Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

1
       2       end
  

Archive 2016 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest

  
 
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #1 · p.1 #1 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


I saw a post somewhere that asserted that if you were taking pictures with old 24-105/f4 v1; why are you bothering to buy a 5dsr. You could achieve similar resolution with 5dii.

This got me to thinking about the age old question about Prime vs Zoom and AA vs non AA and better vs lessor zooms..

Here is DXO measures as reference:
https://www.dxomark.com/lenses/launched-between-1987-and-2016/focal-from-1-to-35/lens_zoom-prime-zoom/sensor_brand-Canon#hideAdvancedOptions=false&viewMode=list&yDataType=rankDxo

What I get out DxO based on comparing 24-105v1 vs 24-70v2 vs 35v2/1.4 is:

5DIII 5DSR
24-105f4v1 = 15 5diii =18 5dsrequiv mpx 20% improvement with 5dsr at 35mm
24-70f2.4v2 = 18 5diii = 32 5dsr 80% improvement with 5dsr (at 24mm)
35/1.4/v2 = 18 5diii = 37 5dsr 15% improvement 35 prime over 24-70v2 zoom

This confirms, ignoring everything but resolution, if you stick with 24-105/f4/v1, you will get significant improvments but you could get similar improvements by either buying a better lens (24-70 in this case) or buying a 5DSR. But the 24-105 really limits the 5DSR gains. [maybe the 24-70 does too at 35mm - Dxo does not say. I suspect so. But looking at the area under the - - - - line it looks like 70mm on 2-70 would be not more than 15% worse
http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=0&API=2&LensComp=787&FLIComp=2&APIComp=2&CT=AVG

add: the digital picture shows 24-105 having noticeable improvement on 24-105 and 35/ii having noticeable improvement on 24-70
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=787&Camera=979&FLI=2&API=4&LensComp=355&Sample=0&SampleComp=0&CameraComp=979&FLIComp=0&APIComp=0]

2)
Zeiss Otus 85/1.4 = 41
Zeiss Otus 55/1.4 = 37
Canon 35/1.4 II =37
Zeiss 25/2 = 32

Canon 24-70/2.8 v2 32 at 24mm.[likey in high 20's at 50mm based on link above compared to 24mm ]

This confirms most good primes are 25% better than really good zooms (eg 24-70v2), So if you have a really good prime use it first over your zoom. Remember that 24-70 is best at 24, thus it is likely to be worse at 35 for example, and you would be better to use the prime. Eg the zoom is not likely to be great across the board. In fact, until there is better data, its not clear that the 24--105 may be better at 35/f8 than the 24-70 is. Or maybe at f8 the gap is not as big.

3)
Nikon d800 otus 55/1.4 =29
Nikon d800e otus 55/1.4 =33

I guess this leads to generalization - if your primary concern is resolution:
1) Buy the 5dsr over the 5ds
2) Use primes where you can
3) Use the better zooms

Anyway, that's what the data from DXOmark suggests.

Obviously there are other important factors such as
1) some zooms have IS some don't
2) longer zooms eliminate carrying lots of lens
3) weight
4) field curvature
5) need fstop
6) cost....

Anyway maybe this was obvious to all but just posting it to test with people here and in case its helpful.


-----------------------------

[after much reading and musing I have concluded below:

dxomark is not a reliable tool to make conclusions because i suspect they have copy/quality issues and focusing near - which they do - is not reliable for landscape use testing.

If you have a really good zoom (e.g. 24-70v2) - its just as good as a prime, and if you have to crop the prime to get your composition its better than a prime

but if you have a good zoom (24-105f4v1) the prime will be better if you just so happen to have a prime that matches your composition otherwise the 24-105v1 will be just as good or better.

24-105 v1 visually and testing basis near infinity is not that far behind 24-70v2 except at 24mm where barrel distortion is heavy. At f8 >30mm they are close.]


-------------------------------

Scott

Edited on Oct 20, 2016 at 01:25 AM · View previous versions



Oct 18, 2016 at 02:27 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


This is a somewhat complex issue. (You somewhat acknowledge this with the list at the end of your post.)

