a stitch in time

and you should never under estimate the benefit of a fresh piece of cloth.

So while you can sometimes make something with stitching things together, you can't always use stitching (or HDRI) to make up for the short comings of gear.

Conventional "wisdom" out there at the moment seems to be use image stitching techniques for making your wide angle lens wider (or your normal a wide). Such as this image ...



This image was made with a mild telephoto and stitched together from 5 images taken in portrait to make a highly detailed wide format landscape.

Sadly however this doesn't always work, as nature often is not as patient as me and spoils things by moving.

So when I tried to align these two images the churning clouds have moved between images and re-orienting the camera to make adjustment difficult / painful / tedious or plain old impossible. The above image was (as it happens) taken with my Bessa I on two sections of 6x9 film. I wish I had of had my 6x12 camera and a 75mm lens.

Sometimes there just is no substitute for one take.

Bessa I with Colour Skopa lens

I have had 2 Bessa cameras now I find that I like them. I found that  I prefer the simplicity of the Bessa I over the Bessa RF cameras and some time ago sold my RF. Sadly I made an attempt to service the shutter on the first Bessa I I bought (blog post on that here) but ended up damaging the threads of the lens when disassembling it. I have been variously distracted in the intervening time with things that some readers of my blog will be familiar.

For some time I have been looking for a Skopar lens version on eBay and recently I picked one up. Its a tidy camera with everything in largely good order (except that the shutter is a bit sticky, but I'm not going to touch this one ... yet).

I put my first film through it this week and was stunned by the results. I uses a roll of Fuji Superia X-TRA 400 through it and scanned the results on my Epson 4870 flatbed scanner.

WOW

The camera is not without issues (who isn't) and light leaks interfered with a number of the images, and operator error interfered with a few more. However where I didn't screw it up (not withstanding the light leaks, which were probably operator error too) the stunning shallow depth of field on what is essentially a mild wide angle showed me just why it was I loved this format.

Further the sharpness of the lens from side to side is (in my view) sharper than the Vaskar lens on my earlier one.

I was so impressed with the sharpness and detail I went and took another shot of the same scene as one of the negatives (without a flaw) with the GH1 so that I could make a comparison. It is this which I wish to show you in more detail.

 Bessa I - overview

GH1 - overview

Anyone who has done much with colour negative knows that scanning colour negative to get exactly what a digital camera gives is nearly impossible, so please forgive the colour differences.

However firstly I'll point out that the 6x9 camera produces an image which is a ratio of 2:3 where the GH1 produces one of 4:5 (well in this setting). I happen to like 6x9 as its the same image dimension ratios that 35mm is.

Methods:
I set the zoom on the GH1 to give the same horizontal width as the angle of view of the Bessa.
I scanned the film at 1200 dpi. Both cameras thus produced an image which was more or less 4000 pixels wide. Of course I could probably get 2000dpi out of the scanner without difficulty and looking at these images I'm inclinded to believe that I would indeed get more detail out of the film ... it would not just be the scanner equuivalent of 'upscaling'

So, lets have a look at the details:

Bessa - segment 1

GH1 - segment 1

Dam that's close ... ok, lets look over to the skyscrapers in the LHS background

Bessa - skyscraper segment2

GH1 - skyscraper segment 2

I think that the Bessa is actually resolving the details in the buildings better than the GH1 is. That's astounding as at this level (100% pixel view) we are clearly at the limits of the GH1s capacity and yet there is scope for better out of the Bessa negative

Bessa - segment 3

GH1 - segment 3

Well again (forgetting tones) the Bessa has done a better job of resolving fine details in the leaf (particularly that fresh frond that is vertical and as yet unspread).

If you disagree that there is enough in it to call, then that means that simply this is a tie. So I will go as far as saying that this digital camera which was released in 2010 has caught up to what 6x9 film cameras with box brownie film could achieve back in the late 1950's. Certainly when I did this with my Canon EOS 10D the Bessa with the Vaska lens was its equal.

This does not mean that I'll stop using the GH1 (or any other digital camera for that matter) but does re-enforce why I keep using the Bessa. For there are times when Negative can capture a better image than digital can (such as high contrast situations). My initial view of what to use the Bessa for was for those times when I wanted to have something light weight as a backup camera and didn't want to drag the large format camera around with me.

