New 400mm f2.8 L IS MkII or current 500mm f4 L IS?

[Chris Brennan]

Sometime later this year, I expect to be able to invest in another Canon "big white". And, in my lifetime, I never thought that I'd ever be able to say, "money will not be an issue." But I can safely say it this time around... :):)

I currently shoot with both a 7D and 40D and my main long lens is a 400mm f4 DO, and frequently use it with a 1.4x converter. I'm sure that I will be hanging onto the DO for those times when weight will be a factor. I enjoy using the 400mm focal length but there have been a few times when having something even longer than the 400+1.4x would be beneficial.

The new 400mm f2.8 L IS Mk II is said to be the same weight as the current 500mm f4 so weight is not a variable here. But since I will continue to use crop bodies, the flexibility to add both 1.4x and 2x converters, and retain full AF exists only with the 400/2.8.

If I choose the new 400/2.8, I would also invest in the MkIII converters.

Ultimately the decision will come down to sharpness and ease-of-use, but I'd like some opinions from all of you...

[Lynn Evans]

I own the current 500, it is a wonderful piece of equipment. Price being no object, in your poistion as described, I would also check the next gen 600 to see if it would meet your requirements.
Lynn

[Doug Brown]

I'm sure the 400 f/2.8 II is going to be an exceptional lens. But you've already got a very good (and not inexpensive) 400mm lens. I'd consider a current generation 500 or 600 if I were you. I'd rather get to 800mm with a 1.4x on a 600 than a 2x on a 400. On a related note, you won't find yourself reaching for the 2x all that often if you own a 500 or a 600; the 1.4x will be your TC of choice.

[Roger Clark]

Chris,
I've been doing some calculations, and attached is a plot of resolution for different lenses assuming the lenses are diffraction limited (they are all close) and the pixel sampling of a particular camera. The attached plot shows the results for the 7D and various super telephotos. In the plot, the increase in f/ratio is due to adding TCs, not stopping down. Lower on the plot is better.

There is more on this subject on my developing web page:
http://www.clarkvision.com/articles/telephoto_reach/
(I'll add this plot to the page soon).

The results show that there is little difference in actual resolved detail between the 400 f/2.8, 500 f/4, and 600 f/4, assuming none of the combinations has degraded image quality. With the new series II lenses and TCs, they should all be top performers, so I believe this plot is a reasonable representation of real lenses.

Roger

[Don Nelson]

Nicely done, Roger. And very interesting to see this.

OT -- Off topic....

Hey, one mistake on your page: (OT)
"The above results show how difficult it would be to resolve one of the Apollo capsules left on the Moon."

Indeed it would be impossible with ANY lens to resolve an Apollo capsule on the moon....even from lunar orbit....or on the surface.

Fortunately there weren't any Apollo capsules left on the moon - they never went to the surface. Only the lunar lander went down.

The bases of the lunar landers are, however, still remaining on the moon along with flags, cameras and lunar rovers. The upper part of the lunar lander (LEM=lunar excusion module) docked with the Apollo capsule then returned to near-earth orbit and were disengaged before Apollo capsule reentry. No Apollo capsule ever touched the surface of the moon although rocks were hauled back in them...in sealed boxes.

My dad worked as a chemist at Rocketdyne during the Gemini and Apollo programs - he was involved with water sampling and quality due to the contamination that occurred from Santa Susanna Rocket testing as well as the test launches from Vandenberg -- looking for fuel traces in all the wilderness streams throughout the Los Angeles, Ventura and SLO counties. Like many proud space program employees from many contracting companies working on the programs, we stood in long lines in Long Beach to get a close view of the Gemini and especially the Apollo capsules as they were disassembled by the Quality Control crews at one of the Lockheed hangers. Of course all this ended with the cancellation of the final Apollo flights in 1969. All the Long Beach buildings have been gone for a long time.

Best regards
Don

[Roger Clark]

Nicely done, Roger. And very interesting to see this.

OT -- Off topic....

Hey, one mistake on your page: (OT)
"The above results show how difficult it would be to resolve one of the Apollo capsules left on the Moon."

Indeed it would be impossible with ANY lens to resolve an Apollo capsule on the moon....even from lunar orbit....or on the surface.

Fortunately there weren't any Apollo capsules left on the moon - they never went to the surface. Only the lunar lander went down.

The bases of the lunar landers are, however, still remaining on the moon along with flags, cameras and lunar rovers. The upper part of the lunar lander (LEM=lunar excusion module) docked with the Apollo capsule then returned to near-earth orbit and were disengaged before Apollo capsule reentry. No Apollo capsule ever touched the surface of the moon although rocks were hauled back in them...in sealed boxes.

My dad worked as a chemist at Rocketdyne during the Gemini and Apollo programs - he was involved with water sampling and quality due to the contamination that occurred from Santa Susanna Rocket testing as well as the test launches from Vandenberg -- looking for fuel traces in all the wilderness streams throughout the Los Angeles, Ventura and SLO counties. Like many proud space program employees from many contracting companies working on the programs, we stood in long lines in Long Beach to get a close view of the Gemini and especially the Apollo capsules as they were disassembled by the Quality Control crews at one of the Lockheed hangers. Of course all this ended with the cancellation of the final Apollo flights in 1969. All the Long Beach buildings have been gone for a long time.

Best regards
Don

Don, thanks for the correction. That was an embarrassing mistake! I changed "Apollo capsule" to "lunar landers from the Apollo missions."

