Reading mode: Light Dark

The point of Lytro

Go to last post
NetMage said:
Actually, they have an exact idea of where they are focusing - a phase detect sensor measures focus distance directly, and a contrast system knows it indirectly from the lens information.

Depending on the camera, however, they only have to be reasonably within DOF to call a picture focused.
Click to expand...
The problem is to decide why you bracket.

The normal reason is to get better focus, by e.g. trying three or five different at some + and - steps.

But ... we were talking about replacing lytro. Then you want to take samples of focussing over a bigger range. How many samples? How far apart? Shall the different samples ne taken at linear distances or linear movements of the lens? Lots of choices that have to be specified or fixed. Not easy.

I think it would be rather hard for the interface engineers to decide how to specify all different wants from users.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
NetMage said:
Actually, they have an exact idea of where they are focusing - a phase detect sensor measures focus distance directly, and a contrast system knows it indirectly from the lens information.
Click to expand...
Actually phase detect sensors can't measure the exact focus distance, only whether the current point is in front of, or behind, the current focus. The camera knows which way to turn the focus motor, not what distance the subject is.

Now being able to tell what the focus distance is relies on the lens knowing. For simple micro-motor focusing lenses all they know is how to move back and forth, so can't give an accurate focal distance readout. For stepper motors or lenses with floating focus elements (eg. Panasonic m4/3 zooms) they could tell the body exactly what distance the focus point is.

TBH the reason DSLRs don't generally have focus bracketing is probably more to do with the fact that any large lens is slow to reverse the direction of focus because it's got to shift large lens elements around. This is why PDAF is so much faster than CDAF on these lenses, because PDAF knows which direction to move while CDAF has to rack back and forth to find the focus point.
 
Roland Karlsson said:
Le Kilt said:
It is dead simple to take a shot, adjust focus forward, take a shot, adjust focus backwards, take a shot. The only question is by how much to adjust, which can be left to the photographer to specify.
Click to expand...
Hmmmm ... that was what I meant. "for the photographer to specify" - how?
Click to expand...
On the Pro1 there are three settings for the amount of bracketing, it doesn't give a precise quantity, just a smaller or larger 'amount'. It works quite well.
Roland Karlsson said:
NOTE that focussing series for Lytro simulation is NOT focus bracketing.
Roland Karlsson said:
The down side for Canon would be if it reveals that many lenses are not focusing precisely.
Click to expand...
--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
Click to expand...
 
Roland Karlsson said:
NetMage said:
Actually, they have an exact idea of where they are focusing - a phase detect sensor measures focus distance directly, and a contrast system knows it indirectly from the lens information.

Depending on the camera, however, they only have to be reasonably within DOF to call a picture focused.
Click to expand...
The problem is to decide why you bracket.

The normal reason is to get better focus, by e.g. trying three or five different at some + and - steps.

But ... we were talking about replacing lytro. Then you want to take samples of focussing over a bigger range. How many samples? How far apart? Shall the different samples ne taken at linear distances or linear movements of the lens? Lots of choices that have to be specified or fixed. Not easy.

I think it would be rather hard for the interface engineers to decide how to specify all different wants from users.
Click to expand...
They would have to integrate a DOF calculator so that all distances are catered for, and that is just some tables and calculations. A sophisticated version could allow all sorts of choices including aperture, shutter speed, max number of shots...

Obviously, in some cases it would mean taking a large number of shots (eg with 85mm f/1.2!), but why not?
 
Le Kilt said:
They would have to integrate a DOF calculator so that all distances are catered for, and that is just some tables and calculations. A sophisticated version could allow all sorts of choices including aperture, shutter speed, max number of shots...
Click to expand...
Exactly my point!

Do you have ANY other functionality in your camera that is so complex?
Le Kilt said:
Obviously, in some cases it would mean taking a large number of shots (eg with 85mm f/1.2!), but why not?
Click to expand...
If the camera had similar functionality to a smart phone where you can run applets I would 100% agree with you. But ... as long as the user interface is based only upon settings in menus ... then focus bracketing of this complexity is outside the scope of a camera.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
NetMage said:
Actually it doesn't quite work that way.
Click to expand...
OK?
NetMage said:
It is more like the microlenses combine sensor pixels into NxN arrays and then they record images from multiple POV but at the same focal length and aperture. Then a lot of processing let's you compute effective DOF for aperture x N (so for the 35mm equivalent (guestimated) f11, you can compute f44)).
Click to expand...
I think that was more or less what I said, but ... nemas problemas :)
NetMage said:
Due to Bayer demosaicing (and I assume off the shelf sensor plus don't reinvent the wheel) N probably has to be even, so not a 5x5 but either a 4x4 or 6x6 array.
Click to expand...
Maybe ...
NetMage said:
They already mention their 11 "megaray" sensor can produce 1080x1080 (1MP) images.
Click to expand...
You can produce any kind of image size. And 1080x1080 is just above the magic I MP so it makes the camera easier to sell. Maybe a coincidence, maybe a marketing choice.

A 4x4 should mean a 11/16 MP = 830x830 pixels.

A 4x4 sounds like VERY small array though if they want good DOF control.
NetMage said:
Of course, the math works best with a 3x3 array, but then they'd need their own demosaic and their own color array.
Click to expand...
Maybe ...

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
Joseph S Wisniewski said:
I also consider an "aimed weapon" stance to be a serious limitation. Whether the Lytro is intended as a tool for casual "snappers" (as someone put it near the beginning of the thread) or more serious photographers, that shooting stance is confrontational, and will provoke reactions from people.
Click to expand...
the lyto looks like a "lipstick camera"

these are widely used in a variety of applications.

critter cams
sport helmet cams
security cams
toy cams (hotwheels has a new toy car with a camera so kids can take photos
from the toy's perspective)
rocket cams (amateur rockets like estes)

some applications make more sense if the lyto also had video
or wieless trigger (including intervalometer) motion detection trigger)

of course the lyto doesn't seem to be marketed for this
 
Roland Karlsson said:
Le Kilt said:
They would have to integrate a DOF calculator so that all distances are catered for, and that is just some tables and calculations. A sophisticated version could allow all sorts of choices including aperture, shutter speed, max number of shots...
Click to expand...
Exactly my point!

Do you have ANY other functionality in your camera that is so complex?
Click to expand...
Yes, AF tracking algorithms
Roland Karlsson said:
Roland Karlsson said:
Obviously, in some cases it would mean taking a large number of shots (eg with 85mm f/1.2!), but why not?
Click to expand...
If the camera had similar functionality to a smart phone where you can run applets I would 100% agree with you. But ... as long as the user interface is based only upon settings in menus ... then focus bracketing of this complexity is outside the scope of a camera.
Click to expand...
All the gubbins are there, it is just some fairly easy programming that is needed.
 
Le Kilt said:
Roland Karlsson said:
Le Kilt said:
On the Pro1 there are three settings for the amount of bracketing, it doesn't give a precise quantity, just a smaller or larger 'amount'. It works quite well.
Click to expand...
For Lytro simulation ?
Click to expand...
Lol, no, just ordinary focus bracketing :-(
Click to expand...
But - we are discussing Lytro simulation ... that might explain the confusion.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
Roland Karlsson said:
NetMage said:
Actually it doesn't quite work that way.
Click to expand...
OK?
Roland Karlsson said:
It is more like the microlenses combine sensor pixels into NxN arrays and then they record images from multiple POV but at the same focal length and aperture. Then a lot of processing let's you compute effective DOF for aperture x N (so for the 35mm equivalent (guestimated) f11, you can compute f44)).
Click to expand...
I think that was more or less what I said, but ... nemas problemas :)
Roland Karlsson said:
Due to Bayer demosaicing (and I assume off the shelf sensor plus don't reinvent the wheel) N probably has to be even, so not a 5x5 but either a 4x4 or 6x6 array.
Click to expand...
Maybe ...
Roland Karlsson said:
They already mention their 11 "megaray" sensor can produce 1080x1080 (1MP) images.
Click to expand...
Click to expand...
Curiously we don't get to see the whole image in 1080x1080, only a 540x540 where we can zoom in, looks like a simple upsizing algorithm.
Roland Karlsson said:
You can produce any kind of image size. And 1080x1080 is just above the magic I MP so it makes the camera easier to sell. Maybe a coincidence, maybe a marketing choice.
Click to expand...
Roland Karlsson said:
A 4x4 should mean a 11/16 MP = 830x830 pixels.

A 4x4 sounds like VERY small array though if they want good DOF control.
Roland Karlsson said:
Of course, the math works best with a 3x3 array, but then they'd need their own demosaic and their own color array.
Click to expand...
Maybe ...

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
Click to expand...
 
Roland Karlsson said:
Le Kilt said:
Roland Karlsson said:
Le Kilt said:
On the Pro1 there are three settings for the amount of bracketing, it doesn't give a precise quantity, just a smaller or larger 'amount'. It works quite well.
Click to expand...
For Lytro simulation ?
Click to expand...
Lol, no, just ordinary focus bracketing :-(
Click to expand...
But - we are discussing Lytro simulation ... that might explain the confusion.
Click to expand...
No, here I was discussing focus bracketing, which is one of the elements needed for Lytro simulation.
 
Le Kilt said:
Curiously we don't get to see the whole image in 1080x1080, only a 540x540 where we can zoom in, looks like a simple upsizing algorithm.
Click to expand...
540x540 hints at a kernel of 6x6 is used per Lytro pixel. That sounds very reasonable.

