Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

1       2      
3
       4              6       7       end
  

Archive 2012 · No interchangeable focusing screens for 5D3

  
 
andrew_rs
Offline
• • •
Upload & Sell: Off
p.3 #1 · p.3 #1 · No interchangeable focusing screens for 5D3


mco_970 wrote:
I wondered if it was a sign to Samyang, Zeiss, etc. to get their scurfy manually focusing paws off of Canon's 35/85/100 lens market.


Yeah, you never really can know all of the reasons for such a decision. It almost reminds me of a few years back when a number of sigma lenses started having problems with newer EOS cameras



Mar 03, 2012 at 12:01 AM
Lars Johnsson
Offline
• • • • • • •
Upload & Sell: Off
p.3 #2 · p.3 #2 · No interchangeable focusing screens for 5D3


wickerprints wrote:
I already said that.

MF shooters using lenses that don't have electronic aperture control through the body already have to cope with metering issues, so they don't really care if a custom screen affects metering further. Those of us who do use EF lenses are already aware of the way in which focusing screens affect the metering without a custom function to compensate.


That's not true. I have 9 non Canon MF lenses for my Canon bodies. And they all have electronic aperture control through the body.



Mar 03, 2012 at 12:07 AM
melcat
Offline
• • • •
Upload & Sell: Off
p.3 #3 · p.3 #3 · No interchangeable focusing screens for 5D3


Dawei Ye wrote:
How does the focus screen affect metering though? Isn't the focus screen only affecting what we see through the viewfinder?


It worked that way on many good film cameras, but on current EOS cameras the meter does look through the screen. You would think the metering would long ago have been integrated into the AF chip in the base of the camera via the submirror - this was where the Olympus OM-4, Nikon F-3, Contax S-1 etc. all had their meters. But it isn't how Canon did it. Now the AF talks to the AE unit and it *still* doesn't work that way.

Canon do still make a split-image screen for the 1 series. The spot meter won't work because the prism's in the way. I'm sure there are studio shooters out there who like the split image and don't need an in-camera meter,.

I have no problem losing the split-image rangefinder devices if AF confirmation works. It was always a pain with slower lenses on film cameras.

But the other functions of the screen - seeing bokeh, the "snap" of an image coming into focus as confirmation of what AF is doing - are important. They are useful even if the screen only shows depth effects to f/2 or so. I've never owned a Nikon body, but I've been reading reports that this is about what the non-interchangeable screens in the better Nikons do, and it is about what I saw when I tried one in the shop around D3/5D time. At that time the Nikon screen was noticeably better than the Canon. I estimate the stock screens on the 1D Mk III and 5D are only showing about f/2.8, with a not peaky enough "parabola". If Canon can get them to about where Nikon is, that might address most of the problem.



Mar 03, 2012 at 12:22 AM
wickerprints
Offline
• • • •
Upload & Sell: Off
p.3 #4 · p.3 #4 · No interchangeable focusing screens for 5D3


Dawei Ye wrote:
How does the focus screen affect metering though? Isn't the focus screen only affecting what we see through the viewfinder?


The metering system is located in the pentaprism housing. It sees only the light that is reflected by the primary reflex mirror, and what it sees depends on how the focusing screen diffuses the light prior to entering the pentaprism. If, for example, you install a split prism screen, spot metering can become inaccurate.

A focusing screen is analogous to a translucent sheet of material upon which a slide projector image is projected. If you tried to project the image onto a sheet of smooth, clear glass, the light just passes right through and you see little if any image. If you project it onto a frosted sheet of glass, you can see it because the focused rays are scattered at the image plane. The amount of scattering depends on the surface, and consequently, the perceived "brightness" of the image. The camera's metering system is therefore also affected by the characteristics of the screen.



Mar 03, 2012 at 12:24 AM
wickerprints
Offline
• • • •
Upload & Sell: Off
p.3 #5 · p.3 #5 · No interchangeable focusing screens for 5D3


melcat wrote:
It worked that way on many good film cameras, but on current EOS cameras the meter does look through the screen. You would think the metering would long ago have been integrated into the AF chip in the base of the camera via the submirror - this was where the Olympus OM-4, Nikon F-3, Contax S-1 etc. all had their meters. But it isn't how Canon did it. Now the AF talks to the AE unit and it *still* doesn't work that way.


