Apo Vario 90-280 sharpness at the long end

[kkochheiser]

Advertisement (gone after registration)

True, but more a function of the camera than the lens.

[osroubek]

As I try more testing, I think that the lens/ camera, when using autofocus, tends to front focus a bit. The images are of an old device I used to micro-adjust DSLR lenses on my Canon years ago

The first one is taken with autofocus (field) and the second one with manual focus with focus peaking, 90 mm, f2.8, 1/6sec. This seemed to be pretty consistent through the zoom range.  Most of the time, the manual focus was more successful. Maybe the upcoming firmware will help?

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[meerec]

It is definitely front focusing — quite obvious from your pics. In those old Canon days, I used the same alignment device, and you would use this data to apply calibration to AF, pity Leica doesn’t give us this option for SL2.

[michali]

7 hours ago, osroubek said:

As I try more testing, I think that the lens/ camera, when using autofocus, tends to front focus a bit. The images are of an old device I used to micro-adjust DSLR lenses on my Canon years ago

The first one is taken with autofocus (field) and the second one with manual focus with focus peaking, 90 mm, f2.8, 1/6sec. This seemed to be pretty consistent through the zoom range.  Most of the time, the manual focus was more successful. Maybe the upcoming firmware will help?

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Have you tried the focus test with AF Mode set to "SPOT " & Focus Mode set to  "AFS" ?  The lens tends to hunt with AFC & I also find that FIELD tends to be less accurate. I use SPOT focus & haven't had miss-focus issues, unless induced by me not holding the camera steady.

 

[Stuart Richardson]

If it is not focusing consistently accurately, I think your only choice is to talk to Leica. Frankly, I am not sure how this would happen, as contract detect focusing systems by their nature do not really need the kind of calibration that phase detect systems do, as the camera is reading the actual point of highest contrast on the sensor. I would talk to Leica service and see what they recommend. Generally if there are any focusing problems with contrast detect it is due to the AF spot not having sufficient contrast to accurately assess focus, or not having enough light. Neither seem to be the case here.

[LocalHero1953]

21 minutes ago, michali said:

Have you tried the focus test with AF Mode set to "SPOT " & Focus Mode set to  "AFS" ?  The lens tends to hunt with AFC & I also find that FIELD tends to be less accurate. I use SPOT focus & haven't had miss-focus issues, unless induced by me not holding the camera steady.

 

+1 for trying it with Spot before reaching a definite conclusion. I don't use Field (either spot or eye/face/body detect), and this is not a problem I have found, though I have never investigated rigorously or compared to similar lenses from other brands.

Edited August 11, 2021 by LocalHero1953

[Jeff S]

Advertisement (gone after registration)

4 hours ago, meerec said:

It is definitely front focusing — quite obvious from your pics. In those old Canon days, I used the same alignment device, and you would use this data to apply calibration to AF, pity Leica doesn’t give us this option for SL2.

The point of mirrorless is that focus is read directly off the sensor, so what you see is what you get.  No alignment should be necessary. There is something else going on here.

My approach is back button AF, with magnified live view by a slight twist of the focus ring for critical focus if needed.  Can’t do that on an SLR.

Jeff

Edited August 11, 2021 by Jeff S

[michali]

The other issue is that you're using long exposures for your focus test, this introduces a number of variables, movement etc. plus shutter slap which is worse at slower speeds, I mentioned this early on in this thread.

Try your focus test as follows:  shutter speed of +125/s -  AF "SPOT" Mode - "AFS" Focus Mode - "Electronic Shutter" & see what results you then get.

Edited August 11, 2021 by michali

[osroubek]

Thanks for your suggestion. I tried raising the ISO and shooting at 125 sec, E shutter, IBIS off, spot focus. It really did not seem to improve the results that much if any. Manual focus was still better in almost all cases. I even updated the lens to version 2.0 today not realizing that is was out of date. Not sure if this would address any of these issues. This is a used lens and not sure if there is an issue. I read this from a previous discussion:  "  The algorithm that sets the focus point of the lens seems to place it at the very front of the DOF area ....which is fine but leaves very little room for error."

I have reached out to Leica in NJ. Waiting to hear back.

[LocalHero1953]

10 minutes ago, osroubek said:

Thanks for your suggestion. I tried raising the ISO and shooting at 125 sec, E shutter, IBIS off, spot focus. It really did not seem to improve the results that much if any. Manual focus was still better in almost all cases. I even updated the lens to version 2.0 today not realizing that is was out of date. Not sure if this would address any of these issues. This is a used lens and not sure if there is an issue. I read this from a previous discussion:  "  The algorithm that sets the focus point of the lens seems to place it at the very front of the DOF area ....which is fine but leaves very little room for error."

I have reached out to Leica in NJ. Waiting to hear back.

I recall that quote from the early launch of the 24-90 zoom, and it was solved with a f/w update. I haven't seen a similar comment for the 90-280.

[jaapv]

This reminds me of one of the Digiloyd-panics, where it turned out the camera tended to focus towards the front of a sloping target. I would advise you to use parallel targets and, as suggested, spot focus for your tests.

[osroubek]

I am using spot focus on the vertical portion of the target only. It seems to be better than the other focusing methods. Does everyone just stick to spot focusing? When would you use the other methods if ever? 

[Jeff S]

14 minutes ago, osroubek said:

I am using spot focus on the vertical portion of the target only. It seems to be better than the other focusing methods. Does everyone just stick to spot focusing? When would you use the other methods if ever? 

See my method above (works best when setting to linear focus throw via latest FW to camera and lens)…. this works well whether spot or field focus, depending on subject matter, and with static subjects using single AF mode rather than continuous.  Nothing beats magnified LV for accurate focus if time permits. A slight press of the shutter button returns to full view.

Jeff

[osroubek]

Do you have the focus throw setting set at maximum? This seems to give the most gradations of focus when doing it manually. 

[Jeff S]

No, at most 360, typically 270 (three quarters turn) suffices.

Jeff

[osroubek]

Thanks!

[osroubek]

After speaking with Leica, they think the lens needs repair. The specialist states the lens should be very sharp wide open as it is optimized this way. The thumbwheel holding the lens foot tight was loose as well, so I was able to return it and will get a new one instead. These are hard to find used, and most of them on Ebay are in Hong Kong!! Hoping for a better experience!! Here's a quick shot with my SL 75 mm using a tripod, spot autofocus, 1/15 sec, 2.2. Extremely sharp! 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Jeff S]

The 90-280 should not have any issues…. it’s a gem when operating normally.

Jeff

[Stuart Richardson]

Still...you may be disappointed if you are critical like this and comparing to an APO Summicron prime. As good as the zooms are, they are not as good as the apo summicrons. Don't let the marketing fool you that they are as good as the primes. They are not. They may be as good as the primes for certain less demanding tasks, but at the end of the day, the APO Summicrons perform better. Obviously the only one that overlaps exactly is the 90mm...

[Advertisement]

Hallo,

Look here: Leica L-Mount

[Jeff S]

58 minutes ago, Stuart Richardson said:

Obviously the only one that overlaps exactly is the 90mm...

Yep, and all my SL APO primes between 91 and 280mm really suck.  
 

But the long zoom is as good or better than comparable zooms, especially with the power of IBIS handheld (and possibly some software tricks).  And output is still fantastic at my modest print sizes, even compared to my longest SL75, which might be technically superior. 
 

Jeff