PDA

View Full Version : F/1.8 Zoom lens!!



arthurking83
19-04-2013, 6:08am
And it's about bloody time too!

Sigma 18-35mm f/1.8 for APS-C (http://www.sigmaphoto.com/product/18-35mm-f18-dc-hsm-art?link=april_m_1_new)

One of the biggest issues with APS-C was that it promised smaller or faster lens designs.
Well at least someone has taken the initiative and provided one of those advantages.

Not sure why they started at this focal length tho .. I would have thought that something a bit longer, in the 24-50 range would have been more appropriate to begin with.

As I suspected they would .. the thirdparty manufacturers would innovate in this area, just like they seem to be doing with lens designs in general. The big two are rehashing old design types with very little or no innovative concepts.

Although it must be said that Olympus has had f/2 zoom lenses for a long time now .. technically they're not quite as innovative as they could have been! The 4/3rds system is even smaller than APS-C so an f/2 zoom lens on 4/3rds is not quite as challenging as an f/1.8 lens on APS-C.

I've finally migrated to the larger 35mm format, so it's come a bit too late for me, but I'm still hoping this one is a great design and produces some good empirical data.

ameerat42
19-04-2013, 8:46am
Saw this on DPR yesterday, AK. It looked well-specificated, I thought. Now I just wonder how much it will be.
:(?m.

Sar NOP
19-04-2013, 10:41am
It's MTF looks impressive : :eek:

Rattus79
19-04-2013, 10:46am
Huh! This could be the start of something beautiful...

Hopefully it performs well too!

ameerat42
19-04-2013, 11:30am
It's MTF looks impressive : :eek:

Blast! I wasn't going to ask you, Sar NOP, but I cannot find the MTF chart ANYwhere!!! (Esp. not on the Σ site.)
:(m.

Lance B
19-04-2013, 11:30am
And it's about bloody time too!

Sigma 18-35mm f/1.8 for APS-C (http://www.sigmaphoto.com/product/18-35mm-f18-dc-hsm-art?link=april_m_1_new)

One of the biggest issues with APS-C was that it promised smaller or faster lens designs.

Don't know how they were going to provide faster lens designs and they won't necessarily be smaller, either. Laws of physics dictate that whatever the focal length, the aperture is going to be the ratio of that focal length regardless of format. I know the initial thought was that APS C was going to hopefully give us smaller lenses but this was only realised at the longer focal lengths where, say a 200mm lens on FF is "equivalent" to a 135mm on APS C. However, if you want to get the same narrow DOF on APS C, you need to have the APS C lens one stop faster so, taking the example of the 200mm lens on FF and it was f4, then you require the 135mm lens to be f2.8 on APS negating the benefit of APS C somewhat.

At the wide angle end, the APS C lenses are generally at least as large as their FF counterparts basically because of the fact that they are using the same lens mount distance from the days of film and they have to have special lens designs which mean that wide angle APS C lenses do not end up being any smaller.

The fact is, this lens has a very small zoom range of only a 2 x zoom ratio and a 18-35 zoom range and it is still relatively large and heavy considering it's small zoom range when all is said and done so, this shows that to make it a larger zoom range would mean a lens that is very much larger and heavier and would introduce other compromises as well, like way more distortion, vignetting, aberrations and of course that large weight and size penalty. An APS C 18-35 f1.8 lens on FF would be "equivalent" to a 28-52 f2.5 on FF, hardly a stressing focal length range or a stressing wide open aperture either and therefore would be at least a similar size if not smaller than the APS C counterpart. Looking at previous Nikon lenses, there is a 28-50 f3.5 lens, unfortunately it is manual focus and only f3.5 so, not really a comparison, but it is tiny and light and does show that it is not a focal length that is used all that much for zooms as there are very few made in that range and certainly none at f2.5 due to this reason. One really wonders at the value of such a lens on APS C, especially considering that where you generally want wide apertures like f1.8 you generally are using it on potraits or other applications where you want a narrow DOF for isolation purposes and at 35mm on APS C, this is not really a portrat length.

As they say, there is no free lunch and for every perceived benefit of a system, there is an equal loss somewhere else.


Well at least someone has taken the initiative and provided one of those advantages.

Not sure why they started at this focal length tho .. I would have thought that something a bit longer, in the 24-50 range would have been more appropriate to begin with.

My thoughts as well. 50mm on APS C is closer to a portrait length and therefore where you would want narrow DOF or some other subjects requiring narrow DOF for isolation purposes.


As I suspected they would .. the thirdparty manufacturers would innovate in this area, just like they seem to be doing with lens designs in general. The big two are rehashing old design types with very little or no innovative concepts.

They rehash old designs because they cater more the professional areas and because they are the popular, most used and most useful focal lengths and zoom ranges. You even said youself that it you wonder why they used this zoom range and I have to ask the same thing.


Although it must be said that Olympus has had f/2 zoom lenses for a long time now .. technically they're not quite as innovative as they could have been! The 4/3rds system is even smaller than APS-C so an f/2 zoom lens on 4/3rds is not quite as challenging as an f/1.8 lens on APS-C.

I've finally migrated to the larger 35mm format, so it's come a bit too late for me, but I'm still hoping this one is a great design and produces some good empirical data.

ameerat42
19-04-2013, 11:43am
I think the main point of this lens is its speed - f/1.8 - and the fact that it is constant over the zoom range.
That means it would have to be looking through a wide piece of glass ALL the time (at max. aperture, that is).
And then it has to be retro-focus to project that wide-ish view into (even the APS-C) innards... So, size and weight.

As for its max. reproduction ratio - 1:4.3 - well, modest, but for the rest of what it can do I suppose it's OK.

I just hope it's price is "right".
Am.

WhoDo
19-04-2013, 5:02pm
Confuses the heck outa me! Here is a quote from Petapixel on the subject:
The lens, officially called the Sigma 18-35mm f/1.8 DC HSM, is geared toward APS-C format cameras. On a 1.5x crop sensor it’ll be the equivalent of a 27-52.5mm f/1.8 lens.


