A correct attribution for the "Flash Effect" photo?

Discussions about the history of the ship, technical details, etc.

Moderator: Bill Jurens

User avatar
wadinga
Senior Member
Posts: 2054
Joined: Sat Mar 12, 2005 3:49 pm
Location: Tonbridge England

Re: A correct attribution for the "Flash Effect" photo?

Post by wadinga » Fri Feb 07, 2020 8:03 pm

Fellow Contributors,

Specifically for you Bill, but open to all.

Can we return to Bild 146-1990-061 and NH69729 please? As A photogrammetrist are you able to say anything about the angle between the camera axis and the rail? It has been suggested that the camera axis might not be very perpendicular to the rail, but to my inexpert eye there is no foreshortening or distortion of the parallelogram which the stanchions and rails create. Limited depth of field (aperture wide open for fastest shutter speed) means that the rails cannot be focussed as well as the distant object. That said, all the rails seem to be equally out of focus, whereas if one side were closer the degree of blur would surely be different. It would need to be much more than plus or minus 10 degrees from perpendicular in order to significantly affect the theory of a 90 degree crossing between the ships.

If one were to assume the stanchions were a given distance apart and this can be determined from plans, now we know where they are, and a standard Leica 50mm lens used (as suggested) can the apparent width of Bismarck compared with this inter stanchion distance to give us another approximate range to Bismarck?

I realise that given your last post you may consider this exercise impossible or irrelevant, but I would like your opinion as to whether it would have any value.

All the best

wadinga
"There seems to be something wrong with our bloody ships today!"

Bill Jurens
Moderator
Posts: 743
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: A correct attribution for the "Flash Effect" photo?

Post by Bill Jurens » Sat Feb 08, 2020 1:33 am

An interesting question. I'm not accustomed to working with such poor quality images, so I may have to dig into old boxes of creativity not opened in a while. There may be something that can be derived from the convergence lines for the stanchions, both vertically and horizontally, i.e. via the angle the tops of the stanchions form with the horizon. My sense of it, though, based on the numbers I've been able to work through quickly, and doing some graphic analysis, is that the angle between the stanchions and the observer is only a few degrees, perhaps two to five, meaning that the camera is pointed almost directly perpendicular to the railing.

But, having said that, that's a quick 'back of the envelope' estimate, and I quite easily could be wrong. It's happened before!

Bill Jurens

hans zurbriggen
Member
Posts: 83
Joined: Sat Dec 21, 2019 8:15 am

Re: A correct attribution for the "Flash Effect" photo?

Post by hans zurbriggen » Sat Feb 08, 2020 8:48 am

Hello,
to Mr. Jurens writing: 'the standard motion picture 35mm frame was essentially 24.9 mm by 18.7 mm in size'
many, many thanks Mr. Jurens for this info. I was wrong in my previous calculations assuming size was the same as photographic one (c 35 mm x 24 mm).
However I have not understood the whole complexity of the calculation you propose.
Focal length is known (the 3 of them being 35, 50 and 75 mm on Arriflex 35, thanks to Mr. Nilsson input). Calculation should be possible in a quite easy way (based from old studies of optic, see diagram below).
Range_Optic.jpg
Range_Optic.jpg (42.64 KiB) Viewed 544 times
'f' (focal = 0.035, 0.050 or 0.075 meters), H (actual height of Bismarck main mast over water = c 45 meters) and Ha (apparent height of Bismarck mast compared to photogram size = c 17 times less than photogram horizontal base of c 24 mm) are known.
For the three lenses I think we can determine R (distance of Bismack from cameraman) with a reasonable approximation. I have got the following results:
35 mm : R = c 1100 meters
50 mm : R = c 1600 m
75 mm : R = c 2400 m
Am I wrong in any way in above approximate method, assuming the original film negative was not cropped ?

hans

Bill Jurens
Moderator
Posts: 743
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: A correct attribution for the "Flash Effect" photo?

