Valid MF criticism or not?

[awofinden]

When I look at those 2 examples they are perfectly indicative of the differences I have seen between my 1ds mark 2 and P21. The problem isn't sharpness, it's the look, there tonal response is just different, you pays your money and you make your choice. ( I know what my choice is everytime).

[samuel_js]

Because its out of focus....
[a href=\"index.php?act=findpost&pid=137841\"][{POST_SNAPBACK}][/a]

So the girl has a deep valley between the eyebrowns?  Because it's on focus on the sides...  

[sundstei]

So the girl has a deep valley between the eyebrowns?  Because it's on focus on the sides... 
[a href=\"index.php?act=findpost&pid=137897\"][{POST_SNAPBACK}][/a]

What are you talking about?

[jjj]

Thats a very long post from someone who doesn't care.
[a href=\"index.php?act=findpost&pid=137602\"][{POST_SNAPBACK}][/a]

Not liking the image in the post, doesn't mean I don't care about the argument, which is actually quite interesting.

[jjj]

I agree, but I would alter that just a bit to say identical physical aperture size, so when the images are sized down to match each other that DOF is identical across the board.

But that wouldn't be realistic as you would use the same aperture in reality to get the same exposure. The image differences between differing film formats and differeing sensor size is a result of that variation in physical aperture size.

And (fairly) basic math shows that the larger the format, the easier it is to achieve shallow DOF.

And that is exactly why images look different from larger sensors/film types

But I'm very skeptical of the claim that such differences are easily discernible in properly processed web-sized JPEGs (600-800 pixel), or even printed typical magazine page size, for most subject matter.

....That's pretty much the claim made by several people earlier in this thread, that the MFDB "difference" is so great that it is easily noticeable even in web-sized JPEGs. And let's just say I'm skeptical of that claim, for pretty much the reasons you mention.
[{POST_SNAPBACK}][/a]

There's a demo online showing the difference bewteen 35mm film imaging and DV imaging and online it is very, very obvious. Both films were shot with the same DV camera but one used a 35mm lens with a convertor so the DV camera filmed the larger image produced by the larger optics.
Just did a quick look for the demo and the site has been overhauled and no longer has the footage shot in a bar with a DV sensor and then with the 35mm adaptor.
A shame as it illustrates exactly what people have been arguing about so very well.
Go here for the website [a href=\"http://www.redrockmicro.com/focus2.html]http://www.redrockmicro.com/focus2.html[/url] the footage may be elsewhere, I don't have time to look all through site at moment. A colleague of mine used this kit or similar and his short looked so much more cinematic than is normal with DV. It looked like a  much bigger budget piece than it actually was. Simply through the [optical] look of the footage due to using a larger efective sensor. And nothing to do with the pixel count!

[samuel_js]

What are you talking about?
[a href=\"index.php?act=findpost&pid=137902\"][{POST_SNAPBACK}][/a]

Hi, look carefully which parts of the image are on focus. I don't think is a focus probem, is more a lens resolution problem. In general, I think the Leaf image is much better. Natural skin color, 3D feeling, very nice and clear.
Just compare the Canon with Frank's shot. It's what I call medium format.
But again, this is just how I see it.

[Frank Doorhof]

The funny thing is.
Look at the eyes and the eye lashes.
On the canon this is flat, on BOTH MF systems you can see much more depth.
Maybe some people don't see depth the way I see it, but for me that's a give away.
Oversharpening and jpeg artifacts with large compression will mask this somewhat but if napier would post a non processed sample I think it will be even better visable.

My wife said when I showed her a close up of an eye, "The eye ball is round, with your other photo this never drew my attention".

[RobertJ]

Hi Frank,
What aperture did you use on the 120mm for that eye closeup?  Wide open?

[Frank Doorhof]

F10
So not wide open.
In the studio wide open is a problem with DOF if I shoot the Macro wide open that close the top of the eyeball will be in focus but the eyelashes will not

[Gary Ferguson]

There has been a lot of discussion of DoF issues on this forum over the years, some folks insisting that DoF is affected only by the FL of the lens, aperture and distance to the subject. My own view is that it is also affected by the sharpness of the lens at the plane of focus and this factor is very difficult to incorporate into the DoF formulas.[a href=\"index.php?act=findpost&pid=137791\"][{POST_SNAPBACK}][/a]

I agree with you.