For me this shows that if you are wanting to use a 6x9 folder for some images, that you don't have to feel that you are not going to get top notch results. If you aren't getting results like this out of it, then you are doing something wrong or you have not set your camera up properly (AKA doing something wrong)


:-)

fixing a floppy Bessa folder

readers of this blog will know that I quite like my Bessa I and Bessa RF folders. They are great cameras but can be less than ideal producing soft images at times. I'm an analytical kind of guy so I quickly applied a little investigation into this camera to sort out why its not so perfect.

It turns out that a little wear had occurred and some play had been introduced in the rigidity of the front element. Carefully folding and unfolding it I noticed that this part of the mechanism which completes the locking down of the 'front standard' needed assistance.



If you grab your camera (assuming your a Bessa RF owner) you can see here the place I mean. The suraces which the red arrows point to meet when unfolding, and the
arm with the white masking tape on it merges in behind that "can opener" shaped one to lock the front standard (when thinking of the bessa as a non adjustable view camera). This had developed some wear in the mechanism and was allowing the front element to swing a little. This would be desireable in a view camera but its not what you want in a rangefinder ;-)


How I've solved it is to place essentially a small shim over the small arm there. In the photo above I've used masking tape to test the idea, but I've now settled on another simple method that is non-destructive and reversable. I have bent a small section from a aluminum beer can over the arm and secured it with a thin bead of blu-tac. I put the bead of blu-tac it on the inside of the aluminum along where it will sit on the arm. I used the edge of a metal ruler to form the section of aluminum shim so that it folds in a neat "3 sides of a box" and sits neatly over the arm.

This rig now alters the Bessa to have a nice solid front and has returned a snappy feeling to the unfolding (rather than the wobbly feeling before). I've checked the alignment of the lens carefully with both film and ground glass and its been working fine ever since.

Here is a sample of the full width of the horizontal middle of a 6x9 frame at f8



I think that you can easily test your front element by

1. unfolding the camera as if taking a photograph
2. opening the film door of the camera exposing two light sealing plates at the back (these will allow the camera to sit on the film plane with the lens facing up
3. place a flat and light spirit level across the front of the lens

using this you can get an idea how far tilted the lens is from the focal plane (measure in X and Y axis to make sure).

My Bessa is now pretty spot on (as can be seen above)

Bessa - film slack is bad

Soon after I started using my Bessa I had some annoying problems occurring like in this image:



All the opinions I've had on this indicate that its light leaking along the edge of the film when changing the film (taking out the old, and putting in the new).

For those who are not familiar with 120 film this may seem a strange thing, but the fact is that 120 film is just wound onto spools (like in the above wikipedia link) with a light proof paper backing. This effectively sandwiches the film between two layers of light proof coated paper. It depends however on being tightly wound or else light can leak in at the edges of the paper and spool at the top.

There is a pinch spring arrangement which is designed to keep the film tight on both the takeup spool and the film supply spool.

It looks like this (I'll get to the foam in a minute)

Many people have suggested that this needs to be bent in a little more to keep the film from becoming unspooled with movement in the camera.

I tried this, but well .. its spring steel, thin and flimsy. So I was less than confident that this was going to work.

Well sure enough every film I took out of the camera needed a final "wind" in my fingers to "tighten it up on the spool".

I thought that what it needed was some 'friction' and something to 'grab' it to prevent it becoming slack. Sure the take up spool can't move backwards, but nothing is holding the supply spool.

I thought I'd add a little window sealing foam tape to the holders of the spools to ensure that it was firmly held and didn't unwind.

I put a little of it to go around the edges of the holders like in this picture. Even with no film on the spool you can see that it holds it nicely against the spring clip.

Its important to not use much tape either, as you only want to make the slightest effect. The tape here bends around over the spool holder so that the paper flow will not then snag on the foam and then drag it into the holder and cause any jam.

The tape is at both ends to keep the roll tensioned evenly as in the image below.



It then neatly drags the paper backing and the self adhesive tape has enough space on the other side so that it attaches well and can't be then dragged under and jam mid roll (this btw is what happened on my first attempt).



