Panasonic Lumix G X 12-35mm F2.8 ASPH Power OIS hands-on preview

The Sony also has vastly inferior optical performance (look at the reviews), has inferior build quality (no weather sealing), has no IS, and is far bigger and heavier. But of course, for camera forum trolls like you, these things are all irrelevant - the only thing that matters is peni.., um, sensor size.

Watch it guys, supeyugin1 is apparently an expert who knows how much it is to manufacture a product just by looking at it and things like performance and quality don't amount to anything.

No, it doesn't matter that this lens has a metal barrel and the 16-50mm is only plastic. Disregard the fact that it has 4 aspherical lenses and stabilization where as the Sony only has 2 aspherical and no stabilization. Ignore the soft corners, complex distortion and chromatic aberration of the 16-50mm because the 12-35mm must be cheaper to produce since it is smaller. The 16-50mm is awesome for it can mount on a NEX via a $400 mirrored adapter with a f/3.5 max limit in video mode.

There is similar Sigma lens, 24-70/3.5-5.6 which costs $220.
But it should be more expensive than Panasonic, which has a constant f/5.6 equivalent. That's why Panasonic cost should be no more than $400, including weather sealing, IS, and all those super-duper elements. Then, there is Sigma 17-50/2.8 with IS, asph, LD, and costs $600. Sorry, but Panasonic is really out of their mind.

Superyugin1, change hobby. NOW.
You're not REALLY multiplying x2 the aperture because of the m4/3 format, are you? ARE YOU? Please tell me you're not.
The aperture is the same across different formats, it's ONLY THE FOCAL LENGTH that changes in other formats in equivalence to 35mm.
f/2.8 is f/2.8, there's no "equivalence". Please study before you state such atrocities.
Geez.

"The Sony also ... is far bigger and heavier"

That is why is costs much more to produce.

supeyugin1, wrong again. This lens is an f2.8 lens and they cost significantly more to manufacture. Do you realize that aperture does 2 things, it affects Brightness and depth of Field. This lens will be a bright (fast) lens regardless of format allowing faster shutter speed and/or lower ISO settings. Most people don't buy zoom lenses for shallow DoF, they buy prime lenses for that. You also miss the point that this is a Super High Grade lens, which costs more to manufacture versus the Standard Grade you present.

street photograhy benefits from fast shutter speed and large dof. 2x crop makes it easier to achive both.

I do street photography mainly, with an APS-C camera and a 31mm f/1.8 lens. I try to stick to apertures no wider than f/2.2 because otherwise the DoF becomes too shallow. With m4/3 I can shoot with the Panny 20mm f/1.7 wide open without having to worry about my DoF not being deep enough, *and* I get faster shutter speeds.

Not everyone is obsessed with razor-thin DoF.

At least Panasonic offers such a lens, what does Sony offer? Nothing! Nothing comparable they can take 1300 $ for.

Sony offers DT 16-50/2.8 at $700. It's Alpha, not NEX, but works through adapter. Also Sigma 17-50 at $650. $1300 for smaller-lighter-simpler (14 elements vs 16 in Sony and 17 in Sigma) lens is about 2x overpriced.

how much does that adapter cost by the way?

Where did you get your numbers? The sensor area difference between these systems (MFT, APS, FF) is not linear.

The total light received by sensor area between MFT and APS-C is between 0,46 EV and 0,64 EV (Canon APS vs. the rest).

Crop factor (calculated by diagonal) between MFT and APS-C goes from 1.25 to 1.33 (Canon APS vs the rest). To get one stop, the crop factor difference should be 1.41 which it is not!

Problem with this that the systems are different by aspect ratio too. If you crop 2:3 system to 3:4, then the difference with sensor area would get even smaller.

For DOF calculations, one should consider the actual lenses available and their focal lengths. The focal length differences might not match the crop factor differences with all lenses.

Slight error, crop factor between APS and MFT varies 1.23X from 1.31X.

4/3 is FF. Also, 4/3 is not a crop format.

And if you used exactly the same FF lens on MFT body via adapter, its brightness would NOT change.

Except for average brightness with large apertures, which will be higher because corner fall off area will be cut off on MFT.

