Hang on- how useful is f/2.8?!

[topcatproduction]

I'm gradually upgrading some of my kit, and I basically just want a UWA and walk around lens to complete my ideal set up. I was thinking of the Tokina 11-16 f/2.8 and Canon 17-55 f/2.8 lenses, the tokina for the large aperture and canon for its amazing reputation!

However- I was out on a stroll the other evening and took some snaps at f/2.8 with my sigma as it was getting dull, (please remember I'm still learning) and when I got them on my laptop I couldn't believe how daft I was- none were useable as all needed more DOF so were mostly OOF. Which made me think- why do I really want the Tokina when f/2.8 is no good for landscapes which is what it is designed for?

The f/2.8 on the Canon will be nice to get shallow DOF as its a walk around lens, but isn't it a waste of money on a landscape lens and even for shooting inside in low light? IS must be more handy where you do not want shallow DOF???

Please don't rip me apart for my understandings! I just want some explanations!!!

Thanks! TC

 

[Arkady]

That extra stop is worth it's weight in Gold when you need to freeze movement and there's little light available.

 

[mini_mad216]

True, I have found the extra stop invaluable but from the sounds of things whar you really need is a good tripod so that you can put the higher apertures (for DOF) to good use without having to worry about shutter speed.

 

[HoppyUK]

I'm gradually upgrading some of my kit, and I basically just want a UWA and walk around lens to complete my ideal set up. I was thinking of the Tokina 11-16 f/2.8 and Canon 17-55 f/2.8 lenses, the tokina for the large aperture and canon for its amazing reputation!

However- I was out on a stroll the other evening and took some snaps at f/2.8 with my sigma as it was getting dull, (please remember I'm still learning) and when I got them on my laptop I couldn't believe how daft I was- none were useable as all needed more DOF so were mostly OOF. Which made me think- why do I really want the Tokina when f/2.8 is no good for landscapes which is what it is designed for?

The f/2.8 on the Canon will be nice to get shallow DOF as its a walk around lens, but isn't it a waste of money on a landscape lens and even for shooting inside in low light? IS must be more handy where you do not want shallow DOF???

Please don't rip me apart for my understandings! I just want some explanations!!!

Thanks! TC

I think you've got it right. F/2.8 is not much use in a wide angle for landscape. But Tokina make it that way because if it was f/4 nobody would buy one with that limited range and price - Canon 10-22 is better. Or Sigma 10-20, because it's cheaper.

I think f/2.8 is mainly for professionals, working in difficult environments when they've got to get the shot. Then, when you've maxed out the ISO and are using the longest shutter speed you dare, the guy with the lowest f/number lens will get the best shot.

Like you I usually find the shallow depth of field you get with low f/numbers is a drawback. I'd rather use flash, or higher ISO, or rely on the IS making a longer shutter speed viable. And I've also got the ultimate option of not bothering to take a picture at all! Pros don't have that luxury.

And also, f/2.8 is only one stop over f/4. Another way of doing it is to have modest f/number zooms like f/4 or so - with the benefits that they have in terms of weight and price - and have a couple of f/1.4 primes for the low light stuff.

F/1.4 to f/4 is three stops, which is massive difference. The DoF is also mega shallow which also opens up a different style of picture taking with big bokeh backgrounds. Looks well nice

 

[AndyElliott]

The APS-C and fourthirds cameras can make use of the f2.8 more - by virtue of greater DOF. It was interesting to note looking at the 300/2.8 lenses on the 1D, 5D, D700 and D3 cameras in the BBC/NHM Wildlife Tog of the Year photos that not one was taken wide open - f4 and f5.6 were the only ones I could spot. One reason is sharpness, the other is the DOF. The f2.8 certainly helps with AF speed. The other thing is that most lenses typically sharpen up at 1-2 stops below max aperture to their maximum sharpness. The IQ you can get from a Canon 300/2.8 @f4 is noticeably better than the 300/4 wide open.

My 300/2.8 is equivalent of 600/2.8 in terms of focal length and light-gathering power, but more like 600/5 in terms of DOF (not f5.6, most calculations fudge the difference in aspect ratio, whereas looking at the vertical crop makes more sense if with the assumptions of prize-size and print aspect ratio). So I can actually make use of the extra light gathering power and get enough DOF. This extra light gathering power also allows me to reduce the ISO and reduce the noise, offsetting some of the difference between 'full-frame' and crop sensors. This argument also follows to a slightly lesser extent for APS-C cameras.

So far as landscape goes, it's a bit pointless - DOF, AF and shutter speed (all the landscape togs I know would hit somebody for not using a tripod) are not an issue.

That sort of focal length and speed is useful for indoor shooting. My 14-35 f2 (yes, f2, not f2.8) is a very useful thing to have when the light is not great...

 

[Yardbent]

,
and I basically just want a UWA and walk around lens to complete my ideal set up. I was thinking of the Tokina 11-16 f/2.8 and Canon 17-55 f/2.8 lenses, TC

I'm just starting too - and enjoying landscapes most

as a beginner most [all?] my shots are in daylight and sunsets using the Nikon kit 18-55 most at 18mm about f5.6 [on a tripod!]
just replaced it with a 17-70 - oh yes i want a 11-18 Tamron - but do i ?

at the moment 17 is wide enough - decided to wait till my experience "needs" less than 17mm ..............just a thought ?

