OK I must be stupid.

Poruchik Rzhevsky said:

For example, let's consider the 14-54 / 2.8-3.5 II on 4/3:

Click to expand...

http://www.slrgear.com/reviews/showproduct.php/product/1239/cat/15

Poruchik Rzhevsky said:

and the Tamron 28-75 / 2.8 on FF:

Click to expand...

http://www.slrgear.com/reviews/showproduct.php/product/49/cat/23

Since when is Canon EOS 20D (the camera used to test Tamron) a Full Frame body? ;-)
Be careful with your comparisons before jumping to conclusions.

Marcin
http://www.flickr.com/photos/lensjoy/

Poruchik Rzhevsky said:

kociasek said:

For example, let's consider the 14-54 / 2.8-3.5 II on 4/3:
http://www.slrgear.com/reviews/showproduct.php/product/1239/cat/15
and the Tamron 28-75 / 2.8 on FF:
http://www.slrgear.com/reviews/showproduct.php/product/49/cat/23

Click to expand...

Click to expand...

kociasek said:

Since when is Canon EOS 20D (the camera used to test Tamron) a Full
Frame body? ;-)
Be careful with your comparisons before jumping to conclusions.

Click to expand...

You have to click on the "Full Frame Results" tab. Thanks for asking for the clarification.

in terms of the ability to blur the background, yes, 50/2 on 4/3 is equiv. to 100/4 on full frame, with an equal field of view.

in terms of light gathering ability,no, f 2 is still f 2, just that when you use the f 2 you will not get the background bokeh as you would expect from a 50/2 lens on 35mm slr.

ch01 said:

in terms of the ability to blur the background, yes, 50/2 on 4/3 is
equiv. to 100/4 on full frame, with an equal field of view.

Click to expand...

The reason this is so is because the apertures are the same (50mm / 2 = 100mm / 4 = 25mm). And if the apertures are the same...

ch01 said:

in terms of light gathering ability,no, f 2 is still f 2, just that
when you use the f 2 you will not get the background bokeh as you
would expect from a 50/2 lens on 35mm slr.

Click to expand...

...so will be the total amount of light (for the same scene, framing, perspective, and shutter speed). The f-ratio, which is the quotient of the focal length and the aperture, tells you the intensity of the light, not the total amount of light.

So, yes, f/2 yields the same intensity of light across formats, but it does not result in the same total amount of light, and it is the total amount of light, not the intensity of the light, that determines the image noise (along with sensor efficiency).

Thus, for equally efficient sensors and the same total amout of light (not same intensity of light) the total image noise will be the same. The way FF sensors get less noise than smaller sensors even for equally efficient sensors is by either using a larger aperture (not f-ratio), which also results in a more shallow DOF, or by using a longer shutter speed, which requires either good light, flash, or tripod.

Ok that argument seems very strange. Sorry but sounds like left field to me. F2 on 4/3 means the same ratio of total light on the sensor as f2 on a full frame sensor. Sure the full frame gathers more light but has to spread it over a larger area so all ends up equal. Noise I would think has more to do with pixel density and individual sensor designs.
--
Thanks, Mike
'Never Lose the Wonder!'

amblepath said:

Ok that argument seems very strange. Sorry but sounds like left
field to me.

Click to expand...

Yes -- many struggle with this concept. : )

amblepath said:

F2 on 4/3 means the same ratio of total light on the sensor as f2 on a full
frame sensor.

Click to expand...

Yes.

amblepath said:

Sure the full frame gathers more light but has to spread it over a larger area
so all ends up equal.

Click to expand...

No, all does not "end up equal". You said it yourself -- "full frame gathers more light" (for the same f-ratio and shutter speed).

amblepath said:

Noise I would think has more to do with pixel density and individual
sensor designs.

Click to expand...

Well, it does take more than a sentence or two to explain in it's entirety, and gets more complicated when we have to take into account different pixel counts and different levels of detail, as well as the amount of NR performed in-camera. Both of these cases can be dealt with by applying NR to the more detailed image to match the detail of the less detailed image before comparing noise. Of course, lens sharpness now plays a role, etc., etc.

However, for the most part, real life results tend to follow the "same total light equals same total noise", as modern sensors don't differ in efficiency by all that much. A very good example is this one:

http://www.seriouscompacts.com/2008/12/panasonic-g1-nikon-d700-iso-shootout.html

Of course, the noise is never going to be exactly equal, but it is close enough that using total light as a guide for total noise with cameras of the same generation is pretty much how it goes.