Has anybody of the "equivalent aperture" folks done pinhole camera ever? What happens if you use the same pinhole, but increase distance between pinhole and film by factor of 2?

Answer:
The image will get 4 times larger but at the same time 2 stops darker.

The thing is, I'd imagine for the majority of people, having the extra DoF is preferred to not having it. I certainly appreciate i, especially in low light where there's always a trade of between stopping blur and DoF.

Also, f2.8 is f2.8, whether on m43, APS-C, 35mm FF, 120, 5x4 or 10x8. To say the lens, purely due to the DoF, is f4.0 is not very helpful to people who don't know better. It just spreads confusion, like seen in this whole thread.

It is not light intensity that matters but total amount collected and yes compact camera's are indeed limited to tiny apertures of around f14 usually as the largest.
You could also say focal length is focal length but it is generally accepted that it is not the focal length that matters but the field of view and thats why we use equivalent focal length. The same is true for equivalent aperture and equivalent iso because what matters is the size of the aperture and the total amount of light collected. Every idiot can understand that the light intensity for a given f-number for different formats is constant, its really funny or perhaps sad that all these posts stating this obvious fact are uprated by 5-20.

@malcolm82

So f1.4 on FF equals total light of f5.6 on 4x5"?

Please do a following test: load both cameras with exactly the same 100 ISO slide film and expose for same time with those apertures (4 stop difference). Then scan both developed images at similar resolution and settings and downscale 4x5" image to the size of 24x36mm scan.

What do you get?

What exactly are you getting at? iso 100 on 5x4 would capture 15 times the amount of light as iso 100 on full frame.

edit: I should correct this with 'can capture 15 times the amount of light' obviously when using f5.6 instead of f1.4 on full frame it would be underexposed, all of the 'equivalence trolls' know this and it is just a basic fact so there should be no need to point it out but ive learned from experience on this forum...
So to make it simple lets assume you have a 144x96mm large format sensor, the equivalent shot of f1.4 at iso 100 on full frame would be f5.6 at iso 1600.
To argue that you need an f1.4 lens on large format to equal an f1.4 lens on full frame would be quite absurd.

A single photocell does not determine image quality, all the photocells combined do.
A 144x96mm sensor with the same number of pixels as a 36x24mm sensor would collect the same number of photons per pixel at iso 1600 as the small sensor at iso 100. Why do you think larger sensor camera's have settings for much higher iso's?
The fact that larger sensors usually have higher pixel counts is a different issue and not necessarily a downside. The sony nex 7 though actually has more pixels than the canon 5d mk3.

That's probably the same what the MF folk thinks about the unwashed kleinbild format masses. What's the point of FF if Mf is o much better?

deniz, with FF they also have an edge in IQ

You'd think FF owners would "envy m43 glass": constant aperture zooms, affordable & available f/0.95 primes, and huge size and weight advantages.

Why does one want super fast primes? Normally for shallow DOF and low light shooting. Larger sensors give you better performance in both those areas so shooting f1.0 on m4/3 is very much like shootng f1.4 on APS-C or f1.8 on FF.

Why would a FF shooter envy $1000 25mm f0.95 when they can just buy a 50mm f1.8 for $200 and shoot similar photos with AF?

I am not saying these lenses are bad, it is just hard to see the lenses go for much when their equivalent performance in FF or APS-C is considerably cheaper. Granted FF bodies are expensive, but the lenses are what we are talking about here.

Having previously owned a mountain of L glass, it's because there are massive size and weight advantages to m43 glass.

And it's because the fast primes are image stabilised (on many bodies), so there's another size and weight advantage of not having to carry a tripod in many lower shutter speed situations.

If m43 glass was simply f/2.8 primes and f3.5-5.6 zooms, then the size and weight advantage would mean much less. But m43 lenses include many fast primes and a number of constant aperture zooms.

If one believes size and wight and image stabilised primes mean nothing, then there's nothing to envy in m43 that isn't available in FF.

f2.0 makes it bigger and probably 3-4x more expensive that just using f2.8 on APS-C. It seems everyone wants to defeat the whole purpose of the m4/3 system.

@tkbslc: for many this is the only system they own. While the compact lenses are great for travel, it would not hurt to have some higher level lenses even if they are big. The issue at this point not so much reduction is size or weight, as having to use (and store, transport) two systems instead of one.