 

[photon]

Shallow depth of field? On the 40D, an 11mm lens at ƒ/2.8 has a hyperfocal distance of 2.26m, when everything from a bit over a metre to infinity should be acceptably sharp.

 

[HoppyUK]

The APS-C and fourthirds cameras can make use of the f2.8 more - by virtue of greater DOF. It was interesting to note looking at the 300/2.8 lenses on the 1D, 5D, D700 and D3 cameras in the BBC/NHM Wildlife Tog of the Year photos that not one was taken wide open - f4 and f5.6 were the only ones I could spot. One reason is sharpness, the other is the DOF. The f2.8 certainly helps with AF speed. The other thing is that most lenses typically sharpen up at 1-2 stops below max aperture to their maximum sharpness. The IQ you can get from a Canon 300/2.8 @f4 is noticeably better than the 300/4 wide open.

My 300/2.8 is equivalent of 600/2.8 in terms of focal length and light-gathering power, but more like 600/5 in terms of DOF (not f5.6, most calculations fudge the difference in aspect ratio, whereas looking at the vertical crop makes more sense if with the assumptions of prize-size and print aspect ratio). So I can actually make use of the extra light gathering power and get enough DOF. This extra light gathering power also allows me to reduce the ISO and reduce the noise, offsetting some of the difference between 'full-frame' and crop sensors. This argument also follows to a slightly lesser extent for APS-C cameras.

So far as landscape goes, it's a bit pointless - DOF, AF and shutter speed (all the landscape togs I know would hit somebody for not using a tripod) are not an issue.

That sort of focal length and speed is useful for indoor shooting. My 14-35 f2 (yes, f2, not f2.8) is a very useful thing to have when the light is not great...

All of which makes perfect sense, except that 4/3rds format cannot make itself four times larger, to match the area of full frame, whereas other formats with equal pixel density can match 4/3rds simply by cropping (eg 7D).

It's not as if those Olympus lenses are cheap either...

Not looking for an argument, just pointing out that there are downsides to smaller formats too

 

[will69]

Have a look at the EXIF data on your twenty favourite shots. If they are all towards the widest aperture of the lens, then upgrade to a f2.8. If they tend towards the middle of the aperture ange then it probably eans this is a want rather than a need.

 

[neil_g]

remember even if youre not using it at 2.8 having that aperture available to the AF system will help with accuracy.

but yes its a godsend in low light when flash is not an option.

 

[AndyElliott]

All of which makes perfect sense, except that 4/3rds format cannot make itself four times larger, to match the area of full frame, whereas other formats with equal pixel density can match 4/3rds simply by cropping (eg 7D).

It's not as if those Olympus lenses are cheap either...

Not looking for an argument, just pointing out that there are downsides to smaller formats too

No - it's just the counter argument (the one you are making) is usually the one that is heard!

Full-frame has generally better noise control and dynamic range. This makes it excellent for shooting in low light (e.g. concerts). The shallow DOF is great for portraiture.

The bigger DOF of the crop formats is better for wildlife, as is the greater reach. It is also better for macro work - although this is arguably a draw because of the lower diffraction limit.

BTW, the only camera that can meet the pixel density (and therefore reach) of a 12MP fourthirds sensor is the 7D. However, when you take into account the anti-alias filter, the 7D is a smidge behind.

Also, most of the fourthirds lenses (certainly all of the pro and top-pro grade) can comfortably achieve pixel-level sharpness at the 12MP. The 7D struggles to muster that many lenses that can do that. The 'good' L glass is designed for full-frame and most of it can't make full use of the 18MP sensor. Some of the EF-S glass can, but mostly not. To paraphrase another user here with a 7D - "think of it as a 40D with better AF and you will not be disappointed". Of course this is only an issue with heavy cropping and/or very large print sizes. But to say the 7D has the same reach as the fourthirds format cameras isn't quite true.

Sure, the Olympus super-high-grade (aka top-pro) glass isn't cheap. But it is bloody good too! The middle-tier glass is really very good, and certainly competitively priced. The ZD 50-200 f2.8-3.5 is a good example. Only a third of a stop slower than the constant f2.8 70-200 lenses, a lot smaller and lighter, weather sealed, as sharp across the frame as the Nikkor 70-200 f2.8 VRII and sharper than the Canon 70-200 f2.8.

 

[puddleduck]

Even for landscape, I still like a nice bright f/2.8 lens - really helps when you start putting on Grads / Pols etc.

 

[AndyElliott]

Even for landscape, I still like a nice bright f/2.8 lens - really helps when you start putting on Grads / Pols etc.

That's a good point - always nice to be able to see through the lens!

 

[Flash In The Pan]

Even for landscape, I still like a nice bright f/2.8 lens - really helps when you start putting on Grads / Pols etc.

I think you've hit the nail on the head, Andy. What you gain with an f/2.8 lens is a brighter viewfinder, no-one is saying just because a lens has a max aperture of 2.8 that it has to be used wide open all the time.

 

[specialman]

....why do I really want the Tokina when f/2.8 is no good for landscapes which is what it is designed for?....

Interesting comment that, the fact that it's commonly accepted that ultra-wides are only for landscape work. Editorial photographers have a lot of use for wide aperture, wide-angle lenses, as they can really give you something that a slower aperture lens can't.