Some time ago Zeiss's Marketing Manager used to post regularly on photo.net, and he made a similar point. His position was that all lenses of the same focal length do not deliver the same depth of field, but that there was a long standing agreement amongst lens manufacturers never to engrave more than a specific amount of depth of field on their lenses, and not to publically claim a DOF advantage. They were free to engrave less (as Leica did, using a DOF scale on their lenses based on a 25 micron circle of confusion instead of the 30 micron COC used by virtually all other 35mm lens manufacturers), but no one could claim or engrave more.

He gave the example that the Hasselblad 38mm Biogon had a particularly deep effective depth of field, which tallies with my own experience of using this lens alongside the 40mm Hasselblad IF. He also said that a number of extremely high resolution modern lenses achieved their outstanding performance in the plane of focus (the only part of the image measured on the MTF charts) at the cost of the image quality in the out of focus areas. Again, something my own observations tends to support.

[samuel_js]

I agree with you.

Some time ago Zeiss's Marketing Manager used to post regularly on photo.net, and he made a similar point. His position was that all lenses of the same focal length do not deliver the same depth of field, but that there was a long standing agreement amongst lens manufacturers never to engrave more than a specific amount of depth of field on their lenses, and not to publically claim a DOF advantage. They were free to engrave less (as Leica did, using a DOF scale on their lenses based on a 25 micron circle of confusion instead of the 30 micron COC used by virtually all other 35mm lens manufacturers), but no one could claim or engrave more.

He gave the example that the Hasselblad 38mm Biogon had a particularly deep effective depth of field, which tallies with my own experience of using this lens alongside the 40mm Hasselblad IF. He also said that a number of extremely high resolution modern lenses achieved their outstanding performance in the plane of focus (the only part of the image measured on the MTF charts) at the cost of the image quality in the out of focus areas. Again, something my own observations tends to support.
[a href=\"index.php?act=findpost&pid=137952\"][{POST_SNAPBACK}][/a]

Talking about Zeiss, I just found this scanned picture. It was taken with a Zeiss CFi 120 makro. Wide open. Pretty good performance I think. The film was kodak trix 400 so a lot of grain. I think the resolution is spectacular but is in some way masked by the grain. I supose it can be worked out a little bit more...  
This is a very advantage in the DB, almost noise free.

Full image:



100% crop 1



100% crop 2



Alot of detail in the skin... but not as clean as a DB in the eyelashes. But this was wide open at f4...

[mmurph]

No offense to anyone, but this has *not* been one of the better LL threads in this forum.    

I think it started off completely on the wrong foot with some marketing hyperbole.  Usually the pros here ignore that type of bait. This time everyone jumped in.

I was scanning 4x5 slides and negatives yesterday while reading this.  I used to use 6x7 film, gave that up, but still find a place/need for the 4x5. We have *always* had multiple tools in our kit.  As pros we balance all of the pros and cons - technical, financial, etc., to choose the correct tool.

I can tell you that 4x5 film is *much* more expensive than I like!      At $4 a sheet, I have spent $1,200 on film and developing over the last 2 months for a personal project. That is almost the depreciation cost of using a 1DsII or 1Ds3 for 1 year.  If I keep going the total will be $5,000 this year. Ouch.    

But sometimes no other tool will fit - not for subtle lens or DOF effects in my case, but for the sheer amount of detail and complexity that I need to capture.

Still, I have to say that the 1DsII is one of my all-around favorite cameras that I have ever owned. The Mamiya 7II is a close second. I like my Toyo AII, and my Cambo Wide.  My old Bronica ETRSi, on the other hand, is a POS.   Worst camera I have owned, very sloppy. I'd rather use my Holga than that thing.

Use the tools. Share your learning - it is financially impossible for all of us to test every tool nowadays.  But don't recreate the old Nikon v. Canon, or PC v. Mac, wars in this forum, of all places.

Best,
Michael

[samuel_js]

I just did another scan at 4800. Just playing....  
You can actually see the flowers that were beside the window in the reflection on the eye. I'll miss my Imacon, just sold it on Ebay  

Here's the 100% crop:

[paul_jones]

I agree with you.

Some time ago Zeiss's Marketing Manager used to post regularly on photo.net, and he made a similar point. His position was that all lenses of the same focal length do not deliver the same depth of field, but that there was a long standing agreement amongst lens manufacturers never to engrave more than a specific amount of depth of field on their lenses, and not to publically claim a DOF advantage. They were free to engrave less (as Leica did, using a DOF scale on their lenses based on a 25 micron circle of confusion instead of the 30 micron COC used by virtually all other 35mm lens manufacturers), but no one could claim or engrave more.

