Larger format

[Plato]

Sure would. If using 35mm full frame lenses, the maximum square image would be approximately 30mm x 30mm.

Since FF is 36X24 why isn't vignetting a problem?

If you draw a circle such that the corners of a 36 x 24 rectangle just touch it and then draw a square by extending the vertical sides of the rectange to 36mm you will see that it would require a significantly larger circle to encompass it.

Edit: I'll leave this here in case anyone else doesn't see the problem instantly. (I had to think about it for a minute - I very nearly made the same post as Plato. )

I edited my post within a minute after writing it but, obviously, not quick enough. Yes, you are correct.

 

[Derrel]

A few months ago, Nikon's Eurpean Manager for Professional products told dPreview why Nikon released the 35mm f/1.8 AF-S G lens in the reduced-frame capable DX format instead of a full-frame capable design. He said,
"The main target is D40/D60/D90 owners. They make up 80% of our DSLR sales..." the rest of the quote is here.35mm F1.8 for DX? What is Nikon up to?: Digital Photography Review

So....the D40/D60/D90 bodies account for 80 percent of Nikon's d-slr sales. Both Canon and Nikon had ,roughly, 40 percent each of world-wide dslr sales in the 2008 year. Given that the D300 and D700 and the EOS 5D-Mark II and 50D bodies from Canon and Nikon represent the pro-sumer/serious enthusiast/semi-pro segment, I would guess that those camera segments represent maybe 10-12 percent of the remaining sales figures, and that the flagship Nikon D3-series and Canon 1D-series bodies make up perhaps six to eight percent of the dslr cameras the Big Two make and sell.

Chasing a market that does not exist, or that is very,very small is not a good strategy. Both Canon and Nikon are making a lot of sales to beginners and to enthusiasts. Leica's S2 system will be a good barometer of just how much demand there is for a larger than FX but smaller than MF digital solution. I personally think that people will soon be up-sampling images more often than turning to larger capture formats. Already the 24 MP D3x is showing the limitations of camera support and critical focusing as well as lens quality and the need to shoot at non-diffraction-affected apertures. There's a big handling gap between FX d-slrs and MF solutions, and the Leica offering doesn't have very sophisticated AF (one,central AF bracket) for handling action or runway fashion or other AF-intensive shooting.

I really do not think the rumoured Nikon MX format will appear until this recession is over, and also AFTER Nikon/Canon have both had a chance to see how well Leica does with its S2. Right now, safety in sales volume seems to be where the camera makers are headed, and Sony is pushing FF prices lower and lower. I do not think a 'tweener format d-slr is imminently forthcoming from Canon or Nikon--especially when D40-D60-D90 priced gear makes up 80 percent of sales.

 

[epp_b]

But how many images end up being square?

Fair point, but I guess it depends on your style.

You'd lose a lot of primary image area when taking shots that would end up rectangular (the vast majority) just to end up with a 4% increase in area when you took a perfectly square shot. I can't see many people going for that particular tradeoff.

Perhaps, but more people than would pay $40,000 for a Nikon

If you draw a circle such that the corners of a 36 x 24 rectangle just touch it and then draw a square by extending the vertical sides of the rectange to 36mm you will see that it would require a significantly larger circle to encompass it.

Edit: I'll leave this here in case anyone else doesn't see the problem instantly. (I had to think about it for a minute - I very nearly made the same post as Plato. )

The diagram here explains it very easily.

 

[Josh66]

Not sure about the size. 36mm would probably have pretty strong vignetting in the corners...

If you used 36mm square you would take the corners outside the current field and would probably have to crop back in post processing.

If you put a square format sensor within the current field you would end up with a smaller 'long' dimension and I doubt that many people would see that as adequate payback for not having to turn the camera on its side.

I realized that immediately after typing that. That's why I added the part about not being sure of the size, and a 36mm square having strong vignetting...

30x30 seems OK though. I think I could live with that.