And I'll make a correction. While the lunar rocks were put in boxes that had seals, all the seals failed and the boxes were at or neat Earth atmospheric pressure when they retrieved them from the capsule after landing. That lead to the idea that all the lunar rocks were contaminated with Earth's water and the Moon was actually extremely dry. That idea was finally blown away when I discovered widespread water on the surface of the Moon announced in September 2009.
Clark, R. N., 2009, Detection of Adsorbed Water and Hydroxyl on the Moon, Science, 326, 562-564, DOI: 10.1126/science.1178105.

Roger

[Don Nelson]

Chris- (going Back on Topic)

I'm a 400DO, 500, 800 user.... If I were in your shoes with only the 400DO, I would not be thinking about adding a 4002.8 (I or II) since it duplicates a focal length you already own. Since the 400DO + 1.4TCII performs quite well, and gives you a ~560/f4, I would not be thinking about adding a 500/4 either. Since you've said money is no object, I'd stronly encourage you to look at the 800. I wouldn't consider the 600 as it will be replaced sometime in the not too distant future with the II model. I think you'll find the 400DO and 800 combination will be quite useful. Or maybe you can find one of the 1200's....anyone know if there is truth to the rumor that canon is taking orders again for these (~120K?). http://www.the-digital-picture.com/r...ns-Review.aspx

[Don Nelson]

Thanks for the additional item on the seals.

Anyway congrats on that article. I recall reading that. Nice to know who you are!

for those that want to read Roger's article - its online at http://www.sciencemag.org/content/32...a-466966865e9a.

Best regards.

[Bill Stubbs]

Don,
One minor correction - a 400 f4 DO +1.4 TC is effectively ~560 at f5.6, not f4 (the 1.4TC costs one stop). This is not critical, unless one is looking for ultimate low-light performance. I'd suggest, like you, that Chris consider the current 800 f5.6L IS, since it's lighter than either the current 400 f2.8 or 600 f4 by over 1 1/2 lbs. That + a 1.4 TC is effectively ~1120 f8 which will still autofocus on 1D series bodies. IMO, that's right at the practical limit anyway, as otherwise, we wind up shooting through too much air, with all the potential problems that entails. Should the new 500 II or 600 II be that much of a weight or optical improvement when they finally come out, he can either get one of those if he decides he really needs an intermediate length, or pick up what will then be the "old model", i.e. the current 500 or 600 f4, at a considerable saving.

Incidentally, I don't know about anyone else, but I don't think I'd really want a 1200, even if I could afford it; it's like that 200-500 f2.8 Sigma - do any of us really want to carry 34 to 36 lbs. of lens, plus the needed support gear (I don't think even Doug can hand hold one of those!:) )?

[Lynn Evans]

If you can buy a 1200, you will have the funds to pay someone to carry it!!:p

[Don Nelson]

Bill

Well, yes and no. A 400DO f4 with a 1.4 has an effective f stop of 5.6 for light purposes, but the depth of field will only be around f4.6. So its not exactly equivalent to a lens made to 560mm f5.6.


Moose likes teleconverters for this depth of field effect.... If you'd like a good explanation, Moose Peterson has it in his book. You can find it on page 35 of http://books.google.com/books?id=6rM...0moose&f=false

And I agree with Lynn's assessment of the 1200.

Best regards
Don

[Roger Clark]

On my plot above, the 800 f/5.6 on the 7D plots above the 600 f/4+1.4x and below the 400 f/2.8+2x. The clear aperture of the 800 f/5.6 is 143 mm. The 600 f/4 is 150mm aperture. The diffraction and the pixel sampling limits resolution, although the apertures are so close, the difference is very small. This does assume that the TCs do not significantly degrade resolution beyond that by diffraction. My experience with the 300 f/2.8 and 500 f/4 with TCs are that the image quality is superb and any degradation is due do diffraction as expected. And each lens has about the same diffraction.

So bottom line, the 400 f/2.8, 500 f/4, 600 f/4, and 800 f/5.6 as have very close to the same ultimate resolution.

If I were making this choice, I would choose between the 600 f/4 and 400 f/2.8, and after seeing the results of the computations, I would choose the 400 f/2.8. After all, there are also advantages of an f/2.8 aperture for isolating the subject.

Roger

[Chris Brennan]

Thank you all for your enlightened opinions! And a big thanks to Roger whose technical explanations has proven to me that my 7D can indeed "play with the big boys"!

After all is said and done, I'm inclined to go with the new 400/2.8 assuming that it meets with the critical reviews that we all hope it will. However, this selection is still a few months off so I've got some time to mull it over...

I can safely say that the 1200 will NOT be under consideration!

[Arthur Morris]

Chris, So far nobody has asked the key 400 f/2.8 question: what do you like to photograph????

[Joerg Rockenberger]

I enjoy using the 400mm focal length but there have been a few times when having something even longer than the 400+1.4x would be beneficial.

As happens so often you already answered your own question. JR

[Chris Brennan]

Chris, So far nobody has asked the key 400 f/2.8 question: what do you like to photograph????

I enjoy shooting raptors pretty much all year! Hawks whenever I can, eagles, mostly in the fall and winter, and osprey in the spring and summer. Although I prefer flight shots, other activity or perched shots are fine under the right circumstances. And I shoot military jet demonstration teams: The Blue Angels, The USAF Thunderbirds, the CAF Snowbirds, as well as individual jet demos... (Sorry, but I'm just not a "prop" warbird kind of guy...) usually in the spring, summer and early fall.

I hope that answers your question, Artie...

[Arthur Morris]

Thanks Chris. What's a "prop" warbird???

