FS7: Sigma 50-100mm DC Backfocus Changes with Zoom on Metabones EF Adapter

basspig

basspig

Veteran
I just got the Sigma Art lens for my FS7. Running it on my Metabone EF adapter (which has no problems with my 12-28 Tokina zoom). The backfocus shifts by 24" when I widen the zoom from 50mm to 100mm. For example, I focus on an object at 100mm, then pull back to 50mm and now the objects 24" behind the original subject are in focus.

Do I have a bum lens that is defective?
 
Cary Knoop said:
Yes this Sigma lens is not parfocal. The other thing with this lens is the heavy focus breathing.

But nevertheless I think it is a great lens!
Click to expand...

Yes it's very sharp and has goodles low light ability. I did note lateral ca when shooting outdoors in snow/sun conditions. Not as bad at 100 as 50.
 
Newshooter had the Cine versions of the 18-35mm & 50-100mm, tested on a collimater by a Lens Tech whose concluded both the zooms were parfocal. The Cine, & Stills, versions of these lenses share the same optical formula, so both should be parfocal.

It's possible the Cine lenses are built to tighter tolerances but suspect your issue has to do the Metabones adapter, which unlike their Speed Boostes, has no provision for adjusting backfocus.

I'm considering picking up 18-35mm T2 Cine, & would like to get it in EF-mount, but it's not shimable, so would probably go for the PL.
 
Bern Caughey said:
Newshooter had the Cine versions of the 18-35mm & 50-100mm, tested on a collimater by a Lens Tech whose concluded both the zooms were parfocal. The Cine, & Stills, versions of these lenses share the same optical formula, so both should be parfocal.

It's possible the Cine lenses are built to tighter tolerances but suspect your issue has to do the Metabones adapter, which unlike their Speed Boostes, has no provision for adjusting backfocus.

I'm considering picking up 18-35mm T2 Cine, & would like to get it in EF-mount, but it's not shimable, so would probably go for the PL.
Click to expand...
I have the lens and it is not parfocal. Sure if you focus zoomed in and zoom out it looks pretty decent but not in reverse.

Even the newshooter article states they are not parfocal!

http://www.newsshooter.com/2016/11/...scribe-their-new-cine-zooms-as-near-parfocal/
 
Cary Knoop said:
I have the lens and it is not parfocal. Sure if you focus zoomed in and zoom out it looks pretty decent but not in reverse.
Click to expand...

Without some form of backfocus adjustement it's a crap shoot whether any parfocal lens will perform so. I tested the stills version of the 18-35/1.8 in EF mount on a Canon 7D & it was most definitely not parfocal, but was that the lens' fault, or the camera's?

Cary Knoop said:
Even the newshooter article states they are not parfocal!

http://www.newsshooter.com/2016/11/...scribe-their-new-cine-zooms-as-near-parfocal/
Click to expand...

Newshooter has a newer article about their tests of the Cine Sigmas on a optical bench, & concluded the lenses were parfocal.

Sigma 18-35mm T2.0 and 50-100mm T2.0 Cine Zoom review Part 1 – A Technical Evaluation
by Matthew Allard

http://www.newsshooter.com/2016/12/...ne-zoom-review-part-1-a-technical-evaluation/

Matthew Duclos recently weighed in too, but only tested the 18-35mm.

Are Sigma’s Cine Zooms Parfocal?
by Matthew Duclos

https://thecinelens.com/2017/02/08/are-sigmas-cine-zooms-parfocal/

Mitch Gross' post on the topic is worth reading.

Mitch Gross said:
I kinda got into it today on Twitter with Matt Duclos and Dan Chung. Here is my essential argument on "near-Parfocal":

It is a distinction without a difference.

No zoom lens is absolutely Parfocal. There is always some slop in the system, and there is an allowable tolerance to this characteristic. It's known as the Zoom Curve, because as it drifts away and back from the absolute perfect spec, the motion as graphed represents a curve. If a lens is within tolerance, then it is Parfocal.

A lens design may not be Parfocal in that even at its absolute theoretical performance it has zoom curve drift. So this lens could never achieve a perfect Parfocal performance of not drifting at all throughout the zoom range. But that lens might have a zoom curve in its mathematical design that still is within tolerance for Parfocal performance. This is what Sigma is calling near-Parfocal.

If it walks like a duck and quacks like a duck, then it's a duck. If a zoom lens performs the zoom curve test on an optical collimator bench to within the standard's tolerance, then therefore by jove it is Parfocal.

Anyway, that's my opinion on the matter. I was up way too late last night trying to explain this one in 140 characters or less.
Click to expand...

http://www.dvxuser.com/V6/showthrea...Lenses&p=1986669259&viewfull=1#post1986669259
 
Last edited:
When I focus at 100 millimeters and then pull back to 50 millimeters it goes pretty badly out of focus. The focal point shifts two feet further in the distance.
 
The best way to tell if a zoom is parfocal is to have it tested on a optical bench. Would love to see how the stills versions of both Sigma zooms stand up.

Would also love Metabones adapters with backfocus adjustment similar to the Speedboosters'.
 
Last edited:
My piles of photographic zoom lenses say that they are almost all parfocal (to varying degrees), but *only* if the back focus is carefully adjusted. Adapters are almost always the wrong length to give you perfect back focus. Even native mount lenses are usually off a little.

If you want a lens to be parfocal, you need some kind of back focus adjustment, in the lens, the camera, or the adapter.
 
balazer said:
My piles of photographic zoom lenses say that they are almost all parfocal (to varying degrees), but *only* if the back focus is carefully adjusted. Adapters are almost always the wrong length to give you perfect back focus. Even native mount lenses are usually off a little.

If you want a lens to be parfocal, you need some kind of back focus adjustment, in the lens, the camera, or the adapter.
Click to expand...

Your stills lenses almost certainly aren't parfocal.


Matt Duclos did an article about this recently

https://thecinelens.com/2017/02/08/are-sigmas-cine-zooms-parfocal/
 
Most lenses I got to check on collimator and projector (probably few thousands in my 30 year career in the business) have been in fact parfocal. Unless they were broken.
The problem 90+% percent of the time is either back focus on lenses or flange depth on camera. That later includes wrong thickness on the adapters.
I'll post a video about it- we just shot it and need to tweak it a little to make it more accessible for general public....
That also seems to be the OP lens problem from his posts....
 
In which case I'd certainly be interested in your video as that flies in the face of a lot of people's experience, as well as Matt Duclos' article above....
 
That does not include vari focal lenses - we have some in the M43 system.
Those are refocused continuously while "zoomed". But that's a different story all together....
 
How do you mean? He says there that the Sigma still based lens is more parfocal than the Angenieux... Do you have a different version of it? ;-)
 
You must log in or register to reply here.
Back
Top