He gave the example that the Hasselblad 38mm Biogon had a particularly deep effective depth of field, which tallies with my own experience of using this lens alongside the 40mm Hasselblad IF. He also said that a number of extremely high resolution modern lenses achieved their outstanding performance in the plane of focus (the only part of the image measured on the MTF charts) at the cost of the image quality in the out of focus areas. Again, something my own observations tends to support.
[a href=\"index.php?act=findpost&pid=137952\"][{POST_SNAPBACK}][/a]

that interesting. when i tested the contax f2 80 against the HC 100 2.2, i would have thought that the longer and almost as fast hc100 would have a smaller depth of feild and more blur than the contax. but the contax was way nicer. the head and shoulders shots i took, on the contax the shelves in the background was complete mush, but you could see all the sheves quite clearly with the hc100.

paul

[Ray]

I'm getting a sense of desperation amongst some of you MF aficionados.  

The thread started off with the suggestion that the 1Ds3 might be able to challenge the image quality from some of the more expensive MFDBs.

We can't really get a resolution of that issue until the 1Ds3 is released and available to those of you who already own MFDBs, so you can make proper comparisons of the identical model in the identical pose with identical lighting using a lens of equivalent focal length and f stop to provide equivalent DoF with both cameras.

The comparisons that I've knocked up from Andre Napier's image and Canon's one and only sample image of the 1Ds3 provide only a rough guideline as to the similarity of quality.

In these examples it's easy to pick out differences because the subject, lighting and circumstances are different. If someone makes the comment the girls forehead in the 1Ds3 shot is not sharp, it's a fair bet that it's due to different focussing between the two images. The girls nose in the 1Ds3 image seems more in focus to me than the model's nose in the Aptus shot.

[attachment=3191:attachment]

Someone made the comment that the eyeball in the Aptus shot had greater depth.  Well, you might have noticed that the model has her face turned (more)away from the camera (in the Aptus shot) and is looking away from the camera instead of at it. The eye is being photographed from a different angle.

Another issue is the pixel count difference between the two crops. We are here comparing a crop from a 33mp image with a similar crop from a 21mp image. It so happens that the models' faces are approximately the same size within the frame, at least close enough to make the comparison reasonably valid.

But the Aptus crop is definitely comprised of more pixels. In the earlier comparison I interpolated the 1Ds3 crop so both crops were the same size. Perhaps I shouldn't have done that. Here are the uninterpolated crops.

[attachment=3192:attachment]

Finally, the difficulty of getting equivalent DoF between different formats with different aspect ratios has to be considered. Which dimension are you going to choose? I get the impression that the 4/3rds aspect ratio of most MFDBs is preferred for portraiture and fashion, so perhaps we should be comparing a 24mm wide format with a 36mm wide format, ie. 1.5x the width.

If we do this, both the focal length and f stop should be multiplied by the same factor for equivalent DoF, so an 85mm lens at f1.8 on 35mm format should be equivalent to a 127mm lens at f2.7 on 48x36mm format. This approach will give an edge to the MF format. If we use the height for comparison, the FL and f stop is slightly different and 35mm gets the edge.

It's easy to see differences between images of different scenes, images which actually are different. What is more difficult is assigning and attributing such differences to a particular camera and format when one doesn't already know which camera was used to take the shot.

I used to be very interested in audio hi fi matters several years ago. I used to read hi fi magazines strong on advertising and subjective impressions but short on objective analysis. I always remember the occasion when one such magazine reported on a serious attempt to test the objectivity of audio experts who claimed they could always tell the difference between an exotic system with oxygen-free copper cables, ultra-expensive amplifier etc, and an ordinary but good hi fi system.

They each brought their own music, every nuance of which would have been familiar to them. They were asked to identify which system was in use as the music was switched between various system, some systems employing expensive amplifiers, some with cheap amplifiers. The speakers were always the same though.

The results were astounding. They were about as accurate as tossing a coin. However, there was one respect in which the results were consistent, but unfortunately consistently wrong. One particular system with a cheap Pioneer amplifier was consistently confused with another system employing a terribly expensive Mark Levinson hybrid amplifier which most hi fi enthusiasts would have drooled over.

[thsinar]

Well Ray,

the initial and starting point wasn't at all about suggesting that the Canon could compete on a quality level or challenge the MF world: there was no word about this in the initial post and published "Canon statement".