As for your choice of the 400 f/2.8, it makes zero sense to me given what you like to photograph. That lens is great for folks who photograph lots of big mammals in low light.... That does not sound like you. As far back as the original ABP I wrote that folks should always go for a longer, slower lens than a shorter faster one....

Are you planning on hand holding whatever new lens you buy?

[David Stephens]

As for your choice of the 400 f/2.8, it makes zero sense to me given what you like to photograph. That lens is great for folks who photograph lots of big mammals in low light.... That does not sound like you. As far back as the original ABP I wrote that folks should always go for a longer, slower lens than a shorter faster one....

Are you planning on hand holding whatever new lens you buy?

Art, I took your advice and bought the 500mm f/4L IS and I'm steadily improving at hand holding it, although I mostly use it on a tripod. Still, the hand-holding, particularly when locked onto a perched raptor waiting on it to fly can be trying. It's an isometric exercise that's significantly added to my left arm strength, but it's also got me thinking about a 300mm f/2.8, plus a 1.4x and 2.0x TC.

Also, Roger says:

So bottom line, the 400 f/2.8, 500 f/4, 600 f/4, and 800 f/5.6 as have very close to the same ultimate resolution.

If I were making this choice, I would choose between the 600 f/4 and 400 f/2.8, and after seeing the results of the computations, I would choose the 400 f/2.8. After all, there are also advantages of an f/2.8 aperture for isolating the subject.

Which has me thinking that a shorter/faster lens, plus TCs could be an answer for much of this shooting. Also, the prospect of the Series III TCs having higher IQ (yet to be seen, but we'll know soon) all have me leaning toward a lighter combo. Of course, there'd still be plenty of use for the 500mm plus the 1.4TC and the 500mm alone, but I might be inclined to get a shorter, ligher, faster combo for hand-holding situations, like shooting from boats and cars, or while hiking.

Is the paradyme starting to shift?

[Josée Normandeau]

I enjoy shooting raptors pretty much all year! Hawks whenever I can, eagles, mostly in the fall and winter, and osprey in the spring and summer. Although I prefer flight shots, other activity or perched shots are fine under the right circumstances. And I shoot military jet demonstration teams: The Blue Angels, The USAF Thunderbirds, the CAF Snowbirds, as well as individual jet demos... (Sorry, but I'm just not a "prop" warbird kind of guy...) usually in the spring, summer and early fall.

I hope that answers your question, Artie...

You don't need a very big lens to shoot those metallic birds. I did it with a 300mm. I think your 400 DO will be just fine.

As a complement to the 400, and it will be a big help with those raptors, I suggest the 800.

If a ''5 feet 4, 125 pounds woman'' can handhold a 500 (Doug convinced me tripods are for wimps), I guess you can handhold a 800 when the situation requires it. :p

[Arthur Morris]

Josée, You are forgetting that Chris is now a rich man :) So he needs to consider both the 500 and 600 f/4 L IS Series II lenses that will be released in 2011. Yes, the 800 is a great lens but the two new ones plus the Series III TCs may very well change the playing field....

[Josée Normandeau]

Josée, You are forgetting that Chris is now a rich man :) So he needs to consider both the 500 and 600 f/4 L IS Series II lenses that will be released in 2011. Yes, the 800 is a great lens but the two new ones plus the Series III TCs may very well change the playing field....

Roger Clark said in another post that the 500mm is already a heck of a lens (well he said that in a different way) and that the new version will only slightly improve the quality of the images. I believe that.

We will know for sure when the new versions are released. If Chris really want to consider those two pieces of glass, then he should wait and see what the reviews are (that's what I would do).

It is harder to give advice on wich lens to buy when the buyer has a lot of money to spend...:p:p:p

[Arthur Morris]

Roger Clark said in another post that the 500mm is already a heck of a lens (well he said that in a different way) and that the new version will only slightly improve the quality of the images. I believe that.

We will know for sure when the new versions are released. If Chris really want to consider those two pieces of glass, then he should wait and see what the reviews are (that's what I would do).

It is harder to give advice on wich lens to buy when the buyer has a lot of money to spend...:p:p:p

I would be pretty much amazed that Roger said exactly that. Can you shoot me a link please. First off, the new 500 f/4 will feature the newer 4-stop IS that is already in the 800 and the 70-200 f/2.8L IS II. It is amazing. Second, it will be much lighter than the current 500. Third, it is extremely likely that the series II lenses with the series III TCs will be quite amazing.

So please show me where Roger said, "... the new version will only slightly improve the quality of the images."

[Josée Normandeau]

I would be pretty much amazed that Roger said exactly that. Can you shoot me a link please. First off, the new 500 f/4 will feature the newer 4-stop IS that is already in the 800 and the 70-200 f/2.8L IS II. It is amazing. Second, it will be much lighter than the current 500. Third, it is extremely likely that the series II lenses with the series III TCs will be quite amazing.

So please show me where Roger said, "... the new version will only slightly improve the quality of the images."

This is what Roger said on your thread ''TCs or not TCs for better...''

Quote: Super telephoto lenses have superb image quality and greatly benefit from adding TCs. I've been continuously impressed with my super telephoto lens sharpness as the DSLR pixel sizes have dropped. It seems that with the 500 f/4 and the 300 f/2.8 (the lenses I have), I can stack TCs, even two 2x TCs and still extract more detail (1D IV body).

So this says the supertelephotos are performing extremely well (so the improvement in the new lenses can't be all that much better as they are already stunningly superb optically).

I hope I understood well! :o

Of course the new 500 and 600 wil be extremely fine. Will that be worth the price increase? A lighter lens is a big plus, I admit. And the new IS.

The 800 already has the new IS and is rather light. If I had the money, I would own one...

[Arthur Morris]

Well, you might have understood what he wrote but you totally misquoted him :)

As I surmised, he said nothing about the Series II lenses....

[Roger Clark]

As for your choice of the 400 f/2.8, it makes zero sense to me given what you like to photograph. That lens is great for folks who photograph lots of big mammals in low light.... That does not sound like you. As far back as the original ABP I wrote that folks should always go for a longer, slower lens than a shorter faster one....

Are you planning on hand holding whatever new lens you buy?

Artie,
Things have changed. I agree with your assessment of longer focal length, but for cameras of 3+ years ago. Today's cameras have smaller pixels, so one does not need as long a focal length. For example, back in 2004, and 2005, the premier Canon cameras was the 1D Mark II (and IIN) with 8.2 micron pixels. Now with a 1D Mark IV we have 5.7 micron pixels. That means, for example that a 500 mm lens on a 1DIV gets the same pixels on a subject as a 719 mm focal length lens would have done on a 1D Mark II. It is like have a built in 1.44x TC but without the light loss (because of a more efficient sensor).

So getting a 400 mm lens today is like a 575 mm lens back in the 1D Mark II days.

And a 7D is another apparent boost in equivalent focal length (like a 1.33x TC over a 1DIV, but with the light loss as both sensors have similar throughput).

With the smaller pixels, we are getting to diffraction limits, which were not much of an issue back in 1DII days (and 1DIII days were similar to 1DII days in this regard).

So the combination of smaller pixels and diffraction limits has changed how much focal length one needs in my opinion.

Now hand holding such a large lens (400 f/2.8) is still the same issue it has always been.:o

Roger

[Alan Lillich]

Following up on the last 2 posts and emphasizing what Artie said - it isn't bare lens IQ that is the (hoped) big story with the new 500 and 600. Reduced weight is THE big point. Imagine what a 600 f/4 at the same weight as the current 500 f/4 will do for BIF. Improved optical matching with the series III TCs could mean a 600 f/4 plus 1.4 TC is an 840 f/5.6 at the same IQ and 2 pounds lighter than the current 800. Lots of speculation, but if I were in a money-is-no-object position I would wait for the 500 and 600 announcements.

[Doug Brown]

Following up on the last 2 posts and emphasizing what Artie said - it isn't bare lens IQ that is the (hoped) big story with the new 500 and 600. Reduced weight is THE big point. Imagine what a 600 f/4 at the same weight as the current 500 f/4 will do for BIF. Improved optical matching with the series III TCs could mean a 600 f/4 plus 1.4 TC is an 840 f/5.6 at the same IQ and 2 pounds lighter than the current 800. Lots of speculation, but if I were in a money-is-no-object position I would wait for the 500 and 600 announcements.

AMEN Alan !!!

[Alan Lillich]

OK, my previous post was about the 2 before Rogers latest. I don't dispute Roger's point about pixels on subject equivalence with denser sensors and shorter lenses. A shorter lens on a new body will match the older bodies with a longer lens, but that isn't the goal. For what I do I'm still often with the subject smallish in the frame, so a longer lens will get a better result by filling the frame more.

[Desmond Chan]

Today's cameras have smaller pixels, so one does not need as long a focal length. [snip]

So the combination of smaller pixels and diffraction limits has changed how much focal length one needs in my opinion.

But, given the same distance between you the photographer and the subject, don't you still need to walk a bit closer in order to fill the frame with the subject if you are using a 400, while with a 500 or longer lens you may not need to??

Or are you saying that you can just crop the final image from the 400 and get the same quality as you could back then with an older camera/lens?

Personally I think it's the f2.8 that some photographers just find it so irresistible that they need to have one, even though it means they have to add a TC to the lens to shoot anyway, meaning not really shooting with the f2.8 after all, while a 500 could be cheaper and easily carried around and handled. :p

[Chris Brennan]

Thanks Chris. What's a "prop" warbird???

As for your choice of the 400 f/2.8, it makes zero sense to me given what you like to photograph. That lens is great for folks who photograph lots of big mammals in low light.... That does not sound like you. As far back as the original ABP I wrote that folks should always go for a longer, slower lens than a shorter faster one....

Are you planning on hand holding whatever new lens you buy?

Artie -

A "prop" war bird is usually an aircraft of WWII vintage like a P-51 Mustang or even a B-17...

In terms of the jet shooting, the new lens would probably NOT be used. The past few years, I've gotten some great images hand-handholding the 400DO, and this year, I look forward to complementing it with my new 70-300L.

I'm pretty sure that the new lens will be used for the "birding" part of my photography, and be on a monopod or tripod 100% of the time... I liked the idea of having the 400/2.8 which "shines" with either the 1.4x or the 2x depending on subject matter....

And, oh yeah, I ain't rich yet but it's very promising!!

[Arthur Morris]

The 400 2.8 with the 2X will put you at 800. The new 600 f/4 L IS II with the 2X will put you at 1200. With the size of the subject in the frame proportionate to the square of the focal length you are looking at 64 vs 144.... That represents a massive edge in magnification for the 600 f/4 over the 400 2.8. Not to mention that the 400 2.8 would duplicates a focal length that you already own. So purchasing either the new 500 or the new 600 would make 1,000 times more sense for you. But it's your money and your choice; I am just trying to help....

