Aptus 22 vs 5DII

[JonRoemer]

I am testing mine now. Concern.

I have the original 1ds3 screen installed in both my bodies. It's marked by "CIV" on the screen, you'll need a magnifying glass.

Testing tethered at f1.2. If I focus at about four feet, it's sharp, but at 15 feet it's backfocused. So now what to do?

Let's not even talk about the CA at 1.2; I almost feel like these super fast lenses is stretching what's really usable. One thing if it's surveillance photography of John Gotti in a Little Italy alley, but quite another thing if you really need to depend on these things to be tack sharp.

The 1ds3 has custom focus tweak options in the custommenus, but when it's focusing differently at different distances, WTF?

So much to keep up with. Details, details.

And the focus not matching the actual file, depth of field, is very apparent with the original stock screen in the 1ds3. So be forewarned. Focus at what looks good, and then stop down two more stops!

Hey Gwhitf -> check this out Canon white paper; pp. 13-15.  It goes over steps for doing the microadjustment.

I did this when calibrating my 16-35 II lens.  It's definitely easier to do tethered but don't worry about reviewing the results while you shoot.  You're going to want to review the files on a large monitor and compare them.  So, do tests at every setting, locked down, wide open and using the center focus point.  Either make notes or place a card in each shot with the settings or both (you can read the MA adj. settings in DPP (command - I on a file) but that's a pain.)  It's easier to take notes.  If you place a card in the shot remember that it won't be legible if it's too close to the camera.

Regarding no back focus close-up vs. having it further out, I don't know.  I do know that using MA helped on my 16-35 II at any focus distance.

If you do end up using the MA feature don't forget to copy it over to your other camera bodies.

[lisa_r]

You guys are really confusing me now.

What's wrong with this??

http://www.adorama.com/ICASECS.html

(this was yielded by searching "canon 1ds precision focusing screen" in google.)

[EricWHiss]

Jumping into the fray here on the DOF discussion....   Image circle size ....  that's what affects the rate at which the lens changes from in focus to out of focus.  It's sort of related to format size since the image circle needs to be larger for different film sizes.  If you like math you can actually calculate this.  The larger the film plane (or magnification) the faster the object goes out of focus over distance.    Lots of variety in image circle size from lens to lens in any given camera mount.  I've read that certain lenses such as some Leica R lenses or the Rollei 150mm apo macro have bigger image circles than necessary for their format - and that also tends to make a lens nice and sharp edge to edge plus have lower distortion.  This might explain why one lens has a different apparent DOF and look than another even though its got the same length or max aperture.  There are other factors too such as low frequency contrast or how a lens renders OOF areas that will affect its apparent DOF.  

I'd agree with the poster who said the apparent DOF on the focus screen seems bigger just because of the small size of the screen/viewfinder.  Easy to check that by reducing your image on screen to a thumbnail....DOF seems to grow.

Regarding the Ees focusing screen - works like a charm for manual focusing.  I wish I could get something like this for my MF cameras.

[Frank Doorhof]

Jumping into the fray here on the DOF discussion....   Image circle size ....  that's what affects the rate at which the lens changes from in focus to out of focus.  It's sort of related to format size since the image circle needs to be larger for different film sizes.  If you like math you can actually calculate this.  The larger the film plane (or magnification) the faster the object goes out of focus over distance.    Lots of variety in image circle size from lens to lens in any given camera mount.  I've read that certain lenses such as some Leica R lenses or the Rollei 150mm apo macro have bigger image circles than necessary for their format - and that also tends to make a lens nice and sharp edge to edge plus have lower distortion.  This might explain why one lens has a different apparent DOF and look than another even though its got the same length or max aperture.  There are other factors too such as low frequency contrast or how a lens renders OOF areas that will affect its apparent DOF.  

I'd agree with the poster who said the apparent DOF on the focus screen seems bigger just because of the small size of the screen/viewfinder.  Easy to check that by reducing your image on screen to a thumbnail....DOF seems to grow.

Regarding the Ees focusing screen - works like a charm for manual focusing.  I wish I could get something like this for my MF cameras.

It all boils down to the CoC

[thsinar]

One remark/issue often forgotten: used CoC for DoF is related/dependent to observing distance of the final image/film/sensor. A larger printed image, a bigger display, a larger sensor have a longer optimal observing distance (agreed on 2 times the diagonal) than a smaller, thus the used/optimal CoC can be bigger on the larger image/display/sensor, for the very same DoF. If one now goes closer as this optimal observing/viewing distance, then the whole is not true any longer.

