The end of tolerances reached ?

[Stefan Steib]

http://www.canonrumors.com/tech-articles/this-lens-is-soft-and-other-myths/

I found this very interesting article, written by Roger Cicala from lensrentals.com- it´s not new, maybe some of you know this already. I think for the others it´s worth reading as it´s contents exactly been discussed here with the 80 Mpix backs and the results.
I second everything he says in there for 100 %, he also states the findings of LLoyd Chambers from Diglloyd.com. I say: There is even more about ISO variations of focal lenghts (+/-10% allowed!!!), focus shifts for stepping down etc.pp. many of these hindrances to be found on LLoyds pages at DAP.

I think all working Pro´s need to understand that we are reaching a limit here and Thesis: if it is not a better approach to improve workflow and handling of cameras instead of megapixels.

BTW: a bad and boring picture will stay a bad and boring picture no matter how many megapixel you throw at the viewer to improve the impression.

Greetings from Munich

Stefan Steib hcam.de

 

[Godfrey]

... I think all working Pro´s need to understand that we are reaching a limit here and Thesis: if it is not a better approach to improve workflow and handling of cameras instead of megapixels.

BTW: a bad and boring picture will stay a bad and boring picture no matter how many megapixel you throw at the viewer to improve the impression.

I've subscribed to this point of view for some time. I find that unless I am intent on making truly enormous prints ... which means for me prints larger than say 17x22 inch printed area ... I really don't need more than the current 10-12 Mpixels easily available with many cameras and can often be quite satisfied with even 5-6 Mpixels. Handling, workflow, and skill in making the exposure, focusing the lens, and then rendering the output properly account for a lot more quality than spending megabux to buy the latest 16, 20, 24, 40 or 80 Mpixel wonder.

For those who's work and craft demand these huge pixel resolution cameras, more power to you! But my work only very rarely indeed goes to such sizings or demands such high-end pixel resolution. I'd love to see much more work go into camera responsiveness, usability, and dynamic range than this constant expansion of pixel resolution.

 

[goesbang]

Roger's article should be compulsory reading for all. This is not rocketscience but Roger does explain it particularly eloquently.
It is for precisely this reason that manufacturers such as Alpa, whose entire ethos centres around the highest achievable quality, insist on such seemingly pedantic practices such as shimming of mounts. It is simply impossible to control all the variables in a photographic system (back, body, lens etc), so it is essential to have some point at which adjustment is possible. It is also one of the main reasons why Alpas, Leicas etc are so expensive - these manufacturers go to extreme lengths to keep tolerances very tight. Keep in mind also that the number of units sold is tiny when compared to Canikon etc. To those who constantly complain about high costs of this level of gear, I'm sorry that you can't afford it, but this level of quality costs dearly and the manufacturers are businesses, not charities. Photographers who demand this level of quality always have and always will stump up the cash, or learn to do without. Having spent nearly a million dollars on gear over the course of my career, I can assure you it is tough. It's one of the reasons I don't drive a flashy car. Yet, I am grateful that there are manufacturers producing 80Mp backs, ultra resolution lenses and cameras with awesome tolerances etc. May they stay profitable and keep innovating.
As to the question of how many MP is enough, I think every shooter has to make the decision as to how much is enough for themselves. There is no universal truth here. Hunting elephants with a .22 Rimfire is as silly as hunting squirrels with an elephant gun. However, whatever back you use, you cannot go wrong by insisting that it hangs off a camera built to the tightest possible tolerances.
Cheers,

 

[Shashin]

Thank heavens someone said this:

And stop looking at 100% images on your monitor: not only will it make you go blind, 50% screen resolution is more resolution than your printer can reproduce anyway.

And they understand the significance of viewing distance.

If my images stink, it won't be because I can notice the effects of diffraction or lens softness at 100%.

 

[Godfrey]

... To those who constantly complain about high costs of this level of gear, I'm sorry that you can't afford it, but this level of quality costs dearly and the manufacturers are businesses, not charities. Photographers who demand this level of quality always have and always will stump up the cash, or learn to do without. Having spent nearly a million dollars on gear over the course of my career, I can assure you it is tough. ...

If you're construing my light reference to how much this gear costs as some kind of stupid whine that I can't afford it ... I've owned and used top of the line equipment for many many years. Linhof, Rolleiflex, Hasselblad, Sinar, various others when I did work that required such equipment.

Bravo for you that you feel you can afford a million dollars worth of equipment. I don't know what your photographic endeavors are comprised of, but if you're spending that much money as part of a photographic business, I hope you've made 10-20 times the value back in your sales. Otherwise, it's a waste of money IMO ... business is after all business, profit is the name of the game. If you're a wealthy or dedicated hobbyist for whom putting this amount of money and energy into your work is your life's mission, to supersede other desires and wants, more power to you. I've been there too.

My comment was directed at the spirit of the article, which is that too many photographers are hooked on fancy equipment and there are limits to what equipment can do to better your work, there are limits to how good equipment can get. Once past a certain point, the equipment is one of the least important parts of producing compelling photographs. Most photographers are well to learn their craft better rather than spend a million dollars on fancy ne-plus-ultra equipment. The few for whom such expenditures are essential are indeed the few.

 

[fotografz]

http://www.canonrumors.com/tech-articles/this-lens-is-soft-and-other-myths/

I found this very interesting article, written by Roger Cicala from lensrentals.com- it´s not new, maybe some of you know this already. I think for the others it´s worth reading as it´s contents exactly been discussed here with the 80 Mpix backs and the results.
I second everything he says in there for 100 %, he also states the findings of LLoyd Chambers from Diglloyd.com. I say: There is even more about ISO variations of focal lenghts (+/-10% allowed!!!), focus shifts for stepping down etc.pp. many of these hindrances to be found on LLoyds pages at DAP.

I think all working Pro´s need to understand that we are reaching a limit here and Thesis: if it is not a better approach to improve workflow and handling of cameras instead of megapixels.

BTW: a bad and boring picture will stay a bad and boring picture no matter how many megapixel you throw at the viewer to improve the impression.

Greetings from Munich

Stefan Steib hcam.de

Thanks for this Stefan.

I can use this link to support my POV in other discussions regarding the impending leap in meg count for some of the popular 35mm DSLRs ... specifically people who are waiting for the Sony meg leap expected soon in lieu of getting a MFD ... and those wishing for a huge leap in meg count in the next Leica M camera.

I've long been stumping for other improvements that add versatility and practical use to my tools ... which vary depending on which camera it is, since all of them need improvements in one way or another.

Unless there is other technology involved that we don't know about yet, I will remain skeptical about the true improvements that can be realized from ever increasing the meg count ... other than the ability to market more gear to unsuspecting buyers.

-Marc

 

[Stefan Steib]

don´t get me wrong, I think the 80Mpix back DO make sense, but to be handled with care and with the knowledge about the reasons of the sideefects that occur.
HCam does make large efforts to help making photography with these extremes possible, yes I think we even shifted the limits of what what was possible with wideangles to over 135 degrees rectilinear and now with the new Canon 8-15mm Fisheye Zoom to 180 Degrees (at 15mm a full circle on 80 Mpix).

But I see our HCam as a tool that is used to make pictures, it is only a helping sideeffect for a creative Pro to get there, we are the typewriter that nobody should think of as he reads a book (Hemmingway). The image needs to be done in your brain, no camera will ever change this. We can just support and give more degrees of freedom in usage.

The classic makers are somehow not yet clear in their statements, although they know they have to do something. Schneider Kreuznach has copied our TS lensconcept (with our help - but without paying for it....).
Rodenstock has silently switched to Retrofocus without speaking openly about it (which would really help!). The classic gear makers in the Viewcamera market still seem to hope this will somehow go away or be solved by a miracle.
And as much as I adore Alpa stuff for their precision, the engineering philosophy to do it more and more precise will finally fail at 100 or 120 Mpix latest, when the tolerances will need to be 1 Micron or less which is simply not to be handled anymore.

Canon seems to be on the right path, the Japanese are very clever people, finding a mixture of usability, keizen, mass market and "good enough quality" for 90 % of all Pros to be able to achieve about anything you need to sell such a picture. Their lenses become better and better, they use Retrofocus and computerized designs which do miracles on optics (17mm+24mm TSE and 8-15mm Fisheye zoom), they have the money for new developments and the long term plan to take over the pro market in all aspects sooner or later (with the exception of those parts that do not give enough profit fitting their business model). I´m not quite sure where Nikon or Sony do stand in this scenario, I think they are not that dedicated and centered, they are more watching and reacting.

If anybody wants to see what Canon is after go to Google and type in Canon World Expo 2010 Paris or Shanghai and you will find plenty of videos including the wonder camera, usabilty concepts and more.

Pretty good and the only ones I see who are targeting the before mentioned points with a clear vision.

Greetings from Munich
Stefan Steib - HCam.de

 

[fotografz]

If you're construing my light reference to how much this gear costs as some kind of stupid whine that I can't afford it ... I've owned and used top of the line equipment for many many years. Linhof, Rolleiflex, Hasselblad, Sinar, various others when I did work that required such equipment.

Bravo for you that you feel you can afford a million dollars worth of equipment. I don't know what your photographic endeavors are comprised of, but if you're spending that much money as part of a photographic business, I hope you've made 10-20 times the value back in your sales. Otherwise, it's a waste of money IMO ... business is after all business, profit is the name of the game. If you're a wealthy or dedicated hobbyist for whom putting this amount of money and energy into your work is your life's mission, to supersede other desires and wants, more power to you. I've been there too.

My comment was directed at the spirit of the article, which is that too many photographers are hooked on fancy equipment and there are limits to what equipment can do to better your work, there are limits to how good equipment can get. Once past a certain point, the equipment is one of the least important parts of producing compelling photographs. Most photographers are well to learn their craft better rather than spend a million dollars on fancy ne-plus-ultra equipment. The few for whom such expenditures are essential are indeed the few.

In the spirit of friendly and diplomatic discussion, I'd have to somewhat disagree with the above.

Depending on what photographic business you are in, and for how long, it is conceivable to spend copious amounts of money on gear ... given the price of pro gear, including lighting and all ancillary stuff involved with professional digital capture. Plus, the "business" of photography is often infected with personal desires that are not exactly conducive to a pure profit motivation ... Guy being the poster child for this concept :ROTFL: (and to be fair, I admit to it myself, since I don't need a lot of this stuff to do what I do, but I want it). So, it most certainly is not a waste of money if you enjoy it and it brings some personal pleasure while making a living. For example, I chose to get a Leica S2 kit in lieu of a new fancy car, and instead just put new brakes on my almost 7 year old Volvo SUV. It's just personal, not business

In addition, well run photographic businesses factor in the cost. For all my "growth years" in MFD, I charged a digital capture fee for each job (as did all the photographers I hired when working as an Art Director). Factored in over a two year cycle, and including the tax write offs, in effect, I never personally paid for any of my professional equipment ... the clients did. Now that I am semi-retired, it is a different story and I am more aware of the cost to me personally, which hasn't made me a lot more prudent, just a lot more aware

I'd also offer a counter point to the notion that "... after a certain point the equipment is one of the least important parts of producing compelling photographs." This is a variation on the well worn chestnut that "it is the person behind the camera, not the camera." ... which no one in their right mind would disagree with. However, if you look at it from the perspective of someone who has mastered their craft and is branching out into new areas or different applications of their talents ... sometimes better gear can make a difference and can be conducive to growth, if for no other reason than it's exciting to master a new piece of more demanding gear, and make it do what you creatively envision.

Will it improve the end product? Again, that depends on the creative intent and skill of the user. There are more aspects to larger capture like 40, 60 or 80 meg MFD than just for large prints. I personally see a difference even with prints up to 17" X 22" ... if others do not, that is not my concern ... my only concern is what I see since it is my work, not theirs.