[Arthur Morris]

Following up on the last 2 posts and emphasizing what Artie said - it isn't bare lens IQ that is the (hoped) big story with the new 500 and 600. Reduced weight is THE big point. Imagine what a 600 f/4 at the same weight as the current 500 f/4 will do for BIF. Improved optical matching with the series III TCs could mean a 600 f/4 plus 1.4 TC is an 840 f/5.6 at the same IQ and 2 pounds lighter than the current 800. Lots of speculation, but if I were in a money-is-no-object position I would wait for the 500 and 600 announcements.

I'd be amazed if the 600 II will weigh two pounds less than the 800, but IAC, I have long realized that the new 600--if it works 1/2 as well with the Series III TCs as we are hoping--may very well render the 800 f/5.6 as obsolete....

Simply put it will give you much more versatility than the 800: 600 and 840 and 1200 as compared to 800 and 1140. That assuming that folks are able to create sharp images with the 600 and the 2X. With past f/4 super-telephoto lenses too, too many folks have shied away completely from using the 2X TCs. I have not and have urged others to master the sharpness techniques needed to work successfully at either 1000 or 1200mms.

[Chris Brennan]

The 400 2.8 with the 2X will put you at 800. The new 600 f/4 L IS II with the 2X will put you at 1200. With the size of the subject in the frame proportionate to the square of the focal length you are looking at 64 vs 144.... That represents a massive edge in magnification for the 600 f/4 over the 400 2.8. Not to mention that the 400 2.8 would duplicates a focal length that you already own. So purchasing either the new 500 or the new 600 would make 1,000 times more sense for you. But it's your money and your choice; I am just trying to help....

Artie -

First of all, I sincerely appreciate your help and that of all the others who've contributed!

I know that either new or current 500 or 600 will do remarkably well with a 2x but my initial "plan" was to not go away from the 7D or similar 1.6x crop body, and unless Canon changes something, I won't be able to AF @ f8 with the 1.6x body. On the other hand, I "suppose" I could get a 1D MkIV and have the best of all worlds!

I'd hoped to be in a position to purchase something before the summer months though and I believe that the release date for the new 500 or 600 is still a ways off. It still may make some sense to invest in a MkIV and a current 500 and one of the new III 2x TCs though in my time frame 'cause it's not like this gear loses much of it's value... My suspicion is that the new 400's will be few and far between for quite some time after their spring release date, and the money I save by not buying the new MkII 400 could be invested in the 1DMkIV and the current 500... sorry folks, I'm just rambling here...

[David Stephens]

...
Simply put it will give you much more versatility than the 800: 600 and 840 and 1200 as compared to 800 and 1140. That assuming that folks are able to create sharp images with the 600 and the 2X. With past f/4 super-telephoto lenses too, too many folks have shied away completely from using the 2X TCs. I have not and have urged others to master the sharpness techniques needed to work successfully at either 1000 or 1200mms.

Thanks for the jab, not aimed directly at me but scoring nonetheless.

That's a logical "next step" for me. I've gotten pretty good with my 500/f4 on the tripod and, more recently, I'm getting sharp handheld images, both free standing and braced, with and without the 1.4x TC. I've debated buying a new Series III 2.0x TC, but thought, "I don't have a Series II lens and my 7D isn't going to AF with that TC." Now I'm thinking of subjects where I could use the 1,000mm and also, now that I consider the potential lighter weight AND more powerful IS, a Series II 500mm or 600mm might be even better handheld and in general IQ.

Oh, well, it's only money.

[Roger Clark]

The 400 2.8 with the 2X will put you at 800. The new 600 f/4 L IS II with the 2X will put you at 1200. With the size of the subject in the frame proportionate to the square of the focal length you are looking at 64 vs 144.... That represents a massive edge in magnification for the 600 f/4 over the 400 2.8. Not to mention that the 400 2.8 would duplicates a focal length that you already own. So purchasing either the new 500 or the new 600 would make 1,000 times more sense for you. But it's your money and your choice; I am just trying to help....

Artie,
I'm attaching a plot from my telephoto reach web page:
http://www.clarkvision.com/articles/telephoto_reach/

The plot shows the resolution limits for the 7D plus various telephoto lenses. It shows a significant resolution increase from the 400 f/4 DO to the 400 f/2.8, and that the 400 f/2.8, 500 f/4, and 600 f/4 all have about the same limits. The 800 f/5.6 plots between the 500 f/4 and 600 f/4. So while you get more pixels on a subject with more focal length, you don't actually get more detail.

Of course this is theory and it would be nice to see some actual images from a 7D and a 400 f/2.8 versus an 500, 600, and 800 mm lens to verify these calculations. (Even though I do theory, I am also an experimentalist and want to see real results.) I have verified my results with the lenses I have (e.g. 300 f/4, 300 f2.8 and 500 f/4).

Roger

[Arthur Morris]

Thanks for the jab, not aimed directly at me but scoring nonetheless.

That's a logical "next step" for me. I've gotten pretty good with my 500/f4 on the tripod and, more recently, I'm getting sharp handheld images, both free standing and braced, with and without the 1.4x TC. I've debated buying a new Series III 2.0x TC, but thought, "I don't have a Series II lens and my 7D isn't going to AF with that TC." Now I'm thinking of subjects where I could use the 1,000mm and also, now that I consider the potential lighter weight AND more powerful IS, a Series II 500mm or 600mm might be even better handheld and in general IQ.

Oh, well, it's only money.

Yeah, and we only live once.... No jab at all intended. All I ever do is give me opinion and tell folks how I make the images I do. What they do is up to them and not my business :)

[Arthur Morris]

Roger, I am really trying to understand this stuff.

