The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Sigma 35mm F1.4

Jan Brittenson

Senior Subscriber Member
The certificate which was packed with the lens was signed by a technician. I can not seriously believe he actually ever looked at the lens.
I don't think anyone is under the illusion that a test certificate means there is never anything wrong with a lens when you get it. It could get banged up in transit, be dropped by a dealer, or have a manufacturing defect that manifests itself sometime later - like from thermal stress during transit. But at least you know it was good at some point. It also means when you exchange it as defective your chances of getting another defective unit are way, way lower.
 

Amin

Active member
I'm posting a few D800 and D600 samples I took today across the street from my nephew's birthday party. Again, I don't use this lens for landscapes, only for people photography, but I'm sharing these in case they may be helpful. These were handheld, but I don't think handshake was an issue.

After staring at them a bit, I've come to the conclusion that the lens has fairly strong field curvature. Ie, the two shots of the trees show pretty good edge sharpness while the central area is backfocused, while the third photo (parking lot) shows great central sharpness and edges which are less optimal.

D800 f/2.8: https://www.dropbox.com/s/9iobcz7yh7o8nf9/D800-F2.8.nef


D800 f/4: https://www.dropbox.com/s/q60sx72h3rsz7ql/D800-F4.nef


D600 f/5.6: https://www.dropbox.com/s/12q6le4u4agz0b4/D600-F5.6.nef
 

nikonf

Member
I absolutely agree with you. That said, I can confirm the Sony/Zeiss ZA 135mm f1.8 lives up to it's sterling reputation. The Sony 135mm STF is also quite remarkable. This lens offers some of the most beautiful BOKEH of any lens!
Remember the 'Old Days" when cameras were so affordable you could cherry pick the optics you wanted and buy a body after the lens?
I had great lenses by Leica, Zeiss, Canon, Nikon, Olympus, Pentax and Minolta. They all had a unique look which I could take advantage of for certain aspects of my photography.
When I ran a Photo Lab unit in the US Army Signal Corp, I used to take images with lenses from Leica, Nikon, Zeiss and view them on the enlarging lens magnifier and I could instantly see the difference in the optical characteristics.
Quite often, the Leitz lenses would look vastly superior, and yet there were some noticeable surprises.
The Leitz lenses were almost always better at the 2 maximum apertures.
The 75mm Zeiss Planar on the Rolleiflex was generally better than the Hasselblad 80mm Zeiss Planar. Mirror slap could also have been a reason for the differences.
I don't think anyone is under the illusion that a test certificate means there is never anything wrong with a lens when you get it. It could get banged up in transit, be dropped by a dealer, or have a manufacturing defect that manifests itself sometime later - like from thermal stress during transit. But at least you know it was good at some point. It also means when you exchange it as defective your chances of getting another defective unit are way, way lower.
 

Amin

Active member
I'm posting a few D800 and D600 samples I took today across the street from my nephew's birthday party. Again, I don't use this lens for landscapes, only for people photography, but I'm sharing these in case they may be helpful. These were handheld, but I don't think handshake was an issue.

After staring at them a bit, I've come to the conclusion that the lens has fairly strong field curvature. Ie, the two shots of the trees show pretty good edge sharpness while the central area is backfocused, while the third photo (parking lot) shows great central sharpness and edges which are less optimal.

D800 f/2.8: https://www.dropbox.com/s/9iobcz7yh7o8nf9/D800-F2.8.nef


D800 f/4: https://www.dropbox.com/s/q60sx72h3rsz7ql/D800-F4.nef


D600 f/5.6: https://www.dropbox.com/s/12q6le4u4agz0b4/D600-F5.6.nef
For anyone who'd prefer to view processed JPEGs to downloading the NEFs, here you go. Click through to view full-res versions:





 

Steen

Senior Subscriber Member

at the moment your processed jpegs are not directly accessible, Amin, I get prompted by a Yahoo login when clicking the images

but thanks a lot for providing the NEF files to play with, much appreciated, I'm downloading them as I write this :thumbup:
 

tashley

Subscriber Member
Ditto Amin, but we also need to know how and where you focussed.

Shots I made with my replacement today confirm very strong field curvature despite claims to the contrary from Sigma and from many reviewers: it is to me quite reminiscent of the Nikkor 28mm f1.8G, a lens which is about equally sharp if slightly lagging the lovely look of the Sigma... but then the Nikkor is also a lot cheaper and a lot lighter...
 

Steen

Senior Subscriber Member

I looked up the focus points in Capture NX2.
Here they are, in the same order.










 

Amin

Active member
Steen, those were all autofocused and recomposed, so the point you indicated are not the actual focus points even though NX shows them to be. Those three are well in front of the focus point.

For the first two shots, I focused on the center of the row of trees across the top. Interestingly the trees at the edges are sharper than the center, where the focus appears to be best behind the trees.

For the third, I focused on the green "Pay to Park" sign near the middle of the frame, and that came out very sharp indeed.
 

D&A

Well-known member
Just a general comment as to why we're seeing a lot more curvature in recently released lenses. It's a way for manufacturers to design a fairly fast lens with increased sharpness at a somewhat lower price point (and associated costs). In other words, unless the manufaturer intended to also correct for this curvature and pass along the costs with a very expensive lens, this now is becoming an acceptable alternative, There are many other factors to consider and it's somewhat of an oversimplification, but compared to flat field general use lenses, these lenses have excellent sharpness but exhibit a fair amount of curvature, some more than others.

Some recent lenses that illustrate this point are the Sigma 35mm f1.4, the Nikon 16-35mm f4 VR, the Voigtlander M mount 35mm f1.2and the Samyang 14mm f2.8. In the past, I think manufacturers were afraid to release such lenses for fear that there was no easy way to correct in the film only era....but with the advent of a relatively easy fix in software post processing, most users are more receptive to these compromises. You generally can't make a lens sharper than it is, in post processing, but you can generally correct curvature with use of software.

Dave (D&A)
 
Last edited:

Steen

Senior Subscriber Member

Steen, those were all autofocused and recomposed, so the point you indicated are not the actual focus points even though NX shows them to be. Those three are well in front of the focus point.

For the first two shots, I focused on the center of the row of trees across the top. Interestingly the trees at the edges are sharper than the center, where the focus appears to be best behind the trees.

For the third, I focused on the green "Pay to Park" sign near the middle of the frame, and that came out very sharp indeed.
Oh sorry, Amin
I have now removed the misleading focus point illustrations
I apologize for the confusion caused
And once again, thanks a lot for providing the NEF files !
 

tashley

Subscriber Member
Just a general comment as to why we're seeing a lot more curvature in recently released lenses. It's a way for manufacturers to design a fairly fast lens with increased sharpness at a somewhat lower price point (and associated costs). In other words, unless the manufaturer intended to also correct for this curvature and pass along the costs with a very expensive lens, this now is becoming an acceptable alternative, There are many other factors to consider and it's somewhat of an oversimplification, but compared to flat field general use lenses, these lenses have excellent sharpness but exhibit a fair amount of curvature, some more than others.

Some recent lenses that illustrate this point are the Sigma 35mm f1.4, the Nikon 16-35mm f4 VR, the Voigtlander M mount 35mm f1.2and the Samyang 14mm f2.8. In the past, I think manufacturers were afraid to release such lenses for fear that there was no easy way to correct in the film only era....but with the advent of a relatively easy fix in software post processing, most users are more receptive to these compromises. You generally can't make a lens sharper than it is, in post processing, but you can generally correct curvature in with use of software.

Dave (D&A)
Dave, I think that field curvature (being a curved rather than planar field of focus) and distortion (being the inability of a lens to project straight subject lines onto the film/sensor) are related in lens design but that whilst software in post can correct distortion usually at relatively minimal cost to IQ, it can only correct the curved field of focus by cropping, which is usually a much bigger cost.

So while I get that digital is something that changes the balance of lens design so as to allow more distortion (a quotidian notion today but look at the horror that greeted Hasselbald's first being explicitly public about this a few years back!) I don't get that field curvature becomes any more acceptable other than as an economic decision.

What is also very clear is that manufacturers are wise to the ways in which labs test resolution, and are very good at designing cost-effective glass with great charts and great central sharpness, but which in real-world use have really soft edges. Of course, the edges aren't really soft: they are just sharp at a different distance from the camera than the central target. Consequently, given the rarity of planar targets in the real world, this often goes unnoticed by the photographer due to the varying shapes of the subject field.

It astonishes me that one routinely reads reviews of lenses that say 'there is no significant field curvature' when in real use it is so strong as to ruin any images that need planar focus abilities. Of course lens design is a lot more complex than this: lenses natural tendencies are to create a curved field of focus because of the fact that the photographer sits at the centre of a virtual sphere, and the equal distances to the centre of his target and the edges describe a circle (the radius) rather than the unequal focus distances that planar focus at different subject ranges present. But we know it can be done, beautifully, as per the Zeiss lens on the RX-1...
 

tashley

Subscriber Member
Please just get the RX-1 already:)
I have! It's brilliant! But I would like for some uses to be able to use the resolution of the D800 sensor with a lens that can cope with it. A not unreasonable wish, given the marketing claims made and the prices involved, I think.
 

D&A

Well-known member
Dave, I think that field curvature (being a curved rather than planar field of focus) and distortion (being the inability of a lens to project straight subject lines onto the film/sensor) are related in lens design but that whilst software in post can correct distortion usually at relatively minimal cost to IQ, it can only correct the curved field of focus by cropping, which is usually a much bigger cost.

So while I get that digital is something that changes the balance of lens design so as to allow more distortion (a quotidian notion today but look at the horror that greeted Hasselbald's first being explicitly public about this a few years back!) I don't get that field curvature becomes any more acceptable other than as an economic decision.

What is also very clear is that manufacturers are wise to the ways in which labs test resolution, and are very good at designing cost-effective glass with great charts and great central sharpness, but which in real-world use have really soft edges. Of course, the edges aren't really soft: they are just sharp at a different distance from the camera than the central target. Consequently, given the rarity of planar targets in the real world, this often goes unnoticed by the photographer due to the varying shapes of the subject field.

It astonishes me that one routinely reads reviews of lenses that say 'there is no significant field curvature' when in real use it is so strong as to ruin any images that need planar focus abilities. Of course lens design is a lot more complex than this: lenses natural tendencies are to create a curved field of focus because of the fact that the photographer sits at the centre of a virtual sphere, and the equal distances to the centre of his target and the edges describe a circle (the radius) rather than the unequal focus distances that planar focus at different subject ranges present. But we know it can be done, beautifully, as per the Zeiss lens on the RX-1...
HI Tim,

Just two small points. It's not that I'm implying that correction of field curvature through software is an a completely acceptable alternative to designing a plannar like lens. On the contrary...it's that with the advent of digital and farily recent advances with programs like PTLens and the like, that more are seemingly willing to accept the "often times" tradeoffs of applying such software to their images in order to obtain somewhat more affordable but highly sharp lens, once all is said and done.

This was generally not the case in the film only era....and it's not just field curvature where some are willing to make compromises in image quality where in the past, it might have not been acceptable. It reminds me of the phrase 'It's the new norm"...where certain standards of acceptability are often compromised for gainging other things.

You're correct, designing a plannar lens in a spherical world isn't easy nor inexpensive but it seems the manufaturers feel out their intended customers base and determine what will be acceptable and sell at a certain price point...espcially when quite a few feel there is at least some way they can address these image anomolies...as opposed to spending hundreds or even a few thousand more on a similarly designed lens albeit without the curvature. Just some personal thoughts of mine.

Dave (D&A)
 
I am not a landscape photographer, and my experience in this field is really
quite limited. It seems that is very difficult to find a lens, expecially for 35 mm format, that is up to the performance required at least for a certain type of landscape. Having in practice only Leica lenses (just a few Nikon zooms for autofocus) I don't have the possibility to make comparisons. So I would like to have your opinions regarding the usability of the elmarit r 28 in that field.
A test shot with D800 (I apologize for the subject)focused on the gray wall, with the full size version available on flickr.


_DSC1322 by sergio lovisolo, on Flickr

link to full size
All sizes | _DSC1322 | Flickr - Photo Sharing!

Thanks.
Sergio
 

drb

New member
My lens has been repaired. Need to go pick it up later today.
Guy:

Sightly OT, but let us know what you think of the Sigma repair facility after you get your lens back. I'm assuming it's Cris Camera Services in Chandler. While I've only taken a few shots with my Sigma 35 f1.4 and it seems good, I do have the relatively new Sigma 150mm macro that seems to have auto focus issues and I was planning on sending it the same facility. It would be great to hear what you think of their work.
 

tashley

Subscriber Member
also really interested to hear back Guy - mine is off to Sigma UK tomorrow and I must say that by email they couldn't be more friendly and helpful.. but the proof of the pudding will be in the snapping not the chatting!
 

Guy Mancuso

Administrator, Instructor
Okay I'm here at shop. First off Nikon has a bug in the firmware. Bob brought this up the other day and just found it too. Apparently putting the lens back on at its saved value was worthless than going back to zero as well. So do this reset back to zero turn camera off than back on than turn AF back in and the values are much better. Weird but it seems so and at first lens seemed like nothing fixed until I did that. But it was still around -15 which is better and a lot of shots looked good in close range but still a little off so I left the lens for the tech to double check it again in the morning. Now most would walk away and say its fine but not me and also we are talking AF here which frankly has been always suspect to me. I still can manually focus better than any AF machine but I'm slower . Lol

The folks here are great and it was a fast turn around also but it's more me here. It felt a little hit and miss. I did find out they where talking to Japan on this lens as well and did do firmware and adjustments as well. To me it's like a doctor never aggressive enough with meds to kill off something but conservative enough to fight something off. I'm a go for it person. So maybe they took baby steps here.
 

tashley

Subscriber Member
Hey Guy, that sounds really useful, thank you! But I am not sure I quite understand the process for getting around the bug. Does the bug only affect this lens? And what exactly is the sequence you used? Sorry to ask for it again but am being thick!
 
Top