Read more at http://petapixel.com/2013/04/18/sigma-drops-bombshell-announces-a-18-35mm-f1-8-lens/#E4P3teqeQgDtrwAA.99
Now that reads to me like it's a lens for FF or why bother to talk about the crop factor in these terms? I would have thought that if it was designed for APS-C then it would be 18-35mm on APS-C and not suitable for FF. Have I misunderstood (I do that a lot these days)?

MattNQ
19-04-2013, 5:33pm
Sigma is on fire of late- the new 35/1.4 from last year is reportedly very nice and this one looks to be a cracker of a lens to be sure.

I'm thinking we'll see some more interesting lenses from them before the year is out.

ameerat42
19-04-2013, 7:29pm
Confuses the heck outa me! Here is a quote from Petapixel on the subject:
Now that reads to me like it's a lens for FF or why bother to talk about the crop factor in these terms? I would have thought that if it was designed for APS-C then it would be 18-35mm on APS-C and not suitable for FF. Have I misunderstood (I do that a lot these days)?

?Makes then sound knowledgeable to the un-knowledged? A red herring? Space filler? Nothing better to say?... etc? etc?
:rolleyes:

WhoDo
19-04-2013, 10:44pm
?Makes then sound knowledgeable to the un-knowledged? A red herring? Space filler? Nothing better to say?... etc? etc?
:rolleyes:
:confused013 :Doh::beer_mug:

Lance B
19-04-2013, 11:14pm
Confuses the heck outa me! Here is a quote from Petapixel on the subject:
Now that reads to me like it's a lens for FF or why bother to talk about the crop factor in these terms? I would have thought that if it was designed for APS-C then it would be 18-35mm on APS-C and not suitable for FF. Have I misunderstood (I do that a lot these days)?

No, it's an APS C lens, not a FF lens. It is saying that if it were to be the equivalent of a FF lens, then it would be a 27mm-52.5mm.

- - - Updated - - -


Sigma is on fire of late- the new 35/1.4 from last year is reportedly very nice and this one looks to be a cracker of a lens to be sure.

I'm thinking we'll see some more interesting lenses from them before the year is out.

The Sigma 35mm f1.4 is a real gem, one of the sharpest lenses I own, almost as sharp as the 85mm f1.4g or my 300 f2.8 VRII. It is a superb walk around general purpose prime with superb IQ, one of my favourite leses!

D800+ 35mm f1.4

http://www.pbase.com/lance_b/image/149667482/original.jpg


http://www.pbase.com/lance_b/image/149667496/original.jpg

http://www.pbase.com/lance_b/image/149311337/original.jpg

See the tiny little frog

http://www.pbase.com/lance_b/image/149311684/original.jpg

Here it is in crop

http://www.pbase.com/lance_b/image/149311686/original.jpg

MattNQ
20-04-2013, 12:04am
Some wonderful shots Lance - That lens is indeed superb on your D800.

Lance B
20-04-2013, 12:30am
Some wonderful shots Lance - That lens is indeed superb on your D800.

Thank you for your nice comments, Matt! :)

WhoDo
20-04-2013, 7:29am
No, it's an APS C lens, not a FF lens. It is saying that if it were to be the equivalent of a FF lens, then it would be a 27mm-52.5mm.
Understood, but what I don't get is why bother to do the comparison on a DX-only lens? Sure, give the equivalent FoV if the lens can be swapped from one to the other, but why bother when it's never going to be other than 18-35mm and APS-C only? How many FF photographers would even bother to buy or use a DX body-lens combo and also need to understand the equivalent FoV they'd be getting? :confused013

arthurking83
20-04-2013, 9:13am
Don't know how they were going to provide faster lens designs and they won't necessarily be smaller, either. Laws of physics dictate that whatever the focal length, the aperture is going to be the ratio of that focal length regardless of format. .......

In this instance the laws of physics have nothing to do with the minaturization of lens designs.

The idea that APS-C format cameras require smaller lenses is simply because of the smaller format. ... In a way, it's a similar notion to lens design for mirrorless(ie. rangefinder) cameras.

That is, especially at the shorter focal lengths, the faster aperture wider angle lenses require less extreme optics, and hence smaller designs.

Measure the front element of your 35/1.4 .. I bet you that the front lens element is going to be a lot larger than 25mm in diameter!!
It's probably more like 50 or more mm which technically indicates that the lens is actually capable of being a 35mm f/0.75 lens as a minimum focal length/front lens diameter.
While the front lens element is not necessarily the entrance pupil in every lens design known, it's a close enough approximation.

The reason the front element needs to be so so large on faster aperture lenses is ensure that as many rays of light are used in the formation of the image at the entrance pupil .. which then transpires that these rays of light on the formed image then create an image of higher quality at the sensor.
(well that's the basic theory).

So to put it into perspective ... your 35mm f/1.4 lens should have a front diameter of only 25mm .. and hence only require a 25mm front filter thread size, but the reality is very much different.
If actual lens designs approached theoretical lens designs requirements of manufacture, then there would be no allowances for APS-C lenses to become smaller either.

Lance B
20-04-2013, 10:44am
In this instance the laws of physics have nothing to do with the minaturization of lens designs.

The idea that APS-C format cameras require smaller lenses is simply because of the smaller format. ... In a way, it's a similar notion to lens design for mirrorless(ie. rangefinder) cameras.

That is, especially at the shorter focal lengths, the faster aperture wider angle lenses require less extreme optics, and hence smaller designs.

Measure the front element of your 35/1.4 .. I bet you that the front lens element is going to be a lot larger than 25mm in diameter!!
It's probably more like 50 or more mm which technically indicates that the lens is actually capable of being a 35mm f/0.75 lens as a minimum focal length/front lens diameter.
While the front lens element is not necessarily the entrance pupil in every lens design known, it's a close enough approximation.

The reason the front element needs to be so so large on faster aperture lenses is ensure that as many rays of light are used in the formation of the image at the entrance pupil .. which then transpires that these rays of light on the formed image then create an image of higher quality at the sensor.
(well that's the basic theory).

