antonio.giacomo

<p>I would take a carbon fibre tripod. Also, a second camera body. But if you are doing interiors, do check that a tripod is allowed. Do not forget the battery charger.</p>

<p>The late Henri Cartier Bresson (please genuflect) used to put black tape on the shiny parts of his camera to make it far less obvious, so that he could get much closer to his prey without arising suspicions (but he was originally a hunter). He also used to tie the camera to his wrist to avoid looking like a tourist.</p>

Hi Kevin,

 

I was told that the "standard" lens for 5x4 would be 150mm, because that is roughly the diagonal of the frame.

However, for architecture and landscape I found a Rodenstock 180mm OK for most situations. If I want something a

bit wider, I use a Nikon 135mm. If it is an architectural interior, I usually use a Schneider 90mm Super Angulon,

which is a massive lump of glass, but it does cover 7x5, so there is a lot of movement possible on 5x4.

Hi Rishi (et al),

 

I use neither Nikon nor Adobe software to convert .NEF files. I use DxO Optics Pro (www.DxO.com). It does far more than just converting raw files. Every supported camera and lens has been calibrated in an optics laboratory. It aims to correct all the defects in a body, and all the aberrations in lenses, and the results are very visible.

 

As a professional in image processing, I can immediately see what it is doing, even though DxO Labs give very little away on their web site (would you expect them to?). It is not particularly fast, even on a fast multiprocessor, but that is because its algorithms are very advanced and processor intensive.

 

In terms of results, it is like the difference between fast 35mm film and slow medium format film.

 

I have produced a 600x800mm print from a Nikon D80 with the 18-70mm lens. At the time, I did not know that the customer wanted a print that size, otherwise I would have digitised from medium format. It does not have the biting crispness that I would have got from pulling a Cibachrome from 5x4, but it is perfectly acceptable, and the customer ordered further work.

 

DxO Optics Pro supports a very respectable range of camera bodies and lenses, including third party lenses.

 

It does have one defect. Its perspective correction facility is no better than Photoshop’s, which means that it is utter crap (I do know a little about perspective control).

 

If you are a serious digital photographer, it would be worthwhile running a free trial. However, do not expect results as spectacular as those illustrated on their web site, because they have obviously chosen very bad cameras and lenses.

Finally:<div></div>

Yet more:<div></div>

More:<div></div>

Hi all,

 

If you are seriously into architectural photography, where converging verticals are a cardinal sin, you are certainly wasting your time trying to correct them with Photoshop.

 

Neither Filter>Distort>Lens Correction nor Perspective Crop use the correct projective geometry. Assuming the camera has been tipped up, all they will do is to squash the image more at the bottom than the top, whereas, to do it properly, the image needs to be stretched vertically as well. Unfortunately, the stretching is not linear ? it cannot be done by resizing the image vertically ? more vertical stretching is needed at the top than the bottom. As a result, these two tools only give a reasonable result if the amount of correction is very small.

 

Edit>Transform>Perspective is much better, but, even then, the result is technically wrong. Worse, and this is a serious defect in Photoshop, the operation has to be set up by eye.

 

Although my main interest is photography is street photography, I do architectural work as well, and I try to get things as perfect as possible. Ideally, I would use a view camera, which is, overall, a lot faster than a digital camera with its fixed lens, but often I have only a digital camera, admittedly with a suitably wide lens. For digital, the software I use is PTGui. It started as a GUI front end for Panorama Tools, written by Helmut Dersch, which stitches images together to form a panorama. Other than an image in the set which is dead centre of the desired panorama, all other images will need perspective transformations. It is also possible to stitch an image to itself. PTGui is now much improved.

 

The way it all works is that you mark control points between the adjacent images, then the panorama creation applies perspective transformations and affine transformations to register the images so that they can be blended. In the early versions of PTGui it was a bit tedious, but it did mean that you were going into the detail, and not having to set up the perspective transformation from the whole of the images by eye. The latest versions of PTGui have a tool to generate control points automatically, which can then be manually edited. PTGui also has a facility for marking horizontals and verticals between the images. To apply perspective transformations to a single image, horizontals and verticals are marked within the image. Then, when you say ?go? it will rotate the image, and apply the perspective transformation. Also, if you have not corrected the special distortion of you lens already (using, for example DxO Optics Pro), it will correct the distortion of your lens.

 

There is a problem with PTGui. It does not lower the basic resolution of the image, so it can generate massive image files if there is a lot of stretching to be done, which need to be cropped. Other software might find its output a bit much.