Let me just take the "primes provide more resolution than zooms" point. In fact, the best primes can generally provide more optical resolution than the best zooms... at least at apertures where the primes are at their best.

But it isn't that simple.

Let's say that I am photographing landscape. I find a subject and I work out an ideal composition — a position defined by my position relative to the subject on three axes (forward/backward, left/right, and up/down) and the subsequent selection of a focal length that frames the composition the way I want. Let's further say that it turns out that my ideal framing after the other axis decisions have been made requires a 35mm focal length.

If I have an excellent 35mm prime in my bag it may provide more (though not a lot more) optical resolution than my best zoom at 35mm. (Or maybe not, depending... but let's say it does.) But at what aperture? If I'm going to photograph at f/11 or f/16, which is not at all unusual for full frame landscape photography, any difference is minimized to the point that it may or not actually be perceptible in even a very large print. So, yes, the prime might be better in this case — where I have the ideal focal length prime in my bag — though the extent of the difference may well be quite (or "vanishingly") small.

However, many of us like to use focal length as a more precise way to control the framing around our ideal composition. Let's say that the ideal framing uses a 41mm focal length. If I use primes, I will pick the prime whose angle of view is a bit larger than what I need for the photograph, with the expectation that I'll crop in post. (Or, I'll compromise my ideal framing, perhaps at the expense of my ideal distance, etc... which is not my preference.) So, let's say I have that 35mm lens in my bag of primes — I'll use that and crop a bit in post to get to what I would get at 41mm with my zoom.

You probably already know where I'm going with this. Assuming I'm using an excellent zoom (which is what I use), the image quality at 41mm is essentially the same as that at 35mm, so I can "crop in camera" with no real image quality loss — I still get to use all the pixels of the sensor, etc. But the photographer using the prime will lose some (or all?) of any resolution advantage that the 35mm prime might have over the zoom, and in some cases the result could end up having less resolution than the zoom.

I can already hear one answer: Simply find a good composition with a 35mm lens or another prime that you have. I have two responses to that idea:

First, that made perfect sense back in the day when virtually everyone only had prime lenses to work with — it was either use the lens you have or don't make the photograph.

Second, in this case we are really just making a choice about which particular compromise we will accept. While it is true that we might make an optical resolution compromise (relative to the most ideal prime result) when we choose to use a zoom, by accepting a less than ideal framing of the subject as a side effect of only using primes one is also compromising. Some my feel differently than I do, but when I consider the truly minimal compromise of image quality (and only in the ideal cases where the prime would be perfect for framing the shot) versus the compromise of not being able to frame my subject in the way I consider ideal, for my part I'll take the zoom over the prime.

I arrived at this perspective over some period of time. I've done this long enough to recall a time when primes were the only option. Long after quite good zooms became available I continued to trust that primes would give me better results. During my transition period I often went out armed with up to nine lenses, of which 3-4 might be zooms and the rest were primes. Over time I began to realize that I was getting equal (or better, if the alternative had been cropping) results from the zooms and I simply found myself using the primes less and less for landscape photography... until I finally (it took years!) realized that for me zooms are simply a better option for this kind of photography. (There are exceptions — for example using TS lenses... but even there you can adapt MF zooms.)

Do I not like primes? That's not the case at all. I own and use a bunch of them... but not because they are optically better but because they sometimes are more functionally useful in certain kinds of photography. For example, I generally prefer primes for street photography because they give me one less option to think about when responding to the rapidly changing street environment. I may also choose them sometimes because I simply need the larger apertures that some of them provide.

Another point of view... :-)

Dan

Edited on Oct 19, 2016 at 06:07 PM · View previous versions



Oct 18, 2016 at 03:27 PM
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #3 · p.1 #3 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


gdanmitchell wrote:
This is a somewhat complex issue. (You somewhat acknowledge this with the list at the end of your post.)

Let me just take the "primes provide more resolution than zooms" point. In fact, the best primes can generally provide more optical resolution than the best zooms... at least at apertures where the primes are at their best.