So to put it into perspective ... your 35mm f/1.4 lens should have a front diameter of only 25mm .. and hence only require a 25mm front filter thread size, but the reality is very much different.
If actual lens designs approached theoretical lens designs requirements of manufacture, then there would be no allowances for APS-C lenses to become smaller either.

My comment about aperture to focal length ratio wasn't specifically directed at 18-35, I was remarking on your thoughts that people believed that APS C lenses would be smaller. I am completely aware that wide angle lenses are a different kettle of fish due to their wide nature and require large front elements in order to encompass a wide view, maybe a bit of point scoring are being had by you? I mean, I do own the 14-24 f2.8 lens which has a bulbous front element probably 100mm wide and obviously has nothing to do with focal length to aperture ratio, please don't insult my intelligence. :rolleyes: The fact is, wide angle lenses for APS C are not really smaller and I alluded to this when talking about the lens mount distance which was a leftover from the film days which meant that they have to have special lens designs so as to cause less vignetting and other factors. This generally makes them large than they need to be. Olympus was one of the few that redesigned their entire format when they went 4/3rds and thus could have a lens mount distance that was suited to the 4/3rds format and could, thoerically, make their wide angle lenses smaller than their FF counterparts. However, the other thing to consider again is that to be a true equivalent, they need to have an aperture which is two stops faster than their FF counterparts and for APS C, this needs to be 1 stop faster than their FF counterparts to be a true equivalent.

zollo
20-04-2013, 12:36pm
Faster lens are said to help the af sensors do their thing especially as light dims. I could see a use for a wide-ish lens with f/1.8

I @ M
20-04-2013, 1:09pm
Image samples (http://lcap.tistory.com/entry/Sigma-ART-18-35mm-f18-Preview) look pretty good, Sigma are on a bit of a roll lately.



How many FF photographers would even bother to buy or use a DX body-lens combo and also need to understand the equivalent FoV they'd be getting? :confused013

I stick my hand up as one, I don't believe that the APSC format is anywhere near dead yet and seeing as we own a couple of capable DX bodies to work with relatively cheap dedicated APSC lenses they make good sense as travel cameras.

swifty
20-04-2013, 1:33pm
I for one welcome this lens spec although its not particularly useful to me as I no longer have any investment in the APS-C format.
Sure its a little large but the usefulness of a constant f1.8 zoom cannot be underestimated.
For a street photographer, the zoom range is incredibly useful. How many threads are there debating 28mm vs 35mm vs 50mm (FF equivalent) as the best street focal length. You get all three and everything in between at f1.8 with this lens.

I hope companies like Olympus and Panasonic sit up and take notice. A similar equivalent FL lens for m43 would sell like hot cakes. Make it constant f1.4 and the internet's gonna go beserk!!

Sifor
20-04-2013, 1:48pm
For a street photographer, the zoom range is incredibly useful. How many threads are there debating 28mm vs 35mm vs 50mm (FF equivalent) as the best street focal length. You get all three and everything in between at f1.8 with this lens.

Very good point - this could easily be the go to lens for street photographers..

Never know, this might just be the start of something new in fast zoom lenses...

Lance B
20-04-2013, 3:18pm
Sigma has also taken a big advancement leap with their Lens Dock technology, so that you can update latest firmware for the lens and adjust things like AF fine tune, but more specifically, AF fine tune at different zoom settings. Quite the innovation for a 3rd party lens maker:

http://www.sigma-global.com/en/lenses/cas/product/usb.html

Even though it doesn't list the 18-35 on the linked site for the Dock, it does mention it at the bottom of the info page for the 18-35.

http://www.sigma-global.com/en/lenses/cas/product/art/a_18_35_18/features.html#features06

WhoDo
20-04-2013, 3:44pm
I stick my hand up as one, I don't believe that the APSC format is anywhere near dead yet and seeing as we own a couple of capable DX bodies to work with relatively cheap dedicated APSC lenses they make good sense as travel cameras.

Not sure these will qualify as "relatively cheap", Andrew, but I do take your point. :th3:

cupic
20-04-2013, 4:05pm
Some samples and results
http://lcap.tistory.com/entry/Sigma-ART-18-35mm-f18-Preview

ameerat42
20-04-2013, 4:10pm
Ta for that, Cupic. So that where Sar NOP got the MTF charts from!
Am.

Lance B
20-04-2013, 6:41pm
Ta for that, Cupic. So that where Sar NOP got the MTF charts from!
Am.

I think he would have got it from here like everyone else:

http://www.sigma-global.com/en/lenses/cas/product/art/a_18_35_18/features.html

ameerat42
20-04-2013, 7:58pm
Ta Lance. That day I couldn't find this. Now I can't get away from it!
:om.

arthurking83
20-04-2013, 10:17pm
...... The fact is, wide angle lenses for APS C are not really smaller and I alluded to this when talking about the lens mount distance which was a leftover from the film days which meant that they have to have special lens designs so as to cause less vignetting and other factors .......

OK ... not really sure what you're referring to here Lance but lets put some figures into perspective here:

Sigma 30/1.4(APS-C only): Dimensions (Diameter x Length) φ74.2mm x 63.3mm
Nikon 35/1.4(Fullframe): Diameter x length Approximately 83 x 89.5 mm

APS-C a lot smaller(mainly in percentage terms)


Nikon 16-35/4VR: Diameter x length Approximately 82.5 x 125 mm
Tamron 17-50/2.8VC: LENGTH 94.5mm (no diameter given but filter diameter is 72mm, and the lens diameter looks to be very close to this .. maybe 2mm more)

Note that the Nikon lens is a much less extreme optic in that it's an f/4 .. one stop slower than the Tammy. Sigma's equivalent is about the same size as the Tammy.

At the ultra wide angle end of the scale, we get:

Tokina's 11-16/2.8: Dimensions 84x89mm
Nikon's 14-24/2.8: Dimensions 98x132mm


Pretty much as far as I can see .. the vast majority of APS-C only lenses are in fact on average about 15% smaller in physical dimensions(except weight) to their full frame counterparts.