It's only for portrait :)

I still don't quite understand. The Olympus 45/1.8 is tiny, lightweight, cheap, and optically superb. It works flawlessly on Panasonic bodies - that's the whole point of Micro Four Thirds. Why deny yourself the opportunity to use such a lovely lens just because it doesn't say Panasonic on the barrel?

Firstly, I put it to you that using 3rd party glass in your example is cheating somewhat, you should have used the Nikon 17-55 at €1300.

Secondly, you overpriced the GH2, its €900. Funny, that makes both camera/lens combos €1900...

Thirdly, you have disproved your own point... "neither smaller and lighter"? Eh, yes, both actually!

Sorry I took the kit price for GH2, not body.
Also Olympus M5 body costs 1100 €.

Price and IQ difference is still immense.

Why using Sigma/Tamron is cheating? Because it doesn't enlarge body parts? That is sad.

ZAnton, we're being a bit selective with our comparison, aren't we?

First, you need to find a better place to shop. The GH2 can be had for €814 on Amazon.de (Nikon is listed at €599).

Second, actual sales prices to the public for the Panasonic lens will only be known when on the shelves and there is almost always a difference between what the Mfg. suggests and the retailer sells for. If you intend to use the Manufacturer's suggested price, then do the same for the Sigma (which I believe was €879).

So now, price wise (using one of the most expensive MFT cameras) you have Nikon close to €1500 and MFT at around €1800 - still an important difference, but nowhere near the extremes you try to represent).

Now let's go on to size/weight. You've picked the biggest MFT you could find. But if you put the combo up against a G3 or GX1 you end up with G3 weighing 40% less and the GX1 42% less. And you ignore size. Length of camera+lens:
Nikon = 169mm
GH2 = 150mm
G3 = 121 mm
GX1 = 113 mm

So you are false, it is not the same weight. Do you really think it is a similar size ?
My question was only about size and weight.
But please, check the GH2 price. You made a "small" error of about 90% (you can get it for less than 700 euros taxes included). With so big mistake, it is not easy to compare...
So perhaps it is not useful to be impolite with panasonic users. I don't know.

About the Pana price - I've already told that I took wrong price from amazon.de.

@ Fiacre
Yes, Pana GH2 + Pana 12-35 are 300 gram lighter, and ~600€ more expensive, and has 2 stops worse DR, and 1 stop worse ISO.

Read my message again please. I said "In this combination.."
which means Oly EM5 + Pana 12-35. Since DXO doesn't have Oly EM5 results I took the best in terms of IQ Panasonic camera.

Also, please tell me where you can find Pana GH2 for 700Euro with taxes.

@ HDF2
I am not intended to use MRP. I look what you can buy at this moment (or in the nearest future in case of Pana lens). Pana lens will be cheaper one day. Maybe in a year, maybe in 1,5 years. But we don't know how much will Sigma/Tamron/Nikon cost in that time. So lets speak about the moment when Pana 12-35 will be availible, which is probably 1-2 monthes.

Second, if you take G3 and other super slim and cheap cameras, then the IQ penalty is even higher.
But again, to be precise, I wrote "This combination..."

Please show me a comparable APS-C 24-70mm equivalent lens, with constant f2.8, weather sealing, and similar weight and dimensions.

ZAnton--We are still waiting for you to give us a example of a smaller and light APS-C camera.

Pana is making crap lenses. They assume that CA , vignietting and distortion are "easily" removed in postprocessing and make lenses with no optimisation on that things at all. After that they charge double price for that crap. Sigma, Tamron and Tokina produce way better lenses than pana, say nothing about Canon and Nikon.

Gentlemans, I invite you to continue our discussion 1 post higher.

@ DarkShift
Theoretically it is not, but I have Oly EPen1 - 12 MP, Canon 350D - 8MP (and now Canon 5Dmk2)
And I can say that ancient 350D + Tamron 17-50 nonIS just destroys Oly EPen1 + Kit in terms of everything (except weight and size).

Ok, I have OM-D and 5d mkII.

OM-D has practically the same DR than 5D2, lags maybe one stop behind noise levels, except for dark noise which is much more uniform than ugly pattern noise with 5D2.