But it isn't that simple.

Let's say that I am photographing landscape. I find a subject and I work out an ideal composition — a position defined by my position relative to the subject on three axes (forward/backward, left/right, and up/down) and the subsequent selection
...Show more

http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=1&API=3&LensComp=917&APIComp=5&CT=AVG

confirms what you say at f8, there is not much difference between zeiss otus 55 and 24-70 at 50mm.

however at wide open there might be 15% difference.

http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=1&API=0&LensComp=917&APIComp=2&CT=AVG

And if you cropping a 55 to 41, you have lost 25% of the theorectical resolution or so.
So right again.

But if 35 vs 35, the differences are significant. And don't forget the zeiss mico-contrast

------------------------

So I revise my rules of thumb above - If you want to maximize resolution:
1) Use the 5dsr over the 5ds
2) Use the 5dsr over all others
3) If the scene works for one of your high quality primes use it first
4) But otherwise use the best zoom you can get

Again ignoring focus speed, fps, dynamic range, filter that you may or may not have, you can only carry so much etc.



Oct 18, 2016 at 03:43 PM
Usagi
Offline

Upload & Sell: Off
p.1 #4 · p.1 #4 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Scott Stoness wrote:
http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=1&API=3&LensComp=917&APIComp=5&CT=AVG

confirms what you say at f8, there is not much difference between zeiss otus 55 and 24-70 at 50mm.

however at wide open there might be 15% difference.

http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=1&API=0&LensComp=917&APIComp=2&CT=AVG

And if you cropping a 55 to 41, you have lost 25% of the theorectical resolution or so.
So right again.

But if 35 vs 35, the differences are significant. And don't forget the zeiss mico-contrast

------------------------


15% difference at wide open doesn't sound much. How visible it would be in the final print of landscape image? Assuming that someone shoots landscapes at wide open.




Oct 19, 2016 at 05:26 AM
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #5 · p.1 #5 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


15% is not a large difference and likely it would only 7% larger prints so 30" wide could grow to 32" wide on a print with equivalent quality.

However the difference between 5dsr and 5diii is very large - e.g. 50" wide on similar basis as 32" above with same quality ( assuming 1.66 for combination of 50mpx vs 22mpx and 1.10 for no AA).

However I am a bit sceptical of my dxo/mtf conclusions of 15% - when I look at the digitalpicture visually the 35/1.4 v2 prime is visually better at f8. I would not expect to see a difference of only 15%. So I suspect it is more like 25%.

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=787&Camera=979&Sample=0&FLI=0&API=4&LensComp=994&CameraComp=979&SampleComp=0&FLIComp=0&APIComp=6



Oct 19, 2016 at 07:48 AM
alundeb
Online
• • • • •
Upload & Sell: Off
p.1 #6 · p.1 #6 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


I agree that upgrading camera resolution from 22 to 50 MP will be much more noticeable than upgrading lenses, generally speaking.

The improvement will depend heavily on where you look though. Near the center of the image, camera resolution is dominant, and near the corners lens resolution is dominant. Unless I have a specific problem with corner smearing, I don't use primes over zooms if I can use a zoom in the situation.

Sometimes I use small primes as gap fillers when I don't bring a full kit of zooms.



Oct 19, 2016 at 07:59 AM
dhachey
Offline
• • •
Upload & Sell: Off
p.1 #7 · p.1 #7 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


I use whatever will give me the image I want. I try to previsualize the shot, and select the best gear to give it to me. Fortunately, I own a lot of prime and zoom lenses to choose from.


Oct 19, 2016 at 08:08 AM
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #8 · p.1 #8 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Scott Stoness wrote:
I saw a post somewhere that asserted that if you were taking pictures with old 24-105/f4 v1; why are you bothering to buy a 5dsr. You could achieve similar resolution with 5dii.

This got me to thinking about the age old question about Prime vs Zoom and AA vs non AA and better vs lessor zooms..