What about round sensors?
It could use 100% of the image circle, and you could just crop it however you want. Or get round paper.

 

[Moglex]

But how many images end up being square?

Fair point, but I guess it depends on your style.

Indeed.

Pick 1000 images at random from the net.

How many are square?

OK, apart from forum avatars, how many are square.

You'd lose a lot of primary image area when taking shots that would end up rectangular (the vast majority) just to end up with a 4% increase in area when you took a perfectly square shot. I can't see many people going for that particular tradeoff.

Perhaps, but more people than would pay $40,000 for a Nikon

$40,000 is a figure that someone pulled out of thin air. I don't think that's realistic. If you scaled the linear dimensions of the sensor by ~10% it would cost nowhere near that. I'm not going to pretend I know how much but I'd be very surprised if it was even double the cost of the cheapest full frame.

If you draw a circle such that the corners of a 36 x 24 rectangle just touch it and then draw a square by extending the vertical sides of the rectangle to 36mm you will see that it would require a significantly larger circle to encompass it.


The diagram here explains it very easily.

If you really cannot form a mental image from the description I gave it might help but it does not actually show a square sensor so you still need to use your imagination.

 

[Josh66]

Pick 1000 images at random from the net.

How many are square?

How many are any ratio other than what the sensor is?

The reason you don't see many square images is because there aren't many square sensors.


edit

Most of the square pictures you'll find are film. Medium format. Square frames.

Most of the 35mm pictures you see are also uncropped...

 

[c.cloudwalker]

But how many images end up being square?

Fair point, but I guess it depends on your style.

I loved the square image from my Hassy. Everything I shot for myself/gallery was shot for square. I also had quite the use of it with bands. Promo shots worked beautifully in a square. Left just enough space on the 8x10 print for band name and contact info. And, of course, perfect for album covers.

But, with everything else, 99% got cropped to a rectangular. Once in a while an image would be kept square on the first page of an article. Usually with text running over part of the image area.

 

[epp_b]

If you really cannot form a mental image from the description I gave it might help but it does not actually show a square sensor so you still need to use your imagination.

No, I fully understand. I'm just saying that it's pretty simple to see that the corners are all barely touching the circle, so making it any taller would expand it beyond the circle.

 

[Moglex]

Pick 1000 images at random from the net.

How many are square?

How many are any ratio other than what the sensor is?

The reason you don't see many square images is because there aren't many square sensors.


edit

Most of the square pictures you'll find are film. Medium format. Square frames.

Most of the 35mm pictures you see are also uncropped...

I'm not quite sure what point you're making here.

Are you saying that people would really prefer to be taking square pictures and only a lack of reasonably priced hardware is preventing them?

I had always assumed that the only reason for a square format was because rotating the camera was impractical with a waist level finder and TLR's and Hasselblads did not have pentaprisms when they were originally brought out.

It's interesting to note that two large format cameras which had (Pentax 6x7) or were designed for (Mamiya 645*) pentaprisms use both moved away from the square format.

Having said that, there was always something very pleasing about getting a shot that 'worked' square when using the Hasselblad.


* The 645 was available with a WLF but would obviously be impractical for portrait mode without the pentaprism option.

 

[Garbz]

Just to throw in my 2 cents, square is not pleasing to look at. Eyes have a tendency to like horizons and not look up or down. When they do look up or down it's to follow a very very specific form (like examine a portrait). I personally think thats why you see so few portrait pictures of landscapes.
Just looking at my house:
4 pictures in the room, 2 are 4x3 the rest very wide panoramas.
screen is 16:10 widescreen.
TVs are now 16x9 (movie cinemas are even wider).

Wide is the norm. Wide is comfortable to look at. I don't see any appeal to a square sensor.

As for the original post, "there is no reason whatsoever why Canon or Nikon could not come up with, say, a 4.5cm x 3cm sensor camera and matching lenses". What on earth were you thinking. There's are 2 very big reasons. Cost and Social Norms.