The 5D2 has better detail level depending on the lens and aperture, but the OM-D kills the latter in speed and usability IMHO.

Ah more digital correction sillyness. Fact is a) digital correction is built into the lens ROM so is a feature of the lens and b) these lenses with digital correction applied automatically give a better end result than optically only corrected lens. Every single time anyone argues these lenses are inferior I ask for an example of an optically corrected lens that can beat a Panasonic lens in the same class, and so far no-one has *ever* been able to suggest one.

@ZAnton

There's several Nikon and Canon high speed primes that produce CA and LOCA near wide open apertures. Some Nikkors even at center of image.

Why don't you complain about those?

Zanton: it is clear that m43s is not for you and it is nice that we have a choice. I have never come across a weathersealed f2.8 lens at 24-70 mm equivalent and this weight. I guess the closest would be 450-500 gram or so, but I wonder if that is weahtersealed.

Most of us are into m43s because of the size and weight with (very) good IQ. We are not sure if the 12-35 will deliver of course, but we'll see. The review of the OM-D over here shows that current m43 sensors are as good as the best APS-c or at least that good that you need an FF to get superior IQ.
So I guess it is not so bad and for most of us it is good enough, for some it won't be. Nice to have a choice I think.

The correction data is stored in the firmware ROM in the lenses themselves. So the moment you attach the lens, the camera reads the lens ROM via the camera mount contacts.

Smart design, like that.

Malcolm82- there's no "Full frame" switch on a light meter. f2.8 is f2.8.

Do you even know what that means? Divide the diameter of the lens by the focal length- that's what an f-stop is. It's not some magical light-making orifice.

The important spec is the aperture size for a given field of view lens and not the aperture number, yes this means an f2.8 lens on m43 is only as 'bright' in terms of full sensor exposure as an f5.6 lens on full frame.

No malcolm82, it would be f5.6 if you used the f2.8 m4/3 on a FF camera, which you can't. The actual *hole* in the lens relative to both focal length and form factor is what makes for the f number. However, the f5.6 would relate to DOF and diffraction. But not in speed, no.

Are you really saying this lens is "worse" than a full-frame 24-70 lens because it is all of 17mm longer than you think it could possibly be?

In the size of the lens you forgot to tell that the sensor to lens distance in canon is 44mm... while in m43 land its 20mm so the lens is smaller than we should expect by halfing sensor size...

from 70-200mm shallow DOF is what you want but from 24-70 its just the low light and if it will be used for group shots and such in weddings f2.8 is the minimum for I would try at say.. 20mm or 25mm (40mm to 50mm equiv.)

Oh boy, another armchair engineer telling us like it should be.. I'm sure the real people who came up with this had no idea what they were doing or only put a little bit of effort into this, probably threw it together over 2 or 4 meetings and to hell with it, say what? It's not perfectly half the size of the FF? Weeeeell we could do that but naaaa

You fool

"For the amount of light it gathers this is not a compact lens at all. The new canon 24-70 f/2.8"

weights almost 3 times as much and not even stabilized.

How can you have a problem with the size and cost of the micro four thirds lenses while you don't seem to care that much about the stupidly priced leica lenses. I would love to see an 8X10 print from an equivalent leica lens that resolves more detail than my 25 1.4. Leica is made for a bunch of nostalgic idiots who seem to live in this delusional world that thier optics are out of this world. They haven't made a single decent digital camera that would AF in less than ten seconds, and how much do they charge for their cameras?? You should really be getting upset at them, and not at micro four thirds.

No, good sir, you are not. ƒ-numbers are not apertures. Variable ƒ-number zooms are closer to being fixed aperture lenses than fixed ƒ-number zooms.

Would this lens really be worse off as a ƒ/2-2.8, as it possibly could have been?

More importantly, from an operation point of view, constant F number zooms have only one reason. With manual or non-TTL auto flashes, constant F zooms make flash work easier. TTL flash removes this last reason. In the end money is spent to make something appear more prestigious.

Constant ƒ-zooms are a waste of glass at the short end. If you have problems with the ƒ-number changing while zooming, use aperture priority mode to set it at the minimum ƒ-number on the long end.

Constant ƒ-zooms are not a waste of glass and yes this lens would be worse off if it were 2-2.8. If you want to zoom when using the camera's video mode, it's useful to not have your shot go dark half way through your shot.

It's a "fly by wire" lens with a stepless (or at least something silly small, like 1/30 stop) electronic aperture. They could have made it f2 at the wide end, and had the aperture compensate in video mode. Set f2.8, and as you zoom, the aperture automatically adjusts to maintain f2.8.

So called "constant aperture" lenses are technological dinosaurs. (And I own three of the best, LOL).

Arenas: Do you understand the principie of aperture priority mode?

My dealer does not like the ergonomics of the OMD. He writes: "it is rather small in my opinion. The buttons and layout are a little cramped in comparison to other m4/3 that I know will be out later this year (one in June & the other in October)."
Now he is under non-disclosure. So I don't know what these bodies are. Or whether they are Olympus or Panasonic. But based on what should be coming out, one would expect these to be the replacements for the G3 and GH2. And based on past releases (the G3 was announced in May and the GH2 was announced in September) I'm hoping that the G3 replacement is the first one. Though the G3 is one year old, and the GH2 is two years old. So the GH2 replacement could be first.

It has an OIS switch on its barrel, so it should work with EM-5.

Using both IS systems at the same time is not wise though.

There's no such thing as "equivalent aperture", Paul. F2.8 is always f2.8, be it full frame, four thirds, APS-C or whatever.

@ ManuelVilardeMacedo

If you take that opinion you also need to apply it to focal length, there is no such thing as "equivalent focal length" 12-35mm will ALWAYS be 12-35mm, be it full frame four thirds, APS-C or whatever

The problem and confusion is dpreview putting the equivalent focal length(24-70mm) right next to the unequivalent aperture.

The fact is this lens would look identical to a 24-70mm F5.6 on full frame, yet there is a constant misinformation campaign waged by the micro 4/3rds crowd to confuse people.

Now you may say well aperture is used to determine exposure and if you "convert" the aperture you get a incorrect assumption of light intensity. Well that would be a good point if it wasn't for the fact that aperture is not a measurement of light transmission(T-stop is).

The fact is F/stop is more useful for determining depth of field than exposure so it's totally reasonable to "convert it" if you are already doing such with the focal length.

@bradleyg5

Really? Then a f2.8 FF lens should magically become two stops faster while attached on MFT. I think not...

Take a light meter reading through f2.8 lens for MFT and FF, and tell us what you did found.

While the aperture is physically larger on FF, the imaging area is also around four times larger. So the light intensity is the same for same area.

ManuelVilardeMacedo, do you also have trouble understanding what people mean when they say "equivalent focal length"?

Focal length, ƒ-numbers, and ISO sensitivities are not really measurements of anything interesting, they only have meaning for a given sensor size. To compare between systems you need to correct for the sensor size, and the 36x24mm has been chosen as the default.

There is an "equivalent focal length", and "equivalent f-number" and an "equivalent ISO sensitivity". So far, most people understand the focal length part, just some of the journalists have started to understand the ƒ-number part (but only when it can be used as a negative), but nobody understands the ISO sensitivity part. And you really need to translate them all to keep the correct ratio between them. I wish this would sink in with the journalists soon.

Bradley, that's just not true. Aperture is an optical property of the lens that's not sensor-related. Crop factor is an entirely different thing altogether. A 20mm focal length (in 35mm FF terms) will be 40mm equivalent when mounted on a 4/3 camera, and 30mm on a Canon APS-C. Aperture, however, remains the same whatever the sensor the lens is going to work with. I repeat, there's no such thing as "equivalent aperture". Otherwise it'd be as if someone's visual acuity would vary according to the size of the brain!!!
And Cax, your findings are firmly placed on the realms of fantasy. They are not sustained by any evidence.

@Cax

Because f-number is calculated as following: focal length of the lens / diameter of aperture.

Where did you find anything that relates to sensor size in that equation?

Here we go again.

"Aperture equivalence" is an unecessarily confusing way of explaining DoF equivalence. DoF changes with sensor size, aperture does not.

Ta-da!

Funnily enough, DPReview do a pretty decent job of explaining this, yet there are *still* people on the forums who are peddling utter rubbish in the form of "explanations".

