EVF lag times: E-M5 sets new record

Started Oct 28, 2012 | Discussions thread
Flat view
Anders W Forum Pro • Posts: 21,466
EVF lag times: E-M5 sets new record

In a recent thread about the future of mirrorless versus SLR technology


it was argued (especially by ljfinger) that the EVF lag time, i.e., the time between a real-world event and its appearance in the EVF or on the rear display, constitutes a major drawback of mirrorless cameras that will never make them competitive with OVF-based cameras like DSLRs.

As to the magnitude of this lag on typical EVFs, it was claimed that it amounts to no less than 1/10 s at a minimum and could be as high as 1/3 s.


While this might possibly be true for typical EVFs (I have no way of judging that), it prompted me to try to test where we actually stand as far as the newest and best mirrorless cameras are concerned, one of which being the E-M5. The E-M5 can refresh the EVF/OLED at 120 Hz (if you set frame rate to high via the menu system) as compared to max 60 Hz on prior MFT cameras. Presumably, it also has higher processing capacity, potentially shortening the time from readout to actual display.

What I had already been able to see, by simply moving my hand back and forth in front of the camera, was that there was no longer any perceptible lag even in pretty low light (e.g., a situation where proper exposure would be f/2.0 and 1/30 at ISO 6400). Nevertheless, some lag must of course remain and I was curious enough to try to find out exactly how short it would be.

So I found an online stopwatch, more specifically this,


zoomed in to 1000 percent to make the figures nice and big, put the camera in front of the computer screen, turned on the OLED, and used my old G1 to take a series of shots simultaneously capturing the time shown on the computer display and the time shown on the camera OLED. The difference between the two times of course indicates the magnitude of the lag.

For a variety of reasons, the lag thus observed is not constant but varies somewhat. This may be due to real variations in the lag period as well as to the rather crude and simple method I used. For example, on a millisecond scale, the computer screen is of course unable to show as many as one thousand distinct values during any single second. It will manage to show only some of them. Nevertheless, the average difference between the figures shown by the computer display and those shown on the OLED computed across a sample of reasonable size should give us a decent estimate of the magnitude of the lag.

The average I could compute based on a sample of 25 shots was even better than I would have guessed: 25 ms or 1/40 s. If the claim I linked to above is correct, that's only a quarter of the 100 ms or 1/10 s lag time that typical EVFs can manage at best.

For the record, I should mention that the lens I used on the E-M5 was the 45/1.8 and that the light level according to the camera meter was such that f/2.0 at 1/40 s would give correct exposure at ISO 200 according to the camera meter (evaluative metering). I don't know at this stage whether the lag is even shorter under better light conditions.

I should also mention that the data points underlying the average I report were essentially three (presumably due to the step-wise way in which the clock on the computer display is updated). In nine cases, the difference was zero, in ten cases it was 33 or 34 ms, and in six cases it was 50 or 51 ms.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
Flat view
Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow