why should folks go fullframe?

Yes, I read that when you first posted it back in December... "the 5D produces a half-stop better noise performance and significantly better detail." That's why I wasn't following your above post because it seemed inconsistant with your previous findings.
I'm not quite sure I'm following your statement. I thought the 5D
has 1-2 stop noise advantage over the crop cameras. If you have
the same FOV on both cameras, why would noise be the same as well?
And if you have the same FOV, wouldn't the camera with more mp
provide more detail?
Read this:
http://forums.dpreview.com/forums/read.asp?forum=1032&message=16107908

--
Lee Jay
(see profile for equipment)
--
Jeremy
 
Yes, I read that when you first posted it back in December... "the
5D produces a half-stop better noise performance and significantly
better detail." That's why I wasn't following your above post
because it seemed inconsistant with your previous findings.
Yeah it is.

What I said was, "20D+200/2.8 = 5D+320/4.5 (to first order). Neither will produce better noise performance or more detail than the other at the same shutter speed."

This was the second part of my original test - trading the extra detail for lower noise. These two are 1 1/3 stops apart - the exact amount that the 5D beats the 20D by when detail is held constant.

--
Lee Jay
(see profile for equipment)
 
I don't think diffraction is the only problem stopping down to achieve deeper DOF. Low-light situations anyone?
 
  • sensor resolution outresolves the current optics we have
Well good lenses, like the 50 mm f/1.4, the 100 mm f/2.8 Macro, the 135 mm f/2 L resolve pretty amazing detail. They have to go a long way to fighting noise, tonal range, sensor blooming, and a host of other digital nasties before they can make pixels small enough to outresolve a good lens.

In any case entry-level cameras outresolve a 5x7 inch print, so I wouldn't worry about it.
  • sensor size gets even bigger (like a medium format for digital)
in order for technology to be able to provide additional resolution
and higher sensivities
Again, I don't think this is very realistic. There's a reason 35 mm has been the dominant format for such a long time. It provides very, very good quality, definately good enough for most uses. Medium format has always been an option when 35 mm isn't good enough, but most people didn't take it. My 5D is unquestionably better than 35 mm films, except maybe the slowest B/W. In other words what we have now is better than has ever been required for most uses; nobody is going to be "forced into" larger formats.
I think about this everytime i go and buy a new lens... but also
think that the day before the world knows there's a big change and
our lenses will be unuseable in a couple months, we as photography
passionates will know that much ahead, enough to get all our gear
sold on eBay or anywhere else... right?
Nah. Canon has sold 30,000,000 lenses. That was the news a month or two ago. Those lenses are what keep most of us married to the Canon system. Those lenses are what keep people who own them from checking out Sony's in-body IS or Nikon's ergonomics. Canon has more incentive than anybody to stick to the 35 mm format.
 
I don't think diffraction is the only problem stopping down to
achieve deeper DOF. Low-light situations anyone?
You only get the full-frame low-light advantage when you can allow DOF to get shallower.

--
Lee Jay
(see profile for equipment)
 
it would not..because you both discard the pixel density in your
comparison and only take into account the crop factor.
No, I account for that as well. The extra detail the 5D captures
is not as much as it would if it had the same pixel density, and
it's noise performance is better at the pixel level precisely
because it has lower pixel density.

20D+200/2.8 = 5D+320/4.5 (to first order). Neither will produce
better noise performance or more detail than the other at the same
shutter speed.
320mm..that'S the focal length alone..now you have the higher res of the 5d vs the higher pixel density of the Xt..

the only way to prove it is to do a test..other wise it is pointless speculation.
--
Lee Jay
(see profile for equipment)
--



http://www.pbase.com/zylen
 
Now that you have amazing quality EFS lenses like the 17-55f2.8IS, 10-22, 60 Macro.

Sure the large FF sensors might be superior, but the limiting factor to image quality now becomes the older EF lenses. Only the best EF 'L' lenses will do. You will run into more vignetting and soft borders even with good EF lenses.

At present, the crop DSLRs use the best part of the EF lens, many inexpensive EF lens are excellent on crop cameras.

If you are big into telephoto lenses, you can forget FF. For example the excellent 70-200f2.8IS becomes a FF equivalent of a 112-320f2.8IS super lens on a crop camera. Just try to get something like that in FF. ;-)

Furthermore, there will not likely be higher MP FF sensors. Again due insufficient quality in the EF lenses. To make lenses of higher quality for FF would be size and price prohibitive. Imagine the size and price of a FF equivalent of a 17-55f2.8IS, ie. a 27-88f2.8IS
I see a lot of folks say they want to go full frame and prefer that
than the current crop cameras. Would this be true for folks who are
more into wildlife photography and the crop actually increases the
reach of your telephoto lenses?

If one had truely professional crop cameras, would folks still go
full frame? If so, why?

I am considering investing in EF-S lenses at this time (10-22 and
17-55) and my friends have been advising me against it saying i
should be L lenses which can fit on FF cameas as well...

Please advise.

Thanks
--
[email protected]
 
Yes, I read that when you first posted it back in December... "the
5D produces a half-stop better noise performance and significantly
better detail." That's why I wasn't following your above post
because it seemed inconsistant with your previous findings.
Yeah it is.

What I said was, "20D+200/2.8 = 5D+320/4.5 (to first order).
Neither will produce better noise performance or more detail than
the other at the same shutter speed."

This was the second part of my original test - trading the extra
detail for lower noise. These two are 1 1/3 stops apart - the
exact amount that the 5D beats the 20D by when detail is held
constant.
but then you put the 300mm lens on the 20D and the 5D is beated again.

bottom line..the 5d is not the right camera for wildlife. to equal focal length..it will not get you any more detail but less detail than a 20d with the same lens.
--
Lee Jay
(see profile for equipment)
--



http://www.pbase.com/zylen
 
I would not see ANY advantage to use a 5D for wildlife.. not a
I have to agree with Digic. When I moved from a D60 to a 5D, I saw an immediate improvement with my wildlife images. Even the ones I had to crop heavily. Better focus, better ISO sensativity, and a severe crop, beats any misfocused shot.

As you would imagine, this guy wasn't letting me get very close, so I had to make up with that after the fact, probably crop away 2/3 of the pixels:



I'm surprised no one has brought up the fact that sometimes 1.4x or 1.5x is handier than 1.6x. Not usually, but sometimes.
 
the only way to prove it is to do a test..other wise it is
pointless speculation.
I did.
what you did is to shoot the same scene with the same FRAMING..not at the same focal length.

that's a different story.

can you do the same test now with the same lens from the same distance to subject..same focal, not equivalent.
--



http://www.pbase.com/zylen
 
I would not see ANY advantage to use a 5D for wildlife.. not a
I have to agree with Digic. When I moved from a D60 to a 5D, I saw
an immediate improvement with my wildlife images.
you went from 6mp old sensor to 12mp..that'S double. if you go from 8mp 1.6x crop factor to the 5d you won't notice the same thing.

I saw the difference in my photos recently..comparing them to shots from a 5d taken at the same focal than I was.

Even the ones I
had to crop heavily. Better focus, better ISO sensativity, and a
severe crop, beats any misfocused shot.
then maybe the focusing was your problem..

--



http://www.pbase.com/zylen
 
So if canon makes a full frame sensor that has the same pixel
density as the crop cameras have the 1.6 crop cameras wont have any
more reach that the full frame with the same pixel density.
and what are the odds of that hapening? :)
Very good odds as the rumored 1DSMK2 camera is supposed to be exactly that !!
you would still end up with a load of wasted pixels if you are into
wildlife and birds. but a higher pixel density would surely be
better for wildlife.

I seriously doubt that Canon will do that..
again the rumored 1DSMKI2 replacement
Right now the pixel density is the highest in the 30D/XT/20D
cameras for now. The next highest is the 1DSMK2 then the 5D/1DMK2
at the same pixel density.

