Future of DO technology and Canon lenses

My reasoning is based entirely on a pragmatic approach to known problems. It is very expensive to produce fast lenses, and especially if they must be sharp wide open to the edges. If I were a manufacturer, I would be looking for alternatives. FF sensors may also fit that description.

My DOF arguments are at best tentative, and I will leave the artistic meirits to those more in the know.

I love my D10 and current lenses. I am pretty sure that in time I will want more resolution, low light focusing and maybe a bit wider than the 1.6 and 17-40 provide. At the long end, I have enough for now at 400 plus a 1.4X. The middle is empty, so the new 70-300 is something to watch.

Of all the things that would attract my photo dollars, fast lenses rank near the bottom. I am more likely to buy a 300 f4 prime or zoom than spend many times the dollars for f2.8.

If others are like me, the manufacturers will follow.

The question I ask is if the fast lens lovers are really trying to buck this trend by convinicing the rest of us we need them?

Ben
 
No argument at all, it is in the wording. You say many photographers want shallow DOF. I don't disagree, but who can deny that it is very expensive, and that most will forego that luxury.

Ben
Well if you could just have every item in-focus there would be no
inadaquacy. But there are things like DOF, using hyper-focus, poor
resolution at edges etc. It takes more expensive lenses to overcome
these things, or, as in my scenareo, using a tripod, high ISO and
limiting to the sweet part of the lens and stopping down while
keeping a high enough shutter speed.
I think what you are asking for is precisely the opporite of what
many photographers are looking for. One doesn't always want
everything to be in focus. One often wants to isolate a subject or
subjects to show meaning or emphasis. One can always, however,
stop down to increase DOF. The opposite isn't true if one doesn't
have a fast lens. For that reason, fast lenses will always be
prized. There will always be slow zooms as well, for those who
don't need or want to think about DOF.
 
There does seem to be a trend towards faster ISOs and more modest
glass speed.

However the faster ISO argument leads us back towards full frame
sensors. My understanding is that noise is highly correlated with
the distance between each photo receptor site. If they are packed
in too tightly together, there will be lots of crosstalk (for lack
of a better term) which results in noise. The big deal about the
Mars rover's digital camera is that its receptors are placed quite
far from each other, resulting in a low res, but incredibly clean
image.

So if we want a Canon CMOS that is more than 6.3MP but have the
same noise characteristics as the 10D/300D, it'll almost have to be
larger. Cramming more receptors in the same footprint will boost
resolution but significantly boost noise. This is likely the reason
why Canon's higher-end sensors are 1.3x FOV - higher res with low
noise requires the larger size.
I'm friends with a photographer who always seems to be testing new equipment. As such, we've played with quite a number of different cameras. One of the was the Contax N Digital, a 6 megapixel full-frame camera. It had awful high ISO performance, with lots of noise. About a year later, he had a Canon 1Ds for the weekend. Despite having almost twice the pixels in the same physical area, the images were better in every respect, including noise. Another example is the Canon D30 and 10D. Same physical sensor size, but the 10D has twice as many pixels AND better noise performance. At a given level of technology, there is a direct relationship between the size of the pixels and the amount of the noise. However, sensor technology is not standing still. Since the cost of making full frame sensors will always be much, much higher than APS sensors, the smaller sensor is here to stay for the foreseeable future.
 
The question I ask is if the fast lens lovers are really trying to
buck this trend by convinicing the rest of us we need them?

Ben
If you dont need it dont buy it but anyone who thinks that faster ISOs is a substitute for a good fast prime is a fool and still has a lot to learn when it comes to photography. There are plenty of conditions that not only require 1600 ISO but f 2.0 due to poor lighting conditions, and shallow DOF last I heard is still used artistically in portraits etc..
 
Slow lenses are inexpensive to develop and produce and I am sure they will always be available to those who want them. “Features” like slower maximum aperture and FOV crop add limits to what you can do with your equipment. As others have pointed out that if you are unhappy with the corners of a shot you can always crop them out but you can’t add more scene to an image taken with a cropped FOV sensor without going to a lot of trouble. The smaller, lighter lenses that a cropped sensor system can produce only occur in short FL lenses. For example the 4/3 300mm f/2.8 is about the same size as a regular 35mm 300mm f/2.8 lens. It is the long lenses that are really big and heavy so the cropped sensor really doesn’t buy you much. Most lens manufacturers already produce fast lenses with sharp corners. As far as I have been able to determine, most people who are serious about photography desire fast lenses and use shallow DOF as a creative tool, not a means to cover up lens deficiencies. I do agree with you that I pick the 300mm f/4 over the f/2.8 but not because of the price. I usually find that the speed gained by the extra stop on that particular lens is outweighed by the longer minimum focusing distance and the huge size and weight increase. I think that fast lenses are desired by many precisely because of the shallow DOF that they are able to produce even more than the faster shutter speeds that they can yield.

