Saturday, January 15, 2022

Looking Forward to 2022

Even as Omicron is surging wildly out of control around the world I can't help but dream, hope, and wish the following events actually get to see the light of day.

These are all motorsport related photo-opportunities.

Retromobile - 16-20 March

Tour Auto - 25-30 April

Vintage Revival Montlhery - 7-8 May

Rallye des Princesses - 14-19 May

Cafe Racer Montlhery - 18-19 June

le Mans Classic - 30 June, 3 July

 

With luck, I will see you there!

la traversee de Pari estivale ~ 2021 redo

Sunday, January 02, 2022

Someone who performs outstanding optical testing...

I've known of the Lensrentals blog for some time, now.  Yet it's only very recently that I came to more fully understand and appreciate what Dr. Roger Cicala has to say. 

I enjoyed reading Dr.Cicala's comments on prime lenses and think his conclusions are interesting.  Of course I'm now considering adding a Sigma 30mm f/1.4 DC Contemporary lenses to my always too long list of optics that I might _have_ to have.

Dr. Cicala's conclusions were based in this article on traditional MTF measurements.  That is, contrast measurements were made along a single 2D line.  No attempts were made to account for optical field curvature in a 3D space.

Then I came across a current article and it really caught my attention.  Roger writes about integrating MTF information in a way that goes beyond simple flat field measurements.  

This is wonderful in the way the information helps one "see" how lenses behave in three dimensions. I think what Dr. Cicala & Co are now doing is ground breaking. I love it!

Additionally, I appreciate is that he clearly states that he is _not_ testing for out of focus rendition.  His tests might at best remotely hint at how out of focus regions can be rendered, but that is an area he is currently leaving to others.

Which leads me to something I read some years ago.  This was something that David Duncan Douglas wrote about when he first encountered Nippon Kogaku lenses.  His story, in general, is very well known.

M.Douglas' colleague made a nicely sharp image and was asked what lenses he was using.  M. Douglas' glass was designed and manufactured by Leitz and Zeiss and he was finding his lenses were rather soft and unremarkable compared with the photo his colleague had made.

On his first visit to Nippon Kogaku (now Nikon) they used a display method where one could clearly see and evaluate the performance of a lens.  A lens projector was the display method.  The way it worked was to have a grid image projected onto a screen through a lens.  As a lenses focus ring was turned a person could judge for themselves not only resolution, but _how an optic transitioned from out of focus into focus and back out of focus_.

There is a standard, albeit subjective, method for considering out of focus rendition of optics as well as "sharpness". Lens projectors are still used today for evaluating lenses used in cinema.  In fact, at least one testing lab has added out of focus rendition information to their web pages. 

Coming back to Dr. Cicala's lens tesing, I wish I had access to the kinds of equipment he uses.  Lens Rentals looks at the latest generation lenses.  I'd like to see how the old manual focus Nikon Nikkors behave.  There would be a lot for me to learn.

Both methods (MTF in 2D/3D and grid image projection) do not take into account the image recording materials.  These methods consider only the lenses.  I will leave the pros and cons of this for perhaps another time.  

Until then, and only if you're really interested in the subject and haven't already encountered Lens Rentals blog, consider what Dr. Cicala has written.

 

Nice Port ~ 2021



Thursday, December 09, 2021

... from a time long before...

I was cruising through my Flickr site and stumbled on an image.  

It was entered in one of David Hobby's Strobist contests and surprisingly to me it won forth or fifth place.

Did I really do this?  Wow.

 

Radiant Bones - All Hallows Eve

Tuesday, December 07, 2021

Do lenses matter as much as we're lead to believe they do?

Here I am, staying for yet another three months in Nice, France.  After this year is said and done we'll have been here 6 months.  Half a year.  Extraordinary year, this.

There's not been much time this trip to consider photography, the craft, the art, the tools, nor the techniques.  Yet I snapped a couple images recently that reminded me of something that I've thought about posting.

I wanted to call into question the little matter of lenses and how much importance we place on them.

