• Log in with Facebook Log in with Twitter
Photo

next PZ lens test report - Sigma 14mm f/1.8 DG HSM ART


  • Please log in to reply
58 replies to this topic

#1 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,451 posts
  • LocationSydney, Australia

Posted 19 November 2017 - 09:36 AM

Depending on how you look at it ... it may be nice ... or not that great ...

 

http://www.opticalli...5-sigma14f18art


  • dave9t5 likes this
Chief Editor
photozone.de

#2 Rover

Rover

    Advanced Member

  • Members
  • PipPipPip
  • 1,604 posts
  • LocationRussia

Posted 19 November 2017 - 10:18 AM

Great stuff!.. However, I should point out that the Laowa 12mm lens mentioned in the intro isn't f/2, just f/2.8. :) If you need an example to replace it, there's the Samyang 14mm f/2.4 XP (or whatever it's called). Both this and the Irix are only 1/2 stop (in practice probably even 1/3, because the camera would probably round that aperture down to f/2.5?) brighter than the "norm" though, not 1 1/3 stop, so I'm not sure they really count at all. This SIGMA is totally unique.

#3 toni-a

toni-a

    Advanced Member

  • Members
  • PipPipPip
  • 1,314 posts
  • LocationLebanon

Posted 19 November 2017 - 11:11 AM

was this one taken on tripod ??

 

0F6A1014.jpg

 

 

if yes whay ISO 2400f1.8 otherwise at high pixel count , unless you haveextremely steady hands and did several tries I go for at least 1/2x focal shutter speed 1/15s is too slow for me at leasf



#4 dave's clichés

dave's clichés

    Advanced Member

  • Members
  • PipPipPip
  • 1,768 posts

Posted 19 November 2017 - 11:25 AM

Good well written review!  

 

   Thanks for the secondary review also, on a 22 Mps sensor, as per usual the 50Mps sensor does not flatter.

 

 

  Strangely the Samyang 14mm F2.8 seems to do better (other than distortion) with only a shortfall of one stop. BTW for those who use Adobe PS CS6/ACR 9.0,distortion correction for the Samyang can be found under the Sony A7III category.

 

   The Sigma here is not my bag, the AF is approximative, (so manual focus is almost a given at larger stops), and price/size/ weight are prohibitive!



#5 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,451 posts
  • LocationSydney, Australia

Posted 19 November 2017 - 09:52 PM

Great stuff!.. However, I should point out that the Laowa 12mm lens mentioned in the intro isn't f/2, just f/2.8. :) If you need an example to replace it, there's the Samyang 14mm f/2.4 XP (or whatever it's called). Both this and the Irix are only 1/2 stop (in practice probably even 1/3, because the camera would probably round that aperture down to f/2.5?) brighter than the "norm" though, not 1 1/3 stop, so I'm not sure they really count at all. This SIGMA is totally unique.

 

Ah, right, mixed the Laowa with their 15mm f/2.


was this one taken on tripod ??

 

0F6A1014.jpg

 

 

if yes whay ISO 2400f1.8 otherwise at high pixel count , unless you haveextremely steady hands and did several tries I go for at least 1/2x focal shutter speed 1/15s is too slow for me at leasf

 

That would have been an exotic tripod ;-)


  • Brightcolours likes this
Chief Editor
photozone.de

#6 Rover

Rover

    Advanced Member

  • Members
  • PipPipPip
  • 1,604 posts
  • LocationRussia

Posted 20 November 2017 - 07:21 AM

Two more things. The Specifications section describes the hood as "bayonet mount, built-in". I thought it's one or the other? This lens definitely has a non-removable hood. :) Also, you said that the MC-11 converter is for using the lens on the Sony and Sigma mirrorless cameras while in reality it's only about the Sony cameras - you can use it to adapt the Sigma / Canon mount lenses to a Sony camera (with whichever version of the adapter), not the Canon mount lenses to the Sigma. :) 



#7 JoJu

JoJu

    Advanced Member

  • Members
  • PipPipPip
  • 2,873 posts
  • LocationSwitzerland

Posted 20 November 2017 - 07:57 AM

Yes, it offers an insane max. aperture at this focal length but at least technically you won't find greatness at f/1.8. the corners are soft. Whether this matters at this aperture is a different question. Astrophotographers may have hoped for more (at least at 50mp) whereas it's a not an issue in shallow depth-of-field scenarios really.

