Automatic Hyperfocal Menu Selection Options

[koenkooi]

Canon doesn't support magic lantern either. My point is nobody ever mentioned problems with magic lantern here, say "don't install it, it will bork your camera", which I take to mean it works rather well. And so could a simple scripting language.

If you check the Magic Lantern forum, you'll see enough people with cameras in a state they can't get out of. It usually revolves around properly reformatting a card and reinstalling ML, but a bridge too far for a lot of people. And this is for people savvy enough to get it manually installed once already.

I have ML on my original M and I like it very much, but you need have an affinity for technical things and the ability to follow instructions. I have problems with the latter, which makes ML an adventure

 

[dcm]

Just wait for a camera virus that holds all of your photos hostage until you pay a ransom…

 

[neuroanatomist]

Sometimes MacOS and iOS apps don’t work properly. While people should know to contact the developer, I bet quite a few of them contact Apple. I know some do, from conversations with ‘geniuses’ that I’ve overheard when picking up items at Apple Stores. Opening up a system to user-driven modifications is going to lead to increased support costs.

Developing and implementing a ‘simple scripting language’ will also incur costs. Making it secure enough to prevent unintentional poorly executed code might be possible, but that adds cost and makes the language less simple.

But what about intentional, malicious exploits? cameras are used in sensitive settings, e.g. journalists in hostile environments (where governments have the desire to control reporting), police and private investigations, etc. even devices touted as closed and secure (iPhones) can be tracked and hacked by nefarious governments. A device with sensitive information, Wi-Fi, GPS, and an open scripting language is a gaping security hole. Fighting the necessary ongoing battle against malicious exploits is neither cheap nor simple.

The fact that you think that this would be easy suggests you haven’t really thought through the ramifications.

As for your earlier argument:

I suspect such a feature was not added because not adding it allows Canon to make a profit on selling items such as a remote shutter release with a timer, rather than letting users write a script that instructs the camera to take a photo every X seconds.

Are you aware that Canon has included an interval timer on models starting with the 80D?

 

[stevelee]

I see a surprising amount of discussion here about autofocus. Maybe I’m just too quick to use manual focus rather than jumping through all sorts of hoops to get the camera to interpret what I want in focus. Even in chasing a bee around with a macro lens, I tend to move my self back and forth to focus rather than twisting a ring and changing magnification.

My main autofocus error I recall was two years ago, and that was operator stupidity, not a camera failure. I was shooting with my G7X II, whose little lens means depth of field is normally huge anyway, so I tended not to think about focus. I was riding in a large glass bubble to the top of a dome in Stockholm. There was light rain. In some shots the scene below was blurry, but rain drops on the glass were tack sharp. If my brain had been functioning, I would have focused manually on infinity except for the shots I did of people inside the ride with me.

 

[Antono Refa]

Sometimes MacOS and iOS apps don’t work properly. While people should know to contact the developer, I bet quite a few of them contact Apple. I know some do, from conversations with ‘geniuses’ that I’ve overheard when picking up items at Apple Stores. Opening up a system to user-driven modifications is going to lead to increased support costs.

Yes, having support answer "we don't support 3rd party scripts" will be massively expensive. Sure. Like developing two new cameras, half a dozen lenses, and three international ad campaigns.

Developing and implementing a ‘simple scripting language’ will also incur costs.

Of course it would.

Making it secure enough to prevent unintentional poorly executed code might be possible, but that adds cost and makes the language less simple.

On the contrary. A simple language would be easier to secure. E.g. you don't have to worry about the a script deleting files, if it can't access the memory cards, beyond the "shutter release" function causing the camera to take a photo and save it.

But what about intentional, malicious exploits?

A scripting language that can do even less than what the photographer can do through the available interface (buttons & menus), only automatically, would make it impossible to create intentional, malicious exploits.

The fact that you think that this would be easy suggests you haven’t really thought through the ramifications.

No, its you using a straw man argument (claiming the scripting language would have more features than I've suggested) for argument's sake.

 

[neuroanatomist]

I see a surprising amount of discussion here about autofocus. Maybe I’m just too quick to use manual focus rather than jumping through all sorts of hoops to get the camera to interpret what I want in focus.

people are lazy. I know I am, I can drive a manual transmission and my car has paddle shifters to simulate that, but it’s easier to let the car’s computer do the work for me (in that case, if I screwed up with the paddles the car‘s computer would take matters into his own ‘hands’ anyway).

I like autofocus! In the SLR days when manual focus was the only option, do you finders had a split prism to make it easier to achieve critical focus. In a DSLR viewfinder, that’s harder to do…especially with lenses faster than f/2.8 where the viewfinder DoF is deeper than the resulting image DoF.

However, with the EOS M and R cameras, I quite like the MF ‘fine tune’ feature where after AF lock if you turn the MF ring the viewfinder automatically switches to a 10x mag view at the AF point. I suspect that will be very powerful when combined with the Eye AF in the R3.

 

[stevelee]

I have not driven in the mountains since I got my current car. I think I would use the left paddle when going down a big hill, but maybe not. The computer applies brakes and power to different wheels as it thinks best, and that gives me the illusion of sports car roadability without my doing anything. It is a fun car to drive on curvy, hilly backroads, of which we have a few.

I use autofocus almost all the time. My point was mostly that I don’t really feel the need to complain about what features my camera lacks or miss them in that regard.

For the Jupiter-Saturn conjunction I used Live View and 10x magnification on the screen for manual focusing. Autofocus might have done fine, but there was no reason for me to find out.