Thierry

Jumping into the fray here on the DOF discussion....   Image circle size ....  that's what affects the rate at which the lens changes from in focus to out of focus.  It's sort of related to format size since the image circle needs to be larger for different film sizes.  If you like math you can actually calculate this.  The larger the film plane (or magnification) the faster the object goes out of focus over distance.    Lots of variety in image circle size from lens to lens in any given camera mount.  I've read that certain lenses such as some Leica R lenses or the Rollei 150mm apo macro have bigger image circles than necessary for their format - and that also tends to make a lens nice and sharp edge to edge plus have lower distortion.  This might explain why one lens has a different apparent DOF and look than another even though its got the same length or max aperture.  There are other factors too such as low frequency contrast or how a lens renders OOF areas that will affect its apparent DOF.  

I'd agree with the poster who said the apparent DOF on the focus screen seems bigger just because of the small size of the screen/viewfinder.  Easy to check that by reducing your image on screen to a thumbnail....DOF seems to grow.

Regarding the Ees focusing screen - works like a charm for manual focusing.  I wish I could get something like this for my MF cameras.

[EricWHiss]

It all boils down to the CoC

Nope. Frank that's not correct.  Think about it ... the CoC is arbitrary and has nothing to do with the lens at all.  The lens is designed to project an image circle large enough to cover a certain sized film plane.    If you use film on the same camera with same lens the CoC might be one value, but if you switch to a digital back changing nothing else, you'd use a different CoC for decisions on what's the minimum aperture or biggest enlargement. People that are more finicky than others will choose a smaller value for CoC.    

The factor here in the rate at how quickly a lens goes from in focus to OOF is image circle.

[Frank Doorhof]

in this case I meant with the diiference in the viewfinder. Coc and dof are connected in the appearance of distance and size.
A small print viewed from the same distance as a large print will show a difference in dof. That could be what you see in the viewfinder as on screen.

The optical brightness also helps because our eyes are very sensitive for contrast.
It's just an idea of what could explain the dof issue mentioned in this thread with the viewfinder
But also between mf and dslr.
Read could.

[paul_jones]

You guys are really confusing me now.

What's wrong with this??

http://www.adorama.com/ICASECS.html

(this was yielded by searching "canon 1ds precision focusing screen" in google.)

only adorama seems to list this model screen working with a 1ds. on others sites, inc canon, this is for a 5d (thats how i remeber it- i spent a fair bit of time researching it a while ago).

i think its a typo.

paul

[paul_jones]

Hey Gwhitf -> check this out Canon white paper; pp. 13-15.  It goes over steps for doing the microadjustment.

I did this when calibrating my 16-35 II lens.  It's definitely easier to do tethered but don't worry about reviewing the results while you shoot.  You're going to want to review the files on a large monitor and compare them.  So, do tests at every setting, locked down, wide open and using the center focus point.  Either make notes or place a card in each shot with the settings or both (you can read the MA adj. settings in DPP (command - I on a file) but that's a pain.)  It's easier to take notes.  If you place a card in the shot remember that it won't be legible if it's too close to the camera.

Regarding no back focus close-up vs. having it further out, I don't know.  I do know that using MA helped on my 16-35 II at any focus distance.

If you do end up using the MA feature don't forget to copy it over to your other camera bodies.

i done this a few times since ive had my 1.2 50 and 1.2 85 with the mk3. those lenses really need it- they are way out without first doing it (my lenses were).

but lately i was finding that my far right and far left focus points were way out- but my center point was spot on (after my calibrating). so i took my camera to canon, luckly on its last month of warrantee, and they confirmed it was way out. the tech apparently spent an entire day calibrating each focus point one by one with a special canon chart.

but its very good now. its amazing ive lived for a whole year thinking i just get inconsitant focus with the canon when shooting wide.

paul

[Carsten W]

only adorama seems to list this model screen working with a 1ds. on others sites, inc canon, this is for a 5d (thats how i remeber it- i spent a fair bit of time researching it a while ago).

i think its a typo.

paul

The Ee-series is for the Canon 5D, with the Ec-series being for the 1D/1Ds-series of cameras (at least from Mark II and up, perhaps even the originals). The -S screens are the ones for manual focusing, and showing more accurate depth of field in the viewfinder. I think the Ed-S is one of those typos which got propagated everywhere by copying on the internet. If you search for it, you will often see Ed-S in the title of the item, but Ec-S in the description.

[Snook]

Wanted to jump in with a quick question:
Will  the Next 1DsM?? or the Next 5D??  have 16 bit?
Is it possible?

S.

[jani]

Wanted to jump in with a quick question:
Will  the Next 1DsM?? or the Next 5D??  have 16 bit?
Is it possible?

Canon may know the answer to both the first question, nobody else can know for sure.

It certainly is possible to use 16 bits to encode the signal, but whether it has any useful purpose is again something that only Canon may answer.

