Reading mode: Light Dark

Warning about the 24-70mm F2.8

Go to last post
MightyMike said:
OS/VC/IS/OIS/VR is doing the same thing as you would see in an imaginary internal shift lens where the shift amount is minuscule. Shifting an element or an entire lens is not the same as decentering. Decentering is a tilting of one element or element group or an improperly glued together achromat. shifting doesn't cause 1 side or 1 or more corners to become out of focus. i suppose if the image circle isn't big enough for there was significant field curvature that could happen in a shift situation on a shift specific lens, however OS shifts by very minute amounts in comparison. So OS/VC/IS/OIS/VR does not cause decentering!
Click to expand...
Shifting an entire lens is not decentering. Just as shifting the sensor is not.

Shifting a lens element, or lens group is decentering though. Probably the dectering is small enough and the actual lens design is such that the image damage is low.
 
KL Matt said:
MightyMike said:
KL Matt said:
KL Matt said:
i feel it shows a clear indication that there are bad copies of the lens, if there are bad copies of the Tamron then there can just as easily be bad copies of its clone made in the same location with the same parts by the same people.
Click to expand...
I hate to disappoint you, but first of all there are bad copies of every lens out there. The question is how many. A sample size of two tells you little.

Tamron is certainly capable of building lenses to a high degree of consistency:

"TheTamron 45mm f/1.8 demonstrates pretty remarkable consistency, one of the most outstanding lenses we've tested as far as low copy-to-copy variation is concerned."

https://www.lensrentals.com/blog/2015/12/tamron-f1-8-vc-prime-lenses-sharpness-mtf-curves

But when you look at fast, complex, standard zooms like the 24-70 tammy and its Canon and Nikon equivalents, nobody is doing that great for consistency:

https://www.lensrentals.com/blog/2015/07/24-70-f2-8-zoom-mtf-and-variation

With Tamron basically holding its own.

Matt
Click to expand...
So you missed the point...
Click to expand...
No, it's that there isn't one to miss.
KL Matt said:
'm simply pointing out there is an above average occurrence of bad copies in technical reviews so its not hard to suggest and therefore expect an above average occurrence of bad copies for the general public.
Click to expand...
I guess that's easy enough to do, sure. What I'm saying is that you can do that, but it's still meaningless. There's a term for that, it's anecdotal evidence. A sample of two (or one, if it was the same lens in both cases) is not significant and does not allow anyone to draw meaningful conclusions.
KL Matt said:
I'm not bashing Tamron or Pentax, just pointing out indications that this lens could have quality control issues.
Click to expand...
Sure it could. Or like others have said here, the one copy that went to multiple reviewers was simply bad or damaged in transit. Sounds like a far simpler explanation than the dark conspiracy of substandard consistency and incompetent workers at Tamron production facilities.

LensRentals I think tests 10 copies, and they were all rather consistent. That says a lot more to me than one bad (and potentially damaged) copy that was shipped a lot, maybe without being packed properly.

Just wait for the Pentax, Mike. It will probably be fine. The chance of consistent quality with a proven design on a proven production line is higher than for an all new design I'm sure.
Click to expand...
I bet you wouldn't put money on it........I wouldn't, not where there's groups of glued couplets moving backwards and forwards in a zoom lens!
KL Matt said:
Matt
Click to expand...
--
Dave's clichés
 
Last edited:
Roland Karlsson said:
Leandros S said:
You're overlooking that one variant has VC, the other does not. Including VC will increase copy variation simply because there are more parts involved, and in the VC version, some parts are moving along axes not available in the non-VC version. So more can go wrong. Go ask Roger Cicala whether it's easier to repair a VC or a non-VC lens, he'll tell you.
Click to expand...
I am sometimes a sceptical guy.
Click to expand...
I was known to be skeptical, but now I've even doubts about that!
Roland Karlsson said:
And VC lenses is on my radar. A lens where one of the components deliberately is decentered? Does that sound OK?

--
/Roland
Kalpanika X3F tools:
https://github.com/kalpanika/x3f
Click to expand...
--
Dave's clichés
 
Last edited:
Roland Karlsson said:
MightyMike said:
OS/VC/IS/OIS/VR is doing the same thing as you would see in an imaginary internal shift lens where the shift amount is minuscule. Shifting an element or an entire lens is not the same as decentering. Decentering is a tilting of one element or element group or an improperly glued together achromat. shifting doesn't cause 1 side or 1 or more corners to become out of focus. i suppose if the image circle isn't big enough for there was significant field curvature that could happen in a shift situation on a shift specific lens, however OS shifts by very minute amounts in comparison. So OS/VC/IS/OIS/VR does not cause decentering!
Click to expand...
Shifting an entire lens is not decentering. Just as shifting the sensor is not.

