[davidmanze]

Hi guys,

             A nice controversial thread here; ( a fill in thread to run us optically into the New Year)

 

     For those who know or have or who have not seen this man's videos......

 

                           ....Ken Wheeler........

                                                          better known as the "Angry Photographer" (also trading under the pseudonym of Theoria Apophasis), a sort of semi Greek style philosophical street version of the famous Ken Rockwell, is currently leading with a new theory backed up by his own opinion that modern lenses are heading in the wrong direction with their increasing number of glass elements. He states that this is leading to poor colour saturation and a list of other optical deficiencies that are degrading modern complex optical designs.

   Here's the link,

   

  anybody who can't make it through to the end of the video need not excuse him/herself.

   

 

https://www.youtube.com/watch?v=0nkmWVEzd5E

 

    Does anyone here feel there is at least an element of truth in his theory especially as far as colour depth is concerned?  Your views!

 

  God what a rant!

[popo]

Repetition, repetition, repetition. So the take away point he is saying (repeatedly) is that new lenses are worse. More elements are bad, mmmkay?

I think he just hates the modern world, living in the tinted glass of the distant past. I watched the whole thing, and I still haven't heard anything in there to say WHY the old glass is "better". The only hint towards the end was some vague talk about rendition, which is usually the argument you take when there is nothing else left.

If he had said he doesn't like big lenses, fair enough. If he had said he doesn't want to pay more for new lenses, that would be somewhat understandable.

I'd file him under time waster. Ken Rockwell makes far more sense than this.

[popo]

Oh, specifically on colour, I don't buy it that modern lenses are somehow lacking there. It is a basic part of physics that at every interface you can get reflection, but modern coatings significantly reduce those. What about glass colouring the image? Again that can be controlled through coatings acting as a filter. At worst, you get some light loss overall. While not all might like to, there is a lot you can do in post colour wise.

Hmm... if I get really bored some time, I could put up a test here. Take as far as is possible the exact same shot with some 50mm lenses and we can have a game of spot the difference.
Sigma 50mm f/1.4 Art - representing a modern design with 13 elements in 8 groups.
Zeiss 50 makro - At 8 elements it doesn't come close to the Sigma. Is this modern or not? The Canon version was only introduced in 2010, but the Nikon version was around before that in 2006. But had they used the same optical formula before then?
Canon FD 50mm f/1.2 (not L). This is mount converted to EF, and as long as I focus close the mirror doesn't try to get intimate with the lens. This has 7 elements and as far as I can find this is the new FD lens introduced in 1980. I think we can safely say this is an old design.

[Scythels]

Glass alters color both by absorption, which is not constant w.r.t. wavelength (color) but also by dispersion, where it spreads the colors out. 

 

Dispersion is to fault for both axial color and lateral color. 

 

Marching up to the 80s there were more and more and more glasses available to designers - peaking at about 450 glasses.  At that point the EPA went on a bit of a "rampage" trying to eliminate lead and arsenic and the number of glasses available fell to just 20-40.  It has rebounded to around 140 total but is still a far cry from what it used to be.

 

The new glasses are mildly different w.r.t. their absorption profiles, but I am extremely doubtful that he knew this. 

 

The idea of "color-matching" lenses is all done with the coatings.  The glass choices are not restricted for that beyond basic checks to ensure the high transmission bandwidth covers all of the visible range.

 

The higher contrast the lens, the more "pure" the colors.  The less axial and lateral, but especially axial, color the lens has the more "pure" the colors.  Older lenses had less contrast because they were not as well corrected and assembled, the result would be "softer" and more "pastel" like colors.

 

However the color fidelity of consumer lenses has never been higher than it is today.

[davidmanze]

Actually this is one of the worst examples of his ranting videos, he usually gives his views in a far more "palatable fashion" given that he is prone to repeating himself ad infinitum..................... Often he mixes in a lot of his knowledge and experience of Nikon lenses and has many videos on his recommendations for the earlier manual and AF-D range of lenses which are disappearing off the review sites. A lot of it is useful.

 

   Ken spent many years dedicating himself to the fight to save his wife who was dying of cancer, here he was in one of his blacker moods and takes out his bitterness on the modern lens world!

