I find split-image focusing (like in the old SLRs) so much faster then anything available to me on my A7r3. I've tried MF Assist and Peaking, but neither of those are as easy as using an a split-image or microprism.
I'm wondering if its possible to incorporate a split-image or microprism into a mirrorless camera?
jeffbuzz wrote:
Fujifilm has always offered real time "digital split image" simulation in their viewfinders. Goes back to the X100s I think.
But many Fuji users don't seem to use it that much.
I think a picture in picture magnification would be the best.
Basically if you take a picture on a Sony body and then press Play on it and magnify in, then you will see a small picture in picture in the left bottom corner of the full scene and also a small box on the full scene image which shows which part of the full image that is magnified. Now imagine that it would work like this as well in Live View when shooting, that is how I would like it to work.
I have the x100f. It tries to implement this, but I don't think its actually useful
With the optical view finder, it pops up this tiny little electronic square in the corner of your view and attempts to provide you a zoomed-in view of the center. In this little square, it does some split-image focusing. The view is small, grainy, dark, and laggy, though. So its not very usable. You pretty much have to switch to the electronic view finder for manual focusing.
In the electronic view finder, the camera takes the center of the of the image and does the same thing. It is larger and brighter, but still grainy and laggy. Its an upgrade from just eyeballing, though.
The biggest issue with the x100 manual focusing is that the lens is focus-by-wire. So you get lag from the lens as well.
The A7's EVF seem to be brighter and higher resolution. Also, there wouldn't be any lag from true MF lenses. So I think that some of the things that Fuji is trying to do, could be implemented even better. It would just be software updates, too. I'd really like to see this work someday because split-image focusing on an SLR is such a pleasant experience.
The lag may be minimal but I would think there is one....
Isn't everything in an EVF digitally processed? I don't see that there would be any more lag versus any other mode of the viewfinder. Given the slow pace of most manual focus work I can't say I ever noticed any lag. It's not like someone would likely be using these methods to capture action. I rarely used them.
agvogel wrote:
I've never found it usable. Highlight peaking does a better job
Agreed, it seemed rather limited to me. Good for scenes with strong vertical line contrast but not much else. I liked the "digital microprism" feature a bit more. Simple highlight peaking works better for me too.
lattesweden wrote:
But many Fuji users don't seem to use it that much.
I think a picture in picture magnification would be the best.
Basically if you take a picture on a Sony body and then press Play on it and magnify in, then you will see a small picture in picture in the left bottom corner of the full scene and also a small box on the full scene image which shows which part of the full image that is magnified. Now imagine that it would work like this as well in Live View when shooting, that is how I would like it to work....Show more →
Fuji has that as well. I have a hard enough time seeing all the EVF without making the displays even smaller. I didn't find the dual display to be helpful but your mileage may vary.
That split-prism has to be in the optical path, but mirrorless only has the sensor plane. I can't envision how that would work, except at a literal single pixel level.
I agree as well, of limited use. But I also don't see the purpose in using MF with the X100 series.
I think a thoughtful implementation would be most useful in an MILC...I'd love something better or faster that I could use in full screen mode with my MF lenses.
jeffbuzz wrote:
Agreed, it seemed rather limited to me. Good for scenes with strong vertical line contrast but not much else. I liked the "digital microprism" feature a bit more. Simple highlight peaking works better for me too.
jeffbuzz wrote:
Isn't everything in an EVF digitally processed? I don't see that there would be any more lag versus any other mode of the viewfinder. Given the slow pace of most manual focus work I can't say I ever noticed any lag. It's not like someone would likely be using these methods to capture action. I rarely used them.
No - you are generally looking at a straight feed off the sensor.
A "split image" would require some degree of interpretation and therefore (I would think) some degree of lag.
Agree that it would not be a big issue in most MF cases today.
Back in the day it was all we had outside of fresnel/matte focusing screens and therefore; the fastest focus method..
Imo an excellent digital implementation for a manual focus aid is Canon's - gives you the same information as the split image, but in a digital interface, more flexibel where you place it and less obtrusive. Only works with lenses that electronically communicate with the camera though.
DaveFP wrote:
No - you are generally looking at a straight feed off the sensor.
Here's how I thought it worked:
The EVF or LCD is displaying a JPG.
It can't be a straight feed as the 24, 36, 42, 50 or 61 megapixels of the sensor won't fit in the 1 - 3 megapixel EVF or LCD.
Remember, the A1s EVF has 9.9 million dots. 3.3 million red, 3.3 million green and 3.3 million blue.
So using the same terminology used for the sensor, the EVF is 3.3MP.
The A1 LCD has only 1.44 million dots, or 0.48MP.
Also, the sensor outputs RAW data. This needs to be processed before it gets to the EVF or LCD.
Evidence of this is picture styles affect the display. eg, select Monochrome and a monochrome image is displayed in EVF or LCD. I set saturation and contrast to -1 as I prefer the EVF that way.
So processing includes:
RAW conversion to JPG
Picture style setting
Downsizing
It can't be a straight feed as the 24, 36, 42, 50 or 61 megapixels of the sensor won't fit in the 1 - 3 megapixel EVF or LCD.
Remember, the A1s EVF has 9.9 million dots. 3.3 million red, 3.3 million green and 3.3 million blue.
So using the same terminology used for the sensor, the EVF is 3.3MP.
The A1 LCD has only 1.44 million dots, or 0.48MP.
Also, the sensor outputs RAW data. This needs to be processed before it gets to the EVF or LCD.
Evidence of this is picture styles affect the display. eg, select Monochrome and a monochrome image is displayed in EVF or LCD. I set saturation and contrast to -1 as I prefer the EVF that way.
So processing includes:
RAW conversion to JPG
Picture style setting
Downsizing ...Show more →
Regardless what kind of processing is or isn’t required EVF lag is typically a two-digit millisecond figure (if I remember correctly around 25ms typically) whereas human reaction time on average is ca 250ms. Likely longer than that for older males, which are the typical photo geek demographic..
Any lag introduced by the EVF is unlikely to be a limiting factor here.
fjablo wrote:
Regardless what kind of processing is or isn’t required EVF lag is typically a two-digit millisecond figure (if I remember correctly around 25ms typically) whereas human reaction time on average is ca 250ms. Likely longer than that for older males, which are the typical photo geek demographic..
Any lag introduced by the EVF is unlikely to be a limiting factor here.
I was not talking about lag, but human reaction time of 250ms is way oversimplified.
As Tyler Durden was very aware.
Put a small scratch on 1 frame of film, displayed at 24fps and even the young girl's dad saw the scratch. Well, it was not a scratch, and reportedly it was big and juicy, but a scratch will suffice for this conversation.
Googling Tyler Durden, I just proved my own point. I watched Fight Club about a week ago and I did notice at least 2 of the 4 times a single frame of him was inserted into the movie before he was introduced. I did not think much of it at the time, but "tyler durden 1 frame" google let me in on the story. They did not cut to a different scene either, he just appeared in the scene for a single frame.
It’s not a question of what you notice but whether the introduced lag (that you may or may not notice) impacts how quickly you adjusted focus. And whether that’s a total of 250ms or 275ms is largely irrelevant. And esp whether a digital split screen takes that to 280ms or not doesn’t matter at all.
And yes that’s simplified. So is your description of the EVF feed above
