20D - improved Katz Eye focsuing screen - Part 1/1

Started Aug 13, 2005 | Discussions thread
Flat view
Doug Kerr Forum Pro • Posts: 20,899
20D - improved Katz Eye focsuing screen - Part 1/1

I have recently installed in my EOS 20D a new type of split-image prism focusing screen, developed by Katz Eye Optics, the Katz Eye Plus focusing screen.

According to Katz Eye Optics, this new screen employs a unique design of the prism set, which allows the full prism to remain visible even at rather small metering apertures. As you will see from the test results below, this promise seems to be fulfilled.

The screen I installed also incorporates their HiLux treatment, intended to provide enhanced brightness to the basic part of the focusing screen.

I recently conducted a basic series of tests on the performance of this new focusing screen. The results are very positive. Here is a summary of my observations. Note that these tests were made without benefit of repeated trials and thus the results may be subject to more experimental error than otherwise.

Your mileage may vary.


The screen incorporates a central split-image prism focusing aid with a horizontal split. Its diameter is precisely that of the partial metering indicator circle in the 20D finder.

The split prim is surrounded by a microprism field ring (sometimes called a "microprism collar"). Its radial thickness is about 25% of the radius of the Partial metering circle.

The screen also provides a "composition circle", which just encircles the left and right outermost AF point indicators.

The screen has the overall mechanical features and dimensions of the OEM screen, with regard to the various steps, tabs, and so forth. The manufacturer recommends that the existing screen position adjustment shim be retained when the screen is installed.

The procedure I recently published for the installation of focusing screens in the 20D seems to work well with this screen.


At issue here is the matter of the "blackout" of the split prism array as the viewing aperture decreases. This is not as simple a matter to describe as it sounds. For one thing, whether both halves of the prism can be seen with usable brightness, at some particular viewing aperture, depends on the exact position of the viewer's eye.

Additionally, with this screen (perhaps more so than with other screens) the onset of darkening is fairly gradual. Because of this, taken in conjunction with the dependence on eye position, it is difficult to establish a simple definition as to "beyond which aperture is the prism no longer usably visible". I have, however, established a way to describe, subjectively, the performance of the screen in this regard.

A related, and complicating, issue is a matter of the sensitivity of the split prism arrangement to focus state. By that I mean how much misalignment of scene lines crossing the boundary between the two halves of the prism occurs for a certain degree of misfocus. I find that in the common "distant subject" situation, the screen will allow both prism halves to be readily seen, without requiring undue attention to eye position, for an aperture so small that the sensitivity of the prism is no longer great enough to provide easy determination of correct focus.

I do not believe that this is a result of the sensitivity declining more rapidly with this screen then for other types. Rather, it appears to be a result of the fact that we can clearly see the prism at smaller apertures than with other screens - apertures at which the prism feature itself is inherently less usable.

I do not at this point have any criterion through which I can characterize the matter of what is usable sensitivity.


Here is a brief summary of my observations regarding visibility of the prism arrangement at different viewing apertures. The tests were conducted with a Canon EF 50 mm f/1.4 lens. Brief tests with other lenses indicate that this aspect of performance is not substantially effected by the focal length of the lens, nor the lens model, so long we are speaking of a "distant subject" context.

In reporting my findings, I speak in terms of performance levels predicated on four different imperatives of attention to eye position. In each case, the criterion is that both halves of the prism can be seen with "usable brightness". Here are the performance levels:

Level 1: Little or no attention required to eye position
Level 2: Minimal attention required
Level 3: Moderate attention required
Level 4: Careful attention required

Level 5: Both halves of the prism cannot be seen simultaneously with usable brightness regardless of eye position

Note that even a "level 4" rating means that visibility of both halves of the prism can be attained.

Here are the results:

Level 1: Down to f/4.5
Level 2: Down to f/9.0
Level 3: Down to f/14
Level 4: Still in force at f/22, the smallest aperture at which I could test.
Level 5: Did not occur for any aperture at which I could test.

[continued in part 2]

Flat view
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow