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

tashley

Subscriber Member
I'm with Jack in half of what he says: shop for good ones yes, but moderate one's expectations, no: define them reasonably and then stick to them, yes!

I tested the new siggy properly this afternoon and it is overall not quite as sharp on centre as the one I returned (best out of bracketed series) and, though it is sharper on the right, it is softer on the left.

My definition of reasonable is that the lens should look reasonably sharp on both left and right of frame at 50% on screen at F5.6 on a D800E, and that if you make a virtual copy of that f5.6 file and look at the right edge of it up against the left edge of the original, they should look similarly sharp. I can't think that that is unreasonable - and it's not a requirement I'd have of older, cheaper, or specialist lenses. But for anything from 24mm upwards, a good modern lens costing in the region of £800 upwards should be able to achieve that. It should also be able to be AFFTuned to within the -20 to +20 range.

I have a standard test scene that I always shoot and over the years I have learned within half a minute to see, when getting a file from a new lens up on screen, if it is acceptable.

So out of curiosity, do other people agree with my definition of what it is reasonable to expect?
 

tashley

Subscriber Member
Apparently, that RX-1 lens is something special. No. 3 in DxO rankings with only 85mm lenses to keep it company.

DxOMark - Camera Lens Ratings

I would keep that RX-1 and think about trading in that D800. ;)
Funny you should say that: in nearly all circumstances, the RX1 gives a better print in terms of perception of sharpness across the frame, especially at F2 thru 4 but also at 5.6, than any lens I own between 21mm and 35mm on a D800 - even if you down-res the D800 file to RX-1 size and even if you up-res the RX-1 file to D800 size. It is the most neutral, well behaved lens in this focal length range I have ever shot.
 

Shashin

Well-known member
So out of curiosity, do other people agree with my definition of what it is reasonable to expect?
Sometimes I agree with you and sometimes not (so I don't think you are entirely crazy). I would say you are probably the most critical person on GetDPI for optics. One thing is for sure, I am glad I am not your sales rep. :ROTFL:
 

Jan Brittenson

Senior Subscriber Member
So out of curiosity, do other people agree with my definition of what it is reasonable to expect?
Yes. Unfortunately, neither Nikon or apparently Sigma agree, which is at the root of the problem. Nikon should for any 'pro' product give it a critical QC pass, then service or tweak it as needed until it passes. This would include critical inspection of files produced on a real-world, actual live camera like a D800E and focus performance on a D4. 1-3 AF tune clicks is fine (the value should be on the test certificate) on most lenses, 15 is not. A zoom that can only be tuned to one end is not. If this adds $100 to the cost of each lens sold (about two-three hours of bench work for a competent technician, averaged) - so be it. Pass it on. This is not the place to cut product cost.
 

tashley

Subscriber Member
Sometimes I agree with you and sometimes not (so I don't think you are entirely crazy). I would say you are probably the most critical person on GetDPI for optics. One thing is for sure, I am glad I am not your sales rep. :ROTFL:
Funny thing is, I sold Jack a lens once and it was a good copy of an interesting but imperfect design and I had to describe how good it was before I sent it to him - and I think he agreed when he received it pretty much exactly. So I think that he and I have pretty similar expectations and make pretty similar allowances.

It's a great irony that about half my work doesn't require great lenses at all :D
 

tashley

Subscriber Member
Yes. Unfortunately, neither Nikon or apparently Sigma agree, which is at the root of the problem. Nikon should for any 'pro' product give it a critical QC pass, then service or tweak it as needed until it passes. This would include critical inspection of files produced on a real-world, actual live camera like a D800E and focus performance on a D4. 1-3 AF tune clicks is fine (the value should be on the test certificate) on most lenses, 15 is not. A zoom that can only be tuned to one end is not. If this adds $100 to the cost of each lens sold (about two-three hours of bench work for a competent technician, averaged) - so be it. Pass it on. This is not the place to cut product cost.
I could not agree more. This is why Leica are good. Even if a sub-par copy escapes the factory, when you send it back they really pay attention because they really care.

I am thinking of trying an experiment with Sigma: Guy, Lloyd Chambers and I have all had one (me two) sub-par copies of this lens and that tells me that the chances of getting a really good one from the retail chain are possibly low. So I might just send them my new copy and say, get this to be good. See what happens. Worth a go I think...
 

Jan Brittenson

Senior Subscriber Member
So I might just send them my new copy and say, get this to be good. See what happens. Worth a go I think...
Well, this has been my approach with Nikon USA... send in whatever doesn't perform out of the box. It has worked very nicely for me, but then I haven't done battle with the 24-120 VR... yet. :) Might be worth a shot.
 

Guy Mancuso

Administrator, Instructor
Over the years I actually have been pretty lucky with getting good copies considering I must have bought over many many dozens( trying to forget that part) but each system has had its challenges for sure. Phase was really the 35mm D and I tested several 55LS and found the best one. On Zeiss everything okay recently but Nikon 24 1.4 I felt somewhat weak on. I have a very good 85mm and also the new 70-200f4 is kick ***. But this Sigma it's weird I actually have 2 things wrong with it . It back focuses past -20 and the right side is soft. I can accept the skew being off a little but at 5.6 it should be good. I'll see what repair does and if they get it right , great if not ill put it on eBay and try again. Ill try until I get a good one. The specs on it are the best in Nikon mount for a 35mm so I want it but need to have a good one.
 

Steen

Senior Subscriber Member

(...) So out of curiosity, do other people agree with my definition of what it is reasonable to expect?

+ 1 more

Another firm yes to that question, I actually think you are far too tolerant when you are willing to downsize to 50 % before judging the sharpness at the left and right side and expecting them to be equal. I always judge such matters at 100 % no less.

I can only judge from the Nikon 4/24-120mm and now the Sigma 1.4/35mm sample pictures you have shown, but I can at least say with certainty that I wouldn't accept such an asymmetry (or I would find myself spending far too much time looking for it).
 

Steen

Senior Subscriber Member

Yes. Unfortunately, neither Nikon or apparently Sigma agree, which is at the root of the problem. Nikon should for any 'pro' product give it a critical QC pass, then service or tweak it as needed until it passes. This would include critical inspection of files produced on a real-world, actual live camera like a D800E and focus performance on a D4. 1-3 AF tune clicks is fine (the value should be on the test certificate) on most lenses, 15 is not. A zoom that can only be tuned to one end is not. If this adds $100 to the cost of each lens sold (about two-three hours of bench work for a competent technician, averaged) - so be it. Pass it on. This is not the place to cut product cost.

Well spoken, Jan, I fully agree :thumbup:
 

Shashin

Well-known member
Yes. Unfortunately, neither Nikon or apparently Sigma agree, which is at the root of the problem. Nikon should for any 'pro' product give it a critical QC pass, then service or tweak it as needed until it passes. This would include critical inspection of files produced on a real-world, actual live camera like a D800E and focus performance on a D4. 1-3 AF tune clicks is fine (the value should be on the test certificate) on most lenses, 15 is not. A zoom that can only be tuned to one end is not. If this adds $100 to the cost of each lens sold (about two-three hours of bench work for a competent technician, averaged) - so be it. Pass it on. This is not the place to cut product cost.
Test certificate??? Nikon does not issue test certificates. It would add a lot more to the cost than $100, unless you are thinking of minimum wage employees doing this. But then who can afford Leica S prices if you are going to do this?
 

tashley

Subscriber Member




+ 1 more

Another firm yes to that question, I actually think you are far too tolerant when you are willing to downsize to 50 % before judging the sharpness at the left and right side and expecting them to be equal. I always judge such matters at 100 % no less.

I can only judge from the Nikon 4/24-120mm and now the Sigma 1.4/35mm sample pictures you have shown, but I can at least say with certainty that I wouldn't accept such an asymmetry (or I would find myself spending far too much time looking for it).
Phew, you come to my rescue again Steen: I was keen to pass on the crown of 'most critical person on GetDPI for optics'...

But I do of course always check at 100%. However, 50% is my 'no brainer for return' procedure otherwise I would be returning 90% rather than 70% and that would just wear me down... :bugeyes:
 

Amin

Active member
Just so's you know Guy, Lloyd Chambers' copy has a slight decentering too... It may be that the design is prone to it.
I hope I don't come across as a Sigma apologist - really I'm not - but it seems to me that most copies of all lenses are more likely than not to have a slight decentering. A quick look at SLRGear blur plots for various lenses indicates as much. That said, I haven't noticed any decentering with my copy of the Sigma 35/1.4.

I have come to the ultimate conclusion that buying glass for ANY high-resolution camera requires a "shoppers" mentality: try on several to find the one that looks and fits you -- or in this case your camera -- the best...

Short conclusion: my advice is to moderate ones expectations -- and then get back to creating images :D
I agree, Jack. I don't have much time or patience for exchanging, so I've lowered my expectations accordingly. I'm fine with an average amount of decentering, a certain amount of AF issues, etc.

So far with the Nikon system I've been very lucky - very good copies of several complex lenses on the first go.
 

Jan Brittenson

Senior Subscriber Member
Test certificate??? Nikon does not issue test certificates. It would add a lot more to the cost than $100, unless you are thinking of minimum wage employees doing this.
Zeiss manages to... and their lenses are priced at or near Nikon's. At Nikon volumes it should be doable. Leica is pretty extreme, no need to go quite that far...
 

nikonf

Member
When it comes to quality control, I have found Leica, Zeiss, Rodenstock and Schneider typically can deliver a manufactured optic which is closer to the original design.
I have encountered some exceptions and yet I am much more confident when I decide to purchase from the aforementioned optics manufacturers.
I am been photographing for more than 40 years and I have only had to return one Rodenstock and one Schneider lens due to defects in manufacture.
Zeiss lenses were much more reliable before they allowed Sony to become involved with the Alpha series of lenses. I had a 24-70 Zeiss Sony Alpha which came brand new in the box with a terrible wobble in the zoom mechanism. 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. This is NOT acceptable for a $2000 lens!
Manufacturers are taking shortcuts with quality control and we are experiencing more lens problems than we should be at this point in time.
It's all about "doing more with less". This has led to disastrous consequences in Information Technology and I fear the same in the manufacture of optics.
Read some of the horror stories Joseph Holmes has posted on his web site regarding finding optics which pass his tests.
He is a world class photographer and his images speak for themselves.
Good luck!
 

tashley

Subscriber Member
DXOmark on the rx-1 lens
"With excellent image quality at maximum aperture becoming outstanding at f/2.8 and on, the Zeiss Sonnar T* 2/35 is likely to become a classic, against which all others are judged."

This is what I hoped for from the Siggy but I will try getting mine tweaked to perfection and report back!
 

Steen

Senior Subscriber Member

(...) It's a USB dock where the lens mounts onto the dock and it's cord plugs into a CPU's USB. I guess with supplied software, one is not only supposed to be able to update firmware of the lens, but also change or re-set the Af fine tune. I presume a lens that say needs -18...could be "nulled" and re-set to zero. I'm not exactly sure if that how it works or not, as Sigma hasn't apparently said much about it. (...)

(...) I got news from Sigma: the dock will be available to end-users but price and date not confirmed...

Imagine what a revolution such a dock would be if it really enables the end-user to AF-fine-tune the lenses and why not at a later stage also cameras ?

A Do-It-Yourself AF-fine-tune kit. That would be awesome.

In theory it has the potential to save the manufacturer a lot of money and the customers a lot of frustration, good reasons why it ought not to be all that expensive.

Just for once the rather worn out Win-Win Situation phrase would make a lot of sense.

Think about it, Nikon :toocool:
 

Jan Brittenson

Senior Subscriber Member
I had only positive experiences with the Sony ZA lenses and think they were outstanding... The Planar 85/1.4 T* clearly was an old design, and more mojo than laser, but I think it compared reasonably to the Summilux-R 80/1.4. More of a Zeiss look though (= high contrast, none of those classic Leica pastels). The only thing I didn't like was that the AF turned the focusing ring, but it lived up to its promise nicely in terms of optics. I think Sony's SLR/SLT lens QC overall is much better than Nikon's though, so it's not just a Zeiss thing.

I still have the A850 (with grip even) since I can't use my M42 collection with Nikon...
 
Top