Apart from the Sigma 50-150mm f/2.8, I doubt that we will ever see any lenses above about 50mm for APS-C sensored cameras(Nikon also has it's 85mm f/3.5 micro), as it appears to be a counterproductive exercise by the manufacturer.
Due to the extreme nature of the designs of wider angled lenses on an SLR format camera it makes sense for manufacturers to produce APS-C only(or 4/3's or mirrorless) lenses .. but once the lens design exceeds the 50mm focal length it seems to require much less in the way of design compromises .. so why limit the longer focal length lens to APS-C only when a larger market bekons!


So one of the not so commonly mentioned benefits of this lens(provided that it does perform as well as the specs say it will), is that a faster aperture zoom lens is possible at the wider end of the focal range and in a size that is relatively small!
Relatively small in comparison to similar focal length lenses with slower aperture values(i.e Nikon 16-35/4 17-35/2.8 , etc .. Canon 16-35/2.8 17-40/4 .. etc, etc) and this lens is an f/1.8!

I'm not sure where you're seeing this point that you've made that APS-C wide angle lenses are not smaller than full frame capable lenses of similar specs? :confused013

Lance B
20-04-2013, 10:40pm
OK ... not really sure what you're referring to here Lance but lets put some figures into perspective here:

Sigma 30/1.4(APS-C only): Dimensions (Diameter x Length) φ74.2mm x 63.3mm
Nikon 35/1.4(Fullframe): Diameter x length Approximately 83 x 89.5 mm

APS-C a lot smaller(mainly in percentage terms)


Nikon 16-35/4VR: Diameter x length Approximately 82.5 x 125 mm
Tamron 17-50/2.8VC: LENGTH 94.5mm (no diameter given but filter diameter is 72mm, and the lens diameter looks to be very close to this .. maybe 2mm more)

Note that the Nikon lens is a much less extreme optic in that it's an f/4 .. one stop slower than the Tammy. Sigma's equivalent is about the same size as the Tammy.

At the ultra wide angle end of the scale, we get:

Tokina's 11-16/2.8: Dimensions 84x89mm
Nikon's 14-24/2.8: Dimensions 98x132mm


Pretty much as far as I can see .. the vast majority of APS-C only lenses are in fact on average about 15% smaller in physical dimensions(except weight) to their full frame counterparts.

Apart from the Sigma 50-150mm f/2.8, I doubt that we will ever see any lenses above about 50mm for APS-C sensored cameras(Nikon also has it's 85mm f/3.5 micro), as it appears to be a counterproductive exercise by the manufacturer.
Due to the extreme nature of the designs of wider angled lenses on an SLR format camera it makes sense for manufacturers to produce APS-C only(or 4/3's or mirrorless) lenses .. but once the lens design exceeds the 50mm focal length it seems to require much less in the way of design compromises .. so why limit the longer focal length lens to APS-C only when a larger market bekons!


So one of the not so commonly mentioned benefits of this lens(provided that it does perform as well as the specs say it will), is that a faster aperture zoom lens is possible at the wider end of the focal range and in a size that is relatively small!
Relatively small in comparison to similar focal length lenses with slower aperture values(i.e Nikon 16-35/4 17-35/2.8 , etc .. Canon 16-35/2.8 17-40/4 .. etc, etc) and this lens is an f/1.8!

I'm not sure where you're seeing this point that you've made that APS-C wide angle lenses are not smaller than full frame capable lenses of similar specs? :confused013

You're not comparing equivalent FOV. You need to compare a 24mm APS C lens with a 35mm FF lens, or a 14mm APS C lens with a 20mm FF lens etc. Unfortunately, Nikon has made ver few DX only lenses and the ones they have made are not really a direct equivalent, ie, some having different max apertures and some don't have VR.

A reasonable comparision is possibly the Nikon 24-70 f2.8 and the 17-55 f2.8, where;

24-70 = 83mm x 133mm 900gms and 77mm filter thread
17-55 = 86mm x 111mm 755 gms and 77mm filter thread.

To me these lenses are not all that different considering that the APS C sensor is 50% smaller than that from the FF sensor. However, we don't really have an exact equivalent for the 24-70 in APS C as it would really need to be an f2 17-55 lens to be a real equivalent and just imagine how big that would be!

Unfortunately, there are not many real equivalent DX (APSC) Nikon lenses to fully compare, but there is one DX lens we can use, the 35mm f1.8 and can compare it to the Nikon 50mm f1.8G:

50mm f1.8G = 72mm x 52.5mm 185gms and 58mm filter
35mm f1.8 DX = 70mm x 52.5mm 200gms and 52mm filter.

As you can see, the DX lens is no smaller than the FX lens for an equivalent FOV and again, to be a real equivalent it should be an f1.2-f1.4 lens!

Luckily, Pentax has made some APS C only lenses for their DSLR's and yu can do a few direct comparisons, like:

FA20mm f2.8 (FF) = 70mm x 44mm 255gms and 67mm filter
DA14mm f2.8 (APS C) = 84mm x 69mm 420gms and 77 filter!

As you can see, the 14mm is quite a bit bigger than it's FF counterpart!

arthurking83
20-04-2013, 11:11pm
You're not comparing equivalent FOV. ......

Well, in this case you'll never be happy then.

Your original comment was:

".... Laws of physics dictate that whatever the focal length, the aperture is going to be the ratio of that focal length regardless of format .... "

focal length and the ratio.

And now you want FOV equivalence too.

As you may well know, this is impossible to achieve on two different formats.


I agree that there are some lenses that have been designed for the smaller format that don't necessarily come in a smaller physical form than their larger format equivalent cameras .. and to be honest the comparison between the Nikon 17-55 and their 24-70 is actually a lame one on two counts;

1. the 17-55 actually can cover the full 135 format at the longer focal lengths
2. the 17-55 should be compared to the more equivalent (old) 28-70/2.8 which are of a similar era of design!


Anyhow, I now have to bow out of a discussion I can no longer comprehend.

Lance B
21-04-2013, 1:40am
Well, in this case you'll never be happy then.

Your original comment was:

".... Laws of physics dictate that whatever the focal length, the aperture is going to be the ratio of that focal length regardless of format .... "

focal length and the ratio.

