Why is f2.8 the "magic number"?

[bmm]

My previous posts show my interest in the development of a bag of good lenses.

 

One question always pops into my head which is why Nikon seems to have

selected f2.8 as the widest aperture for its pro stuff? Is there some

technical limitation that kicks in at that point?

 

I ask as most of my current enjoyment is coming from the really wide end of my

new 50mm f1.4 (although that may be a function of the low depth of field being

new to a previous P&S shooter like me, and the novelty may soon wear off!)

[juanjo_viagran]

Nikon have few big lenses F2 but they cost a REAL FORTUNE..

200mm F2 AFS and 300mm f2 AIS..

about the 2.8 in zoom I'll wait to hear from the knowledgeable..

 

but if Nikon make a zoom with f2 or lower, believe me, only few are going to be able to afford it..:/

[bob_wall1]

Weight versus light-gathering ability, especially with zoom lenses. Take what is their new, and apparently beautiful, 14-24 zoom. You can get prime lenses at least F2.0 in those lengths if not bigger apertures, and those aren't as big or heavy. But to get the best mix of light and weight, you have that monster. Heavy as can be at 4.2 lbs! Just imagine how big, heavy and EXPENSIVE an F2.0 14-24 zoom would be!

[heavyweather]

I see plenty of fast cinema zoom lenses that clock in at f/1.9 and around there. Why can't we

get zooms that groovy in still photography? Is it a technical limitation regarding the Nikon

mount, or would the lenses just be too expensive for Nikon's target market, or what?

[david_dorcich1]

"Is it a technical limitation regarding the Nikon mount"

 

No, f/number has little to do with the mount. Lower the f number the larger the front element of the lens has to be, larger front element heavier and more expensive the lens is.

[bjørn rørslett]

The format decides how big an f/2 lens needs to be. It's relatively easy to make the f/2 zoom lenses for the smaller cine formats. If people demand f/2 zoom lenses for FX systems, they need to accept the associated size *and* cost. The new 14-24 f/2.8 weighs around 1 kg and an f/2 version likely would be 3-4 times that and 8-10 times the cost.

 

I have shot a lot with the Nikkor 300mm f/2, and this is the proverbial boat anchor at 7.5 kg (without the hood and tripod mount), so massive it is almost impossible to carry in the field. It also caused me bad back pain. So, the 300/2.8 models suit me better. For the shorter 200mm, a 200/2 is more practical (I have several of these). A 105/2 is not big. A 50/2 is smaller. A 24/2 is also small. And most shorter lenses with a decent aperture will be bigger again.

[larry h-l]

Take a look at some of the prices for fast zoom lenses for film and video, lenses from

Angenieux and Cooke. An Angenieux 17-80 zoom will set you back about $50,000.

[albertdarmali]

It's not because they decided on the aperture 2.8

 

Like everyone says, it's mostly considering that 2.8 still make some sense in regards to the size and price.

 

I wonder if 70-200mm f/1.4 is possible to create, and whether we have to mount it on a Humvee to use it...

[michael_bradtke]

Caleb

 

Price some of those super speed zooms. People dont usually buy them they rent them. The ones that I have used for 35mm (Panavision) are HUGE!!

 

Michael

[frank uhlig]

Bernhard,

 

which other manufacturers have wider than f/2.8 zooms? Name some, please. I would like to know. (For 35mm film or crop sensor, of course, not in P+S and cinema lenses.) So why bash Nikon?

[lex_jenkins]

It's a compromise between performance, weight and price.

 

BTW, f/2.8 is only nominal. When tested (scientifically using proper bench testing equipment, not sheets of newspaper at home), few lenses are precisely f/2.8, f/1.4, etc. They usually vary a fraction of an EV from every nominal aperture throughout their range.

 

Same with focal lengths. 70mm on my 18-70 DX Nikkor does not match 70mm on my 35-70/2.8 AF Nikkor. In comparison, 70mm on the 18-70 lens is closer to roughly 60mm on the other lens.