@Cax - if you're going to get scientific, can we get properly scientific? The same amount of light *per mm* is still hitting the sensor. The amount of light hitting the sensor *in total* can be inferred from the resulting exposure but is essentially irrelevant photographically speaking.

You see, we are not interested in focal lengths, we are interested in angles of view, and to infer an angle of view from a focal length, we must know the sensor size.

We are not interested in ƒ-numbers, we are interested in the amount of light gathered, and to infer the amount of light gathered from an ƒ-number, we must know the sensor size.

We are not interested in ISO sensitivities, we are interested in image noise, and to infer the image noise from an ISO sensitivity, we must know the sensor size.

Until someone propagates the use of degrees as a measure of angle of view, some sort of total light transmission index as a measure of amount of light, and some sensor size independent measure of light in an exposure, you bozos are just going to have to take our word for it: you can translate these three obsolete photographic measurements to correct for variations in sensor size. 12mm, ƒ/2.8 at ISO 400 on µ4/3 IS THE SAME as 24mm, ƒ/5.6 at ISO 1600 on a 36x24mm system.

Cax, you are not understanding what the review is referring to as equivalent aperture, it has nothing to do with the intensity of the light hitting the sensor. The f-stop is the ratio of entrance pupil diameter to focal length and is a measure of the intensity of the light hitting the sensor. Longer focal length lenses spread the light more as the cone of light heads towards the sensor but the f-stop takes this into account by the lens using a larger hole and thus more total light gets through but it is spread more. The sensor size plays no role in this as the f-stop is measuring light intensity not the total amount of light.

When they are referring to equivalent aperture in terms of DoF, what that means is that as the sensor becomes larger for a given distance and subject size, the focal length of the lens being used is increased. This longer focal length for a fixed camera to subject distance and fixed subject size results in less DoF

Andy pick any optically superior lens and apply the software correction to it. Those who buy optically superior lenses are not fools they can also apply software correction if needed.

How can they apply digital correction? Do you have any idea the exact specifics of the digital correction? No, you're going to best guess it in a PP program which won't be as good because 9/10 you'll over do it and 10/10 it won't be nearly as complex and specific. It's not quuuite as simple as you put it.

@zxaar name a single optically corrected lens /in the same class/ that can beat a Panasonic lens. Obviously you can spend 10x more on a Canon L lens that could beat a digitally corrected Panasonic lens, but show me a single lens in the same class that is sharper even before you apply extra digital corrections manually.

"How can they apply digital correction? Do you have any idea the exact specifics of the digital correction?" ----------- Manufacturers of these lenses have that. Anything else??

@Andy Crowe I am pretty sure my sony R1 Zeiss is better than the panasonic you mentioned even without software correction. And that is 24 to 120 mm lens. There may be many other examples in nikon and canon too. Further speaking of L, this lens is also not cheap.

f2.8 is f2.8.

Are you trolling? Does your light meter have a "full frame" switch on it? Of course not- that's stupid. Do you even know what the "f-stop" actually is, and how it's measured? I'm guessing you don't.

No, you're getting aperture mixed up with depth of field (doF). 1/60th sec @ f2.8 is the same regardless of format. If you had a handheld light meter telling you to set 1/60th @ f2.8 its the same on m43, 35mm or medium format. To get the same depth of field on m43 as 35mm FF then yes, it would need to have a larger aperture.

Personally, I've had more difficulty in getting more DoF than shallow DoF for low light shots, m43 really helps here. But that's me.

Regards,

Colin

http://en.wikipedia.org/wiki/F-number

The F-stop is the ratio of focal-length to the entrance pupil. A 24-70mm F/2.8 lens has a 8.5mm pupil at the wide end, and a 25mm pupil at the long end. The 12-35mm has a 4mm pupil at the wide end and a 12.5mm pupil at the long end. It is not equivalent to a 24-70mm F/2.8, but a 24-70mm F/5.6. The size of the pupil does not change, hence you can't just multiply the focal-length.

As for exposure, a 24-70mm F/2.8 on 135 does have the same light intensity as the 12-30mm, but it has 4 times the image circle, meaning that it gathers 4 times as much light in total. If used with a native sensor, one that utilizes that image circle, you gather 4 times as much light, making it 4 times, or 2 stops, better than a m43 12-30mm F/2.8.