BTW, I disagree with the article regarding prints no matter how well it can be documented logically or mathematically. In most every case, prints look better than their screen version ... and prints from my larger format cameras look better than those from the smaller format cameras ... speaking strictly about the aesthetics of the print itself, not the content, which is the part that the person behind the camera is solely in charge of contributing.

-Marc

 

[Stefan Steib]

As for some samples what I mean about future concepts here some links:

http://www.youtube.com/watch?v=eq8eD9zU7p4&feature=related

now the next is actually a mockup- but wouldn´t we all want this ?

http://www.youtube.com/watch?v=Sn6iO_NGKaw&feature=related

now see this one:
http://www.youtube.com/watch?v=XYgZacQxiks&feature=related

or this:
http://www.youtube.com/watch?v=YB4VUTytx5k&feature=related

There´s a lot more- anybody can find this now - AT THE VIDEO PEOPLE !
I think this shows the idea and the direction - and in my opinion sooner or later the Photo Pro´s will understand that this will also help them.
But then what to do with a mechanical camera ?

And then here - not science ficition- it´s here as a prototype:
http://oh-so-coco.tumblr.com/tagged/Lytro

There is also superresolution - I think this is the next big thing:
try their demo www.photoacute.com

or see some samples here:
http://www.soundsthemovie.com/srtflash/SRT Demo.html

Greetings from Munich
Stefan Steib - HCam.de

 

[jlm]

perhaps we can take hope in the release (finally) by Phase of the IQ screen and interface. The hugely positive response to that feature set alone, larger mpx notwithstanding, may signal a better perception of customer need by the developers

 

[PeterA]

If you want to muck around with good stuff it usually costs- but the best camera always has a beautiful woman in the viewfinder...

 

[Shashin]

And as much as I adore Alpa stuff for their precision, the engineering philosophy to do it more and more precise will finally fail at 100 or 120 Mpix latest, when the tolerances will need to be 1 Micron or less which is simply not to be handled anymore.

Why do you think pixels pitches will reach sub-micron sizes? Tolerances are a factor of depth of focus. Photography is light dependent and a 1/10th of a micron pixel, which is about the upper limit for a 1 micron depth of focus at f/5.6 if you make the CoC equal to the pixel pitch, is not going to collect much light.

Actually, I am not very impresses with a company that makes their cameras too short--you always shim away from the lens plane with an Alpa. Besides, an Alpa camera is just a spacer and helicoid--they don't make the optics and they don't make the back. (and where does the assumption that it is the back that is out of tolerance rather than the lens assembly?) And the tolerances don"t have to be that great for a camera using guestimation focus at mostly f/11.

Don't get me wrong, it is a fine camera. I simply doubt the precision of the camera (in relation to others cameras) and the company's over emphasis of that precision.

 

[Stefan Steib]

Shashin
agreed- this is only one part in the chain, as I mean - you can break a chain by breaking one part, but actually you can´t fix a broken chain with one part if several segments have flaws. The main problem is that there is not a single maker who is doing a complete lens/camera/back/software chain with this technology anymore. There used to be Sinar which was close (but also not making lenses), but I guess this party is over. Hasselblad is going into the closed MF-Body Philosophy, but as soon as you need to use the their backs on Viewcameras you run into the same problems.
Phase One/leaf is doing a pretty good job in documentation and opening the system to 3rd party makers like us, but as can be seen even the small changed holes for the fixation hooks on the IQ180 backs (making these feel loose) have caused some trouble, the communication becomes a permanent control neccessity, there are plenty of complications which stem from a lack of it.
The need for pretesting is growing permanently, adding additional costs to the systems as there are nearly unlimited combinations possible.
I would say if the industry wants to survive there need to be normed interfaces, something like a T2 for Pro lenses and Backs.
These specs need to be defined by a consortium of all makers and shall hold up at least for the next 10 years with forthcoming higher resolutions.
the "old" 4x5" plates and MF mounts for Film cassettes are simply not up to these demands. the lenses need to be locked into a standard interface, that´s something for Schneider and Rodenstock they need to work together if they want to survive- but I think before these 2 will cooperate hell will freeze more likely.

Well..... nobody can say it was not to be seen nor told.