On my 80-200mm f/2.8 is a godsend in the same way that f/1.8 on my 85mm is. On wide-angles I agree that it's obviously not as important for landscape (high DoF) work but there is merit nonetheless

 

[Somelier]

One thing that seems to have been missed here is that a 2.8 lens allows much more precise focusing than a smaller aperture lens.

Grrr, I see the myth about "magnification factor" is creeping in here. Le'ts get it right - a 300mm lens is a 300mm lens, no matter what camera you put it on, only the amount of cropping changes.

 

[AndyElliott]

One thing that seems to have been missed here is that a 2.8 lens allows much more precise focusing than a smaller aperture lens.

Grrr, I see the myth about "magnification factor" is creeping in here. Le'ts get it right - a 300mm lens is a 300mm lens, no matter what camera you put it on, only the amount of cropping changes.

...and the number of pixels covering that cropped area, the how much detail can be resolved (a function of the lens, the AA filter and the number of pixels). My point is about actual reach,e.g. you can crop a 12MP full-frame to Nikon APS-C size and have only 5.3MP covering the area. If you put the same lens on the APS-C body with 12MP, you have 12MP covering that area. The reach is only increased if the lens in question is capable of delivering for the extra demands of the higher pixel density or at least better than the full-frame density.

I think Canon should have made the 7D a 14 or 15 MP beast and reaped the gains in dynamic range and noise control. This would still have given them clear marketing water between them and Nikon. There are very few users who would have found the extra res useful and better (albeit better than already very good) DR and noise control will always be appreciated.

 

[topcatproduction]

Wow I wasn't expecting so many replies- or to be agreed with in general

I shouldn't have said the UWA is designed for landscapes- even I know it's not so not sure why I wrote that but it is a major part none the less!

Now I've decided I dont need the Tokina unfortunately it hasn't made the Canon 10-20 any cheaper- especially as it is about the same price as the Tokina I know the sigma is cheaper but also not as sharp going by other peoples comments...

Thanks for all the posts! I'd still like to see more points to the pro's and con's!

 

[HoppyUK]

No - it's just the counter argument (the one you are making) is usually the one that is heard!

Full-frame has generally better noise control and dynamic range. This makes it excellent for shooting in low light (e.g. concerts). The shallow DOF is great for portraiture.

The bigger DOF of the crop formats is better for wildlife, as is the greater reach. It is also better for macro work - although this is arguably a draw because of the lower diffraction limit.

BTW, the only camera that can meet the pixel density (and therefore reach) of a 12MP fourthirds sensor is the 7D. However, when you take into account the anti-alias filter, the 7D is a smidge behind.

Also, most of the fourthirds lenses (certainly all of the pro and top-pro grade) can comfortably achieve pixel-level sharpness at the 12MP. The 7D struggles to muster that many lenses that can do that. The 'good' L glass is designed for full-frame and most of it can't make full use of the 18MP sensor. Some of the EF-S glass can, but mostly not. To paraphrase another user here with a 7D - "think of it as a 40D with better AF and you will not be disappointed". Of course this is only an issue with heavy cropping and/or very large print sizes. But to say the 7D has the same reach as the fourthirds format cameras isn't quite true.

Sure, the Olympus super-high-grade (aka top-pro) glass isn't cheap. But it is bloody good too! The middle-tier glass is really very good, and certainly competitively priced. The ZD 50-200 f2.8-3.5 is a good example. Only a third of a stop slower than the constant f2.8 70-200 lenses, a lot smaller and lighter, weather sealed, as sharp across the frame as the Nikkor 70-200 f2.8 VRII and sharper than the Canon 70-200 f2.8.

Your arguments are good Andy, but highly selective largely theoretical You dismiss the BIG advanatge of full frame with an almost throwaway line about noise and dynamic range, forgetting to mention sharpness at all, which are huge and fundamental and very real.

Then you replace them with stuff about how even L-class lenses cannot match the resolution of some 4/3rds-specific optics, talking up the benefits of pixel density and then suggesting that the Canon 7D would be better off with a few less!

The whole thing about 4/3rds is founded around a much smaller sensor, and its attendant benefits. And excellent though these sensors are (and getting better) if small was that good, then we'd all be shooting on 14mp compacts which is surely the ultimate incarnation of the concept.

However, when you come to compare the images produced by the best compacts, with the best from 4/3rds and then from full frame, it is not hard to put them in order of merit as far as image quality is concerned. And you don't need to pixel-peep either.

I don't hold with your crop factor theory about using the vertical frame height and aspect ratio to determine depth of field differences either. The logical way to compare is using the frame diagonal, in which case the 4/3rds crop factor is 2.00x relative to full frame.

 

[AndyElliott]

Your arguments are good Andy, but highly selective largely theoretical You dismiss the BIG advanatge of full frame with an almost throwaway line about noise and dynamic range, forgetting to mention sharpness at all, which are huge and fundamental and very real.

It's very easy to see difference in sharpness between FF and APS-C with the same lenses - you're asking a lot more from lens on APS-C than FF - so per-pixel sharpness will look better. However, comparing FF with full-frame lenses and APS-C (or fourthirds) with lenses that have been designed for that format and the sharpness thing goes away. The diffraction limit is higher for FF. So far as noise goes, under ISO 800, the differences are not that great. On a blind test, it would not be easy to see a difference. As far as DR goes, check out the numbers e.g. 5D mk II vs 7D on dxomark as an example. The difference is actually small. Certainly nothing to get overly hung-up about.

