Circular distortion in old race car photos

[AZLF]

This seems as good a place to ask this question as any. I've seen several old (turn of the last century) photos of race cars shot from the side as they raced past. The wheels have distorted into ovals leaning slightly forward. I remember reading years ago the explanation for this but at the moment I cannot recall what it is. I have not ever had the same happen in any photo I have shot and one time I tried to duplicate the effect with negative results. Does anyone know why this distortion was present in the old photos?

 

[Charles Webb]

That is caused by the Speed Graphic's focal plane shutter moving across the film plane as the car is moving in the same direction.

Charlie..............

 

[Clueless]

Do I win a box of Milky-Ways and 5 Silver Dollars?

Just a guess: Horizontal focal-plane shutter?

 

[AZLF]

Thanks. That makes sense to me.

 

[David H. Bebbington]

It's actually a vertical focal-plane shutter. If you imagine a car traveling from left to right, and that you are photographing it holding the camera still without panning, this means its image will be upside down and traveling from right to left. As the focal-plane shutter of a camera like a Speed Graphic travels downwards, it will expose the top of the image (bottom of the subject) first and the bottom of the image last. With a moving subject, the bottom of the image will therefore be recorded later and shifted to the left relative to the top, thus giving a "leaning" effect and oval wheels. This happens because the shutter of a 4x5" press camera is so big. The same thing happens in theory with the focal-plane shutter of a 35 mm camera, but because this is small, has low mass and can be made to accelerate and travel so quickly, the distortion effect is virtually non-existent.

Regards,

David

 

[AZLF]

I just watched the shutter on my Minolta x-700 several times.Looking at the back of the camera with film door open starting from curtain closed the shutter opens starting at the right side of the horizontal frame and pulls completely into the left side. At the end of the time duration the shutter then makes the return trip at a more leisurely pace back from left to right to curtain closed position. I'll have to try shooting some asa 100 film with an nd 2 filter with the camera position in the vertical to see if I can duplicate the effect. Does the fp shutter in the Graflex have a "slit" through which the film is exposed or does it open entirely for the exposure as in the case of the Minolta?Perhaps it does not matter and the effect will be seen in either case.Thanks for the info. This is one of those things that has been nagging at me for some time for no good reason other than a curiosity to know.

 

[Soeren]

It's actually a vertical focal-plane shutter. If you imagine a car traveling from left to right, and that you are photographing it holding the camera still without panning, this means its image will be upside down and traveling from right to left. As the focal-plane shutter of a camera like a Speed Graphic travels downwards, it will expose the top of the image (bottom of the subject) first and the bottom of the image last. With a moving subject, the bottom of the image will therefore be recorded later and shifted to the left relative to the top, thus giving a "leaning" effect and oval wheels. This happens because the shutter of a 4x5" press camera is so big. The same thing happens in theory with the focal-plane shutter of a 35 mm camera, but because this is small, has low mass and can be made to accelerate and travel so quickly, the distortion effect is virtually non-existent.

Regards,

David

Very Precise explanation. Couldn't have done it better.
Cheers Søren

 

[David H. Bebbington]

Does the fp shutter in the Graflex have a "slit" through which the film is exposed or does it open entirely for the exposure as in the case of the Minolta?

Sounds like your Minolta has a horizontal fp shutter, like many other 35s (Leica, etc.). Others have vertical shutters like the famous "Copal Square" shutter (Nikon F, etc.).
I believe I am correct in saying that most 35s have a fixed-tension variable-slit shutter, i.e. different speeds are obtained by using a large or smaller slit. Because 35 mm film film comes in rolls, the shutter needs to "cap" itself (close its slit) when it is moving across the film being wound. The Graphic shutter does not do this, it is therefore a "non-self-capping" type (said by old pressmen to be more reliable) and has 4 separate slits, 3 of which give different timed speeds, the 4th gives a "T" setting (press once to open, press again to close), plus a choice of two tensions. (This relates to the later Pacemaker Speed Graphic model - the "Anniversary" model and earlier types had, if I recall, 4 tensions and 4 slit widths for timed exposures).
It could be fun to try to replicate the distortion of the old-time shutters with your Minolta, the problem is that to get the distortion you need to be in a situation where the shutter is exposing each part of the film for a short time (and thus giving a sharp image) BUT ALSO taking a relatively long time to get from one side of the picture to another. This would mean a low tension and small slit, whereas the shutter of your modern 35 will have high tension and a slit that gets progressively larger for slower speeds.

Regards,

David

 

[egdinger]

Is a focal plane shutter nessacery to achive this effect, or would a shutter in front of the lens the uses a slit cause the same thing? I have an idea for a shutter that I have been toying with and it may work for whay AZLF wants to achive, assuming it works at all.

 

[nworth]

That is caused by the Speed Graphic's focal plane shutter moving across the film plane as the car is moving ....

Charlie..............

The effect is enhanced by the very slow focal plane shutter on the older Graflexes and Speed Graphics. It took the curtain somewhat over 1/30 second to traverse the film at the highest tension. The effect is caused by the tire moving significantly during the time the shutter is exposing that part of the film. With the shutter speed set to 1/1000 (typical for high speed sports at the time) you have a quite narrow slit, and the wheels are still pretty sharp.

 

[127]

Is a focal plane shutter nessacery to achive this effect, or would a shutter in front of the lens the uses a slit cause the same thing? I have an idea for a shutter that I have been toying with and it may work for whay AZLF wants to achive, assuming it works at all.

You need a focal plane shutter to get the full effect. A shutter next to the film exposes each part of the film seperatly, while a shutter at or near the lens is so out of focus, that it exposes the whole film at once, but with different light paths through the lens being used over time.