Post by Bill Jurens » Sat Feb 08, 2020 5:26 pm

Your math seems sound to me. Enjoyed the diagram you posted, which is essentially the same one used in air-photo analysis, except we use it 'upside down', with the subject horizontal at the bottom and the camera pointed straight down.

Note that any 'cropping' would have the result of increasing the apparent focal length, essentially rendering your estimates, if otherwise calculated correctly, as lower limits.

If, of course, the camera was a 16mm instead of 35 mm -- I can think of no other plausible formats for the period -- then the calculations would have to be redone, though overall results would probably be similar.

Bill Jurens

Byron Angel
Senior Member
Posts: 1134
Joined: Sun Mar 06, 2011 1:06 am

Re: A correct attribution for the "Flash Effect" photo?

Post by Byron Angel » Sat Feb 08, 2020 9:50 pm

..... And 1000 - 1500 meters sounds an appropriate interval for two dissimilar ships intending to maneuver in formation.

FWIW.

Byron

hans zurbriggen
Member
Posts: 83
Joined: Sat Dec 21, 2019 8:15 am

Re: A correct attribution for the "Flash Effect" photo?

Post by hans zurbriggen » Sun Feb 09, 2020 8:07 am

Hello,
I agree with Mr. Angel for formation distance. However Bismarck was quite more far (c 2500 meters) at battle start (photo NH69722).
Having already concluded her turn at film start, I suspect a higher distance, 'reduced' by the use of the 75 mm telelens.
To Mr. Nilsson: are you aware which kind of Leica camera was standard used in PK and which focal (if not common 50 mm) ?
To Mr. Jurens: thanks. Assuming photo were taken with a Leica and a focal 50 mm, is photogram difference between movie and photo affecting the apparent distance ? I think a 50 mm photo (35 mm base photogram) should then show a greater view field (see diagram above), 'enhancing' apparent distance versus movie taken with 50 mm (24 mm base).

hans

Bill Jurens
Moderator
Posts: 743
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: A correct attribution for the "Flash Effect" photo?

Post by Bill Jurens » Sun Feb 09, 2020 6:03 pm

Unfortunately, there is very little about the camera that can be reliably derived from an examination of a photo-print. The focal length of the lens has, curiously, nothing to do with the image on the film plane, and even less to do with the image on a photo-print, which is (or at least, was) also filtered through an enlarger lens. The image will appear exactly the same for a 35mm lens with the image cropped and enlarged two times, as it will for a 70 mm lens not enlarged at all. Basically, no matter what the focal length, the lens itself sees only and exactly the same thing; the light rays from the object must all pass through the nodal point of the lens.

It is basically impossible to determine with any reliability what the camera format was from a print, although some clues can sometimes be derived from the 'graininess' of the image -- at a given print size, a greatly enlarged image from a 35mm format negative will tend to look more 'textured' than an image taken from a larger format 'press-camera' which typically has a negative size several times larger, something like 100 mm x 125 mm in size, basically because one is not enlarging the film image as much.

There were problems in the darkroom, too. When the film was placed in the enlarger, it was placed in a 'film-holder', basically a device which enabled one to get the negative in an out of the enlarger conveniently. Most of these negative holders -- basically some sort of metal plate with a 'window' machined near the center -- were built to automatically crop off some of the edges of the negative. (Years ago, there was an affectation -- one could hardly call it more than this -- amongst 35mm 'art' photographers, to re-machine the negative holder so that the entire image would be projected, leaving a black border around the edge of the print, thus 'proving' that the image had not been cropped, and thus represented the photographer's 'complete and honest' rendition of the scene. This effect was so easily 'faked' using other methods, that it fairly quickly faded away. But I diverge...) To get back to the subject at hand, once the image was projected on the enlarging easel, already cropped by the negative holder, metal flaps or bars were used to hold the print paper flat, and these ate away a bit more of the image as well. And, of course, the person doing the enlarging would very commonly change the size of the projected image by enlarging it more, making the composition more pleasing or useful by 'cropping out' features around the edges which were of relatively little interest. Bottom line is that a print tells you almost nothing.