540x540x6x6 = 10497600 = ~ 11M with loss of not used pixels

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
Le Kilt said:
Roland Karlsson said:
Le Kilt said:
Roland Karlsson said:
Le Kilt said:
On the Pro1 there are three settings for the amount of bracketing, it doesn't give a precise quantity, just a smaller or larger 'amount'. It works quite well.
Click to expand...
For Lytro simulation ?
Click to expand...
Lol, no, just ordinary focus bracketing :-(
Click to expand...
But - we are discussing Lytro simulation ... that might explain the confusion.
Click to expand...
No, here I was discussing focus bracketing, which is one of the elements needed for Lytro simulation.
Click to expand...
hehe - you are splitting hairs :P

We were discussing the kind of focus bracketing needed for Lytro simulation - and my conclusion was that normal focus bracketing (which actually seldom exist) in ordinary cameras will probably not work as you cannot configure them to be useful for Lytro simulation.

Puh!! - I hope you agree with that :)

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
Roland Karlsson said:
Le Kilt said:
Roland Karlsson said:
Le Kilt said:
Roland Karlsson said:
Le Kilt said:
On the Pro1 there are three settings for the amount of bracketing, it doesn't give a precise quantity, just a smaller or larger 'amount'. It works quite well.
Click to expand...
For Lytro simulation ?
Click to expand...
Lol, no, just ordinary focus bracketing :-(
Click to expand...
But - we are discussing Lytro simulation ... that might explain the confusion.
Click to expand...
No, here I was discussing focus bracketing, which is one of the elements needed for Lytro simulation.
Click to expand...
hehe - you are splitting hairs :P
Click to expand...
If you say so... ;-)
Roland Karlsson said:
We were discussing the kind of focus bracketing needed for Lytro simulation - and my conclusion was that normal focus bracketing (which actually seldom exist) in ordinary cameras will probably not work as you cannot configure them to be useful for Lytro simulation.

Puh!! - I hope you agree with that :)
Click to expand...
It will easily work with tripod and static subjects.

The tricky bit is getting it to work fast enough for hand-held shots, where the delay in moving the focusing point of the lens may make it impractical in many cases.
 
foot said:
Joseph S Wisniewski said:
I also consider an "aimed weapon" stance to be a serious limitation. Whether the Lytro is intended as a tool for casual "snappers" (as someone put it near the beginning of the thread) or more serious photographers, that shooting stance is confrontational, and will provoke reactions from people.
Click to expand...
the lyto looks like a "lipstick camera"
Click to expand...
Sure, except for the little detail about being 20x the size.

Seriously, it's 1.6 inches square, 4 inches long.
foot said:
these are widely used in a variety of applications.

critter cams
sport helmet cams
security cams
toy cams (hotwheels has a new toy car with a camera so kids can take photos
from the toy's perspective)
rocket cams (amateur rockets like estes)

some applications make more sense if the lyto also had video
or wieless trigger (including intervalometer) motion detection trigger)

of course the lyto doesn't seem to be marketed for this
Click to expand...
We're not quite sure what it is marketed for...

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com
 
The older film EOS line had a focus mode where you aimed and focused twice, and the camera used the distance and DOF to compute the aperature needed for all subjects between the two distances to be in focus.

The digital EOS uses all the focused points to do something similar, though I miss the old function. Focus bracketing doesn't seem like it would be that much harder.
--
-- Please remove the Quote option!
-- Why can't you edit more than once???
-- How about switching to real forum software?
 
Joseph S Wisniewski said:
foot said:
Joseph S Wisniewski said:
I also consider an "aimed weapon" stance to be a serious limitation. Whether the Lytro is intended as a tool for casual "snappers" (as someone put it near the beginning of the thread) or more serious photographers, that shooting stance is confrontational, and will provoke reactions from people.
Click to expand...
the lyto looks like a "lipstick camera"
Click to expand...
Sure, except for the little detail about being 20x the size.

Seriously, it's 1.6 inches square, 4 inches long.
Click to expand...
OMG

good point

this size and shape hasnt sunk in yet...
 
Le Kilt said:
It will easily work with tripod and static subjects.
Click to expand...
No it wont!

You have to add a special Lytro mode to your firmware and settings. Lytro focus settings are NOT focus bracketing.

Then - that it is theoretically possible I have no argue with. But you have to have possibility to rewrite the firmware to implement it. Or have the possibility to run applets on your camera.
Le Kilt said:
The tricky bit is getting it to work fast enough for hand-held shots, where the delay in moving the focusing point of the lens may make it impractical in many cases.
Click to expand...
Not only that. The camera also must have the possibility to fetch the data from the sensor at a high rate. Most DSLR are 5 FPS. Thats by far too slow. I would say that you want all pictures within 1/100 second. For 10 images, thats 1000 FPS. There are some special cameras that can do it - but they are usually VGA resolution.

There is also a theoretical problem. Maybe not important, but it do exist. The lytro is a light field camera, i.e. you get info in 2D on the sensor regarding the 3D object. It can be used for 3D photography and it can theoretically do even more strange stuff. Something that focus stacking cannot do.

--
Roland

support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)

X3F tools : http://www.proxel.se/x3f.html
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top