Basic question: is resolution the same as sharpness?

Possibly the same question in a different form. When you say, "The 800 f/5.6 plots between the 500 f/4 and 600 f/4" are you saying that the old 600 f/4 produces sharper images than the 800 f/5.6?

[Arthur Morris]

Artie -

First of all, I sincerely appreciate your help and that of all the others who've contributed!

I know that either new or current 500 or 600 will do remarkably well with a 2x but my initial "plan" was to not go away from the 7D or similar 1.6x crop body, and unless Canon changes something, I won't be able to AF @ f8 with the 1.6x body. On the other hand, I "suppose" I could get a 1D MkIV and have the best of all worlds!

I'd hoped to be in a position to purchase something before the summer months though and I believe that the release date for the new 500 or 600 is still a ways off. It still may make some sense to invest in a MkIV and a current 500 and one of the new III 2x TCs though in my time frame 'cause it's not like this gear loses much of it's value... My suspicion is that the new 400's will be few and far between for quite some time after their spring release date, and the money I save by not buying the new MkII 400 could be invested in the 1DMkIV and the current 500... sorry folks, I'm just rambling here...

Whatever camera you own choosing the 400 f/2.8 over either any 500 f/4 or 600 f/4 (old or new) or over the 800 makes no sense to me. I do not however, understand Dawes + Pixel Sampling Limits....)

I shall cease trying to beat the dead horse :)

[Roger Clark]

But, given the same distance between you the photographer and the subject, don't you still need to walk a bit closer in order to fill the frame with the subject if you are using a 400, while with a 500 or longer lens you may not need to??

Or are you saying that you can just crop the final image from the 400 and get the same quality as you could back then with an older camera/lens?

Desmond,
I'm saying that compared to the older cameras, we get the same pixels and detail on a subject with a shorter focal length lens. So, for example, with a 16 megapixel 1DIV versus an 8 megapixel 1DII, one can crop the 1DIV to 8 megapixels and get pretty much the same image quality. Of course if you can move closer, that is a real win. But often in bird photography, we can't. But cameras+big telephoto lenses are now approaching a fundamental limit: diffraction. So more focal length won't give more detail on a subject.

Personally I think it's the f2.8 that some photographers just find it so irresistible that they need to have one, even though it means they have to add a TC to the lens to shoot anyway, meaning not really shooting with the f2.8 after all, while a 500 could be cheaper and easily carried around and handled. :p

In most of my bird photography, I will admit that I usually use 1,4x and 2x TCs on my 500 f/4 and 300 f/2.8 (although not 2x with BIF). But now that I am understanding the diffraction limit better, I'm backing off and using the 300 at f/2.8 more. The big test will be my next trip to Tanzania where I'm only taking my 300 f/2.8 and not a 500. So it is a paradigm shift that is hard to embrace because of our years of experience saying focal length matters a lot.

Roger

[Arthur Morris]

The more of this I read the less I understand. I have no clue as to what diffraction is or what it has to do with super-telephoto photography....

What actually is diffraction?

Can we see diffraction in our images?

Is diffraction evident in the image below? (I will gladly post the tech specs after folks chime in here.)

[Roger Clark]

The more of this I read the less I understand. I have no clue as to what diffraction is or what it has to do with super-telephoto photography....

What actually is diffraction?

Can we see diffraction in our images?

Is diffraction evident in the image below? (I will gladly post the tech specs after folks chime in here.)

Artie,
Think of throwing a rock in a lake and seeing the waves generated. If those the waves encounter something like a stick, the waves as they pass by the stick bend around the stick. It is the same with light passing through a lens. At the edge of the lens, the light, which acts like a wave (like the ripples in the water) get bent. The lens is also bending the light, but differently than the diffracted light. So the diffracted light in effect messes up the ideal focus of the lens and limits the fine detail.

Diffraction is affecting most images from most modern digital cameras with most super telephoto lenses. The effect is first reduced contrast in the finest details. How much reduction is dependent on the f/ratio and the pixel sampling. In the image you posted, it is difficult to tell because we don't know what is true. For example, the finest detail in the feathers appears pretty low in contrast in your posted image. That is probably due to the limiting effects of diffraction. But since we don't know the f/ratio and pixel size used to make the image, and how much unsharp mask (or similar algorithm) may have been used to try and correct the low contrast, it is hard to say exactly how much diffraction is affecting the image.

If you tell us that the camera and lens are, I can then tell you how much contrast you lost.

Roger

[Roger Clark]

Roger, I am really trying to understand this stuff.

Basic question: is resolution the same as sharpness?

Possibly the same question in a different form. When you say, "The 800 f/5.6 plots between the 500 f/4 and 600 f/4" are you saying that the old 600 f/4 produces sharper images than the 800 f/5.6?

Resolution and sharpness can be related but are not necessarily the same. One could actually resolve more detail, e.g. fine details in a bird's feathers, bu the image appears what we call soft. Another image might not resolve that same detail in the feathers, but edges have high contrast giving the perception of being sharper. It is complex because of the differing human perception and different people judge different things in images.

Regarding a 600 f/4 and 800 f/5.6, which produces a sharper image depends on the camera. For a camera with large pixels, where the pixels are the limitation, the 800 will produce more detail and can appear sharper. But with a camera with finer pixels (e.g. a 7D with a 600+ 1.4x TC), diffraction is more of a limitation and the images will be almost the same in detail. If the 600 is a really good lens, the 600 could be slightly sharper than the 800 with the 7D because the len's aperture of the 600 is slightly larger (although so close probably only seen in a measured test target).