It was question of compatibility, support in the future (mid- and long term), while suggesting that with Canon it was a "clear future" and with MFDB manufacturers not at all.

The tread has then (logically) switched to a quality comparison/discussion.

Best regards,
Thierry

I'm getting a sense of desperation amongst some of you MF aficionados.   

The thread started off with the suggestion that the 1Ds3 might be able to challenge the image quality from some of the more expensive MFDBs.


[a href=\"index.php?act=findpost&pid=137972\"][{POST_SNAPBACK}][/a]

[Ray]

Well Ray,

the initial and starting point wasn't at all about suggesting that the Canon could compete on a quality level or challenge the MF world: there was no word about this in the initial post and published "Canon statement".

It was question of compatibility, support in the future (mid- and long term), while suggesting that with Canon it was a "clear future" and with MFDB manufacturers not at all.

The tread has then (logically) switched to a quality comparison/discussion.

Best regards,
Thierry
[a href=\"index.php?act=findpost&pid=137974\"][{POST_SNAPBACK}][/a]

You're right Thiery. I jumped in about halfway through the thread and seem to be at least partly responsible for steering the issue towards image quality comparisons. Sorry about that   .

Your concerns are about continuity and compatibility of MF equipment. I guess mine are about the possibility of achieving 22mp MF quality through the purchase of a 1Ds3 which is a camera normally way above my budget...

....and at ISO 800 possibly greater quality than a P45+     .

[thsinar]

We understand each other, Ray!  

I'm obviously following the debate about quality and comparison, won't however jump in it, since too many factors are involved to make a right and fair comparison, if at all possible.

For me, and like many said it here: know what you want, know your gear and how to use it the right way and for the right job, that's the most important.

Best regards,
Thierry

You're right Thiery. I jumped in about halfway through the thread and seem to be at least partly responsible for steering the issue towards image quality comparisons. Sorry about that   .

Your concerns are about continuity and compatibility of MF equipment. I guess mine are about the possibility of achieving 22mp MF quality through the purchase of a 1Ds3 which is a camera normally way above my budget...

....and at ISO 800 possibly greater quality than a P45+     .
[a href=\"index.php?act=findpost&pid=137980\"][{POST_SNAPBACK}][/a]

[AndreNapier]

When I posted my image in this topic my intention was just to illustrate the 3D effect that I see in MF DB and that in My opinion is missing in Dslr. This picture was just a handy example that I had ready in my MacBook. It was post processed, color corrected, PS and over sharpened as it was ready for  Teens magazine cover print this week . On paper all the digital art effects will disappear leaving just sharp enough image also without the exaggerated blue tone.
Since then I went back and reprocessed the Raw with disabled sharpening without any corrections in just as shot form. In PS I just saved it to Jpg Max and resized it. The close up is actual pix. It does not look nothing in the web as it does on the screen in original form , but we all know that. At least I hope you guys are using Safari and not Firefox which even farther destroys the quality. Please do not forget that it was shot freehand wide open with HMI lighting at 1/60sec  and that the intention was not to achieve the sharpest image but rather character and separation.
  I have seen the work of others and shot myself thousands and thousands of frames with Canons and Nikons digital. Some of the work I have seen is superb in all respects. Some of the work is much better than I can dream of achieving with all DB ever. There is so many variables to make a great shot and camera is just a tool. Since I have very extensive background and all my young years as a hair stylist with the international power houses I can easily assume that at anytime I can do a better haircut with a piece of glass from broken beer bottle than any of you with $1500 scissors. Having said that if given a choice I always prefer to cut using the best scissors as they give me the confidence and ability to go beyond the ordinary and would never become a limiting factor in my creativity.  Same thing applies to photography. I know my very limitations and every day I struggle to overcome them. Camera is just as good as we are. Great camera however is as good as we ever desire to become.

Http://AndreNapier.com[attachment=3193:attachment][attachment=3194:attachment]

[Frank Doorhof]

@Ray,
That's what the whole thread is about (or at least for me), people claiming they will get MF quality files with the 1DsIII or something close.

Probarbly the only thing you will get with the 1DsIII is the megapixels that SOME digital backs have

And we all know by now that MP's are not that important for the final quality of a shot.

As Napier also said it's the photographer that in the END will make the picture, however for me and that's personally the 5D was limiting me in what I wanted, near the end of using the 5D I was frustated that it would not give me the lattitude I needed, it did not give me the shadow detail I wanted and it did not give me the smooth gradiants.

I did not switch for the MP's.