The reason why is quite simple. Metering in modern AF bodies is done via the primary mirror because it sees the entire image. The secondary AF mirror behind the primary is small and cannot see the image periphery. Therefore, to put the metering with the AF system would mean you can only meter based on the central portion of the image visible to the secondary mirror, rather than the entire frame. This has the potential to result in metering error, and this configuration does not lend itself to evaluative or center-weighted average metering.



Mar 03, 2012 at 12:28 AM
wickerprints
Offline
• • • •
Upload & Sell: Off
p.3 #6 · p.3 #6 · No interchangeable focusing screens for 5D3


wickerprints wrote:
I already said that.

MF shooters using lenses that don't have electronic aperture control through the body already have to cope with metering issues, so they don't really care if a custom screen affects metering further. Those of us who do use EF lenses are already aware of the way in which focusing screens affect the metering without a custom function to compensate.

Lars Johnsson wrote:
That's not true. I have 9 non Canon MF lenses for my Canon bodies. And they all have electronic aperture control through the body.


Please read what I wrote again. I did not say that all MF lenses do not have aperture control through the body. I said MF shooters who use lenses that do not have aperture control through the body have to cope with metering issues.



Mar 03, 2012 at 12:34 AM
Lars Johnsson
Offline
• • • • • • •
Upload & Sell: Off
p.3 #7 · p.3 #7 · No interchangeable focusing screens for 5D3


Ok,I get your comment.

But a lot of MF users have electronic aperture control through the body. Including all the most popular Zeiss ZE lenses. And it will affect them. For me it doesn't matter because I use the 1 series bodies. But I think Canon at least could have 2-3 extra screens for the new camera



Mar 03, 2012 at 12:43 AM
te4o
Offline
• •
Upload & Sell: Off
p.3 #8 · p.3 #8 · No interchangeable focusing screens for 5D3


Imagine you shoot with a EF 85/1.2 or EF 50/1.2 on the 5D3 with AF well customized and working: you rely on the AF points/fields to get your subject tack sharp. If they say 5D3 = best non-1D AF, is it logical to say:
when AF works on AF lenses, AF confirmation will do the same on MF fast CZ primes?

I mean, imagine the whole branch of Carl Zeiss making ZE lenses as they heard "no M.focusing screen", the same with all other MF alt-brands. There must be a solution which we are not aware of yet, or Canon is really going mad - Apple cannot make MBPs without consulting Intel, Samsung, the myriad of co-producers, I can't believe Canon can leave the MF fast prime users out in the rain.
Probably the new pentaprism with all the overlays does a better job than the Eg-S on the 5D2? I admit I had it on and it improved MFing like 30%. Do you think the 1DX users go MF with 12-14 FPS? There must be something in the new VF BUT I am afraid there is nothing (probably my pessimistic nature...)



Mar 03, 2012 at 12:56 AM
te4o
Offline
• •
Upload & Sell: Off
p.3 #9 · p.3 #9 · No interchangeable focusing screens for 5D3


Lars, is there a 3.2" LCD viewer (Hoodman, Zakuto etc) available? This is the alternative without messing up with metering on a Katzeye / brightscreen... I can't see one yet but they are in the making I guess.

Edited on Mar 03, 2012 at 01:05 AM · View previous versions



Mar 03, 2012 at 12:59 AM
Dawei Ye
Offline
• • • •
Upload & Sell: Off
p.3 #10 · p.3 #10 · No interchangeable focusing screens for 5D3


Dawei Ye wrote:
How does the focus screen affect metering though? Isn't the focus screen only affecting what we see through the viewfinder?

wickerprints wrote:
The metering system is located in the pentaprism housing. It sees only the light that is reflected by the primary reflex mirror, and what it sees depends on how the focusing screen diffuses the light prior to entering the pentaprism. If, for example, you install a split prism screen, spot metering can become inaccurate.

A focusing screen is analogous to a translucent sheet of material upon which a slide projector image is projected. If you tried to project the image onto a sheet of smooth, clear glass, the light just passes right through and you see little if any image. If
...Show more

Ah thanks - interesting stuff!

Must mean that in theory a dirty mirror will affect metering too!



Mar 03, 2012 at 01:03 AM
melcat
Offline
• • • •
Upload & Sell: Off
p.3 #11 · p.3 #11 · No interchangeable focusing screens for 5D3


wickerprints wrote:
Metering in modern AF bodies is done via the primary mirror because it sees the entire image. The secondary AF mirror behind the primary is small and cannot see the image periphery. Therefore, to put the metering with the AF system would mean you can only meter based on the central portion of the image visible to the secondary mirror, rather than the entire frame. This has the potential to result in metering error, and this configuration does not lend itself to evaluative or center-weighted average metering.