Greetings from Munich
Stefan Steib HCam.de

 

[johnnygoesdigital]

Why do you think pixels pitches will reach sub-micron sizes? Tolerances are a factor of depth of focus. Photography is light dependent and a 1/10th of a micron pixel, which is about the upper limit for a 1 micron depth of focus at f/5.6 if you make the CoC equal to the pixel pitch, is not going to collect much light.

Actually, I am not very impresses with a company that makes their cameras too short--you always shim away from the lens plane with an Alpa. Besides, an Alpa camera is just a spacer and helicoid--they don't make the optics and they don't make the back. (and where does the assumption that it is the back that is out of tolerance rather than the lens assembly?) And the tolerances don"t have to be that great for a camera using guestimation focus at mostly f/11.

Don't get me wrong, it is a fine camera. I simply doubt the precision of the camera (in relation to others cameras) and the company's over emphasis of that precision.

It's not just the DOF that's important, nodal points are greatly influenced by tolerances. Lack of a mirror box puts the lens very close to the image plane and therefore tolerances must be a known variable. Alpa brand lenses are mounted in a precision machined framed that take these variables into account...it's not just hype. This is a finely crafted tool and it's this precision that allow 3rd party backs and lenses to exploit every bit of resolution.

Personally, I like DB's in the 30-40 megapixel range. It's just the right amount of resolution and file size for the current lenses I own. After all, it really comes down to the lens to resolve these high megapixel DB's anyway.

Godfrey,

I don't think Siebel's comment was directed at you in any way. If you've seen his photography it reflects a need for "ultimate images" that can go big, with tremendous detail. With today's planned obsolescence, dollar amounts can add up rather quickly. Certain markets drive what gear we use obviously, so as the saying goes..."you get what you pay for", is very accurate.

 

[djonesii]

It's so much not about Mpix or sharp when I shoot

The way I shoot, and what I shoot is all about light, shadows and holding dynamic range, the Mpix are incidental. That said, I still need a MFDB.



My lenses are a bunch of old Mamiya ones that are close enough for me ......

I can see a huge diffidence in post processing a P30+ image and a D300 one. I have touched a few D3X raw files, and MFDB is still better for what I do.

This is clear in an 800 pixel web image.

So all this whoha about ultimate sharpness is a bit of a non issue for me.

In my other shooting work, F8 and be there is just fine!


Dave

 

[Mike M]

Stefan, thanks very much for the links in #9

There's an old saying that "you can never be too rich or too skinny".... and I think that the photography version of that might "you can never have enough speed or megapixels."

I'm not sure if this is the right thread to bring this up or not but here goes... Too many of us are caught up in the Moore's Law mentality and it's making us all slaves to product cycles. We've all been there and I was a slave to technology just like everybody else. But I've found that the best way to understand what is going on in digital might be to apply some McLuhanisms.

Photography started out as a mechanical process and well all know about the film days with lenses, shutters, apertures, film-loading, chemical development, printing etc. The main point to remember is that mechanical processes are linear and sequential. Each step is part of a specialized sequence and there is a clear beginning and end to the process.

But digital imaging is different from mechanical photography in the sense that it is electrical. Electrical processes ultimately become non-linear and non-sequential. For this reason, digital imaging technology is methodically working itself to the point where the entire process of picture taking will be non-linear and non-sequential. Exposure, focus, resolution, framing, lighting etc will eventually be liberated from mechanical processes and handled simultaneously and/or outside of a sequence. We can already see how this has happened in the last decade as the emphasis on picture taking has shifted from capture to post-processing. Many of the processes that previously were done before capture in the mechanical era can now be done after capture in the digital era.

How is this all gonna happen? Nobody knows all the answers yet but we can use our imaginations. The Lytro camera already focuses after capture. How much longer until the equivalent of in-camera-HDR comes along for extended dynamic range that captures detail in ALL possible lighting conditions? What about super-mega-fast ISOs?What about choosing framing and focal length equivalents after capture? What about choosing resolution after capture?