Dont forget difraction sets in with higher pixel densities. Thom
Hogan over on the nikon forum has discussed this as how the D2X has
this effect because of the high pixel density of that camera. So to
high and it starts to have defraction problems and theres no cure
for that.
sure but even with the XT the diffraction is not a problem until
F16. not too bad.
Diffraction isnt a real problem NOW BUT if you want higher resolution it WILL be in a 1.6 crop camera. I was referring to diffraction problems not in current canon pixel density but in the much higher pixel density of D2X.
that kind of make me think that 8mp is pretty much the limit..
maybe 10mp if they push this to the limit.
Agree
So in conclusion full frame is if you want the highest possable
resolution with low noise and less of a problem with diffraction.
again highest possible resolution depend on the subject. if you
take a bird with both cameras with a 400mm lens, you will get more
resolution with the 1.6x crop factor camera than with the 5D.

Yes agree because the pixel density is much higher in 30D than 5D but again it has nothing to do with it being a crop sensor its the higher pixel density.
Other side benifits are better depth of feild control, bigger
brighter veiwfinder.
advantage and inconvenient for the DOF...that is also depending of
the effect you want. if you are in low light and need to open wide
on a subject at close range...you might consider the narrow DOF a
little bit of a problem.
nonsense
Crop sensors will stay because of lower costs. So you can always
sell an EFS lens to somone like that.
--
http://www.pbase.com/dc9mm

--



http://www.pbase.com/zylen
--
http://www.pbase.com/dc9mm

 
the only time the 5D would be better is if you can get closer with
the 5D and NOT with the XT. what are the odds of that hapening?
Uh, so you're saying there's no benefit at all to 13 versus 8 million pixels, to better ISO performance, to using wider lenses, or primes as they were intended ... to being able to use a 135/2 L instead of an 85/1.8 to shoot portraits?

Just from reading this thread, it seems like you have something against the 5D, as if it killed your dog or something.

This was 19 mm through the 5D. I would have been able to stand in the same position with a Rebel XT. Please explain to me why the Rebel would have done a better or at least equal job; what you said above?

 
the only time the 5D would be better is if you can get closer with
the 5D and NOT with the XT. what are the odds of that hapening?
you're out of context..you jump into a conversation without knowing all the facts..I was refering to wildlife.

you cannot get closer most of the time.
Uh, so you're saying there's no benefit at all to 13 versus 8
million pixels, to better ISO performance, to using wider lenses,
or primes as they were intended ... to being able to use a 135/2 L
instead of an 85/1.8 to shoot portraits?

Just from reading this thread, it seems like you have something
against the 5D, as if it killed your dog or something.

This was 19 mm through the 5D. I would have been able to stand in
the same position with a Rebel XT. Please explain to me why the
Rebel would have done a better or at least equal job; what you
said above?

--



http://www.pbase.com/zylen
 
So if canon makes a full frame sensor that has the same pixel
density as the crop cameras have the 1.6 crop cameras wont have any
more reach that the full frame with the same pixel density.
and what are the odds of that hapening? :)
Very good odds as the rumored 1DSMK2 camera is supposed to be
exactly that !!
but then if they put as many pixels as there are in a 1.6x crop camera..you will lose the advantages that you are so liking.

we've already seen that Canon is not about to do that..the 30D came out with the 8mp instead of the anticipated 10.

what are the odds that Canon will put the SAME pixel density on the 5d or 1DII than it is on the 20D?

sorry but I don,t think it will ever happen.

--



http://www.pbase.com/zylen
 
400mm lnes but 400mm was too long..
Bingo. There you go. When the lens is "too long" all benefits of a cropped sensor vanish. And here we have proof that this does in fact happen.

Here are two photos from my 5D at ISO 3200, with a 300 mm f/4 IS shot wide open in both cases:





Somebody is going to suggest that these don't count because they've been downsampled, but that's actually the point. When you're not focal length limited ( ie when "400mm was too long" ) you're working with 13 m instead of 6 m to 8 m pixels. The noise in any one of them is much less visible when you make an 8x10 from both cameras. If you have the same per-pixel noise but twice as many pixels, the end result is a much cleaner image.
 
the only time the 5D would be better is if you can get closer with
the 5D and NOT with the XT. what are the odds of that hapening?
you're out of context..you jump into a conversation without knowing
all the facts..I was refering to wildlife.
No, actually, I'm not. The thread is called "why should folks go fullframe?" and your statement was just as clear. The only time the 5D is ever better than the Rebel is when you're not standing in the same place with both of them.

Since I was working with a tripod, I could have done exactly that. Please, either back up your statement, or think next time before you say things that just aren't true.