At least on the 1DS Mark III the metering doesn't extend to the edges of the screen. See the diagram on p. 17 of the white paper. According to my tests, centre-weighted only covers the area marked by the oval in the viewfinder, which is also approximately the AF area. I was quite surprised to discover this, but went looking after getting some botched exposures and forming a theory as to why. Then I tested it. It is worse than the 5D, but that camera too doesn't meter across the whole frame. I don't know exacty where evaluative is looking, since it's harder to test and I never use it except for flash.

We shall see whether the new RGB meters are any better. Maybe that's actually what drove this change.



Mar 03, 2012 at 01:11 AM
Lars Johnsson
Offline
• • • • • • •
Upload & Sell: Off
p.3 #12 · p.3 #12 · No interchangeable focusing screens for 5D3


te4o wrote:
Imagine you shoot with a EF 85/1.2 or EF 50/1.2 on the 5D3 with AF well customized and working: you rely on the AF points/fields to get your subject tack sharp. If they say 5D3 = best non-1D AF, is it logical to say:
when AF works on AF lenses, AF confirmation will do the same on MF fast CZ primes?

I mean, imagine the whole branch of Carl Zeiss making ZE lenses as they heard "no M.focusing screen", the same with all other MF alt-brands. There must be a solution which we are not aware of yet, or Canon is
...Show more

I'm sure the AF confirmation will work good. But when you get the confirmation or beep, you have do some fine-tuning of the AF. No way you just can shoot after the beep and get good focus on most of your shots.



Mar 03, 2012 at 01:18 AM
Lars Johnsson
Offline
• • • • • • •
Upload & Sell: Off
p.3 #13 · p.3 #13 · No interchangeable focusing screens for 5D3


te4o wrote:
Lars, is there a 3.2" LCD viewer (Hoodman, Zakuto etc) available? This is the alternative without messing up with metering on a Katzeye / brightscreen... I can't see one yet but they are in the making I guess.


I don't know. But there will probably be one



Mar 03, 2012 at 01:19 AM
wickerprints
Offline
• • • •
Upload & Sell: Off
p.3 #14 · p.3 #14 · No interchangeable focusing screens for 5D3


Dawei Ye wrote:
Ah thanks - interesting stuff!

Must mean that in theory a dirty mirror will affect metering too!


You're most welcome In theory, yes, a dirty mirror will adversely affect metering, though it would have to be extremely dirty to do so.

Indeed, some camera owners who do not want to use AF at all, but would like a brighter viewfinder to facilitate the use of a different focusing screen, have proposed replacing the semi-silvered primary reflex mirror with a fully-silvered mirror. This would also adversely affect metering. In this case, the increased light reflection is constant so exposure compensation can be used to offset this error.

However, I believe that when the focusing screen is changed between standard and super-precision mattes, the change in exposure is in part dependent on the f-number of the lens, and so dialing in the same amount of EC may still leave some error. But overall, I think the metering accuracy issue with regard to focusing screens is a minor one, and the error is not really that huge.



Mar 03, 2012 at 01:25 AM
skibum5
Offline
• • • • • •
Upload & Sell: Off
p.3 #15 · p.3 #15 · No interchangeable focusing screens for 5D3


wickerprints wrote:
But overall, I think the metering accuracy issue with regard to focusing screens is a minor one, and the error is not really that huge.


i though that some screens tossed it off by a stop or more
especially with long lenses (?)



Mar 03, 2012 at 01:28 AM
15Bit
Offline
• • • • •
Upload & Sell: Off
p.3 #16 · p.3 #16 · No interchangeable focusing screens for 5D3


The lack of an S screen is a disappointment. It put me off buying a 7D, and i guess in the long run it will put me off the 5D3. They aren't losing a sale to me now though as i can't even remotely afford the 5Dmk3.

I wonder if i'm alone here in that i actually use MF with AF lenses - its the only way i can get selective focus in the right place on larger aperture lenses. And until they cover a lot more of the view with AF points i can't see that changing.