Greg

--

 
Yes, there is no doubt that technology will continue to advance, but think of it this way. The newer technology will be even better when used with larger sensors than it will with the smaller ones. Also the high cost of producing FF sensors will also come down with newer technology. There are a lot of reasons to keep FF other than just the noise aspect. Off hand I can think of over 50 of them and they have EF written on each one of them.

Greg

--

 
Yes, there is no doubt that technology will continue to advance,
but think of it this way. The newer technology will be even better
when used with larger sensors than it will with the smaller ones.
Also the high cost of producing FF sensors will also come down with
newer technology. There are a lot of reasons to keep FF other than
just the noise aspect. Off hand I can think of over 50 of them and
they have EF written on each one of them.
It's a discussion that's been done to death, but the problem with full frame is that it's just a huge honkin' piece of silicon. While costs for a given speed/capacity of computer chip come down, it's always through miniaturization. A given area of silicon chip has actually remained pretty constant over the years. Since you can't make a FF sensor any smaller (because then it wouldn't be FF), the cost of production is always going to be high and the yields are always going to be low. And ultrawide glass doesn't get along well with FF sensors anyway (vignetting due to the smaller acceptance angles of CCD/CMOS sensors vs. film). Yes, it is true that FF sensors will always have more pixels or better noise performance at the same number of pixels compared to APS sensors. However, they will also always command a significant premium. If you can afford it, it's a great choice.
 
You don't need fast lenses for the kind of photogrpahy you do.

I need fast lenses for the kind of photography that I enjoy doing where slow lenses are impractical. I'm struggling to get 1/60th at f/2 with ISO at 1600 at concerts. 1/250, f/2, 1600 only freezes action about half the time in indoor sports.

I'd like to see camera sensors with ISO range of 800-6400 and f/1.2 telephoto lenses.
 
I really think technology will overcome the current limitation of
photo-sensor site size in regards to noise.
The speed of technological advancements is certainly stunning. But every technology eventually reaches its limit - hard drives have been expanding in density at a shocking rate over the last couple years, but the entire industry agrees that they are nearing the limit. New approaches or new form factors will be required to make the next major leap forward.

Same for photo sensors. Whether it be CMOS, CCD, Foveon CMOS, or the next great thing, there will still be certain inherent limitations because of the laws of physics. If the wavelength of visible light can be as large as X, you are stuck with a physical limitation. The smallest possible sensor site will always be expressed as a factor of the largest wavelength of visible light.

The quality of the sensors and processors will continue to improve, but nothing will change the physical properties of light.
 
Yes, there is no doubt that technology will continue to advance,
but think of it this way. The newer technology will be even better
when used with larger sensors than it will with the smaller ones.
Also the high cost of producing FF sensors will also come down with
newer technology. There are a lot of reasons to keep FF other than
just the noise aspect. Off hand I can think of over 50 of them and
they have EF written on each one of them.
It's a discussion that's been done to death, but the problem with
full frame is that it's just a huge honkin' piece of silicon.
While costs for a given speed/capacity of computer chip come down,
it's always through miniaturization. A given area of silicon chip
has actually remained pretty constant over the years. Since you
can't make a FF sensor any smaller (because then it wouldn't be
FF), the cost of production is always going to be high and the
yields are always going to be low. And ultrawide glass doesn't
get along well with FF sensors anyway (vignetting due to the
smaller acceptance angles of CCD/CMOS sensors vs. film). Yes, it
is true that FF sensors will always have more pixels or better
noise performance at the same number of pixels compared to APS
sensors. However, they will also always command a significant
premium. If you can afford it, it's a great choice.
--

 
Yes, it
is true that FF sensors will always have more pixels or better
noise performance at the same number of pixels compared to APS
sensors. However, they will also always command a significant
premium. If you can afford it, it's a great choice.
I'm a huge believer in free markets and I think you're totally right. We'd be fools to ignore the economic realities of such things. If the market demands and supports premium FF sensors, then Canon and the rest will find a way to supply them. If the market as a whole deems FF sensors too expensive and unnecessary, then they may go the way of the dinosaur.

So the old saying is true, "vote with your wallets!"
 
You don't need fast lenses for the kind of photogrpahy you do.

I need fast lenses for the kind of photography that I enjoy doing
where slow lenses are impractical. I'm struggling to get 1/60th at
f/2 with ISO at 1600 at concerts. 1/250, f/2, 1600 only freezes
action about half the time in indoor sports.

I'd like to see camera sensors with ISO range of 800-6400 and f/1.2
telephoto lenses.
My dream would be ISO6400 with the noise of today's 1600, sensor-level anti-shake for another stop or two, and an 85mm f/1.4. Oh yeah, and a AF-assist light. Then I could pretend there was no such thing as a flash!
 
So what you are saying is that technology for reducing noise and increasing signal will get better but the technology for producing large sensors is now at the best it will ever get. I don’t think we will be waiting very long to see you proven wrong. Already Kodak is putting FF sensors in an almost affordable body and it looks like Canon is starting to come down on the price of the 1Ds.