Based on your post "if if if if" I'm sure you would agree that inventing unusual situations that one camera is superior in, and then pretending that applies to all of photography in general, is deceptive.
 
BTW..the 1Ds II does not have the same pixel density as a 1.6x crop factor..not sure where you got that.

rumored? that camera exsit..it is not a rumor. again not sure what you mean here.
you would still end up with a load of wasted pixels if you are into
wildlife and birds. but a higher pixel density would surely be
better for wildlife.

I seriously doubt that Canon will do that..
again the rumored 1DSMKI2 replacement
Right now the pixel density is the highest in the 30D/XT/20D
cameras for now. The next highest is the 1DSMK2 then the 5D/1DMK2
at the same pixel density.

Dont forget difraction sets in with higher pixel densities. Thom
Hogan over on the nikon forum has discussed this as how the D2X has
this effect because of the high pixel density of that camera. So to
high and it starts to have defraction problems and theres no cure
for that.
sure but even with the XT the diffraction is not a problem until
F16. not too bad.
Diffraction isnt a real problem NOW BUT if you want higher
resolution it WILL be in a 1.6 crop camera. I was referring to
diffraction problems not in current canon pixel density but in the
much higher pixel density of D2X.
that kind of make me think that 8mp is pretty much the limit..
maybe 10mp if they push this to the limit.
Agree
So in conclusion full frame is if you want the highest possable
resolution with low noise and less of a problem with diffraction.
again highest possible resolution depend on the subject. if you
take a bird with both cameras with a 400mm lens, you will get more
resolution with the 1.6x crop factor camera than with the 5D.

Yes agree because the pixel density is much higher in 30D than 5D but again it has nothing to do with it being a crop sensor its the higher pixel density.
Other side benifits are better depth of feild control, bigger
brighter veiwfinder.
advantage and inconvenient for the DOF...that is also depending of
the effect you want. if you are in low light and need to open wide
on a subject at close range...you might consider the narrow DOF a
little bit of a problem.
nonsense
Crop sensors will stay because of lower costs. So you can always
sell an EFS lens to somone like that.
--
http://www.pbase.com/dc9mm

--



http://www.pbase.com/zylen
--
http://www.pbase.com/dc9mm

--



http://www.pbase.com/zylen
 
400mm lnes but 400mm was too long..
again you're out of context..you and don't know all the facts. I was refering to the 400mm F5.6.. it was too long because it did not have a zoom possibility.

but we were both 200mm range..my photos ended up much more detailed than his. that's it. he was using a 5d, I was using a XT.

we could not frame the subject completely most of the time..

BUT....BUT...when he could frame the subject..I could get nice close up :)

either way I won. you see..at the end of the photo shoot he was asking the people if he could get closer to the lion to get a face close up..of course it was over and he could not. but I did get the face close up with my XT and the 200mm lens.
Bingo. There you go. When the lens is "too long" all benefits of
a cropped sensor vanish. And here we have proof that this does in
fact happen.
but the bingo does not happen often with wildlife ..that's the problem and that is why most wildlife photographers prefer some crop factor.

it does not "vanis" as you say..because with the same focal length..I will ALWAYS get more detail than you from the same distance to subject. either I will get a close up shot..or a full body shot..either way I am getting more detail.
Here are two photos from my 5D at ISO 3200, with a 300 mm f/4 IS
shot wide open in both cases:
so you're showing me tiny resize web images as a proof? you can't be serious :)))
Somebody is going to suggest that these don't count because they've
been downsampled, but that's actually the point.
what point? that at web size you can't see the difference between a 5d shot and a XT shot? I am glad for you that the web is your ultimate goal..it is not mine :)

When you're not
focal length limited ( ie when "400mm was too long" ) you're
working with 13 m instead of 6 m to 8 m pixels.
when you are not..but when you are into wildlife..and not pet dog..you are focal length limited.

The noise in any
one of them is much less visible when you make an 8x10 from both
cameras. If you have the same per-pixel noise but twice as many
pixels, the end result is a much cleaner image.
I don't care about the noise.. I rarely have to use ISO 800 anyway.

--



http://www.pbase.com/zylen
 

Keyboard shortcuts

Back
Top