 

It is possible to design faster lenses cost effectively, but only for smaller sensor formats. My Olympus C-3040Z P&S digicam has a (35mm equivalent) 35-105mm f/1.8-f/2.6 zoom, and the entire camera, lens and all, cost less than $700 new several years ago. Such a lens in DX format, let alone 35mm format, would cost many thousands of dollars and be very heavy.

[keith_leonin]

<i>So why bash Nikon? -- Frank</i>

<p><p>

Ahhh, Frank... missing the point yet again. Not bashing, asking... And since this is the N forum, it makes little sense to ask about C or P or O or even L with the red dot.

<p><p>

[randrew1]

Back in the olden days of 35mm film, f/2.0 was very common. I have a 1937 Retina IIa with an f/2.0 lens. I have a 50mm f/1.8 Nikon lens. They made f/1.4 and f/1.2 lenses. If you are talking about a digital camera, part of the problem is the mount. The mount is far enough from the sensor that wide angle lenses are "retro focus" lenses. I'm told that it is more difficult to make a wide aperture retro focus lens. (I'm not a lens designer.) With DSLR's, a 35mm normal lens falls into this range of retro focus lenses.

[bmm]

Frank - my apologies if you interpreted me as "bashing Nikon". I was not doing that at all. I am far too new to the hobby to do such a thing and for the record am absolutely delighted with my D80 and 50mm f1.4 - and only slightly less than absolutely delighted with my 18-200 VR.

 

It was just a humble, maybe naive question, based on the fact that I am looking forward to buying more good glass in the future, and also that I am really enjoying the wide open end of the 50mm (and therefore would be interested in any other lenses that get down to say f2 or below).

 

You may have seen my other posts on specific primes which accord with this, and I am looking to get a fast wide-angle (either Nikon 35mm f2 or Sigma 30mm f1.4) plus a fast prime tele (maybe an 85mm, 1055mm or even 135mm) to add to my kit.

 

But this sets me to questioning the speed limitation on zooms as well.

[mark_amos]

Bernard, I suspicion you already know the answer. The type of photography your are enjoying with your 50, a relatively light weight and compact lens, is not the type of photography where Nikon surveys an optimally profitable market. Beginners like slow cheap zoom lenses even if they have horrible distortion as long as they are sharp, and professionals don't seem to mind if their f2.8 lenses weigh 2 lbs because they want zooms also. Nikon is not making fast and compact dx-frame primes, and why should they? Everybody buys the other poop faster than they can make it? They probably figure you will learn your Photoshop "language" and fix your pictures yourself in whatever way that keeps you in their market segment.

[johnw63]

A lens at f2.8 is costly enough, outside the standard range ! I'd call it the 24mm to 180mm range. Even the 180 is not cheap. Optically, to go lower, would make them MORE expensive and apparently MUCH bigger. It's more of a optical/lens size thing than their ability to make one problem.

[bmm]

Mark - firstly I like your cynicism. My profession is as an economist/business strategist in a field where technology lifecycles and the "forcing" of consumer choice are important so yes, from that angle, I do know the game.

 

But... I am a beginner. Yet I do NOT like my "slow cheap zoom with horrible distortion" (the 18-200VR in my case - though that statement probably over-exaggerates its deficiencies!); or should I say it's ok but for the $1000 in retrospect I sometimes wish I'd made a different choice.

 

And thinking blue-sky, even a limited-range fast zoom - say hypothetically 24-85mm at f2.0 - would be a great "advanced" first lens and could be complemented in time with the 17-35, and then the 70-200 or 80-200 to cover the long end.

[woolly1]

I would add that in terms of light gathering advances Nikon appear to be attacking the cause from the opposite direction and improving the low-light capability of their bodies.... ie. the D300 & D3.

Higher ISO performance = faster shutter speeds.

 

Ever larger apertures also has a benefit/penalty of narrower DOF too.

 

In the longer term Nikon's money may better be spent on one body/sensor than several extremely fast and expensive lenses that no-one will buy perhaps. Building bigger lenses doesn't really expand the technology envelope whereas who knows what the limits of sensors and processors are.