 

Did you do a resolution test at  ∞ ? Otherwise I just ask how do you rate the lens for astrophotography? And a very sharp center very wide open is just not great enough? I don't want to talk away the soft borders, but I'd like to discuss ultra wide shots, available lights and the importance of sharp corners  ^_^

The stair house shot could not have benefitted of better corners - due to DoF the corners already are soft.

Emphasiszing something in the corners and a lot of blurred background? Don't know, so far I didn't feel my composition sucks because of weak corners.

 

However, I don't agree with the weak corners at ∞ and I know, you guys usually test at 25-50 × f which in that case would be 50 × 14mm = 700 mm distance between test chart and sensor. That's a bit bold to conclude from this distance to it's behaviour at distances suitable for astro.

 

i-MnSdccX-L.jpg

 

Here's the full res.

 

Typically, I use the lens mostly for this kind of subjects

 

i-N7gWzBv-L.jpg

 

or that one

 

i-VXNSRTn-L.jpg

 

Soft corners? I don't care much, I'm not using this lens for reproductions of flat objects wide open  :D

 

Of course I understand you're using the standards of PZ or OL for your tests and for all lenses, but there's were trouble starts: Not all lenses are needed/bought for the same purposes. Sharp corners are very interesting for a couple of subjects - just not for all and if it comes to wide angle lenses tested not even close to infinity, I scratch my head about the testing method.

 

What if the field curvatures radius approaches ∞ at ∞? You haven't tested that, otherwise you would have told us.

 

And the Laowa 15/2 is only made for Sony E mount...  :P



#8 Rover

Rover

    Advanced Member

  • Members
  • PipPipPip
  • 1,604 posts
  • LocationRussia

Posted 20 November 2017 - 08:58 AM

At any rate, this lens looks great, a vast improvement over the predecessor. I have and love the said predecessor, but it doesn't even come close to being sharp across the frame (APS-H mind you...) until ~oh say~ f/13 or so, and even then it depends on the orientation of the scene (a sign of strong field curvature). I haven't analyzed it formally of course, just had the time (7,5 years and counting) to get used to its quirks. 20 years of lens design haven't passed for nothing...

 

The price / size is another matter entirely. :) I'm not going to upgrade but if Sigma had released a wider prime - say, 12mm - I would've been more willing.



#9 JoJu

JoJu

    Advanced Member

  • Members
  • PipPipPip
  • 2,873 posts
  • LocationSwitzerland

Posted 20 November 2017 - 09:57 AM

I also was thinking about and got a sample to try very briefly the Irix 11/4. Less heavy, less costly and adjustable focus stop for ∞. THis is where I am sceptical - even if the Sigma would be great for astro-stuff, how to focus it at ∞ ?

 

If a lens like this needs AF is something everyone should decide for himself*. For landscape I don't see much of a benefit. For other stuff it's helpful, especially in LiveView ad with electronic shutter (I don't think, with an OVF one gets the focus right at f/1.8) and at this kind of weight, it's easier to manouever with AF.

 

*but if the answer is "yes", it's good to adjust it at more than only one distance. I feel this feature falls very short in most tests of Tamrons or Sigmas. -_-

 

As well as I say: If I have AF, I also have a focus motor and can remote control the lens. Meaning, the camera can be so close at the floor, that I don't see properly the screen, or I need to but it back to the wall to get the full room. Then it becomes handy, controlling the  camera with a smartphone or tablet and setting the focus ring.

 

Oh, and dave: you forgot in your list "no filter thread"



#10 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,451 posts
  • LocationSydney, Australia

Posted 20 November 2017 - 12:25 PM

No, I didn't test at infinity but at f/1.8 there was quite some field curvature anyway. Even in the theoretical event that the lens is better at infinity, the field curvature would hit astro-photographers.


Chief Editor
photozone.de

#11 JoJu

JoJu

    Advanced Member

  • Members
  • PipPipPip
  • 2,873 posts
  • LocationSwitzerland

Posted 20 November 2017 - 12:46 PM

Is the field curvature constant at all distances or does it change during various distance settings? Asking in general. With it's 4 aspherical lenses it's a challenge to design anyway.

 

When the sky will become a bit clearer, I like to verify how much of an influence this flaw has. To me it's better than your rating, but I have different prioirities.



#12 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,451 posts
  • LocationSydney, Australia

Posted 20 November 2017 - 02:03 PM

I don't think that the flaws will be noticeable in real life in the vast majority of use cases. The performance is pretty good at medium apertures and, as you noted, it not that relevant in shallow DoF scenes. The results are, obviously, also much better at 21mp. 50mp is simply a killer.