 

Now for some examples. The first image is the result that we are trying to achieve. It is not digital ? it was shot on a Wista Field, with a 90mm Super Angulon (and I apologise about the tonal range ? the tonal range of Ektachrome, or any silver image, can be way beyond the capabilities of digital technology, in this case a large format scanner). My camera was dead straight (laboriously checked with a large spirit level), so, if you think something is not straight in the image, it is the Cathedral, not me. For example, the roof of the nave has a distinct bend.

 

The second image is a photograph taken by my daughter in Brugge, as it came from the camera. Like me, she cannot hold a camera accurately vertically, so it needs a bit of rotation first.

 

The third image is the result from that image obtained from Photoshop?s Filter>Distort>Lens Correction. What it has done is to squash the image horizontally at the bottom, and, when cropped, the resolution turns out to be much lower. But, worse, the image has not been stretched at the top. This is awful.

 

The fourth image is the result obtained from Photoshop?s Edit>Transform>Perspective. Now there is some vertical stretching, but, unfortunately, there is too much stretching at the bottom. This is bad.

 

The last image was processed through PTGui, with a little bit of enhancement with Photoshop to lighten the shadows. This is the only one that looks right to me.

 

http://www.photo.net/photo/4814809

 

So, my recommendation to anyone who wants to apply perspective control to an image in post processing is not to use Photoshop (which is crap), but to use a tool that is dedicated to the job (or, much better, get yourself a view camera, so you do not have to do anything at all after you have fired the shutter: just send the film to the lab).

 

<div></div>

IMHO T400CN was the very best 400-speed film that Kodak ever made. I still have a few rolls in my refrigerator. If yours has been refrigerated, it should be OK for quite a few years yet. The colours will not go off because it does not have any. For the best results, you MUST process in C41 chemistry, and print it onto traditional bromide, not colour negative. If you process it in C41, you can expose it anywhere between ISO 50 and 800, with standard C41 processing. 50 will give you exceptionally fine grain, but with rather thick negatives. 800 will give you perfectly printable negatives but with more grain. If you want to process it in a black and white developer, the only one that I know of that handles chromogenic film is Speedibrews' 422. 8 mins at 20 centigrade in the neat developer should give you about an EI of 3200, but the negatives will be very difficult to print. Stick to C41.

 

When Kodak discontinued the film they suggested that the Portra BW film was a substitute (but it is intended for printing onto colour negative paper), or Tri-X (!) if you wanted to print on bromide. Absolutely no way. If you are using 35mm, switch to Ilford's XP2. If you were a T400CN rollfilm user. Hard luck.

IMHO T400CN was the very best 400-speed film that Kodak ever made. I still have a few rolls in my refrigerator. If yours has been refrigerated, it should be OK for quite a few years yet. The colours will not go off because it does not have any. For the best results, you MUST process in C41 chemistry, and print it onto traditional bromide, not colour negative. If you process it in C41, you can expose it anywhere between ISO 50 and 800, with standard C41 processing. 50 will give you exceptionally fine grain, but with rather thick negatives. 800 will give you perfectly printable negatives but with more grain. If you want to process it in a black and white developer, the only one that I know of that handles chromogenic film is Speedibrews� 422. 8 mins at 20 centigrade in the neat developer should give you about an EI of 3200, but the negatives will be very difficult to print. Stick to C41.

 

When Kodak discontinued the film they suggested that the Portra BW film was a substitute (but it is intended for printing onto colour negative paper), or Tri-X (!) if you wanted to print on bromide. Absolutely no way. If you are using 35mm, switch to Ilford�s XP2. If you were a T400CN rollfilm user. Hard luck.

I have just bought a Hama Linear Polarising filter. Written on the

back of the box is �not for cameras with autofocus and/or exposure

metering through the lens (TTL)�. I have tried it on a Nikon D70s with

a AF-S Nikkor 18-70 1:3.5-4.5G ED lens, a Fuji FinePix S2Pro with a

Sigma 24-70 D HF lens and a Pentax MZ6 with a Sigma 24mm 1:1.18 EX DG

lens. All the cameras focus perfectly through the linear polariser.

 

The Pentax instruction book says that the camera might not focus

properly through an �ordinary� polariser, and a circular polariser is

recommended. There is mention of polarising filters in neither the

Nikon nor the Fuji instruction books.

 

The polariser should not bother the lens because it is only the body

that does the focussing. Also, TTL metering is a good idea with a

polariser, because the filter drops 2 to 4 (or more) stops depending,

and you can only measure it through the filter.

 

The accompany image was shot on the Nikon D70s with the filter. The

camera could focus on the clouds.