Along similar lines, except in unusual situations, it is again essentially impossible to discriminate between a 35mm film image and a 35mm still. As old 35mm cameras often had settings for longer exposure times, a heavily 'blurred' image of a fairly rapidly moving object would suggest a longer exposure time, and a hand-held still camera. Old film cameras often could not, at least in my recollection, use exposure times greater than about 1/16 of a second, whereas press cameras often had one-second settings. Sometimes, one can see the artifacts of the camera shutter in the image, as some cameras used so-called 'leaf' shutters which open and close radially, whereas others used 'focal plane' shutters, which opened and closed somewhat like a window-blind. But as some 35mm cameras had leaf shutters, and some larger cameras had focal plane shutters -- cheaper and more versatile as the shutter was located in the camera rather than in the lens -- it's often hard to tell.

I'd be interested in knowing how the distance to Bismarck at 'battle start' (Photo NH69722) was actually derived. I seem to recall that there was a great deal of correspondence on this some years ago, but don't recall that we ever were able to reach entirely firm conclusions.

Bill Jurens

hans zurbriggen
Member
Posts: 83
Joined: Sat Dec 21, 2019 8:15 am

Re: A correct attribution for the "Flash Effect" photo?

Post by hans zurbriggen » Mon Feb 10, 2020 7:35 am

Hello,
to Mr. Jurens writing: 'I'd be interested in knowing how the distance to Bismarck at 'battle start' (Photo NH69722) was actually derived ': thanks for the explanations. There is good summary at link viewtopic.php?f=1&t=482&hilit=NH69722&start=45#p3718 with distance estimations provided by Mr. Bonomi, Mr. Jurens, Mr. Nilsson and Mr. Marty, using different methods, all in agreement about a range between 2300 and 3000 meters.
In same thread some detailed explanation how distance in NH69722 was estimated by the participants to discussion:
viewtopic.php?f=1&t=482&p=3470&hilit=NH69722#p3470
viewtopic.php?f=1&t=482&p=3470&hilit=NH69722#p3494
viewtopic.php?f=1&t=482&p=3470&hilit=NH69722#p3525
viewtopic.php?f=1&t=482&hilit=NH69722&start=30#p3535
viewtopic.php?f=1&t=482&hilit=NH69722&start=30#p3680
Has anything changed since then ?

hans

Bill Jurens
Moderator
Posts: 743
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: A correct attribution for the "Flash Effect" photo?

Post by Bill Jurens » Mon Feb 10, 2020 8:16 am

Thank you for digging up some of this old correspondence, which -- reading the associated dates of posting -- is a good demonstration of just how long these particular discussions have been going on.

So far as I know, nothing much has changed since then. The tools for photogrammetric analysis certainly haven't changed much, nor, so far as I can tell, has the relative skill and abilities of those presenting various alternative analyses. What astonishes me, frankly, is that even fifteen or so years later, we still appear to be working with images posted on the internet, i.e. so far as I know, nobody in that period seems to have actually made any effort to obtain access to the original negatives. One really can't do too much meaningful photogrammetric analysis without this, and -- even if such access eventually obtained, it remains quite likely that the quality of the 'original' record will not permit very accurate measurements to be made.

When one adds in the fact that in most cases the relative timing of the various exposures is, to say the very least, somewhat in dispute, the situation, in my humble opinion, remains pretty bleak. That's not to say that one cannot, via the exercise of (sometimes) considerable imagination and conjecture, find some rough correlations between the photos and the existing original track charts, but it remains, again in my opinion, possible to associate the photographic and diagrammatic record in a variety of reasonably consistent ways, with no way of really telling which of the conjectural reconstructions is actually the correct one. Why does nobody bother? It's not worth the effort, as in the end there is no way to determine which of the reconstructions is the correct one. In other words, we are discussing noise rather than signal.