 

  In spite of his unorthodox presentation,  I still think he's a good guy! (maybe it's the rebel in me)  he has given good reviews to such lenses as the Tamron 24-70 F2.8 and the Nikon 20mm F1.8G and other recent glass ....... As has been said, there are rendition differences in modern design but the Otus lenses get round most of those problems by adding even more glass.

 

 

PS. One exception to this is what is turning up more often in modern reviews is "onion bokeh". It annoys me with the Tamron 150-600! I'd be happy if they could get rid of that!

[toni-a]

You can't have your cake and eat it, everything comes with a price.

Yes modern lenses are ususally better performers but this comes with a price. the aspherical elements tend to deteriorate the bokeh, but would you prefer a lens without aspherical elements ? all depends of your preferences.

more elements might have an effect on color, but newer lenses have better coating so the overall result is generally much better.

Would you prefer using an old yashica lens from the 50s over Canon 50mm  f1.2 L ??

[JJ_SO]

30 years ago I bought a Zeiss 35/1.4 Distagon and a Planar 85/1.4 for Contax/Yashica. The Distagon had 9 elements. the Planar 6 and both had T* coating. My Sigma 35 these days has 13, was less costly, weighs 65 grams more, has AF, 1 aspherical elements more and one FLD glass. I would not trade in that that Sigma for an old Distagon. It had compared to a Nikon 35/2 neutral to cold colours. Same goes for Nikkor 85/1.4, 10 elements and meanwhile for my taste too much CA - it will have to go if a newer 85/1.4 appears, be it Nikon or Sigma. But if I see the Zeiss with backlight and wide open, I have to say, T* coating at that time was far away for what contemporary coatings perform.

 

There will always be people clinging to the past, everything was better some time ago, especially films and those old brilliant lenses. So far I haven't seen a comparison of an old and new lens of the same FL and aperture convincing me there were better lenses made years ago. But if those people are happy with old lenses adapted to new cameras - hey, who am I to criticize that? I see it as a matter of personal taste, neither right nor wrong.

[Scythels]

You can't have your cake and eat it, everything comes with a price.

Yes modern lenses are ususally better performers but this comes with a price. the aspherical elements tend to deteriorate the bokeh, but would you prefer a lens without aspherical elements ? all depends of your preferences.

more elements might have an effect on color, but newer lenses have better coating so the overall result is generally much better.

Would you prefer using an old yashica lens from the 50s over Canon 50mm  f1.2 L ??

Categorically speaking the "onion ring bokeh" isn't because of aspheres, but because of CNC manufacturing methods. 


Oldschool traditional methods are what is known as "full aperture" grinding and polishing.  They hit the entire surface at once, evenly, so they do not leave any small / local artifacts.  You can do up to about 6 elements on a single machine and the machinery is cheap and compact, so it's great for volume manufacture. 

 

The downside is that you need master opticians with often 25-35+ years experience to monitor things, since it is a "blind" process, and they are a dying breed.  The surface quality produces can also not be too hot, and you can't make anything but spherical shapes. 

With CNC "anyone" can operate the machine and the quality of surfaces produces is ~4x better.  The cost is also ~2x higher. 

 

Unfortunately periodic error in the servos prints through into the parts, since CNC is "sub aperture" polishing and grinding, meaning it works on only a small part of the surface at once.  The result is a "new breed" of manufacturing errors known as mid spatial frequency errors.  The rings are a form of mid spatial frequency error.  They came first with aspheres when CNC was not popular for spherical parts, but now because they can hire almost any joe schmoe to run the machine instead of finding master opticians, it has become vogue for all parts (which is imo very good). 

 

You can see CNC grinding/generation simulated here -  https://www.youtube.com/watch?v=4_etSKB0ueY

 

And you can see a (lengthy) webinar on optical surface spec with a great deal of MSF error discussion here - https://www.youtube.com/watch?v=mcl3PppDu5o

 

Dr. Degroote Nelson teaches the course OPT 243 - Optical Fabrication and Testing at the University of Rochester and is director of R&D at Optimax, a very large optics manufacture in the US.

[davidmanze]

Thanks for that AiryDiscus, I watched both videos.

                

 I remember you had touched on that in a previous post, but seeing the animated grinding in the first video it certainly shows how these imperfections arise. I take it that my budget  Tamron does not call for high surface specifications.