Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Strange focus on my Sigma 35mm (art) with Dragon

Christoffer Glans

Well-known member
Joined
Jun 11, 2007
Messages
4,993
Reaction score
36
Points
48
Age
42
Location
Stockholm, Sweden
I've been noticing some strange lack of sharpness on my Sigma 35mm so I checked it out with the backfocus on my Dragon. I set it to infinity and calibrated, but it just wouldn't get into focus. I closed the shutter in order to is if it were because it was wide open, but it still performed a bit unsharp even at f8.0.

Tried my Canon 24-70mm 2.8L II set at 35mm and it too were a bit unsharp, but far better than the Sigma. What is going on with my lens? And all lenses experience some kind of directional blur of some kind, most notable on the 35mm. The directional blur change from X to Y when going past an object with the focal plane when focusing.

I recorded it and rendered them out at 100% crop. First one is the 24-70mm (at f2.8) and the second is the Sigma 35mm (at f1.4).

Download

Anyone who can adress this? Is it as it should or is there something wrong with my lens or camera or both?
 
Anyone? :confused1:
 
Did you adjust the FFD of the camera? It my be your camera that is off, not the lens. Also, your ART lens seems to have way too much CA like it's misaligned or something. Have you tried using it on a stills camera to see if its the same result?
 
The FFD, is that the front adjustment (little screw with a screw thing under it) to adjust backfocus? The Sigma looks better on my 7D I think, it's 5K so it's close to the resolution of the Dragon and it doesn't look like on my Dragon. I also don't recall my MX being like this, this is after I upgraded to Dragon. What to do? :001_unsure:
 
The FFD, is that the front adjustment (little screw with a screw thing under it) to adjust backfocus? The Sigma looks better on my 7D I think, it's 5K so it's close to the resolution of the Dragon and it doesn't look like on my Dragon. I also don't recall my MX being like this, this is after I upgraded to Dragon. What to do? :001_unsure:

Yes, the FFD adjustment is the screw on the brain.
 
Set up a focus chart at 6 feet.
light it to about 5.6 at 320 ISO
meausre your camera from the cart to the focus mark, make sure it is exactly six feet.

Set the lens to six feet.

Use the the magnify tool and get a large monitor and look at the center of the chart.
maybe even turn on edge focus.
Rotate the sensor focus until you get the sharpest focus.
Remove the lens, put on the next lens preferably wider and set it to six feet to check if it is focus.

This is will ensure your back fuss is correct to the sensor plane.

Or, use a redfocus and a pl mount. That works too.
 
Sounds like back focus / flange focus distance. I got a red focus if you wan to check it. Takes seconds to set it right.
 
Set up a focus chart at 6 feet.
light it to about 5.6 at 320 ISO
meausre your camera from the cart to the focus mark, make sure it is exactly six feet.

Set the lens to six feet.

Use the the magnify tool and get a large monitor and look at the center of the chart.
maybe even turn on edge focus.
Rotate the sensor focus until you get the sharpest focus.
Remove the lens, put on the next lens preferably wider and set it to six feet to check if it is focus.

This is will ensure your back fuss is correct to the sensor plane.

Or, use a redfocus and a pl mount. That works too.


I did this and calibrated the backfocus. Problem with my Canon 24-70mm 2.8 II is that it's shifting focus when zooming and when I calibrated it at 50mm it doesn't focus well on 24mm.
Then my Sigma 35mm got a little better, but I can still see some shifts in the bokeh. If you scroll through this clip at the moment of focus, the bokeh is more out of focus on the X-angle than the Y-angle and when going past the focus point it switches so that the bokeh is more on the Y-angle. It's like the bokeh isn't uniform.

Sigma 35mm focus test

Also, another strange thing I've noticed. When I mounted the 24-70mm I pointed it straight to the middle of the focus chart. Then, without changing the position, tilt or pan of the camera I changed the lens to the 35mm and then the center isn't in the center anymore? On the pictures you can see the difference.
 

Attachments

  • focus1.jpg
    focus1.jpg
    40.3 KB · Views: 0
  • focus2.jpg
    focus2.jpg
    51.8 KB · Views: 0
You are landing on some of the downsides of using still lenses for motion picture work. Image Focus Shift occurs when racking focus and the image moving due to low tolerances in the lens design. Image Zoom Shift is the same idea except it occurs when zooming.

On purpose designed motion picture lenses you shouldn't experience any of that.
 
Yes. Not all zooms hold focus through the range, especially with still lenses which are optimised for short focus throw to allow autofocus to capture the action. The whole point of still lenses is to get the focus quickly, hence the short throw that is math, shorter distances to get to the "correct" focus is faster, for sports, action, etc. So everything is optimised for the lens to do that. They are not created to zoom in the middle of a shot because you do not zoom in the middle of a still shot.