Let's say, for example, that we had twenty photographs of a ten-minute interval in a boxing match, with perhaps fifteen eye-witness accounts of the action, a few sketches of how the opponents might have moved during the period, and 60 seconds from a 'highlight' reel presented on the evening news. Given that information, how accurately could we really reconstruct the events taking place during that period? My guess is that once a reconstruction was made, it would not stack up to a complete archival film record very well at all.

Bill Jurens

hans zurbriggen
Member
Posts: 83
Joined: Sat Dec 21, 2019 8:15 am

Re: A correct attribution for the "Flash Effect" photo?

Post by hans zurbriggen » Mon Feb 10, 2020 9:02 am

Hello,
to Mr. Jurens: I see your considerations. However, willing to reconstruct a battle, we must use what we have. Even if NH69722 photo has been heavily cropped (to be demonstrated anyway as there are several versions available), than initial distance is even more than 2500 meters, surely not less, therefore we are on safe side assuming 2500.
It may be approximate, it may include tolerances, but to me it is better than saying 'a battle occurred in which Hood was sunk'.
Based from this initial distance, that I understand everybody at last accepted in 2006 discussion (see above links), my answer to Mr. Byron came, supporting a larger than 1000-1500 meters distance for film Bismarck firing sequence.

to Mr. Nilsson: are you aware what was standard PK photo camera used in 1941 and which focal was it mounting? Is there any documentation?

hans

User avatar
wadinga
Senior Member
Posts: 2054
Joined: Sat Mar 12, 2005 3:49 pm
Location: Tonbridge England

Re: A correct attribution for the "Flash Effect" photo?

Post by wadinga » Mon Feb 10, 2020 10:44 am

Fellow Contributors,

The debates of the past raged long and hard around the validity of NH69722 as an accurate record of the vessels' relative positions (and distances) specifically at battle start time. Unlike the present subject and its closely associated photos, there is no doubt as to the orientation of the two vessels relative to one another. I believe Bill's analysis agrees with the speculation that these three photos under consideration here show Bismarck and Prinz Eugen starting on right angle courses to each other, at a time when Bismarck is still firing, and the apparent distance compared with NH69722 is a great deal shorter. For me the Flash picture shows Bismarck on PG's stern quarter, still firing at a very close distance.

Despite guesses about the number of cartridge cases rolling around the deck of PG (including any outside the field of view) and speculations about how quickly salvo smoke dissipates, the only evidence that NH69722 is the very first salvo is Busch's photo annotation in Die Erste. Many of the captions in this book have been disputed subsequently. In The Story of the Prinz Eugen, published after the war, Busch says the German ships were 600 yds apart at the start of the action.
this seems to be a well informed article about the standard issue Wehrmacht Leica, and the painter Schmitz-Westerholdt (part of the PK team) quoted in Die Erste, specifically mentions his Leica being around his neck although he never took any photographs.

All the best

wadinga
"There seems to be something wrong with our bloody ships today!"

User avatar
Herr Nilsson
Senior Member
Posts: 1433
Joined: Thu Oct 21, 2004 11:19 am
Location: Germany

Re: A correct attribution for the "Flash Effect" photo?

Post by Herr Nilsson » Mon Feb 10, 2020 1:03 pm

hans zurbriggen wrote:
Mon Feb 10, 2020 9:02 am

to Mr. Nilsson: are you aware what was standard PK photo camera used in 1941 and which focal was it mounting? Is there any documentation?
Not necessarily just a Leica. The Kine Exakta was also very popular for Kriegsmarine purposes.
Bill Jurens wrote:
Fri Feb 07, 2020 7:17 pm
Technically -- i.e. without quibbling about the third decimal place, the standard motion picture 35mm frame was essentially 24.9 mm by 18.7 mm in size.
As far as I know the format for the Arriflex was 22x16 mm.
Regards

Marc

"Thank God we blow up and sink more easily." (unknown officer from HMS Norfolk)

Bill Jurens
Moderator
Posts: 743
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: A correct attribution for the "Flash Effect" photo?

