Full Frame Fallacy?

[Guy Parsons]

Hi Joe, the best way to understand SLR shutters is to go buy/look at an old second-hand 35mm film SLR, the older the better. The type you see is called a "focal plane shutter" because it lies just in front of the film as close to the focal plane as possible.

When you play with it in the shop at slower shutter speeds and then faster shutter speeds (use manual settings) with the film back open, you can easily see what happens.

Older SLR cameras had cloth type shutters that rolled up each side of the frame. Press the shutter and the first curtain would zip open across the frame and fully expose the frame, then the second curtain would follow at a set distance apart depending on the shutter speed selected.

Now the curtains always travel across the frame at the same constant speed, and the shutter exposure time is determined by the spacing between the shutters so that any one part of the frame does get a constant exposure, even though the right hand side of the frame is exposed at a slightly different time to the left hand side of the frame.

That different exposure time explains the very old photos of racing cars that appeared to be leaning forwards, they held the camera (without panning) so that the curtain was travelling in such a direction that the bottom of the image was exposed first and the top last, that way the top of the car is ahead (later) than the bottom of the car.

In older 35mm SLRs the exposing slot may travel across in something like 1/60 second and all that is too slow for the modern world. The exposing slot may be only a fraction of an inch wide at high shutter "speeds".

To speed things up the curtains were turned into slim metal blades that now do an equivalent travelling slot exposure of the frame but move from top to bottom, or is it bottom to top? Shorter distance to travel, can do it faster, especially if the blades are much lighter than the old heavy curtain material.

Some SLR cameras like an ancient KowaSix 6x6cm camera that I once owned didn't use a curtain type shutter at all, they went for Copal type shutters inside the interchangeable lenses. That yields lighter and faster shutter operation and higher flash synchronisation speeds.

Of course I should have Googled in the first place, just now Googled for "how does slr shutter work" and got this at least.....
http://www.wrotniak.net/photo/tech/fp-shutter.html says it better than I do.

Regards............. Guy

 

[LucaPCP]

I suspect that many SLR camera makers are going FF to differentiate a more expensive market. Now that medium format is no more, FF will be the new medium format.

The basic fact is that sensors are becoming better and better.

Compact and superzoom cameras are biting at the heels of the SLRs, so that, in order to extract more profit, many makers want to differentiate and go FF.

They are simply developing a new market segment.

Luca

 

[Agramer]

You compare systems on DOF. That is ok if the DOF is crutial to you,
but you can't tell that fast lens is slow lens because sensor will
give you DOF like bigger sensor with slower lens. . .

No, I can say it's slower because smaller sensors have poorer
high-ISO performance. That's one reason we all own dSLRs instead of
compacts with 1/2.5" sensors, isn't it?

An f4 lens on full-frame will perform about the same as an f2 lens on
4/3 (2x-crop), both in terms of DOF and noise at the same shutter
speed.

--
Lee Jay
(see profile for equipment)

That isn't quite right.

If someone use 2,8 on 4/3 with 1/125 exp and ISO 800 he/she must be prepared for more noice and bigger DoF than with 2,8 on 24x36 with 1/125 and ISO 800. Nothing would be slower. He/she will gett more noice and bigger DoF, thats all. Many people use ISO 100 or 200 99% of the time (like me) and that is irrelevant. I use ISO 100 - 200 and I can't see the difference. I can see the difference in DoF and if I must choose ...well I did for macro work bigger DoF is present from heaven

People saying that 4/3 is bad because of noice and speed...but if you say that someone will get more noice than speed will be the same.

If you look "right" (my right may be different than yours E3 will be blind before 5D (that will be bad for all photographers, but not crutial for many because many don't work in that kind of light) and it will give bigger DoF with the same aperture than 5D. For some people that is good and fore some that is bad.

I didn't see any difference with 5D (I don't need ISO 800 for 99,9% of my work). I do understand that some people do, but I don't know many of them.
Cheers

 

[Louis_Dobson]

but don't actually state, that there are many factors pushing in different directions, and what suits depends on what you do, and want to do.

I'm currently amusing myself with people shooting with longish lenses. 35mmFF is better at this: it gathers more light and allows more control of the background. But my mainstay is seascapes. 4/3rds is better at this, I can get where I need to shoot without terrible inconvenience. A lot of the time I go wandering out with nothing in mind and see what happens. 4/3rds is better here too: I can carry a wider range of kit before it hurts. But often I go down to known haunts at dawn and dusk. There, 35mmFF is better: I know roughly what I will find, so I can leave kit at home, but not exactly how I will find it, so the extra exposure latitude is hand.

Horses for courses.

--
http://www.flickr.com/photos/acam

 

[Raist3d]

You are shooting at 100mm. I am shooting at 400mm equivalence. That's 4x the focal length equivalence and blur-subjecting. So your comparison is completely invalid.