As I recall, you have heard this all before in an earlier thread of yours:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=31215066

but I was hoping that I might be able to express it more eloquently. Guess not. : )

hi, i was just answering op's questions...I don't see any mentioning about noise...anyway, I was just stating what I know, not comparing the advantages or disadvantages of 4/3 or ff.

but since you mentioned, I guess noise is more related to the flux (or total/area of sensor) of light, instead of total amount.

correct me if i misunderstand.

Poruchik Rzhevsky said:

So, yes, f/2 yields the same intensity of light across formats, but
it does not result in the same total amount of light, and it is the
total amount of light, not the intensity of the light, that
determines the image noise (along with sensor efficiency).

Click to expand...

ch01 said:

hi, i was just answering op's questions...I don't see any mentioning
about noise...

Click to expand...

No worries. When people bring up total amount of light and intensity of light, however, noise is usually the motive, so apologies for my misinterpretation.

ch01 said:

but since you mentioned, I guess noise is more related to the
intensity (or total/area of sensor) of light, instead of total amount.

correct me if i misunderstand.

Click to expand...

The other way around. Image noise is a function of the total amount of light, not the intensity of the light. Please see my post above:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=31235809

if this still doesn't make sense. But first, let me say that I am talking about total image noise, not per-pixel noise, which gets complicated very quickly, since we need to discuss how noise and detail are related.

I thought you were talking about per pixel level noise.

yes you are right, given the F.O.V is same for 2 sensors with different sizes, with all other parameters properly set for this comparison, I agree total noise is indeed a function of total light gathered.

--
http://www.chenghelucky.blogspot.com

ch01 said:

I thought you were talking about per pixel level noise.

Click to expand...

My bad. I really should have made that clear from the start. My apologies for the confusion.

ch01 said:

yes you are right, given the F.O.V is same for 2 sensors with
different sizes, with all other parameters properly set for this
comparison, I agree total noise is indeed a function of total light
gathered.

Click to expand...

Good to see we are on the same page! : )

John King said:

And just to throw in my two-penneth worth ... FF/APS are a different
aspect ratio from 4/3rds, and therefore the AoV, light-gathering etc
are not directly comparable in ANY meaningful way. So, I specially
agree with your point " ... just pick the one that works best for
what you want to do." Shame that this is anathema to some people ...

Click to expand...

He said 'equivalent'. It is time you learned the difference.

-

Azures said:

John King said:

And just to throw in my two-penneth worth ... FF/APS are a different
aspect ratio from 4/3rds, and therefore the AoV, light-gathering etc
are not directly comparable in ANY meaningful way. So, I specially
agree with your point " ... just pick the one that works best for
what you want to do." Shame that this is anathema to some people ...

Click to expand...

He said 'equivalent'. It is time you learned the difference. Tattoo
it on your arm if you can not.

Click to expand...

He did not say "equivalent" anywhere in his post; he did say:

"In terms of exposure, F2 is F2 is F2, no matter what focal length or format you're shooting that day. But when you come to depth of field and related issues, it's all about the physical diameter of the aperture. In this regard, the 50/2 has a 25mm aperture and so offers characteristics similar to a full frame 100/4 (and a little bit 'faster' than an APS-C 60/2.8).

Now some people have decided to use this simple consequence of the laws of physics to launch into heated debates that pretend to prove the superiority of one format over another. This is ludicrous, each format simply has its own set of strengths and weaknesses, just pick the one that works best for what you want to do."

Just for the record ...

So you are absolutely wrong yet again ...

And it is time you learned some manners ...

I do not subscribe to the "tattoos on the arm" philosophy, as you patently do ... Several of my friends had these when I was growing up; but then, I did grow up ...

Azures said:

John King said:

And just to throw in my two-penneth worth ... FF/APS are a different
aspect ratio from 4/3rds, and therefore the AoV, light-gathering etc
are not directly comparable in ANY meaningful way. So, I specially
agree with your point " ... just pick the one that works best for
what you want to do." Shame that this is anathema to some people ...

Click to expand...

He said 'equivalent'. It is time you learned the difference.

Click to expand...

Azures wrote:
{ORIGINALLY}

Azures said:

He said 'equivalent'. It is time you learned the difference. Tattoo
it on your arm if you can not.