So if you have a 600 f/4 and 800 f/5.6 and a 7D, try it and see if you can see a difference.

Roger

[Arthur Morris]

Artie,
Think of throwing a rock in a lake and seeing the waves generated. If those the waves encounter something like a stick, the waves as they pass by the stick bend around the stick. It is the same with light passing through a lens. At the edge of the lens, the light, which acts like a wave (like the ripples in the water) get bent. The lens is also bending the light, but differently than the diffracted light. So the diffracted light in effect messes up the ideal focus of the lens and limits the fine detail.

Diffraction is affecting most images from most modern digital cameras with most super telephoto lenses. The effect is first reduced contrast in the finest details. How much reduction is dependent on the f/ratio and the pixel sampling. In the image you posted, it is difficult to tell because we don't know what is true. For example, the finest detail in the feathers appears pretty low in contrast in your posted image. That is probably due to the limiting effects of diffraction. But since we don't know the f/ratio and pixel size used to make the image, and how much unsharp mask (or similar algorithm) may have been used to try and correct the low contrast, it is hard to say exactly how much diffraction is affecting the image.

If you tell us that the camera and lens are, I can then tell you how much contrast you lost.

Roger

The image is an 800 wide unsharpened JPEG created from the extracted JPEG. 800 with the 1.4X TC and the Mark IV. Does the aperture matter????

[Arthur Morris]

Resolution and sharpness can be related but are not necessarily the same. One could actually resolve more detail, e.g. fine details in a bird's feathers, bu the image appears what we call soft. Another image might not resolve that same detail in the feathers, but edges have high contrast giving the perception of being sharper. It is complex because of the differing human perception and different people judge different things in images.

Regarding a 600 f/4 and 800 f/5.6, which produces a sharper image depends on the camera. For a camera with large pixels, where the pixels are the limitation, the 800 will produce more detail and can appear sharper. But with a camera with finer pixels (e.g. a 7D with a 600+ 1.4x TC), diffraction is more of a limitation and the images will be almost the same in detail. If the 600 is a really good lens, the 600 could be slightly sharper than the 800 with the 7D because the len's aperture of the 600 is slightly larger (although so close probably only seen in a measured test target).

So if you have a 600 f/4 and 800 f/5.6 and a 7D, try it and see if you can see a difference.

Roger

Hey Roger, I am thinking that my brain might not be wired for this stuff. Right now I have only the 800, the 400 DO, and the 70-200 2.8II plus some little stuff. Sold the 600 f/4 and both 500 f/4s....

Here's the thing: I do not have a great eye for fine detail. My keeper's look great to me and pretty much to anyone who looks at them. I am trying to get educated here but sometimes I ask myself why I am doing that :)

[Desmond Chan]

Desmond,
I'm saying that compared to the older cameras, we get the same pixels and detail on a subject with a shorter focal length lens. So, for example, with a 16 megapixel 1DIV versus an 8 megapixel 1DII, one can crop the 1DIV to 8 megapixels and get pretty much the same image quality.

That's what I got from what you said, Roger.

Of course if you can move closer, that is a real win. But often in bird photography, we can't.

Agreed, and all the more reason to pick the lens with a longer focal length to use.

But cameras+big telephoto lenses are now approaching a fundamental limit: diffraction. So more focal length won't give more detail on a subject.

All right.

In most of my bird photography, I will admit that I usually use 1,4x and 2x TCs on my 500 f/4 and 300 f/2.8 (although not 2x with BIF). But now that I am understanding the diffraction limit better, I'm backing off and using the 300 at f/2.8 more.

And try to get closer to the subject, use TCs, or crop the final image in order to get the same subject size in the frame of the final image?

So it is a paradigm shift that is hard to embrace because of our years of experience saying focal length matters a lot.

I think I'd agree with you on that.

Still, I wonder how all these actually affect the image quality of the final photographs as seen with our naked eyes. It looks like it could affect the selection of the tools we'd use. And it sounds like it could make getting a decent image of a distance bird easier (just crop :p).

[Alan Lillich]

I'm saying that compared to the older cameras, we get the same pixels and detail on a subject with a shorter focal length lens. So, for example, with a 16 megapixel 1DIV versus an 8 megapixel 1DII, one can crop the 1DIV to 8 megapixels and get pretty much the same image quality. Of course if you can move closer, that is a real win. But often in bird photography, we can't. But cameras+big telephoto lenses are now approaching a fundamental limit: diffraction. So more focal length won't give more detail on a subject.

Roger,

I'm still missing something here. I am not interested in "only" matching what could be done in the past. I care about getting the most out of the present. Your chart presented resolution in arc-seconds. With the same body at the same location shooting the same subject a longer lens will image the subject over more pixels, capturing smaller physical detail at the pixel level.

I agree that we're at the point where smaller pixels are not the way to go. But longer lenses still seem to make sense to me.

Alan

[Roger Clark]

Roger,

I'm still missing something here. I am not interested in "only" matching what could be done in the past. I care about getting the most out of the present. Your chart presented resolution in arc-seconds. With the same body at the same location shooting the same subject a longer lens will image the subject over more pixels, capturing smaller physical detail at the pixel level.

I agree that we're at the point where smaller pixels are not the way to go. But longer lenses still seem to make sense to me.