You can't release a camera sensor and be done with it. First you need a sensor that has to be ground up redesigned, matching electronics. New sensor usually would mean new body housing too (Hassey are BIG), new Lenses. Note the plural? You can't just release a couple of lenses and convince the world you're serious. You'd need an entire series to make people pay attention. Ok then there's the cost of the larger system, larger sensors are expensive so straight away any camera would need to cost more than their full frame brothers too.

Now with skyrocketing costs, how do you convince professional photographers (who are the only ones left who can afford your new gear), that your new untried, untested format is the format of choice compared to the medium format used by well established companies. Especially that is when you're targeting the said medium format photographers with your ludicrously high megapixels yet excellent performing 35mm options anyway. You could run a massive marketing campaign creating more cost!

I see no barrier to Canon entering the medium format market if they actually made a medium format camera compatible with medium format lenses. But create their own entirely new format? The barrier to entry is HUGE because the cost is high and people on the whole simply don't care. (No offence Molex, it's one thing to convince one person that this is a good idea, but it's quite another to convince a 10s of thousands of pros who have been using 35mm since they were old enough to hold them. Which is precisely what they'd need to do to even come close to recovering the R&D cost)

My opinion.

 

[Moglex]

As for the original post, "there is no reason whatsoever why Canon or Nikon could not come up with, say, a 4.5cm x 3cm sensor camera and matching lenses". What on earth were you thinking. There's are 2 very big reasons. Cost and Social Norms.

Cost is obviously a big issue.

'Social norms' are quite irrelevant. If you are living in the film world then yes, releasing a new format is a big deal (as I said, and for reasons given, in my original post). In the digital world it's no big deal. If you don't believe me, look at the first diagramme Here

You can't release a camera sensor and be done with it.

Of course not. It would be an enormous project.

New sensor usually would mean new body housing too (Hassey are BIG)

I know, I have one. But you are making the same mistake as everyone else seems to be making here in assuming you have to jump straight to medium format size.

The example I gave was a 25% (linear) increase but even a 10% linear increase would give a very useful quality improvement.

new Lenses. Note the plural? You can't just release a couple of lenses and convince the world you're serious. You'd need an entire series to make people pay attention. Ok then there's the cost of the larger system, larger sensors are expensive so straight away any camera would need to cost more than their full frame brothers too.

There is no doubt that to do something like this would be a major, multi-million pound project.

But in this day and age it could be a great deal easier than it would have been in the past because every aspect of the design information will be computerised which would save a lot of work.

Now with skyrocketing costs, how do you convince professional photographers (who are the only ones left who can afford your new gear), that your new untried, untested format is the format of choice compared to the medium format used by well established companies. Especially that is when you're targeting the said medium format photographers with your ludicrously high megapixels yet excellent performing 35mm options anyway. You could run a massive marketing campaign creating more cost!

This isn't really an issue because I was never suggesting that (Can/Nik)on start to compete in the MF market.

All the manufacturers brought out 'untried, untested' formats when they introduced crop sensors and they seemed to sell quite well.

With the freedom manufacturers now have to use any format they choose it seems a pity that you need to leap from 35mm FF up to using 6x6 bodies and sensors that do not make use of the costly, bulky and large (but gorgeous) glass that goes with them.

Yes, it would be a costly exercise but Canon and Nikon are prepared to spend what it takes to come up with their flagship products that will sell to only a relatively tiny number of photographers. How many of their longest, fastest lenses do you suppose they sell? Yet they were prepared to design them and tool up to produce them knowing that they would probably only sell in the low hundreds (or even tens) of their most exotic designs.