There is slightly more resistance on the uptake now when I advance the film, but since then 100% of my images have been perfect and film is wound on as tightly on the uptake spool as it was on the supply spool when I got it from the packet.

fantastic :-)

the Bessa RF

The Bessa RF is a 6x9 120 film format folding camera. You might not be so keen on a 1942 camera (even if it is cheaper) so stay tuned for developments on this front, as it appears that there will be a release of a new Bessa III in 2008. That camera will be by Fuji, and if you find this camera to be interesting that one will undoubtedly address every niggle and problem I have with this camera. Very exciting stuff, but anyway back to this camera, the RF.

A while ago I bought a Bessa I and was impressed with the camera despite its limitations. Although the RF is an older camera I was eager to see if the RF would be a better tool for making images than the I.

Despite being older than the Bessa I it has a couple of features which are not found on the I (although can be found on the more expensive Bessa II)

• focus is by rack
• there is a coupled range finder to assist with focusing the camera

For people brought up on SLR cameras (where a mirror system allows you to see what is focused) these cameras are more like the point and shoot digitals with only a small peep hole on the side of the camera to allow you to point it in the right direction when taking a picture. Focus (since they didn't have auto focus in 1942) is done by the photographer.

If you're not familiar with Range Finders they are a tool for determining distance. If you know the distance you can then set the focus by positioning the lens. You can see the focus knob more clearly in this image. Its that knob on the bottom there (which when holding the camera is the left hand side)


It works by moving the lens (by that chrome arm there) back and forward, just like you do with a view camera. The main difference (aside that its much smaller) is that you can't view the image on the ground glass to confirm focus.

The camera has three small windows along the top. The one on the left (of this picture) is the view finder, just like my Bessa I this is what you use to roughly point the camera at what you're taking. Then left and right of the bellows (second and third along from the left) you can see the two range finding windows that the system uses to help you to focus.



You may notice a tripod mount over on the bottom corner of the front cover (swung to one side) . This is a tripod mount for portrait orientations. The whole thing folds down neatly to a compact package not much bigger than your hand. Its smaller in all dimensions than a VHS cassette (remember them?) and weighs about 800 grams.

Its slim and easily fits into the side pocket of a backpack. I've found so far that it makes VERY good images (when you get the focus right). Unlike many 6x9 cameras you actually get very nearly 6cm by 9cm, now, if you consider that 4x5 sheet film is about 10cm by 13 cm it means that you can take 3 images with this and stitch them together in your favourite stitching program (mine is PTGui) to make the equivalent of scanning a sheet of 4x5.

Well, I took my 4x5 out the other day, and set up with my 90mm lens (about the same as 28mm on a 35mm or Full Frame DSLR camera). This is a little shorter than the focal length of the Bessa, which is 105mm making it slightly wide of normal, but when you combine three images to make one larger image (as I mentioned above) they are almost exactly the same.

The image to the right is a screen grab of the scans I took with both cameras. To me the top one stands out as being sharper than the bottom one. Well folks the bottom one is from the 4x5 and the top one from the Bessa RF.

Hot dam that's sharp!

These are scans at 2400 dpi using my Epson flat bed scanner, which is right at its limits at this sort of detail (some say slightly past them ... but anyway). But still, both are scanned on the same gear, so if you get a better scan with a better scanner, then both will just look better!

Assuming a little overlap on your images when you take them this will make an image that is 14,000 x 8000 pixels in size. Thats around 100 megapixels if you want it!

I hope your PC is grunty.

So if you're a landscape photographer, and presently using a compact digital camera but are not satisfied with its image quality (even though it might say its 10 megapixels). Then keep taking your compact with you as its an excellent light meter (and for the snaps too) and add a Bessa folder to your pack! The other weekend I took this image (which I've even scalled back here from 5200 x 7700 pixels) if you click on it (like all the images on my blog) you'll see a bigger one (though not the full size).



its not all roses though. I've had some issues with the camera, so stay tuned (or if you're in the future search on the Bessa RF key words.

old cameras still work

Just spent a little time with my first roll of colour 120 film in my Bessa (listed elsewhere on this blog, just search for bessa). I can only say that it has passed with flying colours. The image below links to a rather bigger 3832 x 2628 pixel image (which is about 10 MPixels), Yet this is only from a 1200dpi scan of the film. For people used to working with 35mm, that minature format requires 2700 dpi (and preferably 4000 dpi) to get substantial images. Scanning this at even 2400 dpi would give 7864 x 5256 or 41 megapixels.