Then you replace them with stuff about how even L-class lenses cannot match the resolution of some 4/3rds-specific optics, talking up the benefits of pixel density and then suggesting that the Canon 7D would be better off with a few less!

All true. I think Canon would have been better to either reduce the resolution of the 7D or provide it with more lenses that can make use of its resolution. If the 7D was able to make use of its extra resolution, that would be great. Given that there aren't that many lenses that can, reducing the resolution, increasing the DR and noise performance would be a much better option.

The benefits of a covering the image circle with a squarer sensor where the incident light is nearer to perpendicular than full-frame lenses are real. Less vignetting, lower lateral CA and better edge-edge sharpness.

To quote DPReview, from the ZD 9-18mm review, one of Olympus's budget lenses:

"We've long said that the great strength of the E-system lies in the lenses, with Olympus taking advantage of the smaller area of the Four Thirds sensor to produce optics which are simply more consistent in overall image quality and edge-to-edge performance than those of its competitors."

There are other such examples from other reviewers.

The whole thing about 4/3rds is founded around a much smaller sensor, and its attendant benefits. And excellent though these sensors are (and getting better) if small was that good, then we'd all be shooting on 14mp compacts which is surely the ultimate incarnation of the concept.

Firstly, the difference in height of a FT sensor is not much different to an APS-C one. The area is smaller by virtue of it being squarer. In terms of a lot of pictures printed (I've got a Colin Prior calendar on my wall as an example), the aspect ratio is nearer to the 4:3 than 3:2 - medium format cameras are nearer to this ratio too. The extra width, at best may well be lost when you subsequently crop the frame for display. So smaller than full frame, but not much in it with APS-C.

It's also horses for courses, as I said, some things are done better by full-frame, some things aren't. Particularly when you add in the practicality aspect of the size and weight of the lenses.

However, when you come to compare the images produced by the best compacts, with the best from 4/3rds and then from full frame, it is not hard to put them in order of merit as far as image quality is concerned. And you don't need to pixel-peep either.

Oh, I would disagree there... Print something up at A3 and I would be pretty sure good examples of each piece of kit would fox experts as to which was which. Print much larger, then the differences become more apparent. Only pick subject matter requiring the use of 3200 ISO and above and the difference will also be apparent.

I don't hold with your crop factor theory about using the vertical frame height and aspect ratio to determine depth of field differences either. The logical way to compare is using the frame diagonal, in which case the 4/3rds crop factor is 2.00x relative to full frame.

If you look at how the circle of confusion is calculated, it is based on a certain viewing size of a print. Because of typical aspect ratios of print, there is a compelling argument to say that the vertical difference is the relevant difference to use. The reality is the difference between 1.8x and 2.0x. Which really isn't much to make all that difference...


(BTW: this is now getting seriously OT...)

 

[HoppyUK]

<snip>

...However, comparing FF with full-frame lenses and APS-C (or fourthirds) with lenses that have been designed for that format and the sharpness thing goes away.

Erm, no. It doesn't.

<snip>

(BTW: this is now getting seriously OT...)

Quite, so I won't go through it all, other than to say I think your points are largely subjective and debate theories that don't hold up in practise I'm not trying to argue that one format is better than the other, just that they have different virtues.

One of the great things about photography is that when you look at final prints, the debate is concluded. You cannot deny the evidence of your own eyes. Which is what I have done, and it proved to be rather costly with the sale of my 40D and crop-format lenses in favour of a 5D2.

I did the test that you are suggesting, comparing directly a 40D and 7D with a format-specific lens of very high quality (EF-S 17-55 2.8), and a 5D2 with a full frame lens (EF 24-105L). Printing out at A4 and larger, the difference between the 40D (10mp) and 7D with 80% more pixels, the difference was negligible. Yes, it was there, both in terms of noise and sharpness, but nobody in my family except me could really tell them apart.

Not so with the 5D2 - it was clearly and obviously better, to the naked eye and without pixel peeping. Sharper, more detail, more dynamic range (visible in both highlights and shadows), less noise. Not surprising really, as although it has a roughly similar pixel count to the 7D (18mp vs 21mp) the 5D2's pixels are twice the size. That's gallons more light gathering, loads more lovely photons, just much more image forming data for the lens, the sensor and the software to work with. The result is there for all to see.

 

[trencheel303]

I'm gradually upgrading some of my kit, and I basically just want a UWA and walk around lens to complete my ideal set up. I was thinking of the Tokina 11-16 f/2.8 and Canon 17-55 f/2.8 lenses, the tokina for the large aperture and canon for its amazing reputation!

However- I was out on a stroll the other evening and took some snaps at f/2.8 with my sigma as it was getting dull, (please remember I'm still learning) and when I got them on my laptop I couldn't believe how daft I was- none were useable as all needed more DOF so were mostly OOF. Which made me think- why do I really want the Tokina when f/2.8 is no good for landscapes which is what it is designed for?

The f/2.8 on the Canon will be nice to get shallow DOF as its a walk around lens, but isn't it a waste of money on a landscape lens and even for shooting inside in low light? IS must be more handy where you do not want shallow DOF???

Please don't rip me apart for my understandings! I just want some explanations!!!