Here is DXO measures as reference:
https://www.dxomark.com/lenses/launched-between-1987-and-2016/focal-from-1-to-35/lens_zoom-prime-zoom/sensor_brand-Canon#hideAdvancedOptions=false&viewMode=list&yDataType=rankDxo

What I get out DxO based on comparing 24-105v1 vs 24-70v2 vs 35v2/1.4 is:

5DIII 5DSR
24-105f4v1 = 15 5diii =18 5dsrequiv mpx 20% improvement with 5dsr
...Show more

Interestingly - when I look at thedigitalpicture images comparison the 24-105 is not 1/2 the resolution as 24-70.

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=787&Camera=979&Sample=0&FLI=2&API=4&LensComp=355&CameraComp=979&SampleComp=0&FLIComp=2&APIComp=3

So I don't know it this is because dxo images are based on a bad copy - which they sometimes do based on looking at their Samyang 15/2.8 on 5dsr - or there is something else wrong with dxo or my visual perception of a slight difference tanslates into a large gain in resolution.

This might change the outcome.

-----

Further - when I compare lenstip 24-105v1 vs 24-70v2 they are virtually identical at f8; and 24-70v2 is certainly not 80% better,.
http://www.lenstip.com/240.4-Lens_review-Canon_EF_24-105_mm_f_4L_IS_USM_Image_resolution.html
http://www.lenstip.com/358.4-Lens_review-Canon_EF_24-70_mm_f_2.8L_II_USM_Image_resolution.html

Dxo must have had a bad copy or bad test of 24-105 !

Which means:
1) first get a 5dsr over 5ds
2) Next get a 5dsr over any other
3) Next generally if you are not cropping for composition based on a prime, use the prime
4) Next 24-105 and 24-70 do not seem to be dramatically different in resolution. 24-70v2 is a bit better (enough to visually noticeable at edges but not dramatic) at edges but it lacks IS and 70-105. This question is still up in the air. At f8, dxo says it matters a lot but lenstip and the digitalpicture indicate that it does not.

---------------------
follow up: lens rentals.com shows
https://www.lensrentals.com/blog/2014/03/a-24-105-comparison/

that canon 24-105 is almost the same in iq as sigma 24-105

but dxomark show 18 for canon and 25 for sigma - I think this confirms that either dxo had a bad copy or their mpx measure is something quite different than I thought it was.

But definitely 24-70 is not 80% better than 24-105 canon.



Edited on Oct 19, 2016 at 09:06 PM · View previous versions



Oct 19, 2016 at 12:36 PM
alundeb
Online
• • • • •
Upload & Sell: Off
p.1 #9 · p.1 #9 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Somewhere in the middle of Dan's post this thread went italic. What kind of magic is that?


Oct 19, 2016 at 02:59 PM
jcolwell
Offline
• • • • • • •
Upload & Sell: On
p.1 #10 · p.1 #10 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


alundeb wrote:
Somewhere in the middle of Dan's post this thread went italic. What kind of magic is that?


I think it's far too complex to address in an unclassified, online forum, such as this.



Oct 19, 2016 at 03:05 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.1 #11 · p.1 #11 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


alundeb wrote:
Somewhere in the middle of Dan's post this thread went italic. What kind of magic is that?


Sorry. I fixed that.

Dan



Oct 19, 2016 at 06:08 PM
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #12 · p.1 #12 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


https://www.lensrentals.com/blog/2014/03/a-24-105-comparison/

is a a very good read which compares 24-105/f4 sigma vs canon and concludes:
- They are very close
- Both are great
- Canon is sharper at edges some spots/ sigma others
- You cannot assume that mtf or visual tests at 4' (the digital picture images and all of dxomark and photozone) are representative of what you can achieve at near infinity. This conclusion is worth repeating because virtually everyone but lens rentals tests wide and uwa lens at very near which is not where I shoot the majority of my landscapes. I guess this means I can only use lens rentals measures

Why do I go on a tangent - the question is prime vs zoom ? because my original assumption was dxo was right in their measurements, whereas the reality is 24mm measurements are too close focussed to be reliable [and dxomark seems to have some quality control issues with their testing based on 24-70v2 vs 24-105v1 and the low rating they give samyang 14/2.8 on 5dsr which does not make sense given that it achieve great results on 36mpx nikon].