 

 

Clive

[ilkka_nissila]

I think the main problem currently with (short) fast lenses (like f/1.4, f/2) is that they have relatively low MTF. The move towards smaller image capture area (35mm -> DX) has increased requirements for MTF since the pixels are smaller and post-blowup is greater. This means that fast lenses produce relatively poor image quality at wide apertures compared to what they would do on 35mm film.

 

On FX cameras fast lenses will become more exciting again because they'll perform better on them than on DX.

[oskar_ojala]

A fast lens requires a bigger entry pupil than a slow lens, hence it needs larger elements and becomes bigger. Larger elements mean that distortions are harder to correct, necessitating more elements that are more complex. This again adds weight and cost. Fast lenses are likely bought by people who want to use a large aperture, so the performance at those should be decent. Since the lens is big, heavy and expensive, the numbers made are small, so it gets even more expensive.

 

Movie camera lenses can actually be huge, since they are mounted on rather heavy support. They wouldn't actually need the highest quality either, since movies are usually seen from a relatively small projected image and the subjects move.

[robert_clark]

"The format decides how big an f/2 lens needs to be. It's relatively easy to make the f/2 zoom lenses for the smaller cine formats. If people demand f/2 zoom lenses for FX systems, they need to accept the associated size *and* cost. The new 14-24 f/2.8 weighs around 1 kg and an f/2 version likely would be 3-4 times that and 8-10 times the cost."

 

True, of course, but just take a look at what Olympus is doing. On their 4/3 half frame sensor cameras, the E3 etc you can use an F2/ 28-70 equivalent - same weight as the Nikon 28-70 and a stop faster. The same with the Olympus F2/ 70-200 equivalent, a little heavier but smaller and a whole stop faster.

 

Olympus lenses are also generally testing as higher quality than both Nikon and Canon.

 

If you move to their 150 F2, an F2/ 300mm equivalent, it is almost half the weight of either the slower Nikon 300/2.8 or the 200/2, which is also a 300/2 on the 1.5x sensor. The same goes for the Olympus F2.8/ 300, a 600/2.8 equivalent, much, much lighter than the slower Nikon 600/4 or the 400/2.8 (600/2.8 on a D300).

 

Then look at Olympus's F2.8/ 90-250 (180-500/2.8 equivalent zoom) and you have perhaps the most perfect wildlife/sports zoom yet made. Neither Canon, nor Nikon have anything like this. Add that to the very high build quality, leading optical quality, the most effective in-body anti-shake so far (5 stops), superb weather sealing, and what is claimed to be the fastest and most accurate auto focusing so far, and you have something to contend with.

 

If the IQ of the E3 is on a par with the Nikon D300 or the Canon 40D, a lot of photographers shooting sport and wildlife may well be tempted by this system. After all, in the field, weight counts for a lot as does low light shooting ability.

 

No doubt, though, Clive's point - "I would add that in terms of light gathering advances Nikon appear to be attacking the cause from the opposite direction and improving the low-light capability of their bodies.... ie. the D300 & D3. Higher ISO performance = faster shutter speeds." is also significant; but Olympus can also do that as well.

[Brad_]

It's a trade involving speed, quality, size, weight, and cost. Just like everything else in life...

[roland_vink]

As for why lens aperture are f/2.8 and not f/3 or some other similar number, it is part of a series starting with f/1.0 and reducing by the square root of two (one stop) at each step: 1, 1.4, 2, 2.8, 4, 5.6 ... There is really no reason why lenses have to have these aperture settings, they could equally be 1.2, 1.7, 2.4, 3.4, 4.8 ... but it's largely a matter of what people are used to.

[dan_lindsay]

With faster and faster digital SLRs on the market the days of being concerned about f-stops faster than 2.8 are essentially over. The Noct-Nikkor, for example, won't be produced again because you just turn up the sensitivity of the camera and you're there. Anything faster than 2.8 forces you into a tripod situation which defeats much of what 35mm cameras brought with them,--portability.

[lex_jenkins]

I wish it were that simple, Dan.

 

But ISO sensitivity has no influence on DOF. And it doesn't help composition or autofocusing in dim light. Even an f/2 or f/1.8 lens provides significant advantages over an f/2.8 lens in dim lighting.

 

There's no substitute for a fast lens, given the narrow parameters within which it excels.