Click to expand...

You can edit out your more intemperate comments from your own posts (23 minutes after posting ... ?? The rest of us seem to be limited to 15 minutes ... ).

Rotten shame that you cannot edit them out of my post where I quoted your original post ...

You expose yourself for what you really are, when you stated:

"Tattoo it on your arm if you can not."

Like "arm tattoos", do you? BUT we already knew that about you, didn't we?

Changing the subject does not change your original statement one iota ...

Poruchik Rzhevsky said:

You have to click on the "Full Frame Results" tab. Thanks for asking
for the clarification.

Click to expand...

Oh, I've overlooked that tab, my mistake.

However, I've just checked the results for Tamron on FF and the Olympus lens still clearly beats Tamron at wide open F2.8. At optimum apertures (f5.6 for Zuiko, F.8.0 Tamron) both go neck-and-neck. So, sorry, your claim that "at the same AOV and DOF, the Tamron is not only sharper, but flatter, than the Olympus" is not true. In fact, as na overall performer, Olympus is better.

Is (Focal length) / (aperture). So:

Fstop = (FL) / A

If a lens has a 2.8 F/stop and is 200mm long then how wide is the aperture?

2.8 = 200 / A

A = 71.43mm

If you look at the "Crop factor" it does NOT change the actual focal length so there is no change in the F stop.

for example a 28 f/2.8.

On a "full frame": 2.8 = 28/10

If you change the focal length because of "crop" then its

F/stop = (28*2) / (10*2)
28*2 = 56
10*2 = 20
F/stop = 56/20
F/stop = 2.8

Poruchik Rzhevsky said:

Poruchik Rzhevsky said:

Sure the full frame gathers more light but has to spread it over a larger area
so all ends up equal.

Click to expand...

No, all does not "end up equal". You said it yourself -- "full frame
gathers more light" (for the same f-ratio and shutter speed).

Click to expand...

The light per area does indeed end up (approximately) equal. Let's not confuse the argument by going off on a tangent here.

Juergen Hahn said:

Juergen Hahn said:

Juergen Hahn said:

Sure the full frame gathers more light but has to spread it over a larger area
so all ends up equal.

Click to expand...

No, all does not "end up equal". You said it yourself -- "full frame
gathers more light" (for the same f-ratio and shutter speed).

Click to expand...

The light per area does indeed end up (approximately) equal. Let's
not confuse the argument by going off on a tangent here.

Click to expand...

Could you clarify where you disagree at?

--
http://www.pbase.com/sngreen

Honestly razor thin DOF is only an advantage in portraiture or certain artistic styles of photography. Isolating the subject is a concern but I've seen plenty of shots by various 4:3 lenses that isolate the subject just fine. Heck I've seen tons of P&S pics that isolate the subject just fine and that's with a sensor 1/4-1/32 the size of 4:3 and a lens that is usually f2.0 at its brightest.

The vast majority buy fast lenses for their light gathering ability. A macro photographer isn't going to use f2 at 1:2 or 1:1 he's going to use the slowest stop he can where the 50/2 has a two stop advantage (2.8 vs 2, 32 vs 22 means the Zuiko is brighter for focusing and has a one stop advantage in DOF) over the equivalent 105/2.8 Nikon. For landscape it's of no advantage. For sports the lenses are fast to freeze action not because of DOF and at those FL the DOF is so razor thin anyway it doesn't really matter. For architecture, which is my main interest, DOF has NEVER been of any concern to me while handheld and tripod usage usually has me stopping down to f8 or more to get the most resolution out of my lens as well as the most DOF.

You will never win a semantic argument based on physics and going "Oh no! I don't have enough DOF! If only I had the extra 2mm of blurriness that a 35mm FF camera will give me!" Is pure hyperbole at best. Bokeh, AF speed and high ISO noise also don't figure into a discussion of DOF.

kociasek said:

However, I've just checked the results for Tamron on FF and the
Olympus lens still clearly beats Tamron at wide open F2.8.

Click to expand...

You will note that I said "at the same AOV and DOF" as opposed to "at the same AOV and f-ratio (two stops more shallow DOF)".

kociasek said:

At optimum apertures (f5.6 for Zuiko, F.8.0 Tamron) both go neck-and-neck.

Click to expand...