It's stuff like this that quite frankly gets old. I understand exploring potential of other cameras and other options but not when the things posted are clearly so one sided lacking objectivity. You don't even own the E-3 and you talk about what it can and can't do. Moreover, you pretty much changed the subject anyway. Your claim was that it couldn't, and that if it did you would be running to that camera. So what happened with that? Changed your mind?

Finally the quality at ISO 400 on the E-3 will be superior to the 5D at ISO 3200.

Let me repeat that again: the quality of capture of the E-3 at ISO 400 will be superior to the 5D at ISO 3200. I can even see the noise at the resize you did (I don't particularly mind it, I am not that sensitive to that like some others if it looks like grain- but it's worth pointing out because you are making the comparison).

--
Raist3d (Photog. Student & Tools/Systems/Gui Games Developer)
Andreas Feininger (1906-1999) 'Photographers — idiots, of which there are
so many — say, “Oh, if only I had a Nikon or a Leica, I could make great
photographs.” That’s the dumbest thing I ever heard in my life. It’s
nothing but a matter of seeing, and thinking, and interest. That’s what
makes a good photograph.'

 

[Raist3d]

I could easily shoot at ISO 1600 with all the IS advantage fully in place and still get a great shot.

I hate to say this but your whole reply was "let me change the subject when I didn't like the facts." We can play the ISO and aperture game, both. The issue was whether the E-3 could do such a high effective IS and I proved that yes, indeed it can.

If I crank up the ISO the effective IS over the 5D becomes even more obvious if you are going to go "high gain ISO = IS" route - and again, this is not the same point you were arguing about.

--
Raist3d (Photog. Student & Tools/Systems/Gui Games Developer)
Andreas Feininger (1906-1999) 'Photographers — idiots, of which there are
so many — say, “Oh, if only I had a Nikon or a Leica, I could make great
photographs.” That’s the dumbest thing I ever heard in my life. It’s
nothing but a matter of seeing, and thinking, and interest. That’s what
makes a good photograph.'

 

[Lee Jay]

That isn't quite right.
If someone use 2,8 on 4/3 with 1/125 exp and ISO 800 he/she must be
prepared for more noice and bigger DoF than with 2,8 on 24x36 with
1/125 and ISO 800. Nothing would be slower.

You misunderstand. What you said above is right. To get the same image, the larger sensor would need ISO 3200, 1/125th and f5.6. Thus, the 4/3 f2.8 lens behaves like an f5.6 lens on full-frame, or conversely an f2.8 lens on full-frame can be matched by an f1.4 lens on 4/3.

He/she will gett more
noice and bigger DoF, thats all. Many people use ISO 100 or 200 99%
of the time (like me) and that is irrelevant. I use ISO 100 - 200 and
I can't see the difference. I can see the difference in DoF and if I
must choose ...well I did for macro work bigger DoF is present
from heaven

If you don't need the speed of a larger system, you shouldn't pay for it. I don't need it at work and that's why I have an APS-c system and f5.6 zooms. Incidentally, that system doesn't give you more DOF, it gives you the same DOF at the same noise and same shutter speed. I can always get just as much DOF and detail as you can, and that's because both systems can get to diffraction-limited depths-of-field.

People saying that 4/3 is bad because of noice and speed...but if you
say that someone will get more noice than speed will be the same.

Sure. Even my compact is as fast as my 5D if I can accept more noise.

If you look "right" (my right may be different than yours E3 will
be blind before 5D (that will be bad for all photographers, but not
crutial for many because many don't work in that kind of light) and
it will give bigger DoF with the same aperture than 5D.

Bigger DOF with the same f-stop. f-stop = focal length / aperture. With the same aperture (not f-stop) and AOV, all formats have the same DOF.

For some
people that is good and fore some that is bad.
I didn't see any difference with 5D (I don't need ISO 800 for 99,9%
of my work). I do understand that some people do, but I don't know
many of them.

Like I said, if you don't need speed, you shouldn't have to pay for it. This is where the smaller-sensor systems do well. The larger sensor systems afford you the opportunity to get more speed if you need it. I need it at home, I don't need it at work.

--
Lee Jay
(see profile for equipment)

 

[Agramer]

You compare systems on DOF. That is ok if the DOF is crutial to you,
but you can't tell that fast lens is slow lens because sensor will
give you DOF like bigger sensor with slower lens. . .

No, I can say it's slower because smaller sensors have poorer
high-ISO performance. That's one reason we all own dSLRs instead of
compacts with 1/2.5" sensors, isn't it?

That is just one reason...and not the reason no1 for me.

An f4 lens on full-frame will perform about the same as an f2 lens on
4/3 (2x-crop), both in terms of DOF and noise at the same shutter
speed.