Shifting a lens element, or lens group is decentering though. Probably the dectering is small enough and the actual lens design is such that the image damage is low.
Click to expand...
 
MightyMike said:
I disagree, put a teleconverter on a shift lens and you'll see that it doesn't get decentered when shifted,
Click to expand...
Of course it does. It is only matter of if it is important/visible or not.
MightyMike said:
decentering is an act of tilt
Click to expand...
Or shift. I promise you that you cannot shift a lens element all that much before it is a serious problem.
 
Roland Karlsson said:
MightyMike said:
I disagree, put a teleconverter on a shift lens and you'll see that it doesn't get decentered when shifted,
Click to expand...
Of course it does. It is only matter of if it is important/visible or not.
Roland Karlsson said:
decentering is an act of tilt
Click to expand...
Or shift. I promise you that you cannot shift a lens element all that much before it is a serious problem.
Click to expand...
Unless its already been designed as part of the overall design to be shifted without significant effect
 
Roland Karlsson said:
awaldram said:
Roland Karlsson said:
I am sometimes a sceptical guy. And VC lenses is on my radar. A lens where one of the components deliberately is decentered? Does that sound OK?

--
/Roland
Kalpanika X3F tools:
https://github.com/kalpanika/x3f
Click to expand...
And optically stabilisation element angle is altered to keep the image on the sensor stationary

This is within the light plane, no elements are deliberately decentered.!!!!!
Click to expand...
Of course they are. Without decentering you cannot move the image.

So you say that, while decentering, the lens element also is tilted?

Or did I get that wrong?

--
/Roland
Kalpanika X3F tools:
https://github.com/kalpanika/x3f
Click to expand...
Decentering an element does not move the image it just moves the light source off the apex of the concave or convex element leading to non linear magnification across the plane .

This causes blur on that part of the image resulting in big drops in mtf in the centre with peak sharpness moved to the part of the image where the apex of the decentered element is .

The rendered image does not move.

Any OiS mechanism correctly implented has no impact on image IQ do really think it would sell if every sr lens from Canon, Nikon with OiS was marginally blurred in the centre ??

I don't know where you got this strange theory but any research into either optics or examining mtf charts of OiS and non OiS lens would show the fallacy of such a concept.

Wiki is correct in its description pointing out the lens is moved at right angles to the light axis via electromagnetic S

--
My PPG
http://www.pentaxphotogallery.com/artists/andrewwaldram
My Photo Stream
http://www.flickr.com/photos/awaldram/
1x.com
http://1x.com/artist/awaldram/wall
 
Last edited:
DAVID MANZE said:
I bet you wouldn't put money on it........I wouldn't, not where there's groups of glued couplets moving backwards and forwards in a zoom lens!
Click to expand...
WHAT??? A zoom lens with moving elements and cemented couplets??? NOOOOOOOOOO!!!!!

Seriously though what is all this really about? Slandering a lens that does not even yet exist. What a bizarre topic of discussion. Have you ever taken a look at how complex any modern zoom is on the inside? Lots of movement going on in there. Both backwards and forwards! Or do you only buy zooms that move in just one direction???? ;-)

If I were in the market for a fast full frame standard zoom should I ever purchase a K1, I can assure you that the only lens that I would even consider would be the Pentax-branded one. For the seals alone. And because the Tamron won't likely be available due to the licensing of the design.
 
KL Matt said:
Seriously though what is all this really about? Slandering a lens that does not even yet exist.
Click to expand...
I appear to have totally lost the plot here! Which lens doesn't exist?

I have a Pentax D FA 24-70mm f/2,8.
 
