Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
next PZ lens test report: Canon EF 24-105mm f/4 USM L IS II
#11
Sorry for having overwritten the 50mp chart with the 21mp one ... old habits ... 

The 50mp chart is back now.

#12
Frankly I am surprised with the performance of this lens, the days where Canon had no serious competitors and could release low quality products like they did with 17-85 and get away with it are over.

Having their best selling lens so so isn't a smart move.

Canon 24-70f4 would cover the gap however like did 18-135 with the failure of 18-200
#13
How is it possible to get higher resolution values on sensors with lower pixel densities and AA filters?

#14
Quote:How is it possible to get higher resolution values on sensors with lower pixel densities and AA filters?
It is not possible (unless you mess things up as camera manufacturer with wrong thickness sensor glass or something).

It is the testing methodology's fault, where sharpening is applied to the MTF test shots. The lower res. images extreme corners get lifted up quite significantly, and the higher res. extreme corners without AA-filter do not get touched by the lower amount of sharpening applied. 

 

By the way, a lack of AA filter mostly adds fake sharpness to MTF tests, not real sharpness. But at parts where the optics are not thaaat sharp, the optics act as AA filter and you won't see the MTF results being muddied by the fake sharpness.
#15
Quote:How is it possible to get higher resolution values on sensors with lower pixel densities and AA filters?
 

That is explained in the review
#16
I'm not sure about the explanation; especially about the variable sharpening - that would make comparing results with different cameras pretty much impossible, IMO.

But I'd like to hear your opinion (beside the short explanation from the review).

#17
The tests were NEVER cross-comparable - we mention this numerous times throughout the site. They are more or less comparable between cameras of similar resolution. If a review site using a chart-based test system states differently, they are plain lying to you.

 

Honestly I was never comfortable with the numeric LW/PH values because people tend to over-interpret those. These figures have always been "virtual". Again, this applies to all review sites.

 

Anyway, what we are analysing are images that resemble those produced by users. Or in other words - the images are ALWAYS sharpened to some degree. We tune the sharpening (unsharp masking) to a neutral value provided by the imatest tool. This translates comparatively mild sharpening (less than straight camera JPEGs). 


In essence this is similar to the discussion about AA filters - AA filters behave like reverse unsharp masking. So if you have two cameras with the same -say- 36mp sensor but different AA filters, the output is different. Further impacts come from micro-lenses, A/D converters and in-camera RAW pre-processing and external RAW processing (bayer interpolation). Point is - the output will always be different thus whatever you do - the base images are already not cross-comparable.

 

Now about the effect on those LW/PH values:

Imagine a black/white edge that is blurred across -say- 10 pixels.

Then imagine a black/white edge that is blurred across -say- 2 pixels.

What is the effect of your default sharpening in these cases ? Well, in the 1st case there's little sharpening at all because the sharpening radius is pretty much exceeded. In the 2nd case, the sharpening is in effect.

In Imatest this 2nd case is boosted in terms of LW/PH which is what you are seeing in the 21mp MTFs (relative to the 50mp MTFs).

 

Now if you compare a 50mp image with a 21mp image, the 21mp image will appear to be sharper ON PIXEL LEVEL because of the lower resolution. That is exactly the message of those MTF charts. So it is a statement how a lens performs relative to the underlying "sensor system".

 

Of course, we could provide pure figures without sharpening at all but then you wouldn't really see a connection between what you are seeing on screen and those charts. But what would be the point ? Nobody uses images without sharpening. 

(Note: images from a Foveon sensor would only be marginally be sharpened because they have a neutral sharpness to start with).

 

If you want cross comparable results, you have to look for MTFs obtained from an optical bench (without camera). Do these figures help you? Well, in my opinion they don't because it neglects the effect from the image sensor and digital post-processing.
#18
Quote:I'm not sure about the explanation; especially about the variable sharpening - that would make comparing results with different cameras pretty much impossible, IMO.

But I'd like to hear your opinion (beside the short explanation from the review).
Indeed, it gets more difficult to compare results, especially with the 5DSR results. People misinterpret the "bad corner performance" and claim the lenses are not upto the task (which, when you look at actual images made with the lenses and camera, clearly is not true).

 

I don't think it is a viable idea to have Klaus and Marcus start with a new methodology which among other things does not apply sharpening in the MTF workflow. It would make things a bit more comparable, but it would make comparing with the huge catalog of existing tests a big issue. And it would not address things like skewed results from AA-filterless tests, or even the different strengths of AA-filters (or some cameras having an AA-filter for just one direction, like Pentax in the past or the Nikon D610).

 

So, what remains is explaining to people what they actually are looking at, results wise.
#19
The comparing of results is really only FFF - food for fanboys (girls don't do that, maybe there's a reason?)

 

If one wants to decide for a camera system and only is looking for the sharpest lens, I meanwhile call that brainless as there's definitely much more to consider.

 

If one wants to know how good his lens is compared to the "enemy brand", it's not only brainless but straightforward stupid.

 

If one wants to find out which of three given lenses suiting to his camera system is the best package, then we're talking. Cross-comparisons are nothing to be taken seriously. However, testing batches of lenses to see sample variation between them - like lensrentals does - is something to animate manufacturers to invest maybe a bit more into quality - but since that money and a lot customers only buy what they were told to be a "great brand", themanufacturers can lean back and confuse customers by more lenses, more options of standard workhorses to feed the trench war fighters.

#20
Klaus: people are comparing results all the time - and even you did that, sort of, in this review. Put all the warning you want, they will be ignored.

So the sharpening issue is considering pixels, instead of a standardized output size (the old "screen vs. print" dilemma).

 

JoJu, regardless of the fanboys, what I'm talking here is the idea that going for a higher resolution camera would give you worse results. Some people believe that.

  


Forum Jump:


Users browsing this thread:
1 Guest(s)