And now you want FOV equivalence too.

What are you on about? I was always saying that you have to have FOV equivalence, as this whole discussion was about equivalence between APS C and FF. To make it fair on both formats, they have to be seen as equivalent in FOV ie focal length equivalents for FF and APS C, so a 14mm lens on APSC is equivalent to 21mm on FF which will give the same FOV, like a 35mm on APS C will give almost the same FOV as a 50mm on FF etc. I don't know why you would not think that I was somehow differentiating focal length and FOV.

What I said about ".... Laws of physics dictate that whatever the focal length, the aperture is going to be the ratio of that focal length regardless of format .... " is still exactly correct, the aperture is the ratio of focal length to actual aperture size and was relevent when considering that the equivalent APS C lens really needs to have a wide open aperture 1 stop faster than it's FF counterpart, so a 14mm f2.8 lens on APS C need only be 21mm f4 only FF for the same DOF wide open.


As you may well know, this is impossible to achieve on two different formats.

Really? 14mm FOV on APSC isn't the equivalent of 21mm FOV on FF? Last time I looked it was almost exactly the same FOV.



I agree that there are some lenses that have been designed for the smaller format that don't necessarily come in a smaller physical form than their larger format equivalent cameras .. and to be honest the comparison between the Nikon 17-55 and their 24-70 is actually a lame one on two counts;

1. the 17-55 actually can cover the full 135 format at the longer focal lengths

But it isn't because you get vignetting at the wider angles, how can that be construed as being designed to be used on FF???? It is not a FF lens!!! and cannot be considered as one, it is a DX lens, period and it was designed as one.


2. the 17-55 should be compared to the more equivalent (old) 28-70/2.8 which are of a similar era of design!

Rubbish and irrelevent. The Nikon 28-70 f2.8 was designed and first built in 1999. The Nikon 17-55 f2.8 was first designed and bult 2004, so was a whole 5 years later in development. The 24-70 f2.8 was only 2007, just 3 years after the 17-55, both designed in the digital era, so they are a lot closer in design to each other than the 28-70.


Anyhow, I now have to bow out of a discussion I can no longer comprehend.

I am sorry you can't follow simple logic and equivalence.

The Sigma 18-35 f1.8 APS C lens would be equivalent in FOV and focal length and max aperture to a 27mm-52mm f2.5 FF lens. Hardly a stressful design for FF.

arthurking83
21-04-2013, 11:24am
FWIW: the Nikon 35/1.8 Dx lens has the ability to form an image almost over the entire full frame sensor .. even tho it's design is specific to the APS-C format.

If we take into account lenses that are designed more closely for their respective format sizes, we see that the majority of these lenses are smaller than their full frame equivalents.

Where you can quote examples where they are the same or larger, I can equally find more examples where they are smaller by a fair margin .. so the debate is only going to go round in an endless circle.

But what is important in these lenses is how much of an image circle they actually form!

It's obvious that for some manufacturers a specific level of quality is of paramount importance, and they design a specific lens in a specific manner.
This point doesn't mean that the APS-C lens isn't going to be smaller due to the laws if physics of lens requirements

That is, that the APS-C 35/1.8 is not smaller than the full frame 50/1.8 or f/2.0 of f/4 doesn't mean that it shouldn't be. The fact is that they can be(as exampled by Sigma's 30/1.4) and that the manufacturer simply hasn't troubled themselves with the size of the lens for APS-C.


Nikon do this themselves all the time too even with their full frame lenses.
For example again. Look at Nikon's f/2.8 prime lenses in the 24 - 60mm range and they're all a heck of a lot smaller than the Nikon 24-70/2.8 lens
Each one of those prime lenses in the F/2.8 range and between 24mm and 60mm(that I know of) are all a lot smaller than the zoom lens of equivalent focal length .. be that the 28-70 or 24-70mm

Do we at least agree on this?

If so, then what of the 14-24/2.8 vs the 14/2.8 prime lens.
Both of which are in fact quite close in physical size .. where the norm for zoom lenses is that they will generally be 2x or 3x large than a prime lens of similar focal length!
This is obviously a design consideration taken by a manufacturer in developing the lens ... in this instance size must have been of utmost importance to Nikon in the design of the 14-24/2.8.


What I'm afraid you are confusing with the laws of physics and lenses designed for the smaller format has in fact more to do with the design brief of the manufacturer.

As I've already said too tho.. up to a certain focal length point there is no point for a manufacturer to produce a lens specific to the APS-C format .. and by that I'm referring to most manufacturers except Olympus who only have vested interests in the 4/3rds format.

But for Sigma to produce an APS-C only 135mm f/2.8 is a wasted effort, as it seems to be almost as easy to make it encompass the full frame format anyhow .. and they'd be mad not to design it for both formats.

This focal length point appears to be roughly in the 50-100mm range as this is where lenses specific to the APS-C format cease to exist.

Except for the Sigma 50-150/2.8 which is a strange lens. Possibly a marketing exercise by Sigma .. or whatever.. this lens is strangely large for the physical specs it supposedly has.
It is as large as their 70-200/2.8(both lenses with OS) ... which in a theoretical sense makes absolutely no sense.
But if you read the test review for the 50-150/2.8, you can easily analyse the design compromises Sigma has made for this particular lens ...
Comments such as "The Sigma 50-150mm ƒ/2.8 OS is quite simply one of the sharpest zoom lenses we have had the opportunity to test ... " and " .... Stopping down the lens provides practically insignificant improvements in image sharpness: this lens is just excellent at every setting." clearly point to a design where no compromises existed in the ability of the lens!
This obviously precluded minimising the physical size in the design brief for that lens.

So, I think you have confused isolated instances of design compromises that a specific manufacturer has taken into account with the reality of lenses for the real world.

If this wasn't the case then the argument would be much more simple .. no lens designed for specifically for APS-C would exist in a smaller physical size than it's full frame counterpart .. and yet they do!

And this lens is one of those instances.
Why? because we know that to design a wide angle lens with a large aperture is increasingly hard. Forget FOV .. we're now referring strictly with the physical limitations of the lens design and focal lengths!