Post by Bill Jurens » Mon Feb 10, 2020 3:31 pm

Herr Nilsson has given an alternative frame format for German motion picture cameras of the era. My data was based on North American standards, which I assumed -- it appears, in this case incorrectly -- were universal, i.e. international, in nature. I trust his figures. The discrepancy between our figures helps, I think, to illustrate how 'slippery' this sort of analysis can really be.

Mr. Zurbriggen's commentaries along the lines of "we must use what we have" are valid, although I might rephrase it along the lines of "we have no options other than to use what we have", thus removing the imperative. Phrased another way, I am not sure that we 'must' do anything a all Others, certainly of equal or greater intellect and experience, may disagree. The differences in opinion here really seem to revolve around exactly how much interpolation and interpretation is appropriate, i.e. at which point a reconstruction becomes excessively imaginative. There are, of course, no really firm rules on this, and some individuals will legitimately draw the line between forensic reconstruction and imaginative interpolation at different places. This somewhat resembles the situations often occurring in legal cases, where in some situations guilt or innocence can be based upon a preponderance of the evidence, whereas in other cases the situation must be determined 'beyond reasonable doubt'. The question is, really, at what points -- if any -- do conjecture and fact coincide.

This sort of problem may not be resolvable, and certainly seems to have remained somewhat intractable for the last fifteen years or so whilst it has been debated on this forum. In the absence of new material, we may just have to accept the fact that different interpretations can be made, and that in many cases it's difficult to determine exactly which interpretation is superior.

That's discouraging to be sure, but on the bright side, I am encouraged by our newly-recovered ability to discuss the issues in a reasonably respectful manner. Pregress, though perhaps glacially slow, continues.

Bill Jurens

Bill Jurens
Moderator
Posts: 743
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: A correct attribution for the "Flash Effect" photo?

Post by Bill Jurens » Mon Feb 10, 2020 3:32 pm

Duplicate post on my part, deleted.

Bill Jurens

hans zurbriggen
Member
Posts: 83
Joined: Sat Dec 21, 2019 8:15 am

Re: A correct attribution for the "Flash Effect" photo?

Post by hans zurbriggen » Mon Feb 10, 2020 5:57 pm

Hello,
to Mr. Wadinga writing: 'I believe Bill's analysis agrees with the speculation that these three photos under consideration here show Bismarck and Prinz Eugen starting on right angle courses to each other'
Very respectfully, I do not believe so. Can you please show where (here or in his book-articles) Mr. Jurens agreed with this speculation?
I do not agree, believing angle was more 50-60°. Mr. Nilsson confirmed that there is possibility of such an angle as well.

To Mr. Nilsson writing: 'the format for the Arriflex was 22x16 mm'
this would slightly change calculation of the distance of Bismarck in Prinz Eugen film:
35 mm : R = c 1200 meters (instead of 1100)
50 mm : R = c 1750 m (instead of 1600)
75 mm : R = c 2600 m (instead of 2400)
'Not necessarily just a Leica. The Kine Exakta was also very popular for Kriegsmarine purposes.':
Do you know whether PK standard equipment included other than the 50 mm focals, e.g telelenses ?

To Mr. Jurens writing: 'This sort of problem may not be resolvable, and certainly seems to have remained somewhat intractable for the last fifteen years or so'
I beg pardon for using an 'imperative' in order to simplify. I agree with the difficulty to evaluate when very high probability equal fact or is still conjecture.
However, in case of NH69722 I think there was no debate left open at the end of the thread at this link: viewtopic.php?f=1&t=482&hilit=NH69722&start=45#p3718 .
Distance was evaluated by several people, using different methods, in 2300-3000 meters. At least one point seems to me sufficiently ascertained as the only one not in agreement at the end was Mr. Dale, speculating about the ship not being the Prinz Eugen.

hans

Post Reply