Which split image focus screen to get?

bullet1

Veteran Member
Messages
7,345
Reaction score
274
Location
US
I have read through many posts here (did lots of searches) and other places about two leading (maybe the only two) after market focus screen with the split image circle support like we had with the manual focus cameras (ex. Canon A-1).

I am ready to replace the focus screen in my 20D and keep the 350D untouched as doing so would loose the focus point indicators.

From what I gather, the Katz Eye™ “Plus” Split Prism w/Microprism Collar does not dim the circle until f/11 which is a real plus for my 100-400 with or without the Tamron 1.4x TC on. But the evaluative metering could be off by 2 stops with the same 100-400 at 400 f/5.6! CWA is recommended for this screen.

The Haoda focus screen, on the other hand, does not introduce the metering error but it darkens the split image circle at f/5.6 or higher.

I read the AF modes, One Shot, AI Focus and AI Servo are not affected by either screen. Is this a correct assumption?

I also read that the focus indicators will show but not all of them. If I enable all 9 AF points. which ones will be affected? Will all of them still function?

I know the Katz Eye is a bit more expensive at $105. The Haoda can be bought at $72. I am willing to pay a bit more if one is going to work better for me. My slowest lens is 100-400 f/4.5-5.6. I am also considering getting the 17-85 f/4-5.6 for traveling.

Which one would you recommend and why?

Which one is easier to install?

Thank you for your input!
--
Nelson
http://pbase.com/nelsonc

 
I read the AF modes, One Shot, AI Focus and AI Servo are not
affected by either screen. Is this a correct assumption?
That should be correct. The AF sensors are in the opposite direction from the focus screen, and there's no reason the focus screen should affect them.
I also read that the focus indicators will show but not all of
them. If I enable all 9 AF points. which ones will be affected?
Will all of them still function?
All AF points will still function for autofocus - as noted above, they're in the bottom of the mirror box far away from the focus screen. On some cameras (DReb line) you won't have the AF point "box outlines" any more because they're marked on the screen that you took out. On the 20D, they aren't on the screen (as I understand it), so you'd still have the boxes. The red LEDs will not appear, at least not the ones that would appear in the etched area. That supposedly can be at least partially overcome by making pencil dots at the appropriate spots on the etched part of the screen.

I don't have either screen, and I haven't tried this. I'm just reporting what I've read.
 
If I manually select one of 9 AF points, will it light up?
I have the Haoda screen installed in my 20D. It has no effect on the AF points lighting up. They work the same as before. That's because in the 20D, the AF point indicators reside on a seperate screen from the removable focusing screen.
 
I went with the Katz Eye with Plus. It makes using some slow lenses and extension possible.

The best part is the split ring. If forgot how much I missed the ring from my manual focus days. Even when you shoot with AF it is nice to have the split ring as a confirmation that the shot really is in focus.

jerry

--
jerryk.smugmug.com
 
I went with the Katz Eye with Plus. It makes using some slow
lenses and extension possible.
Was the installation straightforward? How long did it take to install it?

Do you find any metering inconsistencies?

Do you find any AI Servo inconsistencies?

How long have you had it?
The best part is the split ring. If forgot how much I missed the
ring from my manual focus days. Even when you shoot with AF it is
nice to have the split ring as a confirmation that the shot really
is in focus.
I am looking forward to get the split hair image circle back. That was the first shock I had when I bought the first DSLR. My previous SLRs are Canon A-1 and AE-1, both with excellent split image manual focus aid.

Thank you for sharing.
--
Nelson
http://pbase.com/nelsonc

 
If I manually select one of 9 AF points, will it light up?
I have the Haoda screen installed in my 20D. It has no effect on
the AF points lighting up. They work the same as before. That's
because in the 20D, the AF point indicators reside on a seperate
screen from the removable focusing screen.
That is good to know. Thank you for the info.

The drawback with the Haoda screen is the darkening of the split image circle when used with a slower lens such as 100-400 zoom to 400 with or without a TC, or the popular 17-85 IS lens zoomed to 85.