 

I would be interested to hear about other cameras that do work with

linear polarisers.<div></div>

Verichrome Pan was a 125ASA black and white film. Develop it in D76 (or ID 11) for 7 minutes in the neat developer at 20�C/68�F or 9 minutes with the developer diluted 1:1. My father showed me how to do that when I was 6.

I started in photography when I was six with a Box Brownie. My father taught me darkroom techniques, but, as my profession is now developing image processing software, I know that a lot of those techniques could be done more conveniently by software. The only graphic design that I need is a facility to put text on an image, to avoid using dry transfer lettering on the print, because my hands are not as steady as they were. These days I use a Bronica for serious work (and the lenses are very serious prices as well). Digitising the negatives at the maximum resolution of a Minolta scanner gives about 100 megapixels, which comes out at 0.6 Gbytes raw per image. That is adequate for exhibition prints at reasonable at quality. I do not think that I am asking for the earth. If I did want the earth, I would want a CCD back for the Bronica that had smaller than 10 micron pixel pitch, and 32-bit resolution for each plane, plus some software that could handle it. Even then, it would only give me a dynamic range of 10:1.

 

I stand by what I said about crap software.

The reason I ask is this. I bought a 35mm scanner. It is very nice.

It can scan a whole reel of film, while you have a cup of coffee. Its

maximum image depth is 48 bits at 19 megapixels, which is even nicer

if you are serious about digital images (but obviously you can do

better the old-fashioned way). It came bundled with Photoshop

Elements. However, Photoshop Elements refuses to open 48-bit images.

Other programs, such as JASC Paint Shop Pro and Serif PhotoPlus will

open 48-bit images, but immediately convert them to 24-bits.

 

I looked at Photoshop 7. It would handle 48-bit images. So I bought a

license. When I got it, I was very disappointed. Its facilities for

48-bit images are very limited. It would not be adequate for

professional photographers. So I then bought the Photoshop 7.0 SDK,

so that I could write my own plug-ins. What I got was the most un-

professional load of crap. The documentation is absolutely atrocious.

How Adobe can charge for it I do not know. However, it did explain

the big problem with Photoshop on a PC. It cannot open images beyond

a certain size on a PC. Quite a few gigabytes of RAM that I have,

plus getting on for a terabyte of disc, do not help. It tries to

force a MAC memory management strategy onto a PC, which is the most

amateurish software implementation that I have ever come across (I am

a professional programmer). It might be OK for a MAC, but not on an

Intel platform.

 

I have wasted a lot of money on really crap software.

 

IS THERE ANYTHING MUCH BETTER THAN ADOBE PHOTOSHOP FOR A PC?

Further results

 

Thank you all for your answers. John Cook is right about conventional pushers. Underexposing and over developing just increases the gamma (and makes the grain clump), but it does not alter the ISO rating of the film appreciably. All it means is that you can use a higher meter setting, and still have �printable� negatives, albeit at the expense of shadow detail.

 

John is also right about the �speed� of the fast films. In both Kodak�s and Ilford�s technical documentation, P3200 and Delta 3200 are rated at ISO 1000. Neither manufacturer prints this on the carton. They both suggest an EI of 3200, which is printed on the outside of the carton (Ilford), and the inside (!!!) of the carton (Kodak). But, if you are interested in shadow detail, you have to stick to the ISO rating to use these films conventionally.

 

What I want is a reasonably fine grain film, and a developer that has the effect of shifting the curve to the left, not upwards. I know, of course, that there is a minimum level that the film is sensitive to, which cannot be altered by a developer, only by hypersensitising the film first. I know the techniques that astronomers use, and, whilst I am trained to carry them out, I do not have a suitably equipped laboratory. Anyway, keeping the hypersensitised film perfectly dry is impracticable for casual photography. Damn! It would be rather nice to have Technical Pan going 10 to 20 times faster in low light.

 

Perhaps I started looking in the wrong place, using a chromogenic film with an astronomical �go faster� colour process. I should have been looking for an equivalent monochrome developer. So, the first one I came across was Speedibrews� Celer-Stellar. What I got from the literature was that it gives the bottom part of the curve a boost, and gives the maximum possible speed and highest acutance with no increase in grain or contrast relative to a conventional fine grain developer used normally. A development of Celer-Stellar is Speedibrew 422 (pun not intended), which has similar characteristics, but additionally compensates for reciprocity failure.

 

It is a strange developer. A pre-soak is needed, and most films have the same recommended development time (4�15�). Temperature is optional � anywhere between 15° and 30°C. Control is through dilution. 1+4 gives normal speed, 1+2 gives one stop extra, 1+1 gives two stops extra, and undiluted (should) give three stops extra. There is, obviously, some very clever chemistry going on here.

 

1+1 with Tri-X gives results at a true ISO 1600 at least as good as the previous best I found (T400 CN in Speedi 41), and rather better than Tri-X in Tetenal Emofin. Undiluted, it gives no further speed increase to Tri-X, but just thickens the negatives.

 

The 422 instructions say that P3200 developed undiluted for the normal time gives ISO 3200, which I can confirm, but the grain is much worse than Tri-X. Probably, this is intrinsic in the film, rather than the developer. Development of the same film in undiluted developer for eight minutes is supposed to give an EI of 12800, but I think that Speedibrews are being somewhat optimistic. I find it useable at 6400. If you want a fast film, and do not mind grain the size of golf balls, it would be OK, but the grain is too much for me. I am sticking to Tri-X at 1600 for my high speed film. A further refinement of the shooting technique is the odd glass of muscle relaxant to help with slower shutter speeds. It is funny how the oldies often turn out to be the best.

 

There is an interesting note in the 422 instructions. They say that Fuji Neopan 1600 comes out at an EI of 1600 in undiluted developer, and Neopan 400 undiluted comes out at 3200. It then says that Neopan 400 at 1600 (ie a dilution of 1+1) gives better results than Neopan 1600. Obviously, they are certain about their trials.

 

It also works with colour films. I tried some Agfa Ultra 100. Developed at 1+1 to the silver image, I got printable negatives at any EI between 25 and 400. Bleaching the silver, then finishing as a C41 film gave an EI of 200 only, with saturated, but not very accurate colour. Speedi 41 is a much better with this film, but there is no point. Kodak�s Portra VC, NC and UC will give much better results at ISO 400, and with standard C41 processing.

 

422 does nothing more for chromogenics over the standard C41 processing. However, I will now be using T400 CN as my standard medium speed monochrome film (at ISO 400), because my pro minilab can process it while I have a pint (or two) in the local pub (which is 25 yards from the minilab).

 

I am not being very scientific in all this. I do not have a densiometer (nor could I justify buying one), and I am using the camera meter, rather than my separate incident light meter. However, I am getting practical quick shooting conditions, and practicing using the camera meter where light sources will be included in the frame.

 

You could, of course, say why am I spending so much time with equipment, films and developers, rather than going out and taking photographs. The answer is pragmatic. Once I have got a reliably reproducible technique sorted out, then I can do the serious photography with total confidence, and sod the imponderables. Hopefully, I will be posting some serious photographs next month.

 

All the above answers are good. However, you might like to go onto the film manufacturers� web sites to see what they say.

 

Agfa says that you should not put any sort of undeveloped photographic material in checked baggage. Kodak also says that any film that is rated at ISO 1000 or higher, or will be processed to achieve ISO 1000 or higher, should be segregated from your other hand baggage (which might mean unloading the camera), and offered for manual inspection. Apparently, you can do this in the US, because you have the right. You might have a problem at a foreign airport.

 

But, me being me, I was not convinced. So I ran some trials at Southampton Airport, with the kind assistance of the Chief Security Officer. I had several films that I shot at ISO 3200, and processed at that speed. They were run through the hand baggage X-ray machine twice (once for outbound, once for the return). I had no sign of fogging whatsoever, even though I had processed all the films down to the fog level. Obviously, the Chief Security Officer was happy to hear of the result.

 

One thing that did interest me was that the X-ray machine was so sensitive that it could see straight through a lead bag. I was, of course, on the �security staff� side of the machine, so I could see.

 

The Chief Security Officer did tell me that the radiation level from the X-ray machine was far less than the cosmic radiation that you will get inside the cabin of an aircraft at 30,000 ft. So, as a precaution, I would suggest that you get some lead bags to carry you film in when you are in the aircraft. It sounds like a good tip.

 

Another tip � perhaps you should buy a roll of Ilford Delta 3200, and keep the carton with your other films (sod the film itself � it is not the best film at that sort of speed, being polite to Ilford). However, the carton has written on it �DO NOT X-RAY�, in eight languages.

 

Finally, if you want to make sure that your film will not get X-rayed, how about this. Buy some plastic re-loadable cassettes (they cost as much each as a film), and re-spool all your films into them. You then put them in your pockets to go through the airport�s metal detector. Providing that you have absolutely no other metal on you (belt buckles, penknives, etc.), you should get straight through. Do not forget to put them back into your lead bag before the �plane takes off.