Dang it all, blogspot resizes it to 1600! Oh well...





Now, this is a 1950's camera. It has no mod cons of any type (no metering, no focus assist, primitive film advance).

Used right, this goes close to giving the sort of detail that you can get with a 1Ds MkIII. Certainly its not as fast to use, or have any features that would make it popular to people photographing Wimbledon or Pro golf, but for an amateur photographer who has a 5 or 6 megapixel compact digital and is seeking better 'landscape' images ... well look seriously at a 6x9 film camera!

I personally like the the Bessa as its a folder, meaning it folds down to a little bit smaller than a video cassette and slips into a side pocket of my backpack. For those who don't feel like an old folder like this you should seriously consider a Fuji GS690 as these are superior optically and have range finders (for focusing the camera).

Size maters (for film that is ;-)

I thought I'd put a few more samples from my flickr account



and a detail segment



click here for full size

notice how the tree I chose for focus stands out? This short depth of field can be an asset that you just don't get with smaller formats (like 35mm or digital).

Bessa useful lens limits

To add a little more to flesh out this camera to a potential buyer (for using not for those who simply collect), I thought I'd explain that this camera will simply not give sharp images from edge to edge at less than f11.

There is debate about the Vaskar lens on the camera, and while some say that the lens isn't sharp other swear that it is. Here I'll say that it can be and that it depends on the photographers choices.

Here's why.

For large format camera users looking at the film plane with ground glass is a normal and even essential thing to do. Large format cameras even have the corners cut off on the ground glass, making it easier to actually do this.

The reason you'd want to look here is to see if you've moved your lens too far, and run out of the area where the image from the lens actually falls (you see, large format cameras move the lens around to give control over focus and perspective see this explanation). So I popped open the back of the Bessa early on in the piece to see what I can see. I've done it again here to explain my assertion above.

In this image, I've taped a bit of translucent (foggy) plastic sheet to allow you to see the image formed by the lens, and the lens through the corner of the frame.



You can just see the lens through the torn off corner of the palastic and you can see its not quite perfectly round. Looking at the image below, taken at f4.5, you can see that the shape made by lens is not perfectly round. This will mean that less light will fall on the edges of the film than in the middle. This will make the image darker at the edges (vignetting) and will also reduce the image quality too.



This image is taken at f8 and you can just see that the the lens is still slightly obscured in the bottom third at the left.


its not until you get to f11 that the lens circle is now nice and round


So basically this means that the lens needs to be stopped down to about f11 or lower if you want edge to edge uniform lighting, or even edge to edge sharpness. You'll notice we're not right at the corner of the film plane here, so to be on the safe side stopping down a little more is needed. In practice I've found that you need to stop down at least f16. So, essentially this means that this is a "sunny 16" kind of camera or use a stable support (Eg tripod or solid base) and then use f16 (preferably smaller) and 1 second or longer so you can get reasonably accurate exposures.

Further, there is no accurate markers in setting the aperture (no dent to feel the position of the aperture setting as in more modern cameras) and no 1/2 or /1/3 stop markers. Meaning you've got to 'guess' the adjustment and being more accurate than full f-stop is challenging (meaning that trying to set 1/2 or 1/3 stops is guess work). This is ok with black and white, and possibly colour negative, but I think it makes a challenge for slide or 'positive' film.

However, if you can set it at f16 or f22 (and use a tripod) it gives very nice results. Like this close focus example at f11:


You can see (if you click on the image) that the edges are starting to fuzz ... this was exposed at f11. Below is and a segment from the center of the negative. Please click on either to see larger versions.


So, if you can put up with the difficulties of using this camera:

• limited useful f-stop range
• single focal length
• no light meter
• focus by number (needing some way to correctly pick the distance)
• no 'dent' to be reliably sure of f-stop selected
• no partial f-stop marks

then it can provide you with excellent images.

Bessa testing part 2

To further answer my questions about the usefulness of my Bessa 6x9 camera I took another couple of images and compared them to the theoretical 5D (actually my 10D with a similar angle of view lens to what the Bessa actually sees).

The image at left was taken today with some slight sleet happening (to obscure the distance) and generally dreadful dull lighting conditions.

I've scanned the film on my Epson scanner and presented the results here.

Firstly lets have an overview of the image taken with the 10D.


As you can see it represents a much smaller segment, and the target area is not in direct sunlight (making the overall contrast similar to what I took with the Bessa above).

S0, lets get to the details:


This (above)is a 100% crop of the image taken from the 10D, while the image below is a 100% crop taken from a 2400dpi scan from the film (ADOX CHM 120 roll). Exposure was 1/10th of a second at f8


Please click on this to see the full detail (as this is sized down by the blogger software, you can even see some of the snow fall in better detail).

Quite stunning if you ask me, but I'll leave you to make your own conclusions.

From the same roll (but no digitals taken for comparison) is this image. I took this propped on a log lying around (no tripod) for support at 1/25th of a second @ f22.


Please click on it for a larger preview. I've then taken a 100% crop from a 2400dpi scan of one of the logs poking out of the stack there. Again this is resized by blogger, so please click on it for a the full 100% view.


Amazing stuff. Using this camera it is important to keep the aperture small. Even though this lens is rated to open up to f4.5 it is nowhere near good at the edges at this. I consider that f16 is the maximum working aperture for clear images (and f22 its best).

So I am now comfortable that the combination of my 10D and the Bessa for images on my trips will give me the combination of versatility and ease (the 10D) and high quality images if I so desire or need (the Bessa).

Testing the 6x9 - Bessa vs DSLR

I've been very keen to see if my new 6x9 folder will make an effective backup to my large format camera, and if its close to my Canon 10D DSLR.

Timing of arrival of the camera coincided with running out chemistry. I've finally got some developer but not any 120 film, so I decided to shove a sheet of cut down 4x5 ADOX into the camera and see what I see. I goofed on the first cut of the sheet, so the second 'slice' needed to go in at an angle to ensure the plate held it firmly. This is the over all view of the 'test shoot'.




The 'angle of view' is very slightly wide, more or less its 'standard' (the lens is a 105mm which on 6x9 gives 44° which is about the same as a 45mm lens on a 35mm camera).


This is an image taken with my 10D with an EF24 f2.8 which as you can see is a little wider. I should point out at this time is 'effectively' a 38mm due to the APS sized sensor recording only a segment of the image.

You can see the extra width here to the right of the chest of drawers, basically the ironing board is in the scene.

To get the image from film I've used an older Epson 3200 scanner. If you don't know them, they are not the hottest item on the shelf, and are regarded by many (including me) as running into their limits at about 1200 or 1800 dpi.

Well, to see as much as I could potentially see, I've scanned the film at 2400 dpi but then down sampled to 1200 for presentation here.

Anyway this is the image that I got from the film:



Quite surprising, certainly a little soft looking, but still one can see the marks on the ruler. The ruler in the picture is (as can bee seen) 1:75, which means that 1 full graduation ~= 1.25cm

The smallest marks are less than 1mm apart, but these aren't quite distinguishable. The next largest markers represent just over 1mm and are distinguishable.

However looking at the image closely , something doesn't quite look right.

Here is a segment from the original 2400 dpi scan. It seems to be showing signs of motion blur when you look carefully at the graduations on the ruler.

Since the camera did not mount securely to the tripod and the exposure time is 8 seconds its entirely possible that I introduced some vibration in the picture!

If so then this image could be significantly shaper than it is, perhaps nearly as sharp as that from my 'hypothetical full frame' image presented later.

Anyway, moving on, next is the 100% segment from the 10D using the EF24. You'll notice that the match box is oriented differently. I only thought of comparing to the 10D AFTER I had taken and developed this shot. (Dumb I know)



Not quite as large and certainly not as sharp. Yet this image was taken with the lens set to f11, I used a Manfrotto 190B tripod, cable release and mirror lockup using RAW (so no in-camera artifacts are working here to lessen the image potential). So its unlikely one would get a sharper image from the camera.