KL Matt said:
KL Matt said:
I bet you wouldn't put money on it........I wouldn't, not where there's groups of glued couplets moving backwards and forwards in a zoom lens!
Click to expand...
WHAT??? A zoom lens with moving elements and cemented couplets??? NOOOOOOOOOO!!!!!
Click to expand...
I thought you would like that one!....
KL Matt said:
Seriously though what is all this really about? Slandering a lens that does not even yet exist.
Click to expand...
Slander!....ho ho ho! no just a little betting shop talk!
KL Matt said:
What a bizarre topic of discussion. Have you ever taken a look at how complex any modern zoom is on the inside?
Click to expand...
I have taken many a lens apart, zooms included, motor problems, cleaning/dust/fungus etc.
KL Matt said:
Lots of movement going on in there. Both backwards and forwards! Or do you only buy zooms that move in just one direction???? ;-)
Click to expand...
Yeah you know like one way streets!
KL Matt said:
If I were in the market for a fast full frame standard zoom should I ever purchase a K1, I can assure you that the only lens that I would even consider would be the Pentax-branded one.

For the seals alone.
Click to expand...
The Tamron version also has dust and humidity seals!

KL Matt said:
And because the Tamron won't likely be available due to the licensing of the design.
Click to expand...
The Tamron won't be available...... but there "will be" the inevitable run of lens problems de-centering, AF inaccuracies etc. probably par for the course within the normal run of modern zoom lenses, à là Pentax DA 16-85mm which had it's fair share of returns. Zooms are just more prone to centering issues. "All" brands included!

Given you get a good copy, I've no doubt that that it will be a very nice lens!

--
Dave's clichés
 
Last edited:
MightyMike said:
Unless its already been designed as part of the overall design to be shifted without significant effect
Click to expand...
Of course, and I assume the designers have done just that.

Still, it is shifted and it is decentered, even if the decentering adds neglect-able degradation.

Everything is a compromise.
 
awaldram said:
Decentering an element does not move the image it just moves the light source off the apex of the concave or convex element leading to non linear magnification across the plane .

This causes blur on that part of the image resulting in big drops in mtf in the centre with peak sharpness moved to the part of the image where the apex of the decentered element is .

The rendered image does not move.
Click to expand...
We have a matter of semantics here. Decentered means literally "not in the center", i.e. moved out of the center. And that is what I am talking about.

You are probably right though that when you talk about problems with "decentered" lenses, then it means that some lens elements are skewed. This will, of course, lead to bigger problems. The term "decentered" is then questionable, as the lens element still is in the center. But calling it "decentered" is common practice anyhow.
awaldram said:
Any OiS mechanism correctly implented has no impact on image IQ do really think it would sell if every sr lens from Canon, Nikon with OiS was marginally blurred in the centre ??

I don't know where you got this strange theory but any research into either optics or examining mtf charts of OiS and non OiS lens would show the fallacy of such a concept.

Wiki is correct in its description pointing out the lens is moved at right angles to the light axis via electromagnetic S
Click to expand...
Yes, OIS mechanism moves the lens element sideways, without tilting it. For small enough movements, and carefully designed lens, this is probably not a practical problem.

But, I promise you, that if you move the lens element far enough, you will start to see problems. You will then get degradation.
 
Roland Karlsson said:
But, I promise you, that if you move the lens element far enough, you will start to see problems. You will then get degradation.
Click to expand...
The same is true for IBiS you move the sensor far enough you will have issues the trick is to move the element or sensor to improve the image not degrade it.!

Remember the 5 axis SR now allows tilt and yew correction

another feather in Ibis cap that OiS cant replicate is rotational correction it doesn't matter how much you rotate a lens element it does nothing :D
 
Although less successful with long telephotros .. the fact that Lens stabilisation seesm to reduce optical p[erfoprmance is a great reason to go wit Olympus and Pentax who are putting increasingly good IBIS in body. Whilst not perfect for long telephoto for other lenses it gives room in theory for cheaper and better and simpler lenses
 
awaldram said:
The same is true for IBiS you move the sensor far enough you will have issues
Click to expand...
If you move it outside of the sharp image circle you might get unsharp or vignetting corners, yes. But, if you move a lens element too far you will degrade the entire image.
awaldram said:
the trick is to move the element or sensor to improve the image not degrade it.!
Click to expand...
Of course.
awaldram said:
Remember the 5 axis SR now allows tilt and yew correction
Click to expand...
Tilt is dangerous, yes.
awaldram said:
another feather in Ibis cap that OiS cant replicate is rotational correction it doesn't matter how much you rotate a lens element it does nothing :D
Click to expand...
They day lenses do rotational correction, I would be surprised :)
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top