Alan

Our experiences have been colored by the cameras that are available, e.g. film and large pixels in the Canon 1DII, 1DII, and 5D2, Nikon D3, etc. But when a new camera comes out with smaller pixels, it gets closer the resolution limits of the lens. When we reach that, more focal length does not help get more detail on the subject. And where that point is is dependent on the lens. For example, a 300 mm f/2.8 versus a 300 f/4 versus a 300 f/5.6. Each lens coupled with a given camera will reach its fundamental limits at dirrefernt focal lengths. What determines the resolution is aperture (assuming excellent design for the lens). So the 300 f/2.8 has the largest aperture of those three and will produce the finest detail and highest contrast in the fine detail.

The 600 f/4 has the largest lens aperture diameter of the super telephotos (150 mm), with the 400 f/2.8 and 800 f/5.6 close at 143 mm. These all have the very close resolution limits.
We are now able to reach these limits with modest focal lengths (e.g. 7D with 600 + 1.4x is very close to the lens limit, or 7D + 400 f/2.8+2x). With the 1DIV with slightly larger pixels we still need use a 2X on a 600 f/4, or stack TCs to reach the limits on a 400 f/2.8.

So to get the most detail on a subject with a 1D Mark IV, the 600 f/4 II will, I bet< do the best job. With a 7D the 600 f/4 still wins by a hair, but the 400 f/2.8 and 800 f/5.6 are almost the same, and the 500 f/4 is only slightly behind.

So focal length is still important, but less so than it used to be. And if the next generation of cameras comes out with yet smaller pixels, focal length will be even less important. If the pixels get small enough, there would be no need for TCs.

Roger

[Alan Lillich]

Roger,

I think I understand what you are saying now. I was looking through the lens the in the wrong direction. Let me know if this restatement is accurate:

The detail that can be resolved with a lens is related to the angular width of the subject features from the viewer's location. This angular width does not change with greater focal length, it is only dependent on the physical size of the features and the distance to them. Different lenses can resolve different subject angular widths, as shown in the charts in earlier posts.

This subject resolution gets projected on the sensor as a different angular width that is dependent on the focal length. Longer focal lengths have a larger projected angular width.

If the pixels on a sensor are larger than the projected angular width then you're not getting everything out of the lens. Multiple pieces of lens-resolved subject detail are projected onto each pixel. In this case a longer focal length lens might help by magnifying the lens-resolved subject detail to better match the pixel pitch.

If the pixels on a sensor are smaller than the projected angular width then you're getting everything possible out of the lens. You are limited by the lens resolving power. If a shorter lens and a longer lens are both out-resolved by the sensor, then the longer lens has no advantage for resolving subject detail. You might as well use the shorter lens and crop.

End of restatement, and follow-on questions. I'm still wondering about other effects. It will be very interesting to try to test these ideas and see what you learn on your trip to Africa. The pixel spacing is discrete and the lens-resolution fall-off is continuous. Are there secondary benefits to having smaller pixels? E.g. better perceived sharpness due to better edge detection, less impact from the anti-alias filter, de-mosiacing working at a smaller scale?

Alan

[Roger Clark]

Roger,

I think I understand what you are saying now. I was looking through the lens the in the wrong direction. Let me know if this restatement is accurate:

The detail that can be resolved with a lens is related to the angular width of the subject features from the viewer's location. This angular width does not change with greater focal length, it is only dependent on the physical size of the features and the distance to them. Different lenses can resolve different subject angular widths, as shown in the charts in earlier posts.

This subject resolution gets projected on the sensor as a different angular width that is dependent on the focal length. Longer focal lengths have a larger projected angular width.

If the pixels on a sensor are larger than the projected angular width then you're not getting everything out of the lens. Multiple pieces of lens-resolved subject detail are projected onto each pixel. In this case a longer focal length lens might help by magnifying the lens-resolved subject detail to better match the pixel pitch.

If the pixels on a sensor are smaller than the projected angular width then you're getting everything possible out of the lens. You are limited by the lens resolving power. If a shorter lens and a longer lens are both out-resolved by the sensor, then the longer lens has no advantage for resolving subject detail. You might as well use the shorter lens and crop.

End of restatement, and follow-on questions. I'm still wondering about other effects. It will be very interesting to try to test these ideas and see what you learn on your trip to Africa. The pixel spacing is discrete and the lens-resolution fall-off is continuous. Are there secondary benefits to having smaller pixels? E.g. better perceived sharpness due to better edge detection, less impact from the anti-alias filter, de-mosiacing working at a smaller scale?

Alan

Alan,
Your re-statement look fine. I do not think there are other benefits to smaller pixels, in fact, just the contrary. While smaller pixels potentially resolve more detail if the lens can deliver that detail, smaller pixels hold fewer electrons (thus saturate with less light). So you can't get the signal-to-noise ratio that you can with larger pixels (this is the main reason that P&S camera images are so noisy: small pixels). Bigger pixels with a bigger lens will deliver better images, but the lenses get large, expensive and heavy, so it is a compromise on how much we can afford and carry, over image quality.

There is a reason astronomers are building the LSST telescope: a 9,900 mm focal length, f/1.18 telescope with a 3.2 gigapixel camera!
http://en.wikipedia.org/wiki/Large_S...rvey_Telescope
Great bird camera but it is not portable.:confused:

Roger

[subhrashis]

Alan,
There is a reason astronomers are building the LSST telescope: a 9,900 mm focal length, f/1.18 telescope with a 3.2 gigapixel camera!
http://en.wikipedia.org/wiki/Large_S...rvey_Telescope
Great bird camera but it is not portable.:confused:

Roger

If only the Earth was flat and the air less polluted, portability would not matter much for this camera!
More portable would be the Zeiss 1700/4 on a medium format!