Thanks! TC

Mate, if you think f/2.8 is bad, you should try shooting from mere feet away your cat, or something, at f/1.8 . His nose can be in focus while his eyes are out, his eyes can be in while his nose is out.. etc. The extra stops make a huge amount of difference in low light though and at f/2.8 and lower, indoor photography without a flash does become feasible. If you want to get more in focus, you have to stop down, which means less light in the lens which probably means tripod time. Similarly, stopping up to take handheld pictures indoors means you are dealing with a very small DOF. Facts of life.

 

[Amp34]

That extra stop is worth it's weight in Gold when you need to freeze movement and there's little light available.

This.

I think you've got it right. F/2.8 is not much use in a wide angle for landscape. But Tokina make it that way because if it was f/4 nobody would buy one with that limited range and price - Canon 10-22 is better. Or Sigma 10-20, because it's cheaper.

I think f/2.8 is mainly for professionals, working in difficult environments when they've got to get the shot. Then, when you've maxed out the ISO and are using the longest shutter speed you dare, the guy with the lowest f/number lens will get the best shot.

Like you I usually find the shallow depth of field you get with low f/numbers is a drawback. I'd rather use flash, or higher ISO, or rely on the IS making a longer shutter speed viable. And I've also got the ultimate option of not bothering to take a picture at all! Pros don't have that luxury.

And also, f/2.8 is only one stop over f/4. Another way of doing it is to have modest f/number zooms like f/4 or so - with the benefits that they have in terms of weight and price - and have a couple of f/1.4 primes for the low light stuff.

F/1.4 to f/4 is three stops, which is massive difference. The DoF is also mega shallow which also opens up a different style of picture taking with big bokeh backgrounds. Looks well nice

I disagree quite strongly with this TBH. With regards to wide angles and fast apertures my 12-24 Tokina at f/4 will focus from around 4ft to infinity at 12mm, the 11-16 will probably be around 8ft to infinity at 11mm, not exactly an unusable aperture with wide angle lenses. You seem to be forgetting the wider the lens the greater the DOF for an aperture. You also have the creative ability of a fast wide angle lens, not all shots on a wide ange lens are going to be vistas, close up shots which may not need a huge DOF for a start.

I regularly use my Tokina 12-24 at f/4 when the light is failing and I don't have or want my tripod. The weight issue is a bit of a fallacy as well, some lenses are a lot heavier when comparing a f/2.8 and f/4 lens but the 11-16 is about the same weight and size as the 12-24.

EDIT: And also when the light gets to a point where an extra stop will stick an extra 30 seconds or more onto the exposure and possibly "ruin" the shot you are trying to achieve.

There is also the "sweet spot" of the lens, no lens is perfect at its highest aperture so most people who want the sharpest shot will drop down a couple of stops, which when comparing two lenses with a stop different could mean the difference between f/5.6 and f/8.

You also have the ability to take low light shots that wouldn't be possible with a slower lens and i'd love to know of any other lens (or prime) that can get near the 11-16 (or even the 12-24/10-22/10-20) in terms of width and aperture speed (especially at a good price)! As an example of your f/1.4 lenses, the widest one I know of is the Canon 24mm f/1.4, nowhere near as wide as any of the zooms.

There are obviously some drawbacks in some situations for using a wide angle lens at its widest aperture but just because it has it doesn't mean you should use it.

This is from a person that uses his a wide angle lens a lot and uses it wide open in quite a few situations. I'm also planning to swap it with the 11-16 in the future as that has some great benefits over a slower lens.

 

[Amp34]

Wow I wasn't expecting so many replies- or to be agreed with in general

I shouldn't have said the UWA is designed for landscapes- even I know it's not so not sure why I wrote that but it is a major part none the less!

Now I've decided I dont need the Tokina unfortunately it hasn't made the Canon 10-20 any cheaper- especially as it is about the same price as the Tokina I know the sigma is cheaper but also not as sharp going by other peoples comments...

Thanks for all the posts! I'd still like to see more points to the pro's and con's!

Well the obvious choice in that is the Tokina 12-24. Similar price to the Sigma 10-20 but similar quality (and better build quality) to the Canon 10-22.

 

[HoppyUK]

This.



I disagree quite strongly with this TBH. With regards to wide angles and fast apertures my 12-24 Tokina at f/4 will focus from around 4ft to infinity at 12mm, the 11-16 will probably be around 8ft to infinity at 11mm, not exactly an unusable aperture with wide angle lenses. You seem to be forgetting the wider the lens the greater the DOF for an aperture. You also have the creative ability of a fast wide angle lens, not all shots on a wide ange lens are going to be vistas, close up shots which may not need a huge DOF for a start.

I regularly use my Tokina 12-24 at f/4 when the light is failing and I don't have or want my tripod. The weight issue is a bit of a fallacy as well, some lenses are a lot heavier when comparing a f/2.8 and f/4 lens but the 11-16 is about the same weight and size as the 12-24.

There is also the "sweet spot" of the lens, no lens is perfect at its highest aperture so most people who want the sharpest shot will drop down a couple of stops, which when comparing two lenses with a stop different could mean the difference between f/5.6 and f/8.

You also have the ability to take low light shots that wouldn't be possible with a slower lens and i'd love to know of any other lens (or prime) that can get near the 11-16 (or even the 12-24/10-22/10-20) in terms of width and aperture speed (especially at a good price)! As an example of your f/1.4 lenses, the widest one I know of is the Canon 24mm f/1.4, nowhere near as wide as any of the zooms.