so going back to thedigitalpicture mtf (comes from lens rentals testing focussed at near infinity rather than close):

the canon 24-70v2 achieves pretty close to zeiss 25/2 (not quite as good but likely not visually noticeable)
http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=0&API=3&LensComp=780&APIComp=4&CT=AVG
the canon 24-70v2 achieves pretty close to zeiss outus 55 (not quite as good but not visually noticable)
http://www.the-digital-picture.com/Reviews/MTF.aspx?Lens=787&FLI=1&API=3&LensComp=917&APIComp=5&CT=AVG
the canon 24-70v2 achieves pretty close to zeiss outus 85 (not quite as good with significantly more astigmatism)

So I would conclude from this and grant's advice (if you crop the prime you don't achieve what is promised) that the 24-70 is just as good as a prime.

I would also conclude from lens rentals that the 24-105 is not that far behind the 24-70 e.g.
lens rentals shows 24-105 being about 20% behind the 24-70 at 24mm and 70mm and 24-105 wupping the 24-70 but infinity at 105mm.
https://www.lensrentals.com/blog/2013/01/canon-24-70-f4-is-resolution-tests/

So if you have a really good zoom (24-70v2) - its just as good as a prime, but if you have a good zoom (24-105) the prime will be better if you just so happen to have a prime that matches your composition otherwise the 24-105v1 will be just as good or better.

I feel like I am having a conversation with myself but hopefully my musings are helpful. I am done and leave you to challenge this conclusion.



Oct 19, 2016 at 09:42 PM
jcolwell
Offline
• • • • • • •
Upload & Sell: On
p.1 #13 · p.1 #13 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


I use manufacturer's MTF curves to get an idea of what to expect from a lens. Both Canon and Zeiss MTF curves are based on physical tests for infinity focus, both wide open and at a single stopped-down velue, usually f/8 for Canon and f/5.6 for Zeiss.


Oct 20, 2016 at 06:15 AM
alundeb
Online
• • • • •
Upload & Sell: Off
p.1 #14 · p.1 #14 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Jim, I believe Canon MTF charts on the web site and in the EF lens book are calculated and even don't account for diffraction. I think they are calaculated at infinity though, and in my experience they give a fair indication of what to expect, when adjusted down for diffraction.

Or do you know about another set of Canon MTF charts?



Oct 20, 2016 at 07:25 AM
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #15 · p.1 #15 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


I posted this elsewhere but will post here to have a complete thread:

According to Dxomark (which is tested at near which is problematic for 24mm but not as problematic at 105mm), the old 24-105f4 is not that different than 24-70v2 at f8 or f11 where the majority of landscape shots occur. In fact it is not that bad at 24mm (see lens rentals above that concludes perhaps 20%).

24/f8 very small
https://www.dxomark.com/Lenses/Compare/Side-by-side/Canon-EF-24-70mm-F28L-II-USM-on-Canon-EOS-5DS-R-versus-EF24-105mm-f-4L-IS-USM-on-Canon-EOS-5DS-R__886_1009_164_1009

35mm / f8 - very small
https://www.dxomark.com/Lenses/Compare/Side-by-side/Canon-EF-24-70mm-F28L-II-USM-on-Canon-EOS-5DS-R-versus-EF24-105mm-f-4L-IS-USM-on-Canon-EOS-5DS-R__886_1009_164_1009

50mm/f8 very small
https://www.dxomark.com/Lenses/Compare/Side-by-side/Canon-EF-24-70mm-F28L-II-USM-on-Canon-EOS-5DS-R-versus-EF24-105mm-f-4L-IS-USM-on-Canon-EOS-5DS-R__886_1009_164_1009

70mm/f8 only different on edges by modest amount
https://www.dxomark.com/Lenses/Compare/Side-by-side/Canon-EF-24-70mm-F28L-II-USM-on-Canon-EOS-5DS-R-versus-EF24-105mm-f-4L-IS-USM-on-Canon-EOS-5DS-R__886_1009_164_1009



Oct 21, 2016 at 09:55 PM
jcolwell
Offline
• • • • • • •
Upload & Sell: On
p.1 #16 · p.1 #16 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


alundeb wrote:
Jim, I believe Canon MTF charts on the web site and in the EF lens book are calculated and even don't account for diffraction. I think they are calaculated at infinity though, and in my experience they give a fair indication of what to expect, when adjusted down for diffraction.