The optimal center thing can also shift slightly between lenses and lens manufacturers. This is why people work so hard at getting matched sets of lenses or get sets os Cinema glass.






I did this and calibrated the backfocus. Problem with my Canon 24-70mm 2.8 II is that it's shifting focus when zooming and when I calibrated it at 50mm it doesn't focus well on 24mm.
Then my Sigma 35mm got a little better, but I can still see some shifts in the bokeh. If you scroll through this clip at the moment of focus, the bokeh is more out of focus on the X-angle than the Y-angle and when going past the focus point it switches so that the bokeh is more on the Y-angle. It's like the bokeh isn't uniform.

Sigma 35mm focus test

Also, another strange thing I've noticed. When I mounted the 24-70mm I pointed it straight to the middle of the focus chart. Then, without changing the position, tilt or pan of the camera I changed the lens to the 35mm and then the center isn't in the center anymore? On the pictures you can see the difference.
 
I understand that still lenses aren't optimal, I'm just curious if it's a problem with my camera, like if the OLPF or the mount isn't aligned perfectly.

The most odd thing though is that the Sigma 35mm worked perfectly on my MX, on that sensor it showed no X/Y shift Bokeh at all, now I've shot some things and doing a rack focus where it's visible even on a non-cropped render from the 6K or 5K resolution.
 
WHen doing this on PL, I get the widest/fastet/"good" lens i have (like an 18 Master) and set it to infinity, and see that anything beyond hyperfocal is in focus wide open.

Then verify it with other lenses.

It's a really bad idea to do this at say 70mm and at a measured distance, unless you trust the marks 110%, which you cannot on stills glass.

Focusing at infinity is a really good giveaway. Remember the Art goes beyond infinity.
 
That's what I did out the window in my first post. It's why I'm curious about the issue. As seen there the lens can't get the 35mm into focus. Even though I work both the focus and back focus adjustment it won't get right.
 
Take Bjorns advice and look at the null target. IF that is out of focus (the sensor plane) then everything else will be. Set a reference. then you can determine what is messed up.
 
Even though I work both the focus and back focus adjustment it won't get right.

this is the root of your problem. You can\t manipulate both and have a reference. The measuring tape is a known reference, a specific distance. the mark on the lens is not. Gunliek is correct. But you can have those things line up and they should.

Do it on the widest lens you have. But the camera bay in most every rental house on earth doesn't have a window on the wall, it has a chart at a known distance from the focal plane.
 
Reason I suggest the infinity route, is that it is hard to actually trust the marks on a still lens to be absolutely right (or even securely and indiscussably hit them) , which is a necessity when you go closer than infinity to set the back-focus right.

Otherwise we agree :)

And David is 100% right, you must have constant parameters and NOT fiddle with the lens while you adjust backfocus.

It can lead to hell to start with a long lens, BTW. Trust me :)
 
this is the root of your problem. You can\t manipulate both and have a reference. The measuring tape is a known reference, a specific distance. the mark on the lens is not. Gunliek is correct. But you can have those things line up and they should.

Do it on the widest lens you have. But the camera bay in most every rental house on earth doesn't have a window on the wall, it has a chart at a known distance from the focal plane.

But they do have longer corridors than my apartment :biggrin: For the chart test I used a laser measurement and the internal numbers for focus. I put the Sigma at the 2m mark and calibrated. Both lenses in the first clip were set to infinity. The 24-70mm seems alright, but the Sigma is handling sharpness very strange.


Reason I suggest the infinity route, is that it is hard to actually trust the marks on a still lens to be absolutely right (or even securely and indiscussably hit them) , which is a necessity when you go closer than infinity to set the back-focus right.

Otherwise we agree :)

And David is 100% right, you must have constant parameters and NOT fiddle with the lens while you adjust backfocus.

It can lead to hell to start with a long lens, BTW. Trust me :)

Actually I didn't fiddle with it. The focus on the clip in the first post was calibrated on the infinity mark for the Sigma 35mm. As you can see the focus never hits actual focus; it goes from blur to less blur to blur. After that I did change the focus a little, trying to see if I could get it at all into focus, but it looks like that all the time and that's pretty unsharp for the Sigma lens, closing it down to f8.0 didn't improve the infinity focus much either. So whatever I do with the backfocus, the Sigma still is quite unsharp at infinity, so I'm pretty sure I'm doing it right, because the problem isn't really that it's not in focus on a certain distant measurement, it's that it's unsharp at that distance even after calibration.
 
Back
Top