indicate that, around the time of the late 1800s, film was still being hand-coated in Rochester on this 200-foot table.

[John Sprung]

That looks pretty damned smooth for a hand-processed print of a hand-cranked, hand-processed nitrate negative from 104 years ago. ....

I'm unfamiliar with how that pitch compares to current day (0.1866 inches?), but even that seems similar, with 3/16 being 0.1875" What I did notice .... was that the perfs seemed by eye to be narrower, more of a sqquare than a modern neg. or print perf.

 

Having shot with my Ensign Cinematograph, I can say that it really isn't all that hard to crank smoothly enough to avoid visible motion and exposure variations. Today, camera pitch is 0.1866" and print 0.1870", and modern neg runs just fine in the Ensign, which was designed for about a thousandth of an inch longer pitch.

 

You're right that the perfs then were narrower and more square. That made them easier to tear. The Bell & Howell perfs that we use today have an arc of a circle rather than straight sides, which gives them a little room to flex instead of ripping. That's why they had to reduce the width of the aperture to 0.980" instead of a full inch.

 

 

 

 

 

-- J.S.