We know that a 18-35 with an aperture of f/4 is going to have a set physical dimension witnessed by the Nikon 16-35 .. and estimated with the Nikon 18-35/3.5-4.5 .. also the Nikon 17-35/2.8.
We know the sizes of these lenses they exist.
We also know from previous discussions that as the aperture is designed to be faster these lens types grow significantly.

Yet Sigma's new lens is physically smaller in every aspect. That the lens is not significantly smaller is a moot point .. the size of the aperture here is the important point .. it's supposedly 1 1/3stops faster than the slowest of the Nikon lenses mentioned.
Look at any other lens type where the aperture speed increases by 1 and 1/3 stop and the faster lens grows in size by at least 2x .. usually more.

I'm hoping that you're getting the picture here.
I fully understand what you're trying to express. But as already said .. you can't compare the lens designs in a strict manner the way you have due to the nature of the two formats.


Put simply if lens design was as simple as the theory of physics .. a 16-35mm f/4 lens should have a physical size of only 35mm length x 8.75 mm diameter.
Whereas the 18-35/1.8 should be 35mm length x 10mm diameter as maximums .. but this doesn't exist in the real world and the lenses real sizes are much more different as we know.

Yet the theoretically larger Sigma 18-35/1.8 is actually smaller than the Nikon 16-35/4 ... once again this is due to what the manufacturer determines to be of importance in the properties of the lens for it to be successful as a marketable product .... and this lens succeeds in proving that APS-C lenses can be made smaller if the manufacturer tried harder.

Going back to your argument that FOV equivalence needs to be factored into the equation, a quick look at the Olympus 300/2.8 compared to the Nikon 600/4 .. which by all accounts are theoretically equivalent.
The Olympus is significantly smaller than the 600 Nikon which provides FOV equivalence. But the two lenses can't really be compared as they are different lenses .. providing different levels of magnification.

You need to keep the argument restricted to one variable only when it comes to lenses for use on different formats. Once you jumble the argument up with two different variables, it becomes confusing and you get nowhere with it.



So what you are saying with respect to focal length FOV and aperture ratio is that a 14mm lens on APS-C provides the same magnification as a 21mm lens on full frame as well?

..... or is FOV the only possible measure that a lens is judged on? :confused013


Anyhow like I said.... with the introduction of multiple variables, it's difficult if not impossible to keep up with what you want to argue about .. it has nothing to do with logic.

Your logic will dictate that a 14mm lens on APS-C will provide not only the same FOV, but that it also has the same magnification as the 21mm lens on full frame.

This is logic that I don't understand then.

Lance B
21-04-2013, 1:41pm
FWIW: the Nikon 35/1.8 Dx lens has the ability to form an image almost over the entire full frame sensor .. even tho it's design is specific to the APS-C format.

If we take into account lenses that are designed more closely for their respective format sizes, we see that the majority of these lenses are smaller than their full frame equivalents.

Only if we take the tele end into consideration.

In my initial post in this thread, if you care to re-read it, I actually said: "but this was only realised at the longer focal lengths where, say a 200mm lens on FF is "equivalent" to a 135mm on APS C", so, I don't know why you are bringing this up. I did say a caveat, though, and this was that you actually need to factor in the 1 stop faster requirment of APS C to have the same wide open aperture DOF consideration as APS C has inherent 1 stop extra DOF due to the format size, I said: "However, if you want to get the same narrow DOF on APS C, you need to have the APS C lens one stop faster so, taking the example of the 200mm lens on FF and it was f4, then you require the 135mm lens to be f2.8 on APS negating the benefit of APS C somewhat. ". I don't think that is too difficult to understand, surely.


Where you can quote examples where they are the same or larger, I can equally find more examples where they are smaller by a fair margin .. so the debate is only going to go round in an endless circle.

I haven't seen too many, if any real equivalent AOS C lenses at the wide end that are smaller by any margin than their FF counterparts.


But what is important in these lenses is how much of an image circle they actually form!

Exactly! Now you're getting it!


It's obvious that for some manufacturers a specific level of quality is of paramount importance, and they design a specific lens in a specific manner.
This point doesn't mean that the APS-C lens isn't going to be smaller due to the laws if physics of lens requirements.

At the wide end, they basically are at least as big as their FF counterparts due to the design constraints of the old film camera lens mount that Canon/Nikon/Pentax etc used so as they didn't have to design a specific lens mount for APS C and therefore adopted the 35mm film lens mount. It is this very decision that has meant that they had to design the wide angle lenses in a specific way so as to avoid issues likelight drop off at the corners of the image circle and other desing considerations. For this reason, these wide angle lenses therefore did not realise the benefit of being smaller due to the sensor size. Do an internet serarch and you will see that I am correct in this matter. Olympus, however, did design their 4/3rds lens mount system from the ground up and they do have an advantage by generally beinfg able to desing smaller wide angle lenses than their FF conterparts (and APS C for that matter).


That is, that the APS-C 35/1.8 is not smaller than the full frame 50/1.8 or f/2.0 of f/4 doesn't mean that it shouldn't be.

Sorry, that is just plainly wrong. Why on earth would they desing the 35mm f1.8 lens so that it was bigger than required? There was no need to make it bigger to suit any other format as they already have a FF 35mm f1.8. No, they deigned the 35mm DX lens that size because it needed to be that size.


The fact is that they can be(as exampled by Sigma's 30/1.4) and that the manufacturer simply hasn't troubled themselves with the size of the lens for APS-C.

Really? The Sigma 30mm f1.4 is a APS C lens only and has a FOV of 45 degreees on APS C is 77mm x 59mm and weighs in at 430gms, the Nikon 50mm f1.4G (the closest equivalent for FF) has a FOV of 46 degrees (in otherwords almost the exact same FOV) on FF and is 74mm x 54mm and weighs in at 280gms. Seems as though the Sigma is larger to me.