I can't see 2 stop difference at ISO 100 and 200. I can't see 1 stop difference. . .or should I say, I'm pleased with ISO 100-400 and I don't need bigger sensor.
Cheers

 

[joe mama]

You are shooting at 100mm. I am shooting at 400mm equivalence. That's
4x the focal length equivalence and blur-subjecting. So your
comparison is completely invalid.

That has nothing to do with anything. I shot at 100mm, 1/10 s -- just over 3 stops per the 1/FL rule. You shot at 400mm, 1/15 s -- just under 5 stops per the 1/FL rule. That makes the difference between our shots a bit under two stops.

Can't compare 400mm with 100mm? Take another read of what I wrote (if you even bothered to read the first time -- second to last paragraph):

http://forums.dpreview.com/forums/read.asp?forum=1022&message=26662985

"I'd really like to see a systematic comparison of the effectiveness of IS of different systems at a variety of FLs. For example, it's oft claimed that IS is less effective at shorter FLs than longer FLs. Let's say that for a particular shooter, the 1 / FL rule applies at 100mm. Some claim that this rule is too generous at, say, 400mm, but overly cautious at 24mm."

It's stuff like this that quite frankly gets old. I understand
exploring potential of other cameras and other options but not when
the things posted are clearly so one sided lacking objectivity.

You're tweaked, or tweaking. Link and quote something I posted that was "clearly so one sided lacking objectivity". Are you saying that I was able to get a sharp shot at 100mm, 1/10 is an example of being "one sided" whereas you posting a shot at 400mm, 1/15 is not?!

You don't even own the E-3 and you talk about what it can and can't do.

Link and quote, buddy boy, link and quote. I'll quote myself from the same link:

"But 5 stops seems rather generous. That means that I could take a shot of a static subject at, say, 100mm, 1/100s without IS (which I know I can easily do), and do just as well at 100mm, 1/3s with IS. Seems too incredible to me."

You, yourself, said:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=26663341

"Obviously not all shots will be in this high of a result but they do exist and the camera can do it. Olympus also clearly advertises "up to 5 stops" not "you get 5 stops no matter what."

But I get "up to" 3 stops past the 1/FL rule without any IS whatsoever, so that makes Olympus' claim only two stops better than what I can do handheld, per the posted images.

Moreover, you pretty much changed the subject anyway. Your claim was
that it couldn't, and that if it did you would be running to that
camera. So what happened with that? Changed your mind?

You know, you just like to try to twist everything I say into some anti-Olympus BS. I never claimed that Oly's in-camera IS couldn't get 5 stops past the 1/FL rule for a particular shot, I said it would be "amazing" to me if it could do that well on average, and want to see a scientific demonstration of that feat.

Finally the quality at ISO 400 on the E-3 will be superior to the 5D
at ISO 3200.

Talk about changing the subject! Holy cr@p, man, you have some brass huevos to accuse me of something you do so brazenly! I see now how your misguided interpretation led you to such a BS conclusion -- because my shot was at ISO 3200 you thought I was trying to substitute high ISO for IS. Did you even bother to look at the FL and shutter speed in the EXIF I gave? That pic just happened to be at ISO 3200. The ISO played no role in the fact that I took that shot, unsupported handheld, at over 3 stops below the 1/FL rule.

Let me repeat that again: the quality of capture of the E-3 at ISO
400 will be superior to the 5D at ISO 3200. I can even see the noise
at the resize you did (I don't particularly mind it, I am not that
sensitive to that like some others if it looks like grain- but it's
worth pointing out because you are making the comparison).

Who said that ISO 3200 on the 5D is better than ISO 400 on the E3?! Who even implied that?! Even my essay on equivalence says that there will be only two stops difference for the same efficiency of sensor.

What's wrong with you, trying to twist everything I say into some sort of anti-Olympus comment that is simply a reflection of your own insecurities?

Grow up.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

 

[Agramer]

That isn't quite right.
If someone use 2,8 on 4/3 with 1/125 exp and ISO 800 he/she must be
prepared for more noice and bigger DoF than with 2,8 on 24x36 with
1/125 and ISO 800. Nothing would be slower.

You misunderstand. What you said above is right. To get the same
image, the larger sensor would need ISO 3200, 1/125th and f5.6.
Thus, the 4/3 f2.8 lens behaves like an f5.6 lens on full-frame, or
conversely an f2.8 lens on full-frame can be matched by an f1.4 lens
on 4/3.

so, you say that D5 have the same kind of noice on ISO 3200 like E3 on ISO 800?