Do you have any slower lenses at f/5.6? Do you find the darkening of the split image circle a problem for MF (can't see the split line) or AF (can't see the center AF light)?
--
Nelson
http://pbase.com/nelsonc

 
Do you have any slower lenses at f/5.6? Do you find the darkening
of the split image circle a problem for MF (can't see the split
line) or AF (can't see the center AF light)?
Regarding the Haoda screen: I just tried it out by using the DOF preview button to stop the lens down while looking at the split screen. It's good up to f/5.6. At f/6.3 and smaller, half of the split screen circle starts getting really dark. I noticed that, depending on which direction I pointed the camera in relative to my room window, the half-circle either goes from mildly dark (still useable) to completely black (completely unusable) at f/6.3. When I'm pointing it in a direction that gives me a mildly dark half-cycle, I can stop down further to f/8 and still get a usable half-circle. But if you're pointing it somewhere else, it'll be very, very dark or even completely black at f/6.3

So if Catz-Eye's claims are true, I'd recommend getting their screen if you're going to need the split screen at smaller apertures.
 
did u find any inconsistencies?

like u'd have the split prism perfectly alligned for focus and either the focus confirmation light didn't turn on or the resulting image was OOF?

if not, don't u think changing the screen was a waste of time cuz your camera was giving accurate AF/MF in the first place?

--
I see dead pixels
 
did u find any inconsistencies?

like u'd have the split prism perfectly alligned for focus and
either the focus confirmation light didn't turn on or the resulting
image was OOF?

if not, don't u think changing the screen was a waste of time cuz
your camera was giving accurate AF/MF in the first place?
I haven't found any inconsistencies. It's a good tool to have. AF isn't always perfect. And frankly, manual focus isn't always perfect either (ie. when you're in a hurry). But between the two, they can complement each other nicely. I got it because it's just nice to have that extra bit of visual autofocus confirmation precision that the plain focus screen and the focus confirmation light lack (both are pretty imprecise). Plus, I have a couple manual focus lenses (24mm TS-E and 45mm TS-E) that I use it with.
 
where the AF has locked in but the center split screen showed minor adjustments were still needed? I'm thinking with FTM lenses this might be a useful tool to have ..

--
I see dead pixels
 
Hi, Nelson,
I have read through many posts here (did lots of searches) and
other places about two leading (maybe the only two) after market
focus screen with the split image circle support like we had with
the manual focus cameras (ex. Canon A-1).

I am ready to replace the focus screen in my 20D and keep the 350D
untouched as doing so would loose the focus point indicators.

From what I gather, the Katz Eye™ “Plus” Split Prism w/Microprism
Collar does not dim the circle until f/11 which is a real plus for
my 100-400 with or without the Tamron 1.4x TC on.
Indeed.
But the
evaluative metering could be off by 2 stops with the same 100-400
at 400 f/5.6! CWA is recommended for this screen.
Indeed.
The Haoda focus screen, on the other hand, does not introduce the
metering error . .
Not necessarily so. Almost inevitably, the expsure metering sensitivity for metering zones near the center is affected by the presence of a slit image prism feature of any kind. However, I think that the Haoda screen does not introduce as much metering shift with respect to the other metering zones as does teh Katz Eye with the extra bright feature (whataver that is called these days).
but it darkens the split image circle at f/5.6 or
higher.
I would expect.
I read the AF modes, One Shot, AI Focus and AI Servo are not
affected by either screen. Is this a correct assumption?
Bascailly yes. The focusing screen is not involved in the AF process at all.
I also read that the focus indicators will show but not all of
them. If I enable all 9 AF points. which ones will be affected?
Will all of them still function?
I don't know what you mean "show". On a 20D, the focusing point indicators seen in the finder are unaffected by the use of an alternative focsuing screen.

Note that the screen has no effect on AF, for any of the AF points.

What you may be referring to is that if one preselects the center AF point, then evaluative exposure metering will "emphasize" the metering results from metering detectors located at the center of the frame, and the use of any screen with a split prism will disturb the calibration of those metering detectors. Thus if the center AF point is preselected, or in fact if AF point selection is set to "Automatic" ("all AF points"). there will be a metering shift. Thus the only relatively-unaffected metering modes are:
  • Evaulative with an AF point preselected, but not the center one
  • CWA
I know the Katz Eye is a bit more expensive at $105. The Haoda can
be bought at $72. I am willing to pay a bit more if one is going
to work better for me. My slowest lens is 100-400 f/4.5-5.6. I am
also considering getting the 17-85 f/4-5.6 for traveling.

Which one would you recommend and why?
I have only used the Katz Eye and had good results with it. I have heard from many members who have used the Haoda screens with good results.
Which one is easier to install?
I think the basic installation process is about the same for either.

Best regards,

Doug

Visit The Pumpkin, a library of my technical articles on photography, optics, and other topics:

http://doug.kerr.home.att.net/pumpkin

'Make everything as simple as possible, but no simpler.'
 