There are obviously some drawbacks in some situations for using a wide angle lens at its widest aperture but just because it has it doesn't mean you should use it.

This is from a person that uses his a wide angle lens a lot and uses it wide open in quite a few situations. I'm also planning to swap it with the 11-16 in the future as that has some great benefits over a slower lens.

Not sure what you are disagreeing with there. All good points, well made

 

[Amp34]

Not sure what you are disagreeing with there. All good points, well made

Pretty much all of your first post. (or at least how I understood it )

Mainly f/2.8 not being much use for landscapes and the use of f/2.8 mainly by professionals when they need to get a shot.

Also some of the other comments I disagreed with as well.

 

[HoppyUK]

Pretty much all of your first post. (or at least how I understood it )

Mainly f/2.8 not being much use for landscapes and the use of f/2.8 mainly by professionals when they need to get a shot.

Also some of the other comments I disagreed with as well.

You can certainly disagree with me, or the way you read it, but I'm not disagreeing with anything you have said I didn't say f/2.8 was no use at all, it's always nice to have, I said it wasn't of "much" use. I guess what I meant there is that I'd rather have the extra focal length range of other lenses - both the extra mm at the short end and much more at the long end.

DoF at 11mm and f/2.8 extends from 5.7ft to infinity (crop format only) so the opportunity for differential focus effects is pretty limited, but it's there of course if that's what you want.

 

[Amp34]

You can certainly disagree with me, or the way you read it, but I'm not disagreeing with anything you have said I didn't say f/2.8 was no use at all, it's always nice to have, I said it wasn't of "much" use. I guess what I meant there is that I'd rather have the extra focal length range of other lenses - both the extra mm at the short end and much more at the long end.

DoF at 11mm and f/2.8 extends from 5.7ft to infinity (crop format only) so the opportunity for differential focus effects is pretty limited, but it's there of course if that's what you want.

I agree with that pretty much. The one thing that does worry me about the 11-16 is the lack of range, however treat it like a prime and I guess it's not as bad.

My main reasons for choosing it over other lenses is because it is the best lens out there for shooting night/sky landscapes but I guess that is a big niche, however f/2.8 is pretty much perfectly fine for standard landscapes as well if needed. It doesn't have the drawbacks of something like the choice between a 70-200 f/2.8 and f/4 with regards to weight and price so if you can deal with the focal lenght (for example you have a 17-55) then there is no real reason not to buy it.

 

[AndyElliott]

Erm, no. It doesn't.

Quite, so I won't go through it all, other than to say I think your points are largely subjective and debate theories that don't hold up in practise I'm not trying to argue that one format is better than the other, just that they have different virtues.

One of the great things about photography is that when you look at final prints, the debate is concluded. You cannot deny the evidence of your own eyes. Which is what I have done, and it proved to be rather costly with the sale of my 40D and crop-format lenses in favour of a 5D2.

I did the test that you are suggesting, comparing directly a 40D and 7D with a format-specific lens of very high quality (EF-S 17-55 2.8), and a 5D2 with a full frame lens (EF 24-105L). Printing out at A4 and larger, the difference between the 40D (10mp) and 7D with 80% more pixels, the difference was negligible. Yes, it was there, both in terms of noise and sharpness, but nobody in my family except me could really tell them apart.

Not so with the 5D2 - it was clearly and obviously better, to the naked eye and without pixel peeping. Sharper, more detail, more dynamic range (visible in both highlights and shadows), less noise. Not surprising really, as although it has a roughly similar pixel count to the 7D (18mp vs 21mp) the 5D2's pixels are twice the size. That's gallons more light gathering, loads more lovely photons, just much more image forming data for the lens, the sensor and the software to work with. The result is there for all to see.

I can get the whole dynamic range and noise thing - although with appropriate processing and not doing anything silly like trying to make the comparison at ISO 1600, at A4 or even A3, the difference shouldn't be that marked (progs like noise ninja will clean up a pretty dirty image very well). I'm not trying to dismiss this difference in a single line, clearly it is there. If you are shooting in available (low) light at high ISOs, the 5D mk2 is going to be a real winner.

Regarding the sharpness thing. Looking at the charts here: http://wlcastleman.com/equip/reviews/17-85compare/index.htm, if you look at the results for the 17-55 compared with the prime, the 17-55 (sharp though it is) still isn't approaching the nyquist limit on even on the 400D. So it isn't that surprising it doesn't look as sharp as the 5D/24-105 when used on the 7D.

My 14-35 is well on the way at f2 and looks like it has plateaued at f4, hitting the nyquist at around f4 (Diwa Labs tests, referenced in http://brandoneu.blogspot.com/2009/09/my-personal-review-of-zuiko-digital-ed.html, and compared with the Nikkor 24-70 used on full-frame). I wouldn't expect the sort of difference in sharpness you are seeing in your comparison in the (seemingly reasonable) comparison made in the test I reference.

The MTF values at f2 even exceed the Zuiko 50mm @f2 - a lens that DPReview describes as:

"The Olympus Zuiko Digital 50mm F2.0 Macro is a lens which we've held in high regard ever since we started using it for our reviews of Four Thirds camera bodies, and subjecting it to our full optical test suite shows precisely why."