A equivalent m43 lens to a 135 24-70mm F/2.8 is a 12-30mm F/1.4.

"...used with a native sensor, one that utilizes that image circle, you gather 4 times as much light, making it 4 times, or 2 stops, better than a m43 12-30mm F/2.8."

Sigh... Go take a photo with your full-frame CaNikon. Then take the same photo with your P&S, using exactly the same aperture, shutter and iso. The exposures will be IDENTICAL. It would still be identical if you used MFT, APS or Medium Format.

Try it, I dare you.

Of course the exposures will be identical. But the poster is still correct that four times the total light (not light per area) will be collected on an FF sensor in the larger image circle.

Of course that's what gives the better noise characteristics of the FF: total light.

So what needs to be disentangled are three things:

Depth of field: function of aperture, distance from subject, angle of view of lens and used image circle.

Total light: (and hence noise level): function of aperture, and used image circle (holding light levels and length of exposure fixed)

Exposure: function of aperture (holding light levels and length of exposure fixed).

Much of the discussion of these matters conflates the three.

Thanks for the sensible answer.

If only other people on this site could be so useful and insightful without getting so sensitive about a simple search for clarity.

@dbm305

Better noise may make sense only if using exactly 4 times larger photosites on a FF sensor. In practise tech advance makes this issue more complicated.

F-stop is focal length divided by aperture diameter. To keep it simple, f2.8 lens is the same regardless of format or imaging area. This is basic photography really.

Mr. Roboto is right. f2.8 is f2.8. It's a purely optical measure that doesn't vary with crop factor.

'mister_roboto'...Trolling?

Let's see...

Trolling: unleashing one or more
cynical or sarcastic remarks on an innocent by-stander.

Sound like you are the only troll on this website. You think you have some justification to belittle other people asking honest questions. Get back under your bridge.

Okay, so, the total light is 4x greater for a 24-70/2.8 lens in 135-format. But... wait for it... the sensor is 864mm^2 vs. 225mm^2, (26%), so the *luminous flux per unit area of the sensor* is only 4% different - in fact, it's (slightly) in favour of 4/3.

Consdier this: if I cut a 17.3 x 13mm section out of a 24x36mm negative (yes, film), the exposure is (obviously) still the same. The degradation in image quality is related to other factors (relatively coarser grain structure, tonality, etc.), not "total light gathered". The difference in sensor tech makes such a direct comparison between different format (or manufacturer or generation) digital cameras a dubious exercise.

The only *real* difference worth worrying about for actual shooting comes from DOF considerations, which cannot be wangled away.

Exactly.

Those who keep insisting that f2.8 on m4/3 equals f5.6 on FF should take some photography classes.

2.8 is 2.8 when it comes to exposure.

2.8 is 5.6 when it comes to DOF at the same focal length, as compared with a 35mm sensor.

People talking about total light gathered are comparing apples to oranges. It's not like you gather a bunch of light, put it in a barrel, and then sell it at the market.

But it is exactly like that. An FX camera at f/5.6, 1/100s, ISO 400 gives approximately the same brightness, image quality and depth of field than a MFT camera at f/2.8, 1/100s, ISO 100. The FX camera (assuming a modern one like D800) will be still better in resolution but if we only are looking at tonal quality, color separation, SNR this kind of stuff, as well as depth of field then this equivalence holds. Thus you get more for your money by buying a Nikon 24-70/2.8 than this 12-35, greater resolving capability, more options for depth of field control, and more light => better image quality, but only pay about twice as much as this small lens. What you're doing essentially is paying extra for the small size of the 12-35. Which of course is a perfectly legitimate thing to do.

"An FX camera at f/5.6, 1/100s, ISO 400 gives approximately the same brightness, image quality and depth of field than a MFT camera at f/2.8, 1/100s, ISO 100. "

That would be true if the same FF lens was used on m43 and just wasted all the light around the sensor. But m43 lens concentrates the light (the same amount) on the smaller surface of the smaller sensor.

Thanks Ilkka, what a nice twisting of words from a true Nikon follower.

Half the money becomes suddenly more if not spend on holy Nikkor glass. :D Which I own too, but that doesn't mean its the right tool for every situation.