If the shutter is some place else then you'll get some combination of the two effects - the whole film will be exposed for the whole time, but with different emphasis, so while the whole film IS exposed, at the begining the of the shutter time the bottom of the film is exposed more while the top receives more light at the end of the shot.

This effect should look nice - it's a form of shutter efficiency, and in the simulations I've done, low efficiency shutters produce better images. However I suspect the geometry involved will mean that to get an effect other than that of a conventional leaf shutter the falling plate would need to be a LONG way infront of the lens (totally inpractical) or some way behind the lens (say 3/4 of the way back in the camera).

Ian

 

[David A. Goldfarb]

J-H Lartigue was probably using a 5x7/13x18 camera like my 5x7" Press Graflex in his famous race car photo. Travel time on this shutter depends on the spring tension, which is set separately from the slit size, and a table is used to calculate the effective shutter speed, but it could be as slow as 1/5 sec. The slower the travel time, the stronger the effect. With an 1/8" slit and maximum tension, by the way, this shutter can have an effective shutter speed of 1/1500 sec.

[Edit--Misinformation deleted, corrected below].

 

[KenM]

For another demonstration of a similar kind of distortion caused by mounting a scanner on a large format camera (!), check out Dead Link Removed.

It's the only time you'll see an Audi TT stretch limo

 

[David H. Bebbington]

As the basic for experimentation in trying to get extreme distortion, I would think this item would be great:
http://cgi.ebay.co.uk/UNUSUAL-GRAFL...597876624QQcategoryZ15248QQrdZ1QQcmdZViewItem

 

[127]

If anyone has access to online journals (or a library even!) you could check out:

Glassner, A 1999. "An open and shut case", IEEE Computer Graphics and Applictions 19,3 (may).

It's a pretty interesting read. HOWEVER Andrew demonstrates the great folloy of the scientist - he's never actually seen some of the things he's talking about, so when he talks about the geometry of leaf shutters its TOTALLY WRONG!. The rest of the paper is worth the read though.

Ian

 

[Jim Jones]

It's actually a vertical focal-plane shutter. If you imagine a car traveling from left to right, and that you are photographing it holding the camera still without panning, this means its image will be upside down and traveling from right to left. As the focal-plane shutter of a camera like a Speed Graphic travels downwards, it will expose the top of the image (bottom of the subject) first and the bottom of the image last. With a moving subject, the bottom of the image will therefore be recorded later and shifted to the left relative to the top, thus giving a "leaning" effect and oval wheels. This happens because the shutter of a 4x5" press camera is so big. The same thing happens in theory with the focal-plane shutter of a 35 mm camera, but because this is small, has low mass and can be made to accelerate and travel so quickly, the distortion effect is virtually non-existent.

Regards,

David

When shooting a race car with a horizontal moving shutter in a stationary camera, the car will be compressed or expanded laterally, depending on which way the shutter and car are moving. With a vertically moving shutter it will appear to lean. If the camera is panned, the background will lean. To duplicate the leaning effect with today's cameras, the shutter must move perpendicular to the subject's movement. A downward moving shutter will make the subject lean forward; an upward moving shutter will make it lean back. The effect can be enhanced by panning against the subject's movement.

 

[David H. Bebbington]

When shooting a race car with a horizontal moving shutter in a stationary camera, the car will be compressed or expanded laterally, depending on which way the shutter and car are moving. With a vertically moving shutter it will appear to lean. If the camera is panned, the background will lean. To duplicate the leaning effect with today's cameras, the shutter must move perpendicular to the subject's movement. A downward moving shutter will make the subject lean forward; an upward moving shutter will make it lean back. The effect can be enhanced by panning against the subject's movement.

Absolutely, which is why (I presume), after some initial dalliance with upward-moving shutters like David A. Goldfarb's Press Graflex, makers decided to make their shutters move downwards, on the grounds that if you're going to have distortion, leaning forward looks more pleasing and suggests speed better. Generations of cartoon and animation artists have agreed!

 

[David A. Goldfarb]

Correction--I think shooting LF makes it so I can't tell up from down anymore. I should have double checked the camera before posting, as indeed, the Press Graflex shutter moves downward rather than upward, so the beginning of the exposure is the top of the frame or the bottom of the picture. In the Lartigue photo the wheel leans forward, because the car is moving faster than the shutter, and the spectators lean backward, because the young Mr. Lartigue is panning.

 

[Ole]

... but I believe the young M. Lartigue used a Goerz camera, not a Graflex...

 

[David A. Goldfarb]

Could have been. I've been trying to find out. Did Goerz make a FP shutter camera? The 5x7" Press Graflex was a pretty common camera at the time among photojournalists. Lartigue's father had some sort of 13x18 camera, which J-H used, but I gather that the family was wealthy, and he had access to various cameras.

 

[Dan Fromm]

Could have been. I've been trying to find out. Did Goerz make a FP shutter camera? The 5x7" Press Graflex was a pretty common camera at the time among photojournalists. Lartigue's father had some sort of 13x18 camera, which J-H used, but I gather that the family was wealthy, and he had access to various cameras.

Goerz Anschutz

 

[Ole]

Looks like there's a bit of doubt about that - and since most of the information on the net was written by art curators and not camera collectors, it's difficult to find out. One page stated it was an Ica Reflex. Could have been, I guess?

Anyway he had a number of cameras and used them all. At least one of them had a vertical focal plane shutter, as we see from that famous picture.

The 13x18 camera he started with was a typical tailboard "Reisekamera" (like mine), and was certainly not the camera with the focal plane shutter.

 

[Sethasaurus]

Easiest way to bend a helicopter blade?
http://en.wikipedia.org/wiki/File:Focalplane_shutter_distortions.jpg