..even in the review of the Canon 100mm f2.8L IS USM Macro

"The 100mm F2.8 L IS USM Macro performs very well even on the resolution-hungry APS-C format. Sharpness is high and fairly even across the frame, chromatic aberration is low, and distortion and falloff minimal. Despite not being on its 'native' format, the results are close to the excellent Olympus Zuiko Digital 50mm F2 Macro, and clearly better than the older Canon EF 100mm F2.8 USM Macro."

The more reasonable comparison might be made with the Zuiko 12-60 zoom - more in the bracket of the 17-55 - and even this manages to hit the nyquist limits. As does the 50-200 f2.8-3.5...

So, I don't see any reason not to get the sharpness you see with the 5D on smaller formats - it would just appear that the 17-55 isn't going to be the lens that delivers it. I think that Olympus are concentrating all their efforts on the smaller sensor without having to worry about stealing customers from their own full-frame offering - I think that shows.

Andy

 

[HoppyUK]

I can get the whole dynamic range and noise thing - although with appropriate processing and not doing anything silly like trying to make the comparison at ISO 1600, at A4 or even A3, the difference shouldn't be that marked (progs like noise ninja will clean up a pretty dirty image very well). I'm not trying to dismiss this difference in a single line, clearly it is there. If you are shooting in available (low) light at high ISOs, the 5D mk2 is going to be a real winner.

Regarding the sharpness thing. Looking at the charts here: http://wlcastleman.com/equip/reviews/17-85compare/index.htm, if you look at the results for the 17-55 compared with the prime, the 17-55 (sharp though it is) still isn't approaching the nyquist limit on even on the 400D. So it isn't that surprising it doesn't look as sharp as the 5D/24-105 when used on the 7D.

My 14-35 is well on the way at f2 and looks like it has plateaued at f4, hitting the nyquist at around f4 (Diwa Labs tests, referenced in http://brandoneu.blogspot.com/2009/09/my-personal-review-of-zuiko-digital-ed.html, and compared with the Nikkor 24-70 used on full-frame). I wouldn't expect the sort of difference in sharpness you are seeing in your comparison in the (seemingly reasonable) comparison made in the test I reference.

The MTF values at f2 even exceed the Zuiko 50mm @f2 - a lens that DPReview describes as:

"The Olympus Zuiko Digital 50mm F2.0 Macro is a lens which we've held in high regard ever since we started using it for our reviews of Four Thirds camera bodies, and subjecting it to our full optical test suite shows precisely why."

..even in the review of the Canon 100mm f2.8L IS USM Macro

"The 100mm F2.8 L IS USM Macro performs very well even on the resolution-hungry APS-C format. Sharpness is high and fairly even across the frame, chromatic aberration is low, and distortion and falloff minimal. Despite not being on its 'native' format, the results are close to the excellent Olympus Zuiko Digital 50mm F2 Macro, and clearly better than the older Canon EF 100mm F2.8 USM Macro."

The more reasonable comparison might be made with the Zuiko 12-60 zoom - more in the bracket of the 17-55 - and even this manages to hit the nyquist limits. As does the 50-200 f2.8-3.5...

So, I don't see any reason not to get the sharpness you see with the 5D on smaller formats - it would just appear that the 17-55 isn't going to be the lens that delivers it. I think that Olympus are concentrating all their efforts on the smaller sensor without having to worry about stealing customers from their own full-frame offering - I think that shows.

Andy

Blimey Andy, you can write almost as much as me!

The theory is all very well, but just get hold of a 5D2 and shoot side by side with your Olympus. Do a similar test to the one I did and print out to A4 or larger. If the full frame images are not clearly and obviously better in every aspect of image quality then I'm a Dutchman. And have just wasted £2k.

You cannot dismiss my tests as invalid because the lens I used wasn't as good as Olympus. The Canon EF-S 17-55 2.8 (shot at f/5.6, viewed in the centre) is universally acclaimed as being extremely good. It is as sharp (in the centre) as the EF-S 60mm macro I used to own. It smashes through the Nyquist ceiling, not that I think that's actually very relevant, and if that lens cannot deliver on a smaller format, then the whole concept is doomed.

Of course, there is always a reason for why bigger formats are better, always a valid theory. I'm not going to bang on about it and the whole imaging process is a complex and multi-faceted thing but I think the hub of it lies in lens MTF and sensor MTF - both analogue devices - and the relationship between resolution and contrast.

Basically, the imaging chain is imperfect and what you put in one end is not what you get out of the other. Not by a long way. If a lens has to work at 50% higher resolution to deliver detail on a smaller format like 4/3rds, then contrast will have dropped a lot (maybe 30%, as a guess) and even Olympus cannot dodge that fact of physics.

Then there is the sensor. If you present a 10mp sensor with an image with resolution equal to 10mp, you will be lucky to get 6-7mp out of it. Smaller sensors with higher pixel density are much worse, maybe only 5mp and less. They are not the neat lines of individual pixels and micro lenses we see in diagrams, under the microscope they irregular splodges with light spilling out all over the place, signal-to-noise being lost all the time, detail eroded. Bigger sensors are just better, more efficient.

That's where I think the difference between formats lies, but as I say, the theory is all very well, it's practise that counts. Comparitive tests are very easy to do, and the differences clear to see.

 

[woof woof]

In the real world it's often best to forget about high mp counts and sensor sizes unless you are going to crop like crazy or print something that'll cover a wall. I have an A4 print that I love that was taken with a £50 3mp fixed focus pas and at a normal viewing distance it could have been taken with almost anything.

