sharpness issues with the Sony 20-70mm f4

Hi all,

I recently bought a 20-70mm f/4 G lens,

and I was wondering if this is to be expected or did I just end up with a bad copy.

I mostly use the 35mm 1.8, but when I got this lens, I instantly noticed that the picture quality was lower than my 35mm. Of course this is a comparison between prime and zoom lens, but even with that, I was really surprised of the quality difference between non-G prime lens vs G zoom lens



This above photo is taken from the new zoom lens, 200% magnification in the center of the image. with f/4, iso 100, 35mm, 1/250, the distance to the subject is about 9 ft. (3m)



This is the same settings, but taken with my prime lens (35mm), at the same aperture f/4.

The quality was so different that it actually surprised me quite a bit. This sample images were not taken from the extreme corners, but are from the center of the image, but to me, it felt like too much details were missing.

I've seen many sample images taken with the 24-70, but none seems to have image sharpness issues as mine.

Is this expected?

Thank you!

Hi,

I have multiple small primes and the 20-70. I've compared them very carefully. I wouldn't expect to see this level of difference in the center.

There's no EXIF. You've said that they were both shot at f4. I see two images with slightly different DOF. The main subject is sharp in the second image shot with the 35mm prime but the background vegetation seems to me to be more OOF than in the image from the zoom. The backgrounds should be very similar. I know full well that lenses have different bokeh, but the difference here is enough to say that there's more at work.

These two images suggest to me that the focus point was moved or that a different aperture was used or both. You've confirmed that the same aperture was used, so I would suggest that the focus point may have been moved very slightly between shots, probably moving the focus point for the zoom slightly behind the subject (which would render it OOF and the background vegetation less OOF, which it is). How was the focusing done? Was it confirmed with EVF magnification?

I haven't found a significant difference between my primes and the zoom in terms of the 20-70's central sharpness. So my experience is different from yours. I observe the primes to be a tad sharper at the borders and in the corners at the widest shared aperture of f4. The differences diminish as I stop the 20-70 down. The primes still retain an advantage, but it's incremental.

That analysis is all from pixel peeping at high magnifications - 200%-400%. That scales to prints 2m to 4m wide - sizes I will never make. I've concluded that for the photography I actually do - online viewing on a 27" Retina screen and prints to 400x600mm max - those incremental differences are barely visible (if at all). So for me, the minor differences in the zoom's IQ aren't a disadvantage,

Regards,

Rod

Rod McD said:

Hi,

I have multiple small primes and the 20-70. I've compared them very carefully. I wouldn't expect to see this level of difference in the center.

There's no EXIF. You've said that they were both shot at f4. I see two images with slightly different DOF. The main subject is sharp in the second image shot with the 35mm prime but the background vegetation seems to me to be more OOF than in the image from the zoom. The backgrounds should be very similar. I know full well that lenses have different bokeh, but the difference here is enough to say that there's more at work.

Click to expand...

Yeah to me it looks like a clear focus issue, the background shouldn't be more in focus on the 1st shot if the lens/camera had properly focused in the subject, it's clearly focused behind it. Test needs to be re-run to figure out if it's a testing issue or not... I've used AF at times myself when testing lenses just to make things faster but this is a good argument for why MF will eliminate variables.

Rod McD said:

These two images suggest to me that the focus point was moved or that a different aperture was used or both. You've confirmed that the same aperture was used, so I would suggest that the focus point may have been moved very slightly between shots, probably moving the focus point for the zoom slightly behind the subject (which would render it OOF and the background vegetation less OOF, which it is). How was the focusing done? Was it confirmed with EVF magnification?

Click to expand...

Even shifting around after acquiring focus could throw things off depending on the distance etc... Another argument for testing at father distances first which are less likely to screw up a test.

Rod McD said:

I haven't found a significant difference between my primes and the zoom in terms of the 20-70's central sharpness. So my experience is different from yours. I observe the primes to be a tad sharper at the borders and in the corners at the widest shared aperture of f4. The differences diminish as I stop the 20-70 down. The primes still retain an advantage, but it's incremental.

That analysis is all from pixel peeping at high magnifications - 200%-400%. That scales to prints 2m to 4m wide - sizes I will never make. I've concluded that for the photography I actually do - online viewing on a 27" Retina screen and prints to 400x600mm max - those incremental differences are barely visible (if at all). So for me, the minor differences in the zoom's IQ aren't a disadvantage,

Regards,

Rod

Click to expand...

The 20-70 is much better than what you picture shows.

I compared mine side by side at 35mm with my FE 35 F1.4 GM and Loxia 35/2 and the FE20-70 realy held up very well, it was hard to tell a differente at almost all apertures.

Try side by side test on tripod with manual focussing first, to rule out any AF issues. If you keep getting bad images with AF then there is something wrong with the AF.

I can only agree about the good IQ of the 20-70/4, i compared it with the Loxia 21 and to my eyes the 20-70 was even better or at least par.

Massinissa said:

I can only agree about the good IQ of the 20-70/4, i compared it with the Loxia 21 and to my eyes the 20-70 was even better or at least par.

Click to expand...

Same here - sterling performer which stands up to pixel peeping at 60Mp even at F4 - rather surprised me TBH

--
** Please ignore the Typos, I'm the world's worst Typist **

Any time the first result seems off, I will absolutely re-test.

I've encountered a few lame lenses over the years. But more often than not, what's really defective is me!

I find it's a real good habit to zoom in to confirm I got the shot, whenever time permits.

AF is very handy sometimes, but it sure isn't foolproof!

--

chipmunk12 said:

Hi all,

I recently bought a 20-70mm f/4 G lens,

and I was wondering if this is to be expected or did I just end up with a bad copy.

I mostly use the 35mm 1.8, but when I got this lens, I instantly noticed that the picture quality was lower than my 35mm. Of course this is a comparison between prime and zoom lens, but even with that, I was really surprised of the quality difference between non-G prime lens vs G zoom lens



This above photo is taken from the new zoom lens, 200% magnification in the center of the image. with f/4, iso 100, 35mm, 1/250, the distance to the subject is about 9 ft. (3m)



This is the same settings, but taken with my prime lens (35mm), at the same aperture f/4.

The quality was so different that it actually surprised me quite a bit. This sample images were not taken from the extreme corners, but are from the center of the image, but to me, it felt like too much details were missing.

I've seen many sample images taken with the 24-70, but none seems to have image sharpness issues as mine.

Is this expected?

Thank you!

Click to expand...

Your focus mis hit on the 20-70 as can clearly be seen by the background being more in focus then on the prime… you need to be absolutely sure that focus is at the exact same point or a comparison is worthless