Greg

--

 
I almost always want the scene to look like what my eyes see. That is, > everywhere I look is in focus, this is what I will work on.
That's not how my eyes work. If I focus on my screen, the plan behind it is out of focus. So, fast camera lenses behave more like my eyes than slow ones.
 
In my 35mm days, I shot Kodachrome 25 slide film usually at F32. I was always unhappy because I had no control over the development, and never wanted to do the dark room stuff myself. I always thought that when I could afford it, I would get a view camera and maybe build a dark room, ug!!. Fast lenses were never on my wish list in the last 22 years.

Now we have digital. I have total control from shot to print. The quality is far better in what I can produce now than what I was able to produce with film. That is in spite of the fact that film is probably still better, but only at a price in my personal time/lifestyle and money that I am unwilling to make.

You can say that most people want to take pictures in dark auditoriums. If you are right, then people will pay several thousand dollars for fast zoom lenses. Not me, I will pass on that sort of photography as being out of my reach.

If I am right, and serious amateurs will spend the time to wait for good light, or use a tripod or whatever it takes in lieu of spending a fortune to get from f4 to f2.8, then that is what the manufacturers will make in abundance.

I am all for FF, bring it on. I might pay $4000 for an 11 MP ff camera someday, but probably never $8000. I might settle for 1.3 crop and 8 mp if it is cheaper and allows me to use a cheaper lens. I am not unhappy with 6MP and 1.6 crop, yet.

If price were no object, I would buy the best. In fact, I buy the best I can afford just like anyone else does.

The argument here is not what is absolute best, but what the general consumer for these high end cameras are looking for, and what trade-offs they are willing to make. Some will buy fast because they read fast is better on these forums. Others will have a genuine need for the special situations that absolutely require it. IMO, no part of photography is more exorbitantly expensive than fast lenses.

All of you could be right, I said my position was speculative, all I am doing here is making sure you understand the argument, not if the argument is correct or not. The technical/artistic reasons for FF, or fast lenses don’t address my argument, or maybe my argument is still not clear.

Ben
 
I almost always want the scene to look like what my eyes see. That is, > everywhere I look is in focus, this is what I will work on.
That's not how my eyes work. If I focus on my screen, the plan
behind it is out of focus. So, fast camera lenses behave more like
my eyes than slow ones.
Well, it depends on the light too. In bright sunlight, your pupils stop down and depth of field increases. At night, your pupils open wide to capture as much light as possible depth of field becomes rather small.
 
So what you are saying is that technology for reducing noise and
increasing signal will get better but the technology for producing
large sensors is now at the best it will ever get. I don’t
think we will be waiting very long to see you proven wrong.
Already Kodak is putting FF sensors in an almost affordable body
and it looks like Canon is starting to come down on the price of
the 1Ds.
Oh no, I'm not saying that at all. Full frame sensors will continue to improve and they also should come down in price over time. While the improvement in quality will happen relatively quickly, in all likelyhood, the price decrease will be much slower. And, at the same time, APS sensors will always be much, much cheaper. As for the Kodak, they have their sensor in a body based on a $300 film SLR and it still costs big money. That should give you an indication about the cost of the sensor relative to the rest of the camera. And it still doesn't work very well with a lot of existing lenses. I have a friend who has a D100 and a 14n. Guess which one he uses most of the time? There will be a $2,000 full frame camera sometime in the next five years. About the same time the APS cameras hit $500.
 
My eyes also have shallow DOF, but they refocus so fast I never notice it. I am constantly aware of the entire scene. My driving is just a bit dangerous because I cannot avoid seeing everything, and not just the road ahead.

The only thing I bring to photography is this total emersion in the scene, and a desire to capture it. No doubt being able to focus on specifics is just as artistically valuable. I am not asserting that everyone sees the world the way I do, or ought to.

I have made an argument in favor of trading the extreme cost of fast lenses for other valuable factors. If the camera makers can produce equipment that allow duffers like me to get sharp images of what they want to capture, at an affordable price, then that is what they will produce.

If most people want fast, they will figure it out. If most people can live without it, or will never pay the steep price, it will always be a luxury.

Ben
 
That on the 14n and 1Ds that the actual cost of production has VERY little to do with the cost that people end up paying. This is especially the case with the 1Ds since it is really the only game in town. People have an amazing way of figuring out how to do what “can’t be done.” My prediction is FF prices will fall more quickly than you think.

Greg

--

 
I have made an argument in favor of trading the extreme cost of
fast lenses for other valuable factors. If the camera makers can
produce equipment that allow duffers like me to get sharp images of
what they want to capture, at an affordable price, then that is
what they will produce.
There already are lots of lenses that are cheap and light and sharp when stopped down. So what is it exactly that you are asking for that isn't already on offer?
 

Keyboard shortcuts

Back
Top