Anyhow, (to get back to the OP's main post) the main reason that lenses have become such a source of frustration in relation to the quick advancements occurring in electric/digital imaging is because optics are currently still a necessary leftover from the mechanical/film photography era. Digital just hasn't liberated itself from that part of the sequence yet...

 

[TomSteele]

Been Following this thread with avid interest and thought I'd add my humble opinion

I honestly don't think you can be overly prepared or equipped for the task in hand as a pro, coming from a wildlife background. I've always found that by putting the effort in to hike 1800ft up with a 500mm f4 and a wide breaking my back Ive always got the shots where others on the same hike have failed as they haven't wanted to put the effort in to carry/invest in the 500. I agree with Marc its all about mastering what you have to work with and getting the best from it.

By god though does having the right tools for the job help. I'd never think of working on my Norton 650ss without a good quality set of tools that are right for the job.

For me pro photography is the same. Our industry is saturated with tens of thousands of people who produce good images giving them away for free or very little shooting on yet another d3 or 1d and In my opinion if theres a bit of kit that can help distance us as pros from that part of the industry and help up create not just a good but a great to a ultimate image then amen to that.

(it reminds me very much of what I have seen in the lithographic printing industry in the uk where everyone undercut and undercut each other till nearly all the companies were just working not to make a profit but at a loss and over the past few years thousands have gone out of business leaving only the ones that offered the quality level and service and stuck by there prices.)

for me 40 60 80 is enough resolution for now till we see a massive leap again akin to going from a d3 or 1d Mark IIn up to a h4d-40 or phase IQ a difference in image quality thats a sight to behold. And I look forward to yet again investing in equipment and taking my first steps into MDF in the following weeks. I agree with Siebel Id rather drive my old focus and spend as much as my profit as I can back into my business and create the best photography I possibly can.

I look forward to more advances akin to true focus and the iq screens this really is a step in the right direction to a much more feature filled system.

slightly coherent ramble over

Tom

 

[Shashin]

It's not just the DOF that's important, nodal points are greatly influenced by tolerances. Lack of a mirror box puts the lens very close to the image plane and therefore tolerances must be a known variable. Alpa brand lenses are mounted in a precision machined framed that take these variables into account...it's not just hype. This is a finely crafted tool and it's this precision that allow 3rd party backs and lenses to exploit every bit of resolution.

Actually, the lens to sensor distance is not important. Depth of focus (not Depth of Field) is related to the angle of the image cone, which is dependent on the aperture only--focal length does not matter. This means if the the lens is off by x mm, the degree the image is out of focus will be the same at a given aperture regardless of focal length or lens design.

As I said, Alpa makes really nice cameras, but then do a whole host of other companies. (Oddly enough, it is much harder to make a camera with a moving mirror than something as simple as an Alpa.) I have yet to see any evidence that Alpa is producing higher resolution photographs than other cameras.

 

[Stefan Steib]

Mike

EXACTLY ON TARGET !
We are in an intermediate status caught between analogue and Digital. If you want, call it hybrid Photography.
One more example:
Kodak has recently launched a new interline chip with 29 MPix. Now this is certainly not a sensor which is capable of Fashion Photography, but - it does not need a shutter ! It is also about 40 % less light sensitive than a normal CCD, but as we all know if there is a market there will be devellopment, after this comes improvement and competition and after some years we will have a valid all purpose solution for highly integrated cameras which can do everything we want with a minimum of parts- thus becoming incredibly cheap.
Now imagine to put such a beast into a module with a firmly attached lens(Ricoh!) and calibrate this unit at manufacturing, adding a dedicated firmware (like a sharpness,transmission,vignetting and distortion fingerprint) and use this with standardized, open SDK interface, modular software (see iPhone camera iOS 4 and soon 5)............

It could be very close, just someone needs to make it.

Greetings from Munich
Stefan Steib HCam.de

 

[johnnygoesdigital]

Shashin,

You mention that tolerances are a factor of depth of focus. Being that depth of field is measured in macroscopic units and depth of focus is measured in microscopic units. It's my understanding that depth of focus increases with longer focal lengths or closer subject distance while depth of field decreases. I'm new to this, so bear with me. Thanks for the inputs.