Nikon do this themselves all the time too even with their full frame lenses.
For example again. Look at Nikon's f/2.8 prime lenses in the 24 - 60mm range and they're all a heck of a lot smaller than the Nikon 24-70/2.8 lens
Each one of those prime lenses in the F/2.8 range and between 24mm and 60mm(that I know of) are all a lot smaller than the zoom lens of equivalent focal length .. be that the 28-70 or 24-70mm.

Do we at least agree on this?

No, because it is a completely absurd comparison. You cannot compare a zoom with a prime.


If so, then what of the 14-24/2.8 vs the 14/2.8 prime lens.

Again, an absurd comparison.


Both of which are in fact quite close in physical size .. where the norm for zoom lenses is that they will generally be 2x or 3x large than a prime lens of similar focal length!
This is obviously a design consideration taken by a manufacturer in developing the lens ... in this instance size must have been of utmost importance to Nikon in the design of the 14-24/2.8.

Again, really? The Nikon 14mm f2.8 is 87mm x 87mm and 670gms, the Nikon 14-24 f2.8 is 98mm x 132mm and 10000gms! Quite a large difference if you ask em.


What I'm afraid you are confusing with the laws of physics and lenses designed for the smaller format has in fact more to do with the design brief of the manufacturer.

It seems as though I have pointed out that you are confusing the issue, not me. I have refuted each of your arguments.


As I've already said too tho.. up to a certain focal length point there is no point for a manufacturer to produce a lens specific to the APS-C format .. and by that I'm referring to most manufacturers except Olympus who only have vested interests in the 4/3rds format.

But for Sigma to produce an APS-C only 135mm f/2.8 is a wasted effort, as it seems to be almost as easy to make it encompass the full frame format anyhow .. and they'd be mad not to design it for both formats.

At the tele end, we are talking a different kettle of fish. If we think of the wide end, we need large fron elements in order to encompass the wide FOV. At the tele end, the laws of physics come into play with regards to focal length over aperture, so, if we have a 200mm f2.8 lens, it will require a front element of 200/2.8 = 72mm and this is regardless of format, ie whatever the size of the sensor, a 200mm lens will need a front element of at least 72mm. Therefore, to make a 200mm lens with an f2.8 aperture, it will be the same size regardless of format. The benefit for APS C, however, is that you can use a 135mm f2.8 lens to sort of be the equivalent of a 200mm f2.8 lens on FF. However, as I stated previously, to be a true equivalent, the 135mm APS C lens really needs to be a 135mm f2 lens if we consider the sytem as a whole and this means things like the DOF being narrower by 1 stop on FF and FF sensors having about 1 stop better ISO and DR when equivalent time frame release of sensors is considered (that's another discussion).


This focal length point appears to be roughly in the 50-100mm range as this is where lenses specific to the APS-C format cease to exist.

Because the laws of physics show that you can't make the lens smaller as the front element has to be what it has to be as I pointed out. The "advantage" is that you get to use a 135mm f2.8 lens on APS C as compared to 200mm f2.8 on FF, discounting the DOF considerations as also pined out above.


Except for the Sigma 50-150/2.8 which is a strange lens. Possibly a marketing exercise by Sigma .. or whatever.. this lens is strangely large for the physical specs it supposedly has.
It is as large as their 70-200/2.8(both lenses with OS) ... which in a theoretical sense makes absolutely no sense.
But if you read the test review for the 50-150/2.8, you can easily analyse the design compromises Sigma has made for this particular lens ...
Comments such as "The Sigma 50-150mm ƒ/2.8 OS is quite simply one of the sharpest zoom lenses we have had the opportunity to test ... " and " .... Stopping down the lens provides practically insignificant improvements in image sharpness: this lens is just excellent at every setting." clearly point to a design where no compromises existed in the ability of the lens!
This obviously precluded minimising the physical size in the design brief for that lens.

Not really relevent to the discussion.


So, I think you have confused isolated instances of design compromises that a specific manufacturer has taken into account with the reality of lenses for the real world.

LOL. It is not me that is confused. I have refuted all your theories. Do an internet search and you will see I am correct.


If this wasn't the case then the argument would be much more simple .. no lens designed for specifically for APS-C would exist in a smaller physical size than it's full frame counterpart .. and yet they do!

Basically only at the long end, which is what I said from my opening post in this thread (see my first responses in this post), and only if you take the equivalent in FOV/Focal Length disregarding DOF considerations with wide open aperture, ie, a 135mm f2 lens will not be smaller thana 200mm f2.8 lens. If we look at Canon's 135mm f2 lens it is 82.5mm x 112.0mm, 72mm filter size, 750g and their 200mm f2.8 lens is 83.2 x 136.2mm, 72mm filter size, 765g - so, you can see that they are almost identical in size! This means that on APS C, the 135mm f2 will have identical FOV and identical DOF when used wide open or one stop down etc to the 200mm f2.8 lens on FF.


And this lens is one of those instances.
Why? because we know that to design a wide angle lens with a large aperture is increasingly hard. Forget FOV .. we're now referring strictly with the physical limitations of the lens design and focal lengths!

You can't forget FOV as this is the resultant of focal length on a given sensor size!!


We know that a 18-35 with an aperture of f/4 is going to have a set physical dimension witnessed by the Nikon 16-35 .. and estimated with the Nikon 18-35/3.5-4.5 .. also the Nikon 17-35/2.8.
We know the sizes of these lenses they exist.
We also know from previous discussions that as the aperture is designed to be faster these lens types grow significantly.

Yes, and your point being? This is irrelevent unless you are comparing the APS C lens to the FF lens of the same FOV/focal length.


Yet Sigma's new lens is physically smaller in every aspect.

Compared to what? I do not know of a FF lens with the same zoomFOV/Focal Length with which to compare it to. It is an APS C lens with a variable focal length/variable FOV of 18-35 f1.8/76.5-44.2 degrees - this would need to be a FF equivalent of 27mm-52mm f2.5 and they are a rarity for FF in this range.


That the lens is not significantly smaller is a moot point .. the size of the aperture here is the important point .. it's supposedly 1 1/3stops faster than the slowest of the Nikon lenses mentioned.
Look at any other lens type where the aperture speed increases by 1 and 1/3 stop and the faster lens grows in size by at least 2x .. usually more.