Back to OP. If f2.8 gives oof shots I think it's time to review shooting technique. I think that wider apertures are very useful for creative effect, speed and low light use and that it's much better to have a wider aperture available than not as even at f1.4 and 50mm unless you're right on top of your subject you'll have a dof to choose from of anything from a few inches to 60 feet and that's more than enough to give you options to play with and more than enough for many subjects. A 12-24mm f2.8 will give lots of dof, at 20mm and f2.8 from less than 5 yards to infinity.

Also it's worth remembering that dof is often largely an illusion.

 

[AndyElliott]

Blimey Andy, you can write almost as much as me!

The theory is all very well, but just get hold of a 5D2 and shoot side by side with your Olympus. Do a similar test to the one I did and print out to A4 or larger. If the full frame images are not clearly and obviously better in every aspect of image quality then I'm a Dutchman. And have just wasted £2k.

You cannot dismiss my tests as invalid because the lens I used wasn't as good as Olympus. The Canon EF-S 17-55 2.8 (shot at f/5.6, viewed in the centre) is universally acclaimed as being extremely good. It is as sharp (in the centre) as the EF-S 60mm macro I used to own. It smashes through the Nyquist ceiling, not that I think that's actually very relevant, and if that lens cannot deliver on a smaller format, then the whole concept is doomed.

Of course, there is always a reason for why bigger formats are better, always a valid theory. I'm not going to bang on about it and the whole imaging process is a complex and multi-faceted thing but I think the hub of it lies in lens MTF and sensor MTF - both analogue devices - and the relationship between resolution and contrast.

Basically, the imaging chain is imperfect and what you put in one end is not what you get out of the other. Not by a long way. If a lens has to work at 50% higher resolution to deliver detail on a smaller format like 4/3rds, then contrast will have dropped a lot (maybe 30%, as a guess) and even Olympus cannot dodge that fact of physics.

Then there is the sensor. If you present a 10mp sensor with an image with resolution equal to 10mp, you will be lucky to get 6-7mp out of it. Smaller sensors with higher pixel density are much worse, maybe only 5mp and less. They are not the neat lines of individual pixels and micro lenses we see in diagrams, under the microscope they irregular splodges with light spilling out all over the place, signal-to-noise being lost all the time, detail eroded. Bigger sensors are just better, more efficient.

That's where I think the difference between formats lies, but as I say, the theory is all very well, it's practise that counts. Comparative tests are very easy to do, and the differences clear to see.

Well I'm glad you're happy with your investment.... I think we shall agree to differ.

 

[Kiwi Paul]

I must admit I am curious of the real world differences between FF and 4/3 i.e. various different types of shots taken with both systems of the same scene at the same time, shot in RAW, processed in LR and printed on A3. Mostly landscapes, but any other stuff (abstract, architecture i.e. castles etc).
I am contemplating hiring a D700 (or D3) with the 24-70 f2.8 lens (which is the Nikon equivalent of my Zuiko 14-35 f2 lens) one weekend and doing the comparison just to satisfy my curiosity.

Paul

 

[HoppyUK]

I must admit I am curious of the real world differences between FF and 4/3 i.e. various different types of shots taken with both systems of the same scene at the same time, shot in RAW, processed in LR and printed on A3. Mostly landscapes, but any other stuff (abstract, architecture i.e. castles etc).
I am contemplating hiring a D700 (or D3) with the 24-70 f2.8 lens (which is the Nikon equivalent of my Zuiko 14-35 f2 lens) one weekend and doing the comparison just to satisfy my curiosity.

Paul

Nooo! Don't do that unless you are fully prepared for the consequences.

I was very happy with crop format until I did

 

[AndyElliott]

I must admit I am curious of the real world differences between FF and 4/3 i.e. various different types of shots taken with both systems of the same scene at the same time, shot in RAW, processed in LR and printed on A3. Mostly landscapes, but any other stuff (abstract, architecture i.e. castles etc).
I am contemplating hiring a D700 (or D3) with the 24-70 f2.8 lens (which is the Nikon equivalent of my Zuiko 14-35 f2 lens) one weekend and doing the comparison just to satisfy my curiosity.

Paul

That would be very interesting - more or less repeating the comparison in http://brandoneu.blogspot.com/2009/09/my-personal-review-of-zuiko-digital-ed.html, but with real examples. There are at least 3 good friends who have moved from 40D (and actually one from a 20D) to the 5Dmk2 and have been very happy (apart from the autofocus...). That said, even up to EAF-standard maximum of 50cm x 40cm, there aren't that many shots where the 5D has been able to give an edge. I'm sure they're there, but not ubiquitous. I think at least one of them is wondering whether the 7D might have been a better choice as she does a lot of natural history - the reach, AF and fps are the things that are attractive to her.

Andy

 

[AndyElliott]

Nooo! Don't do that unless you are fully prepared for the consequences.

I was very happy with crop format until I did

You would have to make Paul very happy indeed with FF to give up the 14-35, 35-100 and 90-250 - they are in a class of their own.

 

[Kiwi Paul]

Well this is my 100th post on this forum, so at least if I do decide to switch to ff I'm now eligible to sell stuff on here ;-) lol.

Paul

 

[AndyElliott]

Well this is my 100th post on this forum, so at least if I do decide to switch to ff I'm now eligible to sell stuff on here ;-) lol.

Paul

I've clearly got into too many arguments - where did 308 (sorry, now 309) come from?