But we have the problem...All the time you sitting in the dark room and look from that point of wiev. Turn the light on I live during the day...on the sunshine

He/she will gett more
noice and bigger DoF, thats all. Many people use ISO 100 or 200 99%
of the time (like me) and that is irrelevant. I use ISO 100 - 200 and
I can't see the difference. I can see the difference in DoF and if I
must choose ...well I did for macro work bigger DoF is present
from heaven

If you don't need the speed of a larger system, you shouldn't pay for
it. I don't need it at work and that's why I have an APS-c system
and f5.6 zooms. Incidentally, that system doesn't give you more DOF,
it gives you the same DOF at the same noise and same shutter speed.
I can always get just as much DOF and detail as you can, and that's
because both systems can get to diffraction-limited depths-of-field.

I don't need sensitivity of the larger system. I'm creature from the bright side of the planet

People saying that 4/3 is bad because of noice and speed...but if you
say that someone will get more noice than speed will be the same.

Sure. Even my compact is as fast as my 5D if I can accept more noise.

that is my point

If you look "right" (my right may be different than yours E3 will
be blind before 5D (that will be bad for all photographers, but not
crutial for many because many don't work in that kind of light) and
it will give bigger DoF with the same aperture than 5D.

Bigger DOF with the same f-stop. f-stop = focal length / aperture.
With the same aperture (not f-stop) and AOV, all formats have the
same DOF.

OK. that's right...but that is too long, I put that short

For some
people that is good and fore some that is bad.
I didn't see any difference with 5D (I don't need ISO 800 for 99,9%
of my work). I do understand that some people do, but I don't know
many of them.

Like I said, if you don't need speed, you shouldn't have to pay for
it. This is where the smaller-sensor systems do well. The larger
sensor systems afford you the opportunity to get more speed if you
need it. I need it at home, I don't need it at work.

I think that we talk abut sensitivity. Don't we?
Cheers

 

[joe mama]

I could easily shoot at ISO 1600 with all the IS advantage fully in
place and still get a great shot.

Who said otherwise? Who implied otherwise? Here's what I said:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=26662985

"And, man, if that Sony FF DSLR has two stops worth of in-camera IS, then, shoot, it's gonna be hard to stay with Canon! : )"

Why do you think I said that? Wait, don't answer, or do, in either case, I don't care.

I hate to say this but your whole reply was "let me change the
subject when I didn't like the facts." We can play the ISO and
aperture game, both. The issue was whether the E-3 could do such a
high effective IS and I proved that yes, indeed it can.

That's way out from left field. I never played an ISO and aperture game. I posted an image that demonstrates that I got just over three stops past the 1/FL rule. The pic just happened to be at ISO 3200, I never brought noise into it, you did, 'cause you're a freakin' fanboy trying to twist everything I say into some anti-Olympus comment.

If I crank up the ISO the effective IS over the 5D becomes even more
obvious if you are going to go "high gain ISO = IS" route - and
again, this is not the same point you were arguing about.

When did I ever say, or imply, "high gain ISO = IS"? You're such a little twerp trying to twist everything I say. Want to know what I feel about high ISO vs in-camera IS (rhetorical question)? Take a read:

http://forums.dpreview.com/forums/read.asp?forum=1032&message=26603910

Grow up.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

 

[ewelch]

No one who is well informed shares your view. Because it's a fact that a bigger sensor will always be better at absolute image quality than a smaller sensor for one very good reason, all other factors being equal.

Physics. Bigger photo sites allows more efficent translation of light into digital data. You can't change the nature of light. A bigger pixel will be able to gather light more efficiently. That raises the signal and allows for lower noise.

Now, mind you this is all theoretical. Full frame will always be better theoretically than a cropped sensor. You cannot make a smaller sensor better than a bigger one using the same basic technology. (Pixels and the mechanisms to convert the signal to digital data).

BUT!

The smaller sensor can become so good that the theoretical has no practical value. So, once smaller sensors (in Olympus' situation their 4/3 sensors) can generate photographs that are good enough for any practical standard, then it's no longer an issue.

Of course, even before that point is reached, and it won't be reached for a long time, there are other compelling reasons to get Olympus anyway. Some reasons are why the 4/3 system was created in the first place. Many bodies can use the same lenses - across manufacturers. The light paths in lenses is straighter so the lenses can be smaller and lighter and faster.

So if one does not rely on shooting at higher than ISO 800, then the E3 for all practical purposes is a fine camera to get today. It won't match the Nikon D3 (which of course costs at least three times as much) or the 1Ds Mark III - which has no peer. But one has to ask one's self, is it worth a whale of a lot more money for the Nikon or Canon (the Canon 5D might be a compelling reason to say yes since it only costs $600 more, but it's quite dated now) to get better high ISO performance?

The answer to that question is unrelated to the fact that full frame sensors are better in principle in terms of specific performance criterion. One just has to identify if that criterion is theirs or not. Denying the basic facts of physics makes your argument the creationism debate of cameras. (If you get my point, I'm not picking on the merits of creationism itself to make this point.)
--
Eric

All cats are mortal.
Socrates died.
Therefore, Socrates was a cat.