Realistically, the FF equivalent is 27-52 f2.5, you have to remember the DOF considerations of APS C having 1 stop more DOF for each aperture. There is more to this discaussion that just the DOF considerations as it also ventures into APS C vs FF on things like high ISO noise, DR etc, but that is for another discussion. So, the aperture of the FF lens really only needs to be f2.5, ie f1.8 x 1.414 (square root of 2). To design a FF lens like this would be easy as it has basically already been done with many lenses from many manufacturers, 24-70 f2.8 is close enough to the f2.5 and yet covers a larger zoom range.


I'm hoping that you're getting the picture here.

:lol: That is really funny. I have a complete grasp of the situation, it is you that is missing the poiny. I have refuted all your arguments and proved them incorrect. As I say, do an internet search of this topic andyou will see I am correct.

quote]I fully understand what you're trying to express.[/quote]

If you understood, you wouldn't need to argue.


But as already said .. you can't compare the lens designs in a strict manner the way you have due to the nature of the two formats.

Well, actually you can and you must to have a level playing field. You cannot take one supposed advantage of one sytem yet ignore the disadvantages of that system.


Put simply if lens design was as simple as the theory of physics .. a 16-35mm f/4 lens should have a physical size of only 35mm length x 8.75 mm diameter.

I've said it once and I will say it again, it's a zoom. You cannot compare a zoom with a prime. SHEESH!! Bows head and shakes.


Whereas the 18-35/1.8 should be 35mm length x 10mm diameter as maximums .. but this doesn't exist in the real world and the lenses real sizes are much more different as we know.

Yet the theoretically larger Sigma 18-35/1.8 is actually smaller than the Nikon 16-35/4 ... once again this is due to what the manufacturer determines to be of importance in the properties of the lens for it to be successful as a marketable product .... and this lens succeeds in proving that APS-C lenses can be made smaller if the manufacturer tried harder.

I cannot believe that you are comaring a APS C lens to a FF lens of the same focal length when it is the FOV that needs to be compared. Quite unbelievable!! You have to compare the APS C 18-35 f1.8 lens to a FF lens of the same focal Length/ FOV and that would be a 27-52mm f2.5 lens!!! How many more times do I need to spell it out!!


Going back to your argument that FOV equivalence needs to be factored into the equation, a quick look at the Olympus 300/2.8 compared to the Nikon 600/4 .. which by all accounts are theoretically equivalent.
The Olympus is significantly smaller than the 600 Nikon which provides FOV equivalence. But the two lenses can't really be compared as they are different lenses .. providing different levels of magnification.


No, they do not they have the exact same FOV and almost the same magnification on their respective formats. Magnification difference only comes about by due to the different close focus of each lens and this is a design cinsideration by each manufacturere and nothing to do with the differing sensors. The difference, however, is very minimal to the point of being considered basically the same. The Oly 300 f2.8 has a minimum focus distance of 2.4mts, whereas the Nikon 600 has a close focus of 5mts, to be equal, the Oly would need to be 2.5mts due to the 2x conversion factor, and then you would find the magnification exactly the same. The thing you also need to remeber is that the 300 f2.8 lens from Olympus needs to be considered the equivalent of a 600mm f5.6 on FF due to it's 2x crop factor which is a 2 stop difference of the 4/3rds sensor.

This is why Olympus went with their top flight zooms as f2 lenses (which really should be f1.4 lenses) to equal the DOF considerations of a FF f2.8 lens. The Olympus f2 lenses, for DOF considerations (which bring into play the sensors for high ISO noise and DR), is really an f4 lens equivalent for FF and an f2.8 lens for APS C. DO the calculations yourself on the On-Line DOF calcualtor if you don't believe me:

http://www.dofmaster.com/dofjs.html

So a 100mm lens set at f2 and 10mts on an Olympus 4/3rds camera = .59mt total DOF. the "equivalent" 200mm lens set to f4 on FF at 10mts = the same .59mt total DOF.

You need to keep the argument restricted to one variable only when it comes to lenses for use on different formats. Once you jumble the argument up with two different variables, it becomes confusing and you get nowhere with it.


[quote]So what you are saying with respect to focal length FOV and aperture ratio is that a 14mm lens on APS-C provides the same magnification as a 21mm lens on full frame as well?

Magnification is a slightly different thing as that is a factor of subject to camera distance and/or close focus ability. What I am saying, and have been saying all along, is that FOV and focal length are linked for their respective formats. A 14mm APS C lens has the same or alomst identical FOV as a 21mm lens on FF. This is what you have to compare when comparing formats as well as maximum wide open aperture.


..... or is FOV the only possible measure that a lens is judged on? :confused013

FOV is a function of focal length on a given sensor size.



Anyhow like I said.... with the introduction of multiple variables, it's difficult if not impossible to keep up with what you want to argue about .. it has nothing to do with logic.

Seems logical to me and all the discussions I have read on the internet, a 14mm lens on APS C has the same FOV as that of a 21mm les on FF and yet the 14mm lens is physically larger. I pointed out many wide angle lenses that this applies to here as well, but they are just some of those that are out there when comparing equivalents. I'll go back to my oiginal point and that is, wide angle APS C lenses rarely give you a smaller lens when they equivalent FOV/focal length and aperures are compared. I also think you need to add the fact that they also need to be 1 stop faster for true equivalence. They are generally as big and in some cases larger than their FF counterparts.

Your logic will dictate that a 14mm lens on APS-C will provide not only the same FOV, but that it also has the same magnification as the 21mm lens on full frame.

Never said they give the same mginification as that is a slightly different thing as this is a factor of camera to subject distance and/or minimum focus, but if they are the same distcance form the respective cameras, then manification should be the same or near as damn it to each other.


This is logic that I don't understand then.

Then read what I have said and you will understand.

ameerat42
21-04-2013, 5:10pm
The latest news...

I've just heard that Σ are withdrawing this lens from sale due to the controversy it has generated.
(Apparently, L Tolstoy is seeking copyright infringement damages.)

...from the perpendicular universe