In the situations that I want the film for, most of the image will be in the shadows. The highlights will be the light sources themselves, so I can let them burn out. This is an example. ¼ sec at f/1.7 hand held. T400CN rated at ISO 1600 developed in Speedi 41. The main light source is the 40W bulb on the wall on the left.

 

I am also trying Tri-X in Speedibrews 422. First results are promising.<div></div>

Years ago, my brother was an engineer working for Agfa reprographics. He had a reprographic business on the side, specialising in the restoration of old photographs. As I understood it, the technique was to make an internegative using lithographic film, which was then processed in a �soft� developer, rather than a lith developer. Ideally, prints from the internegative should be sepia-toned, to make them look �old�.

 

If Technical Pan does not do the job, you could try Kodalith Ortho Type 3. In 35mm, it comes 100� at a time. The advantage it has is that, being Ortho, it can be handled and developed with a darkroom safelight, which is useful, because you are going to have to experiment, and find the development time by inspection. I would start developing it with D76, then possibly a paper developer if more contrast was needed.

Whilst I know a lot about the physics and chemistry of photography

(and digital image processing software, because that is my

profession), when it comes to taking photographs, I have a very

simplistic view. If I can see it, and I like what I see, I want to

photograph it. The problem is that, often, I am looking at night, and

very often indoors. In those circumstances, I would never use flash,

because then I would not be photographing what I see, and a tripod in

normally out of the question. Usually the light sources will be in

the photograph as well, but, because the human eye has an enormous

dynamic range, it can accommodate the highlights, but still see the

shadow detail.

 

I still have a steady hand, and all the 35mm cameras that I currently

use have additional battery packs, which make them heavier, and

therefore easier to hold still at slow shutter speeds, down to 1/8

sec.

 

But then there is the question of film. In low light conditions, the

human eye is far less sensitive to colour, so monochrome images are

quite acceptable. In dim conditions, I would use Kodak Tri-X 400

Professional, developed in ID11, which I have been doing for the last

40 years. It gives me the type of gritty images that I like. But, in

the conditions that I am describing, it is not really fast enough.

 

I have tried both Kodak Professional T-MAX P3200 and Ilford Delta

3200 Professional. Whereas I can see that both these films would be

very useful for sports photography in less than ideal lighting

conditions, for my requirements (bearing in mind that I need the

shadow detail), a meter setting of 3200 is far too high. 1000 or even

less would be needed, and that is also with avoiding the light

sources.

 

In comparing P3200 against Tri-X processed in Tetenal Emofin (a two-

bath developer that can give an effective speed increase, but with

very little increase in grain), to my surprise, I was getting better

results from the Tri-X pushed two stops, than the P3200. This is the

exact opposite of Kodak�s technical advice. The only problem is that

the Tri-X comes out very �punchy�. With an enlarger and multigrade

paper, it would not be a problem � just dial a softer setting on the

enlarger. These days, though, I am rather lazy, and I prefer not to

have to spend any more time in the darkroom than I have to. I have a

scanner which will take a whole roll of uncut 35mm film in one go by

itself. While it is doing it, I can have a cup of coffee, then come

back and use Photoshop to produce a �contact sheet�. But,

unfortunately, the dynamic range of the Tri-X processed that way is

far too large for any scanner.

 

At the same time as the P3200/Tri-X trial, I also ran some Kodak T400

CN, which was processed in Speedibrews� Speedi-41. I was amazed at

the results. Not only was the contrast very manageable, but the grain

was noticeably finer, and the sharpness noticeably better, than the

other two films.

 

Speedi-41 originated in astronomy, and is intended to counter

reciprocity failure. The colour reproduction is �creative�, but

astronomers would not worry about that, because they would be using

spectrometers to measure colour accurately. Obviously, it makes no

difference to chromogenic film, because there are not multiple curves

to get crossed.

 

The Speedi-41 process is rather strange. It runs at 20°C, rather than

the 38°C or so of the usual C41 processes, so you do not need

thermostatic temperature control (although I have that). After the

first developer, the rest of the process can take place in daylight.

Even more interestingly, if some of the frames are a bit thin, they

can be cut out and processed again, more than once, if necessary. In

return, though, the instructions say film used in normal conditions

needs to be underexposed between and three and five stops, otherwise

the developer will solarise it!

 

With T400 CN and Speedi-41under low light conditions, I get a

reliable meter setting of 1600 to retain shadow detail in low light

situations, and I can adjust the contrast of the negatives on the fly

during processing.

 

I know that there are other processes that can be employed. Years

ago, I used uranium intensifier, but I can no longer find a supplier.

There was also pre-sensitising the film using mercury vapour, but I

never tried that.

 

Any more ideas?