 

Cimiez Nice ~ 2021

 

Sure, most people who make images couldn't work without at least one lens.  As we know, and without cutting things too fine, there are plenty of photographers who use pinholes and zone plates.

No.  That's not my point.  My point is lenses and the minutiae of optical performance and image rendering are largely subjective and up to the image maker.  

Outside of ourselves as photographers, who really cares what lens made which image?  The viewer?  I think not.

Without looking at the EXIF data, here's what I mean.  These ancient olive trees are found in the park where they used to play jazz into the wee-hours of the morning during a Nice summer give me pleasure.

I'm happy I had a camera and a lens.  I don't think these could be any better to me had I used my "favorite" optics that stayed home back in Paris.

And yet I fret and look and explore and ponder and search and am ready to buy yet another little optic that will somehow make my photographic life "perfect."

I'm a strange being.  Really.  I am.

 

Cimiez Nice ~ 2021

Saturday, October 02, 2021

Image Creation ~ "Ecosystems"

It looks like The Online Photographer may be changing the frequency of Mike's posts.  Putting food on the table is, of course, the most important thing, right?  Sounds like Mike is taking up writing.  It'll be interesting to see what he comes up with.

I have followed Mike's blog for years and I've been very happy to have learned as much as I have from reading his posts.  His insights into how digital files can be made to look good in black and white are particularly instructional.

Alas, the knowledge of what it takes to convert color into black and white, how to make very large prints from small digital files, how lens designs that influence out of focus rendition, all these things and much much more, when done properly, have been subsumed by and "baked into" the underlying software in current digital systems.

Color grading selection defaults can be set to ensure consistency from image to image.  Human perception modeling color to black and white conversion might be the very thing software is based on (and not based on simple color de-saturation).  Image stacking for noise reduction, image "up-rez"ing, and resolution improvement now can happen without us even knowing nor caring about what it takes to make the magic happen.

Over on Thom's website, he shared some thoughts on how small the standalone camera photography "ecosystem" has become.  The points he makes are, to me, quite valid and relate directly to the "baked into" comment I just made.

Back in film photography days the "ecosystem" was very large.  We had the camera and lens makers. We had the film manufacturers.  We had photo-labs and camera shops that sold chemistry for film development.  We had print paper manufacturers and companies that made enlargers and lenses for enlargers.  We had negative and print processor machines we could buy if we could afford them.  We had a gallery system and a vast book and magazine publishing engine.

The "ecosystem" was not dominated by a single manufacturer.  It couldn't be.  The "ecosystem" was simply too large.  For all its size and weight, the infrastructure each piece of the system integrated into was complete.  Everyone needed everyone else for the whole thing to work.

Today?  Things are, quite obviously, different.

Cell phone cameras have become the epicenter of a no-film, no-chemistry, low-overhead, all-electronic "ecosystem."  Cameras.  Lenses. Network connectivity.  Image sharing platforms. Communications.  Sharing.  Everything is well integrated, even down to having direct connections with the original film-era hold-overs of print making and book and magazine publishing.  The number of suppliers that it takes to create this "ecosystem" are substantially fewer than during the film era.

Several short years ago I watched as a tethered camera captured images of a fashion model along the Seine River here in Paris.  The images were edited by a small team of people on a laptop computer, color-graded on-site and posted directly to the 'net.  All in realtime.

Today I no longer see tethered cameras.  Sure, you can take just about any digital stand-alone camera and tether it to a power laptop computer, but why take on that level of overhead when you can do everything from a single device?

Everything can now be shot with a cellphone, by-passing the need for a laptop computer.  The team of people is downsized to just one or two people.  Everything is transmitted straight across the network to its destination.  The distance from idea to final results and distribution can now be incredibly short.

All this leaves me wondering what to do, if anything, with the old, traditional, stand-alone camera/optics systems?

Over a decade ago I mused that wouldn't it be interesting if someone integrated Linux as the base operating system into a camera?  Samsung tried this using the Linux derivative, Android, but the project was something of a disaster.

