[Klaus]

Sounds good, ta....sweet soft bits too....I was thinking the edges difference 21Vs50mp's might work out this way....i like to think that Canon are putting all the extra pixels in the centre of the sensor :-)


.....and just out of interest - ("We are also calibrating the RAW processing to achieve "neutral" sharpening meaning that the processed image is neither over- nor undersharpened as far as highest resolution spot is concerned") -
Still in C! but not at their standard settings?

No, we (try to) calibrate for an identical sharpness perception on pixel level.

However, we don't use straight JPEGs anyway so the camera defaults are irrelevant.Well, except for the vignetting figures.

[Rainer]

Seem PZ has lowered its standards ...

a lens with such mediocre borders and corners wide open

would not have got more than three stars some years ago.

(just compare with the 24/1.4 II)

But eventually there is an ART bonus somewhere.

 

Rainer

[JJ_SO]

Eventually you should consider to name another fast 20 mm with better borders? or give the lens a try and find out for yourself if it's worth the recommendation?

[soLong]

Deleted

[Rainer]

Eventually you should consider to name another fast 20 mm with better borders? or give the lens a try and find out for yourself if it's worth the recommendation?

 

I have no doubts the lens is worth the recommendation ... I just recalled, when the

review of the 24L II came out and the review of a comparable Zeiss (21/2.8) which

have absolutely comparable performance at comparable f-stops ... the 3 stars for

the Canon were justified by the mediocre border performance wide open ... but as

it seems the thinking within PZ has changed.

[Klaus]

I have no doubts the lens is worth the recommendation ... I just recalled, when the

review of the 24L II came out and the review of a comparable Zeiss (21/2.8) which

have absolutely comparable performance at comparable f-stops ... the 3 stars for

the Canon were justified by the mediocre border performance wide open ... but as

it seems the thinking within PZ has changed.

 

There's no corner performance of the Canon 24mm f/1.4 II to speak of at f/1.4 and the corners are also substantially worse at f/2.

I also have to recognize that this a 20mm lens where the optical problems are a magnitude more difficult than at 24mm. 

It's also a world first whereas today's 24mm f/1.4 are me-too lenses.

 

I concur, however, that the Canon should deserve a half star more though given the peak performance.

[Guest]

Fwiw, I wanted to get a Sigma 35 Art for my D750, tried 3 copies in two stores, all have significant front/back focusing. Really don't feel like sending camera and lens to sigma for calibration, or use the USB dock, so I got a 35 1.8 nikkor in the end

[Rainer]

There's no corner performance of the Canon 24mm f/1.4 II to speak of at f/1.4 and the corners are also substantially worse at f/2.

I also have to recognize that this a 20mm lens where the optical problems are a magnitude more difficult than at 24mm. 

It's also a world first whereas today's 24mm f/1.4 are me-too lenses.

 

I concur, however, that the Canon should deserve a half star more though given the peak performance.

 

Regarding the corner performance ...

... same applies for the Sigma 20/1.4 @1.4 and @2

 

Regarding the difference of 20mm and 24mm ...

... admitted ... but my original comment was the (in my eyes) bad judgement of the 24L in

comparison to the Zeiss 21/2.8 ... I foresee that you point out that 21mm again is something

completely different.

 

Regarding "worlds first things" ...

... That had also applied to the old Sigma 20/1.8 ... which you gave 1.5 stars ...

where was the "worlds first" bonus then? (ok .. the old 20 was testet on APS-C only).

 

I daresay that I foresee the day, when the Sigma 24/1.4 tested on 50mpix will receive

its 4 stars (deservedly though) ... besides the fact that it basically has the same

performance than the 24L.

 

No offence meant Klaus ... I'm just stating the obvious ... you seem to have adjusted

your judgement. I neither find this good nor bad ... just "good to know" ... eventually

you finally arrived in the reality of what is economically producable (rather than

optically desirable). Its just that with that, your judgment of newer tests is then no

longer comparable to that of older tests.

 

[Klaus]

Regarding the corner performance ...

... same applies for the Sigma 20/1.4 @1.4 and @2

 

Regarding the difference of 20mm and 24mm ...

... admitted ... but my original comment was the (in my eyes) bad judgement of the 24L in

comparison to the Zeiss 21/2.8 ... I foresee that you point out that 21mm again is something

completely different.

 

Regarding "worlds first things" ...

... That had also applied to the old Sigma 20/1.8 ... which you gave 1.5 stars ...

where was the "worlds first" bonus then? (ok .. the old 20 was testet on APS-C only).

 

I daresay that I foresee the day, when the Sigma 24/1.4 tested on 50mpix will receive

its 4 stars (deservedly though) ... besides the fact that it basically has the same

performance than the 24L.

 

No offence meant Klaus ... I'm just stating the obvious ... you seem to have adjusted

your judgement. I neither find this good nor bad ... just "good to know" ... eventually

you finally arrived in the reality of what is economically producable (rather than

optically desirable). Its just that with that, your judgment of newer tests is then no

longer comparable to that of older tests.

 

 

 

It is notoriously difficult to judge a lens with an extreme variations between the worst and the best spot.

Just taking the line integral of the curve is pointless (and the MTFs aren't the sole aspect of truth anyway).

 

It may be that I got a bit wiser in the sense that the extreme corner performance is not exceedingly important for most use cases - that is is for high speed wide-angles. The same logic doesn't really apply to say a 70-300mm f/4.5-5.6 zoom lens which should be sharp at max. aperture really.

 

PS: The Sigma 20mm f/1.8 may have been a world first but honestly ... it is crap even at f/4.