Or do you know about another set of Canon MTF charts?


Hi Anders,

I'm going by this,

"Canon tests lenses consistently, so that to the greatest degree possible, MTF
results from two or more lenses from similar lens categories can be meaningfully
compared. The curved lines you see on the MTF charts display the results a given
lens delivered.

Canon performs MTF tests at two different lens apertures: wide-open ...And, a second
test is performed with the lens stopped-down to f/8..."


Ref. http://www.learn.usa.canon.com/resources/articles/2013/reading_MTF_charts.shtml



Oct 22, 2016 at 07:10 AM
alundeb
Online
• • • • •
Upload & Sell: Off
p.1 #17 · p.1 #17 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


jcolwell wrote:
Hi Anders,

I'm going by this,

Ref. http://www.learn.usa.canon.com/resources/articles/2013/reading_MTF_charts.shtml


That was an interesting read, Jim, and especially this part:

"A completely hypothetical “perfect” lens, which doesn’t exist in real life, would render MTF results with all its lines running as a completely horizontally straight line across the very top of the chart from the left to right edge. All thin and thick, and solid and dashed lines, would be stacked on top of each other. Unfortunately, no lens is likely to deliver this type of performance in an MTF test anytime soon!"

It is clear that diffraction is ignored here, and we also see in the MTF values for 30 lp/mm, that for some lenses exceed 0.98 at f/8. That is theorethically impossible far beyond any margin of error or misinterpretation.

The theorethical limit for MTF at 30 lp/mm at f/8 is about 0.85. It will vary with the wavelength you put into the calculations, but here is a reference that is realistic:

http://photo.net/learn/optics/mtf/

Either Rudy Winston doesn't use the term "test" as in "actual measurement on a physical lens", or he doesn't know what he is talking about. You cannot measure a lens and not get the effect of diffraction in your results, that is a fact.



Oct 22, 2016 at 11:25 AM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.1 #18 · p.1 #18 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Diffraction is "ignored" because it is not the result of the characteristics of a particular lens being tested or rated. (Ignoring diffraction in a lens test is sort of like ignoring gravity in a test of vehicle traction.)

Dan



Oct 22, 2016 at 11:29 AM
alundeb
Online
• • • • •
Upload & Sell: Off
p.1 #19 · p.1 #19 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Dan, are you suggesting that the theorethical effect of diffraction is removed from the measurement results before publishing? That the theorethical drop in MTF due to diffraction is added to the measured value? That would be a very strange thing to do, as the published result will then not be something the lens can do in the real world, but a theorethical value.

I can understand that a calculated MTF ignores diffraction since it simplifies the calculation and represents the lens design, but it wouldn't make sense for actual measurements.



Oct 22, 2016 at 11:40 AM
Scott Stoness
Offline
• • • • • •
Upload & Sell: On
p.1 #20 · p.1 #20 · Prime vs Zoom? [and 24-105 vs 24-70] Dxo reading suggest


Here is wide comparison from Canon re 24-105v1 vs 24-105v2 (~70% vs ~80% at edges) - given they both start at the same point (near 100%) the means about 10% difference (0 averaged with 10%/70% averaged up) at f8 (higher thick dashed and sold lines). However "micro contrast" (= smaller lower lines) and astigmatism (gap between dark and dashed) in 24-105 are considerably better in 24-70v2)





24-105/f4/v1 Wide 24mm







24-70/f2.8/v2 Wide 24mm




Oct 22, 2016 at 01:14 PM
1
       2       end




FM Forums | Canon Forum | Join Upload & Sell

1
       2       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.