So for me this indicates that the trusty old Bessa is able to produce an image that can compete with that from a modern DSLR in terms of outright quality if nothing else. Not bad for a 50 year old technology.

The next question that comes to mind is, how would it stack up against a full frame DSLR? Well, I don't have one, but I do have a 50mm lens :-) While a 5D is not quite the same pixel dimensions as the 10D (so the density of recording for a given area is not as 'high' as the 10D) it is close! So, by putting my EF50 f1.8 onto the 10D I can take a segment from this to 'simulate' a hypothetical full frame camera with nearly the same angle of view as the 6x9 folder.

Taking the shot again (not moving anything this time ;-) I can get to see what a full frame digital would probably give in this segment.

Here is that image:



Definitely this is cleaner than what I was able to get from the Bessa and Epson 3200 and completely resolves the ruler,with the 1/2 graduations now clearly distinguishable (click on the image to see the full detail). The better depth of focus (and perhaps focus itself) shows the ruler to be clear.

Stop press

A friend of mine (who actually has a 5D and an EF50mm lens) has taken a picture of a similar ruler at 130cm just as in my test (with slightly different lighting).

This at left is a segment of that image.

So no doubt about it, actual 5D looks remarkably like that of my theoretical 5d (that is a 10D with the same lens mounted just therefore capturing less view) and very very sharp indeed.

So at $2500 will out perform a $100 1950's Voigtlander Bessa with a Vaskar lens in image clarity.

Hmmm ... I would love to see how a Fuji GW690 would do.

What does this mean?

So for me at this point I'm happy with my purchase. I am comfortable that by using my 10D for most of my trip photograpy and using the Bessa for a 4x5 replacement that I can get a good combination of lower costs and higher quality images. I can't justify the money for a 5D at this point, as well I'm sure that the digital cameras will improve (we've all been waiting for a 5D replacement for some years now). With film at about $3 a roll, and developing around the same I'm sure this will 'get me by' for at least another 2 years. Perhaps then a full frame DSLR will be affordable for me.

Lastly, I am sure that taking 2 or 3 images with the Bessa and stitching them together will give me results as good as my 4x5 :-)

Experiments in 6x9 cameras

Today I got my latest addition to my image making tool set. Its a Voigtlander Bessa I.

For those who aren't familiar with it, this is a 120 roll film camera, which is essentially the same paper backed roll film from the good old box brownie (minor changes).

This film is still popular with professionals today, and has been in use for since around 1900. Unlike 35mm (more or less killed off or at least badly wounded) by digital the larger image area provided by this film means that you can comfortably get 12Megapixels from even a modest flat bed scanner, or as much as 26 Megapixels from higher quality film scanners (like the Nikon LS 9000).

Of course its the lens which makes the image, and while all the other bits of the camera help as well you just can't get a good image with a crappy lens or poor focus.

The lens on mine is the Vaskar, which I'm told is no "killer".



However looking at the image on a ground glass (pressed against the film rails) seems to show images that are quite sharp (as long as you have them in focus that is) but a quite narrow DoF.

This last point (focus) is where this camera suffers in use. Focus is by adjusting the lens with nothing more than some markers on it for predicting the distance.

You can see the distance scale on the black ring around the lens (white numbers), as well as the black arrow just visible over there on the left side.

With no way to preview or confirm your focus this is all you have. I've checked it by making a bit of ground glass and putting it on the film area at back of the camera, and its not very accurate. But what you can see is that when its focused, its quite sharp. Nearly as sharp as anything else in my camera outfit (which includes 90 and 180mm Fujinon large format lenses, as well as Canon EF series lenses).

Worse the depth of field focusing guides (black on the chrome) suggest that when focused on infinity that things as close as 20 feet (6 meters) will be in focus too. Sadly this is quite optimistic, and I'm sure lends to the reasons why these older cameras were dismissed as being rubbish by a generation.

The shutter is a bit sticky on the example I have, so I'm not confident that it will operate consistently at shutter speeds of longer than 1/10th of a second. However, for daylight use I think this will be fine.

Anyway, no pictures yet as I've got no developer at home just now, but watch this space for some soon :-)