The rest of us can only speculate.

[Panopeeper]

Wanted to jump in with a quick question:
Will  the Next 1DsM?? or the Next 5D??  have 16 bit?

Probably even the 14th bit is useless at the moment. More would be necessary only if Canon (or another DSLR manufacturer) changed to ISOless mode, like most MFDBs are working, thereby increasing the DR by two stops. Now, that would be great!

[Snook]

Probably even the 14th bit is useless at the moment. More would be necessary only if Canon (or another DSLR manufacturer) changed to ISOless mode, like most MFDBs are working, thereby increasing the DR by two stops. Now, that would be great!

Panopeeper, Sorry but why is it useless...?

Thank you
Snook

PS. I notice a Big difference between my 1DsMII files amd my P30 files when retouching. A Big difference that is. And I am pretty sure that is all to do with the 16 Bit..:+]

[lisa_r]

Is there anyone in the house who can measure DR of recent MF and Canon offerings and state what the differences are in terms of usable DR?

[Doug Peterson]

Panopeeper, Sorry but why is it useless...?

Thank you
Snook

Panopeeper is an ultra-tech when it comes to raw formats and signal processing. He can explain more, but I'm comfortable assuming he's saying that the encoding can only be as good as the signal. (warning I'm going to arbitrarily make up numbers to illustrate the concept). If the end-to-end signal (from capture to writing the final file and every material and mechanism inbetween) only accurately captures 3,000 shades of each color channel than 14 bit encoding will not be meaningfully better than 12-bit encoding.

In other words making a dSLR with mid-tier electronics 16-bit won't necessarily improve the final quality towards the quality of a Digital back.

Panopeeper would be the expert on this so I'll let him explain further.

Doug Peterson,  Head of Technical Services
Capture Integration, Phase One & Canon Dealer  |  Personal Portfolio

[Panopeeper]

why is it useless...?

More bit depth is useful if it represents finer transitions of light intensity between (near)adjacent pixels.

When I got my 40D, I made a really huge effort with targeted shooting, like slightly curved, very smooth surfaces, which created the finest transition. I wanted to "justify" the extra bits; no success. I have not seen anyone with a proof, that the more levels represent real transitions.

I did the same on a few 5D2 images, no success. (I did all that with non-demosaiced raw data.)

However, extreme adjustments (like brightness change by six stops) can show differences between 12bit and 14bit versions of the same image; thus I can't say the 13th bit is wortless. I always suggest using the max bit depth and avoiding lossy raw data.

If you happen to have a 14bit file (or any other, but not lossy), which you believe would suffer under lesser bit depth, upload it and I create you -1 and -2 (or more) bit versions in DNG format and you can experience with them.

[Panopeeper]

I notice a Big difference between my 1DsMII files amd my P30 files when retouching

That is not surprizing; the question is, if you can qualify the difference (i.e. what exactly is different).

I am pretty sure that is all to do with the 16 Bit..

Please upload such a P30 raw file and I will surprize you. A contrasty one please.

[Snook]

That is not surprizing; the question is, if you can qualify the difference (i.e. what exactly is different).

They just hold up MUCH better to abuse and under and over exposing, tweaking, Dodging and Burning etc...
actually in all aspects really.
Now that being said I have never worked on a 1DsMIII file yet either.

Please upload such a P30 raw file and I will surprize you. A contrasty one please.

They just hold up MUCH better to abuse and under and over exposing, tweaking, Dodging and Burning etc...
actually in all aspects really.
Now that being said I have never worked on a 1DsMIII file yet either.
Surprise me in what respect?

S.

[Panopeeper]

Is there anyone in the house who can measure DR of recent MF and Canon offerings and state what the differences are in terms of usable DR?

I compiled a bunch of data, see in Excel form

However, the data re the 5D2 is very few and *very* unreliable; I did not have raw images suitable for accurate measurement. Unfortunately, the raw files provided by Imaging Resources are not really suitable either. The D3 data is based on Stouffer wedge shots, which offers the highest degree or reliability possible on this level (i.e. without special equipment).

What I can say re the 5D2 is, that the DR (maximal @ ISO 50 and 100) is about 1/3 lower than that of the D3 (maximal @ ISO 100 and 200); the difference vanishes with higher ISOs, @ 1600 I don't see any difference.

I can only compare the dynamic ranges; I don't give any absolute number, for that is the question of acceptance (how much noise one accepts). Typically, one accepts much more noise in low light situation than in ideal illumination.

I am waiting for the arrival of a Stouffer wedge myself, and then I will try to find a willing 5D2 owner (and others) in my area to make shots targeted towards noise/DR measurement.

Then I will update the Excel chart.