I see no barrier to Canon entering the medium format market if they actually made a medium format camera compatible with medium format lenses. But create their own entirely new format? The barrier to entry is HUGE because the cost is high and people on the whole simply don't care. (No offence Molex, (Hey, I have connections you know) it's one thing to convince one person that this is a good idea, but it's quite another to convince a 10s of thousands of pros who have been using 35mm since they were old enough to hold them. Which is precisely what they'd need to do to even come close to recovering the R&D cost)

Again, you're almost certainly 100% correct here but it doesn't relate to what I was talking about.

Coming up with a slightly bigger sensor that would only require a slightly bigger camera would be much less of a problem. The main issue I see would be sensor yield. The relationship between sensor size and cost is certainly not linear!

Anyway, this was not an idea that I came up with off the top of my head. It was mentioned in a blog by a professional photographer whom I suspect may have certain information that we are not party to. At least, I hope that's the case.

If anyone remembers the introduction of the film APS system they may recall a great deal of wailing and gnashing of teeth in the photographic press that they did not take the opportunity to increase the negative size a little. At that time there certainly seemed to be plenty of people who would be prepared to buy new glass in order to get the quality boost that a slightly bigger negative would provide.

 

[usayit]

Chasing a market that does not exist, or that is very,very small is not a good strategy. Both Canon and Nikon are making a lot of sales to beginners and to enthusiasts. Leica's S2 system will be a good barometer of just how much demand there is for a larger than FX but smaller than MF digital solution.

Not so sure I agree. Market does exist just not in the eyes of the typical consumer. At least in the fashion district of NYC, there are lots of studios working digital backs on medium format cameras. Hassy's are still high in demand as with third party digital backs. It is the niche market that demands the high prices..... same with Leica and its rangefinders.

Leica's S2 system will not be a good barometer for the market .. at least not in the short term. I surmise the success or failure of this system would be contingent on Leica's ability to bring to the table features that set them apart from those already established.. namely Hasselblad. At least on paper, the S2 brings to the table a medium format digital that is field capable... able to be out in the field with it's 35mm/crop sensored competitors.

At a more general stance, a market that does not exist doesn't necessarily mean a particular company cannot create it and pave the way headstrong... there are loads of examples in technology. It is a huge risk.

 

[Dwig]

One square:

The square format is an artificial construction originally created to allow cameras with waist level finders and non-revolving back to function well. Prior to waist level finders, no cameras shot a square format. Graflex resisted square and introduced a revolving, or more accurately a reversing, back on their SLRs with waist level finders well over 100 years ago (the Mamyia RB67 was a copy of a design over 3/4's of a century old). Square roll film formats were added to the original rectangular formats as afterthoughts to allow for cameras with non-rotating waist level finders. The idea is that you can crop to either a pleasing vertical or horzontal format at you whim.

On a larger format Canon/Nikon:

From a user point of view (not that of an equipment nerd who want cool toys even if they don't have a need for them) Nikon and Canon already make excellent "medium format" DSLRs. Adding a line of cameras using the chip format used in Hasselblads, Phase One backs, and the forth coming Leica S2 would be entering the "large format" arena.

At matching image sizes, today's digital cameras blow film out of the water. There is not question about it. It takes full frame 35mm film to come close to competing with DX (~1/2 frame, or "single-frame", 35mm). Today's so-called full frame DSLRs (~24x36mm sensors) can easily deliver the image quality of yesteryears "medium format" cameras. "Full frame is the new medium format."

The only real need for a larger format is to do the work of yesteryear's large format cameras, 4x5 and up. This has been a rarified market for the last half century and there is no sign that this niche will ever grow. Over the last two years, the number of brands in this market has decreased and several have failed, some being absorbed by others, and announced development of cameras has been discontinued. This is true even when you factor in Leica's forthcoming introduction.

Historically, Canon has been a camera manufacturer who, after a few years began making their own lenses. They've been in 35mm and smaller markets exclusively.