 

[neuroanatomist]

Yes, having support answer "we don't support 3rd party scripts" will be massively expensive. Sure.

Support is not just call center labor. It’s also ensuring features work on new firmwares and new cameras. Sure that sounds trivial, but when the 1D X came out, I found a bug that borked AFMA settings only when the orientation-linked AF point option was enabled. Prior cameras had both features and no bug. A firmware update was needed to correct it. More features means more development and more support costs.

To the extent that those features help drive sales, including them makes sense.

A scripting language that can do even less than what the photographer can do through the available interface (buttons & menus), only automatically, would make it impossible to create intentional, malicious exploits.

Apologies, I overinterpreted your ask. Python and Java are scripting languages, and that’s what I thought you meant.

Sure, something that is restricted to only a limited subset of camera functions could be simple and secure. But to the original point of this topic, automatically setting a specific focus distance is not something a user can make the camera do currently (turning the manual focus ring notwithstanding). So your ‘solution’ isn’t one for the OP, it would require a more complicated language to do that.

Canon has recently implemented features like an intervalometer and focus bracketing. I question whether a very limited scripting capability would be of sufficiently broad appeal to matter.

Canon obviously knows about Magic Lantern and have chosen not to implement any sort of scripting capability.

 

[stevelee]

I think I might charge up my T3i battery and play with the ADEP tomorrow. I hadn’t even thought about it in the last decade. I will need to read the instructions first. I wonder if it would work with the 100-400mm lens that has never been on that camera.

 

[Antono Refa]

Support is not just call center labor. It’s also ensuring features work on new firmwares and new cameras. Sure that sounds trivial, but when the 1D X came out, I found a bug that borked AFMA settings only when the orientation-linked AF point option was enabled. Prior cameras had both features and no bug. A firmware update was needed to correct it. More features means more development and more support costs.

The scripting language would require just a front end, as all the functionality behind is already implemented in hardware. It would have a cost, but not that high.

To the extent that those features help drive sales, including them makes sense.

Of course.

Apologies, I overinterpreted your ask. Python and Java are scripting languages, and that’s what I thought you meant.

I have nothing as complicated in mind.

Sure, something that is restricted to only a limited subset of camera functions could be simple and secure. But to the original point of this topic, automatically setting a specific focus distance is not something a user can make the camera do currently (turning the manual focus ring notwithstanding). So your ‘solution’ isn’t one for the OP, it would require a more complicated language to do that.

Canon has recently implemented features like an intervalometer and focus bracketing.

To the best of my understanding, there's no UI for it, but the camera can tell the lens how far away to focus. Otherwise, how could it focus bracket?

I question whether a very limited scripting capability would be of sufficiently broad appeal to matter.

If magic lantern invested the effort of making it, there's some demand.

 

[neuroanatomist]

To the best of my understanding, there's no UI for it, but the camera can tell the lens how far away to focus. Otherwise, how could it focus bracket?

Focus bracketing starts from the initial focus acquisition point and moves in essentially undefined steps toward infinity. The user specifies a number of steps and a nonspecific increment (fine to coarse on a 1-10 scale). So the firmware is using some internal algorithm to determine that. But there’s no need for a command like ‘set focus distance to 2.4 m’ for focus bracketing.

If magic lantern invested the effort of making it, there's some demand.

Magic Lantern was initially for video. The development of ML has not progressed beyond support for the 5DIII and 60D (the rationale that the R series uses a modified PowerShot firmware doesn’t apply to the 5DIV). The fact that no camera since 2017 can run ML suggests there’s no longer a meaningful demand for it.

 

[Antono Refa]

The fact that no camera since 2017 can run ML suggests there’s no longer a meaningful demand for it.

AFAIK, its a technical problem (Canon changing HW & OS making it hardware to replace firmware), not a matter of motivation.

 

[stevelee]

I think I might charge up my T3i battery and play with the ADEP tomorrow. I hadn’t even thought about it in the last decade. I will need to read the instructions first. I wonder if it would work with the 100-400mm lens that has never been on that camera.

I didn’t charge the battery, but I did read up on the feature. It is a scaled down version of a feature from some earlier cameras. I was shocked to find that the T3i has just 9 autofocus points. I don’t recall having particular focusing problems, and that configuration sounds so primitive. It turns out that all ADEP does is to see how many of the spots covered by those 9 it can get in focus, and lights them up in red. Then for the shot, it stops down the lens as needed. The rest of the settings may or may not make sense. I can see why it didn’t catch on and why I forgot about the feature.

These days when lenses might not even have distance scales it is harder to implement hyperfocal distances any better than just eyeballing it.

 

[koenkooi]

I didn’t charge the battery, but I did read up on the feature. It is a scaled down version of a feature from some earlier cameras. I was shocked to find that the T3i has just 9 autofocus points. I don’t recall having particular focusing problems, and that configuration sounds so primitive. It turns out that all ADEP does is to see how many of the spots covered by those 9 it can get in focus, and lights them up in red. Then for the shot, it stops down the lens as needed. The rest of the settings may or may not make sense. I can see why it didn’t catch on and why I forgot about the feature.

These days when lenses might not even have distance scales it is harder to implement hyperfocal distances any better than just eyeballing it.

I have the distance bar enabled for RF lenses, that has been a great help. It even has a white bar that shows the useable range at that focal length for zooms. I'm consistently 10cm too close when using the 100-500 at its minimum focussing distance, the scale helps me to fine tune it after taking a step back.

I wish it would also work on EF lenses, but alas.