More recently, as in something less than ten years ago, I commented that Sony, the electronics company who should know well how to do this and how to market it, should take the Android OS, throw away the old embedded OS they currently use in their Alpha-series cameras, and create a fully networked, fully integrated, fully connected camera system.

I feel the opportunity is still there for the taking.  Except, as Thom notes, there might not be the "vision" of stand alone camera manufacturers to make the logical (to me, at least) leap into the cellphone present.


Musee des Arts et Metier ~ Paris 2021


Wednesday, September 01, 2021

RIP "Don't Take Pictures"

I'm sure everyone already knows, but "Don't Take Pictures" has shuttered its publication.  Perhaps you've read their announcement.

"Dear Readers,

Don’t Take Pictures will cease publication on September 1, 2021. When we launched in the fall of 2013, my Editor’s Letter explained the magazine’s mission as a space to celebrate the thoughtful, creative act of making photographic art. The following year, we launched online to include more timely features about the photographic community. We found that our online and print components needed one another in order to survive. 
 
The world has changed a lot since 2013 and our consumption of print media has evolved as well. In print, our writers and designers spent six months diving deep into the issue’s theme and each artist’s work. Without the distractions of a screen, our readers were encouraged to focus on photography on paper. Online publishing has changed as well. It is time to rethink the purpose of a magazine in the age of social media. For all that is gained by the ability to find photographers with a few clicks of a mouse, much of the grandeur and magic of their artwork is lost when viewed in slideshows, as thumbnails, or as part of an infinite stream of images. The demands of the internet era require a churning of content that is both unfeasible for Don’t Take Pictures’ staff and antithetical to our mission.
 
The decision to close Don’t Take Pictures was difficult, but its ending should not be mourned—change is not a bad thing. As the art world continues to evolve, our team is constantly rethinking how best to engage with photography in print and digital spaces. We are very proud of our past eight years, 16 beautifully designed print issues, and hundreds of online articles that exemplified what Don’t Take Pictures is about—great photography and great writing. We would like to extended a special thanks to the 90 artists we were honored to feature in print. 
 
Readers whose print subscriptions included an issue this September have been refunded. The Don’t Take Pictures website will remain online as an archive and resource for the foreseeable future, with all previous issues available as PDFs. It has been an absolute honor to share the work of emerging photographers and arts writers in print and online for the past eight years. We appreciate everyone who read our articles online, engaged with us on social media, or read Don’t Take Pictures in print.
 
Sincerely,
Kat Kiernan, Editor-in-Chief
"

I feel at the bottom of all this is an important point. The purpose of a photograph has changed dramatically in recent time. How we consume images, how we look at them, why we look at them, and what they mean to us has little relation to how we engaged images just a decade ago.

This is why I've scaled back my online presence. I now have just this blog and my Flickr page.  That's it. Nothing more.

Us older folk who tend to "remember when" are dying off. Photography had its day in the sun. Now the experience of image making is, for me, reduced to sharing what I still find interesting and pleasurable. I do this regardless of if there is an audience or a group of photographers I can share with, or not. 

Someday I may feel differently, but for now, that is how it is. 

 

Château de Fontainebleau ~ 2021
 

Tuesday, August 31, 2021

Super Resolution ~ research

Super Resolution, where you take a small image and increase its spacial dimensions and try to retain as much detail as possible, has been an interesting topic for photographers over the years.

Google recently published something related to two different approaches.  

They used an artificial intelligence to learn and then apply results.  

 "...diffusion models, originally proposed in 2015, have seen a recent revival in interest due to their training stability and their promising sample quality results on image and audio generation. Thus, they offer potentially favorable trade-offs compared to other types of deep generative models. Diffusion models work by corrupting the training data by progressively adding Gaussian noise, slowly wiping out details in the data until it becomes pure noise, and then training a neural network to reverse this corruption process. Running this reversed corruption process synthesizes data from pure noise by gradually denoising it until a clean sample is produced. This synthesis procedure can be interpreted as an optimization algorithm that follows the gradient of the data density to produce likely samples..."

I wonder how well it might work for amateur and professional photographers?  Perhaps we'll get a chance to see some day.


Château de Fontainebleau ~ 2021