Nikon, on the other hand, is historically a lens maker that added their own camera line to strengthen the market for their lenses when their main customer, Canon, started to make their own. Nikon as often made lenses for other formats, but not cameras. It was rumored in the late '60s that they were developing a medium format SLR line when Bronica, a customer of Nikon's, began making their own lenses. It may or may not be true that they sold their designs to Mamyia who polished them up and released it as the M645 (it does have a very Nikon-like mount design and an extremely Nikon-like meter coupling). Either way, Nikon has stayed away from other format cameras as a rule (their have been a number of scientific models in a range of formats).

I'd be surprised if either Nikon or Canon ventured into "large format" digital, though I wouldn't be surprised if Nikon partnered with another manufacture as the lens supplier for one (one of the sweetest lenses I ever shot with was my Nikkor-W 180mm f/5.6 that I had on my 4x5 field camera).

 

[Garbz]

I know, I have one. But you are making the same mistake as everyone else seems to be making here in assuming you have to jump straight to medium format size.

The example I gave was a 25% (linear) increase but even a 10% linear increase would give a very useful quality improvement

...

If anyone remembers the introduction of the film APS system they may recall a great deal of wailing and gnashing of teeth in the photographic press that they did not take the opportunity to increase the negative size a little. At that time there certainly seemed to be plenty of people who would be prepared to buy new glass in order to get the quality boost that a slightly bigger negative would provide.

Yeah I made that mistake. But then a slight increase in sensor would make even less sense from an "all or nothing" perspective. It would be crazy to have to release a new line of lens for a small increase in sensor size, and would leave a lot of people asking what they are chasing. Consider the three options at the moment. Need high megapixel? Canon's top, Need high ISO performance, Nikon's top, need a mix of both, why not switch straight to MF. I still don't seen any substantial customer base biting on a new format.

I'm not sure what you're saying with APS. Yes as a film it was a failure because it was smaller with no clear benefit (in fact the quality downside). But as a digital it had a very clear advantage, cost. The same clear advantage 35mm has over any larger format. The economics of cutting silicon wafers simply means that any larger format makes a far more expensive camera. That wasn't the case with APS film so the anger was somewhat justified.

 

[Moglex]

Yeah I made that mistake. But then a slight increase in sensor would make even less sense from an "all or nothing" perspective. It would be crazy to have to release a new line of lens for a small increase in sensor size, and would leave a lot of people asking what they are chasing.

Well, you are obviously very confident in your analysis.

Not having access to any market research I wouldn't be confident in making a guess one way or another. I just think it's an interesting idea.

Consider the three options at the moment. Need high megapixel? Canon's top, Need high ISO performance, Nikon's top

It's weird how these things chop and change. Last time I studied the specs it was the other way around.

, need a mix of both, why not switch straight to MF. I still don't seen any substantial customer base biting on a new format.

At the moment there is a massive jump in price between the top 35mm FF system and the MF offerings. Really massive. The jump in quality is nowhere near as pronounced.

I can imagine that there would be a lot of 35mm pro's (and a few amateurs) who would like to move up to MF but cannot afford the premium but who could afford to go to some sort of 'half-way house'. I can similarly imagine a lot of pro's who use MF and have been watching the increases in 35/FF quality and might like a saving in equipment cost/weight/bulk.

Again. though, without market research it's very hard to realistically predict the take up of any such system.

It does seem odd, though, that for anyone who wants 35/FF quality or better that they will be stuch forever more with formats that were set up before the vast majority of current day photographers were even born.

I'm not sure what you're saying with APS. Yes as a film it was a failure because it was smaller with no clear benefit (in fact the quality downside). But as a digital it had a very clear advantage, cost. The same clear advantage 35mm has over any larger format. The economics of cutting silicon wafers simply means that any larger format makes a far more expensive camera. That wasn't the case with APS film so the anger was somewhat justified.

What I'm saying is very simply that there were a lot of people who expressed themselves prepared to pay a premium for a format that was a little better than 35mm.

And if such people were in evidence then, I'm not sure I can see any good reason why there would not be people similarly inclined around now.