Yes, they do. Once again, the pros of the Tamron are two stops more shallow DOF for those that want it at half the cost. The advantages of the Zuiko being a longer focal range.

kociasek said:

So, sorry, your claim that "at the same AOV and DOF, the Tamron is not
only sharper, but flatter, than the Olympus" is not true.

Click to expand...

There's a decent difference in favor of the Tamron on FF at f/2.8 with the Zuiko and f/5.6 for the Tamron (same DOF), and further up the DOF scale they are so close as to render any differences meaningless.

The charts for reference once more:

Tamron 28-75 / 2.8 on FF:

http://www.slrgear.com/reviews/zproducts/tamron28-75/ff/tloader.htm

Olympus/Zuiko 14-54 / 2.8-3.5 II:

http://www.slrgear.com/reviews/zproducts/olympus14-54f28-35II/tloader.htm

kociasek said:

In fact, as an overall performer, Olympus is better.

Click to expand...

On the other hand, I'm not saying that FF is "better than" 4/3. I merely gave one example each of where I saw the FF option outperforming the 4/3 option (28-75 / 2.8 on FF vs 14-54 / 2.8-3.5 II on 4/3), and another example where I saw the 4/3 option outperforming the FF option (24-70 / 2.8L on FF vs 14-35 2 on 4/3) to support my comment that flatness of frame depends on the lenses being compared.

lloyd007 said:

Honestly razor thin DOF is only an advantage in portraiture or
certain artistic styles of photography. Isolating the subject is a
concern but I've seen plenty of shots by various 4:3 lenses that
isolate the subject just fine. Heck I've seen tons of P&S pics that
isolate the subject just fine and that's with a sensor 1/4-1/32 the
size of 4:3 and a lens that is usually f2.0 at its brightest.

Click to expand...

Aboslutely true. Check these out:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=28970330

are they awesome or what?

lloyd007 said:

The vast majority buy fast lenses for their light gathering ability.
A macro photographer isn't going to use f2 at 1:2 or 1:1 he's going
to use the slowest stop he can where the 50/2 has a two stop
advantage (2.8 vs 2, 32 vs 22 means the Zuiko is brighter for
focusing and has a one stop advantage in DOF) over the equivalent
105/2.8 Nikon.

Click to expand...

Couple of points. f/2.8 to f/2 and f/32 to f/22 is one stop, not two. Secondly, the 105 / 2.8 collects one stop more light than the 50 / 2. 'Tis true. I discuss it here:

http://forums.dpreview.com/forums/read.asp?forum=1022&message=31235634

Lenses focus wide open, so the actual aperture being used to capture the image is irrelevant in terms of AF, by the way. What does matter a lot more for macro in terms of focus is viewfinder size and/or liveview performance.

lloyd007 said:

For landscape it's of no advantage.

Click to expand...

But the larger pixel counts of FF are an advantage.

lloyd007 said:

For sports the lenses are fast to freeze action not because of DOF and at those
FL the DOF is so razor thin anyway it doesn't really matter.

Click to expand...

It is a mattter of opinion, but let's go with yours on this. It depends on the lens, of course, but let's take the Olympus 90-250 / 2.8 on 4/3 vs the Nikon 200-400 / 4 on FF. The Olympus has more range (big plus), whereas the Nikon gathers one stop more light which results in one stop less noise (big plus). Which is better? Depends on whether you value more range or less noise, in this particular comparison.

lloyd007 said:

For architecture, which is my main interest, DOF has NEVER been of any
concern to me while handheld and tripod usage usually has me stopping
down to f8 or more to get the most resolution out of my lens as well
as the most DOF.

Click to expand...

Absolutely.

lloyd007 said:

You will never win a semantic argument based on physics and going "Oh
no! I don't have enough DOF! If only I had the extra 2mm of
blurriness that a 35mm FF camera will give me!" Is pure hyperbole at
best. Bokeh, AF speed and high ISO noise also don't figure into a
discussion of DOF.

Click to expand...

While bokeh and AF speed do not figure into a discussion of DOF, background blur and noise do, since they are all intimitely related. However, it's important not to try to say that System A is "better than" System B without being specific about how it is better and under what circumstances.

As I hope I demonstrated with the 90-25 / 2.8 on 4/3 vs 200-400 / 4 on FF comparison, "better than" depends a lot on a person's particular needs, so that what is better for one person is simply not so for another. Saying which is better without saying why and for who just makes for bitter arguments.