The Haoda focus screen, on the other hand, does not introduce the
metering error . .
Not necessarily so. Almost inevitably, the expsure metering
sensitivity for metering zones near the center is affected by the
presence of a slit image prism feature of any kind. However, I
think that the Haoda screen does not introduce as much metering
shift with respect to the other metering zones as does teh Katz Eye
with the extra bright feature (whataver that is called these days).
I am somewhat puzzled by this metering issue. Does the TTL metering actually go through the focus screen? For some reason, I thought the focus screen sits underneath the prism for the viewfinder use only.
Thus the only relatively-unaffected metering modes are:
  • Evaulative with an AF point preselected, but not the center one
  • CWA
I was wondering about whether the off center AF points are affected by the metering shift. I frequently enable the top AF point or one on the side when I shoot a portrait (although sometimes I do focus and recompose with the center AF point). The Evaluative metering could then be used with an off center AF point with little or no metering shift, correct
Which one is easier to install?
I think the basic installation process is about the same for either.
With your excellent installation guide and higher usability with the slower lenses, I am inclined at this point to order the Katz Eye Plus screen. But before I do, I wan to verify the f/5.6 and f/8 darkening effect on my old trusty Canon A-1 first.

Have you experience any metering shift with a slower lens at f/5.6 or f/8?

Thank you!
--
Nelson
http://pbase.com/nelsonc

 
I am somewhat puzzled by this metering issue. Does the TTL
metering actually go through the focus screen? For some reason, I
thought the focus screen sits underneath the prism for the
viewfinder use only.
Here's a diagram of the innards of a D30:
http://www.imaging-resource.com/PRODS/D30/D30A3.HTM

I believe that the 20D is the same basic configuration. The TTL sensor is located just under the hotshoe and looks at the focus screen through the pentamirror just like your eye does.

Here's a cutaway diagram of the 20D:
http://www.dpreview.com/reviews/canoneos20d/page2.asp
 
where the AF has locked in but the center split screen showed minor
adjustments were still needed? I'm thinking with FTM lenses this
might be a useful tool to have ..
Yes, sometimes there isn't enough contrast or detail for the AF to get a good read on, so the autofocus can be a bit off. The split screen allows you to see this and fine tune it accordingly. This is when FTM comes in handy because you just grab the focus ring and turn it however much you need to. You'll need to keep the shutter button half pressed to make sure the autofocus doesn't kick in. Or you if you're using Custom Function 4, where the AF activation is on the ( ) button instead of the shutter button, you don't have to worry about the AF kicking in when you hit the shutter button again.
 
Hi, Bullet,
I am somewhat puzzled by this metering issue. Does the TTL
metering actually go through the focus screen? For some reason, I
thought the focus screen sits underneath the prism for the
viewfinder use only.
The metering detectors regard the top surface of the focsuing screen, and from an angle. (They can't regard it from "directly overhead", or the detectors themselves would obstruct the view of the screen through the finder eyepiece). Anything that changes the angular distribution of the light emitted from the top of the focsuing screen will affect the reletive sensitivity of the metering detectors.
I was wondering about whether the off center AF points are affected
by the metering shift. I frequently enable the top AF point or one
on the side when I shoot a portrait (although sometimes I do focus
and recompose with the center AF point). The Evaluative metering
could then be used with an off center AF point with little or no
metering shift, correct
First a little sermon: Please note that we are not speaking of AF points as such - we are speaking of the metering detectors for the scene areas surrounding the AF points (I know that is probably what you mean, but sometime people are misled by the reference to "the AF points" in connection with metering. Note that there are 35 metering detectors, not all of them "near" any AF point.

The affect of an alternative focsuing screen on the sensitivity of the metering detectors that are not at the center of the frame comes from differences in the angular distribution of the light from the top of the screen. The screens that are "specially brighter" have a distribution that is more concentrated in the direction toward teh eyepiece. Thus there is less of the light emitted toward the metering detectors (recall that they regard the top of the screen screen from "off to the side").
Have you experience any metering shift with a slower lens at f/5.6
or f/8?
I wouild have to check my original test notes on that. I don't exactly recall my findings. (I'm an old guy, and, as Carla says, "I've slept since then"!)

Best regards,

Doug
 
I think one difference is in the AF system. In the D30, the path from the secondary miirror goes, via a third miror, forward to the AF detectors. I think in the 20D, the third mirror folds the path back (so the AF detectors are toward the rear).

I think in the 5D, it is different yet - I don't remember what I concluded when I compared the 20D and 5D phantom drawings. I think in the 5D the AF detectors are in about the same plane as the main sensor.

I think this evolution is all to simplify manufacture.

Best regards,

Doug
 

Keyboard shortcuts

Back
Top