Mar 03, 2012 at 01:30 AM
wickerprints
Offline
• • • •
Upload & Sell: Off
p.3 #17 · p.3 #17 · No interchangeable focusing screens for 5D3


melcat wrote:
At least on the 1DS Mark III the metering doesn't extend to the edges of the screen. See the diagram on p. 17 of the white paper. According to my tests, centre-weighted only covers the area marked by the oval in the viewfinder, which is also approximately the AF area. I was quite surprised to discover this, but went looking after getting some botched exposures and forming a theory as to why. Then I tested it. It is worse than the 5D, but that camera too doesn't meter across the whole frame. I don't know exacty where evaluative is looking,
...Show more

In center-weighted metering, your statement is more or less accurate. However, the same white paper explicitly states that there are 63 metering zones, shown in cyan, yellow, orange, and red. The cyan zones are actual metering zones, and they are used in calculating exposure. Although none of the zones extend to the edge of the frame, many of these cyan zones extend well beyond the AF region. Evaluative metering will weight the exposure toward the active focus point but it takes into account the input from all 63 zones.

The takeaway remains the same: if you put the metering sensor with the AF sensor, you would entirely miss about the top 1/3 and bottom 1/3 of the image.



Mar 03, 2012 at 01:37 AM
melcat
Offline
• • • •
Upload & Sell: Off
p.3 #18 · p.3 #18 · No interchangeable focusing screens for 5D3


The -S screen seems to be an odd bird, in that, when fitted to the camera, it doesn't need to know the aperture of the lens to adjust for the exposure. At least with the lenses I've used on an adaptor, none faster than f/2, none with an EF chip, the only exposure error I see is due to fall-off in a lens, i.e. a hotspot. So it would be more accurate to say that the camera has to consult a table of exposure corrections when an -S is *not* fitted. I'd be fine with a screen requiring such a table now I've retired my last adapted lens.

Zeiss say to use a split-image screen and not believe the AF confirmation. I have never understood that advice. As others have said, the fast Canon lenses AF correctly. The AF confirmation agrees exactly with the -S screen and the final image for my one Zeiss and the 1DS k III. But I wouldn't try it with my 5D, whose AF points are offset down from where the screen marks them. Maybe that's what Zeiss mean.



Mar 03, 2012 at 01:45 AM
wickerprints
Offline
• • • •
Upload & Sell: Off
p.3 #19 · p.3 #19 · No interchangeable focusing screens for 5D3


I would guess that Zeiss recommends using a split prism because AF confirmation when focusing manually is subject to a certain degree of error due to the depth of focus precision of the AF sensor. Note that AF precision is achieved to within 1/3 the depth of focus for high-precision f/2.8 sensors, and within 1 depth of focus for f/5.6 sensors. Hence focus confirmation might be given at the boundary of this margin, leading to some inaccuracy. This doesn't tend to be an issue when the lens is driven by the AF system because by racking the lens through the depth of focus, it can see the boundary of the region better, and choose a point in the middle. But this is pure speculation as to why Zeiss might make such a recommendation.

That said, the use of split prism screens is itself fraught with issues, the major one being the accuracy of its position. The focusing screen may be very slightly different in thickness, or the camera's mirror box tolerances are not quite tight enough. Hence Canon makes available those thin brass shims to adjust the position of the focusing screen as needed. But once shimmed, it need not be redone unless a different screen is installed.



Mar 03, 2012 at 02:01 AM
saneproduction
Offline
• • •
Upload & Sell: Off
p.3 #20 · p.3 #20 · No interchangeable focusing screens for 5D3


wickerprints wrote:
The LCD overlay feature is only indirectly related to the focusing screen.

Canon's logic is this: the LCD overlay eliminates the need for a grid-type focusing screen. Most users will not require a super-precision matte screen. So why make the focusing screen interchangeable in future non-flagship bodies? Now, I'm not saying that this is correct thinking--I am merely pointing out the likely reasoning behind their decision.

For manual focus users, the focusing screen is an issue, but it's not insurmountable--after all, people have installed custom screens in the 7D.

For those, such as myself, who primarily rely on AF but tend to shoot
...Show more

Exactly why I use EG-S screen in my 5D II


...and I like to see if my AF is on target or not. I feel like I can see focus much better at 1.2-2.8 with the precision screen. This is a super big letdown for me. I don't like the 7D VF at all and won't consider 3rd party screens due to metering issues. Looks like I will need to wait around for a camera with superior AF and interchangable screens. Darnit!

Maybe I will get over my annoyance once the glowing reviews come out or it turns out RAW is 1 stop+ better.


Edited on Mar 03, 2012 at 01:49 PM · View previous versions



Mar 03, 2012 at 02:44 AM
1       2      
3
       4              6       7       end




FM Forums | Canon Forum | Join Upload & Sell

1       2      
3
       4              6       7       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.