Chief Editor
photozone.de

#13 JoJu

JoJu

    Advanced Member

  • Members
  • PipPipPip
  • 2,873 posts
  • LocationSwitzerland

Posted 20 November 2017 - 02:21 PM

45 MP is not much less, but I was a bit surprised the corner weakness was relativley low at the shot with the woman at the river - and that shot was taken at f/1.8...



#14 toni-a

toni-a

    Advanced Member

  • Members
  • PipPipPip
  • 1,314 posts
  • LocationLebanon

Posted 20 November 2017 - 03:05 PM

It's not about the lens, but could you compare 50MP pictures downsampled to 21 MP versus native 21 MP ?



#15 Rover

Rover

    Advanced Member

  • Members
  • PipPipPip
  • 1,604 posts
  • LocationRussia

Posted 20 November 2017 - 05:03 PM

I don't think many people are going to use 5DSR for astro, as this camera has (relatively) low top ISO anyway.

Besides, it eats through lenses like a Langolier. :)



#16 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,451 posts
  • LocationSydney, Australia

Posted 20 November 2017 - 10:39 PM

I don't think many people are going to use 5DSR for astro, as this camera has (relatively) low top ISO anyway.

Besides, it eats through lenses like a Langolier. :)

 

Fully agreed. The Sony A7S II should be awesome for this ... if it wasn't for its star eater issue ;-)


Chief Editor
photozone.de

#17 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,849 posts
  • LocationThe Netherlands

Posted 21 November 2017 - 12:01 AM

Field curvature certainly does not have to be the same at different focus distances.

 

Case in point:

http://forum.optical...-55mm-f2/page-2

 

Hence,  the Sigma showing noticable field curvature at testing distance, but less or no noticable field curvature at infinity is a possibility.



#18 Klaus

Klaus

    Chief Editor

  • Moderators
  • 5,451 posts
  • LocationSydney, Australia

Posted 21 November 2017 - 10:33 AM

Well, the field curvature is somewhat eaten by the increased DoF towards infinity.

But again let's be fair - 3-4 years ago, the Sigma would have been cheered regarding its performance.

I'm pretty much convinced that the D850 will produce similar figures. Most Sony FE lenses have also issues on the A7R II.


Chief Editor
photozone.de

#19 JoJu

JoJu

    Advanced Member

  • Members
  • PipPipPip
  • 2,873 posts
  • LocationSwitzerland

Posted 21 November 2017 - 11:00 AM

Well, the field curvature is somewhat eaten by the increased DoF towards infinity.

But again let's be fair - 3-4 years ago, the Sigma would have been cheered regarding its performance.

I'm pretty much convinced that the D850 will produce similar figures. Most Sony FE lenses have also issues on the A7R II.

 

If 3-4 years ago Sigma would have been cheered, then who's cheered today?  :huh: Maybe it's just me, but I'm not aware of any other 14 mm lens with that quite high center and close center performance. If I need sharp borders. I'll stop dwon anyway, but I really want to know a single case or subjectwhen you need high res corners wide open? In my book there's none of this subjects.Not even in theory.

 

Okay, once I used the Zeiss 12 mm on the Fuji to get some shots in an exhibition just with less reflections.

 

This one can do with a Fuji, with a 2.5 kg FF combination the guards usually jump quickly - but even for this purpose - no reproductions but some memories - that lens will do just fine, as most pictures/paintinings are lit sufficienty to view them.

 

I want to know, Klaus, when you need high res corners wide open. To me, your verdict appears not realistic and at the same time you did say only one single word ("It's not a wonder lens in terms of flare/glare but that applies to many lenses in this class anyway.") about flares - these matter more and compared to the Nikon 14-24, that Sigma IS a wonder lens (most others are so, too  ^_^ ).

 

Edit: Oh and by the very way: The Nikon got 4 stars at it's time. At 14 mm and f/2.8, it is in no respect better than the Sigma - not on 20 and especially not on 45 MP!



#20 Brightcolours

Brightcolours

    Advanced Member

  • Members
  • PipPipPip
  • 4,849 posts
  • LocationThe Netherlands

Posted 21 November 2017 - 11:27 AM

It is not just that increased DOF masks field curvature, field curvature can be gone or worse at different focus distances (see the images from the UMN 55mm f2 in link above for a clear example, for e lens which does not even have moving elements). I do not know how to test that reliably, and obviously I do not have that lens even if I did.  ;)

 

Other issues also can vary depending on focus distance with lenses, like bokeh character, astigmatism.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users



© by photozone.de