Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > Digital Photography > Reichman on EVF's and the future of optical viewfinders

Reply
Thread Tools

Reichman on EVF's and the future of optical viewfinders

 
 
Wolfgang Weisselberg
Guest
Posts: n/a
 
      02-27-2011
Outing Trolls is FUN! <(E-Mail Removed)> wrote:
[Bullshit snipped]

Show us your rare moth photographs!

-Wolfgang
 
Reply With Quote
 
 
 
 
Ofnuts
Guest
Posts: n/a
 
      02-27-2011
On 02/27/2011 05:15 PM, Outing Trolls is FUN! wrote:
> On Sun, 27 Feb 2011 14:41:28 +0100, Ofnuts<(E-Mail Removed)>
> wrote:
>
>> On 02/27/2011 11:10 AM, Outing Trolls is FUN! wrote:
>>> If you have compressed the available focal range from 2cm to
>>> 10cm with a close-up filter (the 10cm now being your relative infinity,
>>> optically), then for all practical purposes you have changed the step rate
>>> to only 0.001mm per step. This is far finer precision than most microscope
>>> stages.

>>
>> This is a hasty conclusion. The engineers use 65536 values because that
>> conveniently fits in two bytes. Whether the electronics and lens
>> mechanism that make use of this data have an equivalent accuracy is a
>> quite different matter.

>
> Not at all. I'll let you think through all on your own why this highly
> ignorant TROLL of yours is as full of bullshit as Puppygang Trollberg, when
> it comes to the subject-distance focal-step accuracy being greatly, and
> proportionally, amplified as you reduce the available focal range
> optically.


I'm not questioning the increase in accuracy matching the "compression"
on the focus range, but the assumption that because you have 65536
distinct focus values produced/used by the firmware you also have 65536
accurate and reproducible lens positions.

--
Bertrand
 
Reply With Quote
 
 
 
 
Ofnuts
Guest
Posts: n/a
 
      02-27-2011
On 02/27/2011 10:55 PM, Alfred Molon wrote:
> In article<4d6a7d49$0$26078$(E-Mail Removed)>, Ofnuts says...
>> I'm not questioning the increase in accuracy matching the "compression"
>> on the focus range, but the assumption that because you have 65536
>> distinct focus values produced/used by the firmware you also have 65536
>> accurate and reproducible lens positions.

>
> Wouldn't you agree in any case that is information available to a camera
> manufacturer, not to common mortals? This discussion is meaningless in
> this newsgroup, since nobody has the necessary information and I doubt
> some camera manufacturer would jump in and share their camera design...


This information is available to any common mortal who as ever looked at
a camera.

--
Bertrand
 
Reply With Quote
 
Ofnuts
Guest
Posts: n/a
 
      02-28-2011
On 02/28/2011 11:50 AM, Better Info wrote:

>
> Right in the firmware of every camera. Get out your disassembler and have a
> look. Then you can all have arguments over which cameras have higher
> focus-step bit-depths implemented by which ultrasonic stepper-motors (yes,
> they are capable of the precision, and more, that Numbnuts-the-Troll
> doubts).


Same error... some steppers could be that accurate, but not the ones you
find in a $300 camera (which, btw, isn't always fitted with an
ultrasonic motor). And we haven't considered the build precision of the
lens barrel and mechanism yet.
--
Bertrand
 
Reply With Quote
 
Ofnuts
Guest
Posts: n/a
 
      02-28-2011
On 02/28/2011 03:14 PM, Outing Trolls is FUN! wrote:
> On Mon, 28 Feb 2011 15:04:06 +0100, Ofnuts<(E-Mail Removed)>
> wrote:
>
>> Same error... some steppers could be that accurate, but not the ones you
>> find in a $300 camera

>
> Yet they are accurate enough to accurately focus on 2Ám pixels, smaller
> than the depth of many bacteria. Precision oil-immersion lab microscopes
> have a rough time focusing that accurately with a micrometer stage.


1) focusing on 2Ám pixels doesn't require 2Ám moves. Otherwise one
wonders how they took pictures before the invention of the stepper
motor. And the 2Ám in your pixels are transverse to the light travel
unlike the depth of the bacteria.

2) You are mixing sensor photosites sizes and subject size... In a
microscope the DOF follows the same rule as in a camera. It shortens as
the subject/image ratio increases. And while the camera barely reaches
an actual 1:1 subject/image ratio, the microscope has the equivalent of
100:1 or more...

> Now, let's talk about how DSLRs can't even focus accurately on a 3" deep
> subject, shall we?


They focus as accurately, or even more accurately than a compact camera
anywhere on a 3" deep subject. Things that are outside of the plane of
focus get de-focused faster, but that's a different matter.

*rattle*, *rattle*

--
Bertrand
 
Reply With Quote
 
Wolfgang Weisselberg
Guest
Posts: n/a
 
      03-05-2011
Alfred Molon <(E-Mail Removed)> wrote:
> In article <(E-Mail Removed)>, Wolfgang
> Weisselberg says...
>> So you break the camera into separate parts and point each
>> one? As in 'I have this staggered 3 person group, and I want
>> them all to be in focus'?


> Note that also with an OVF and focusing aids you can only focus on one
> spot at the same time (the one in the centre of the frame).


It's perfectly possible to put focussing aids on more than one
spot, even though that isn't very common. And a proper manual
focussing matte screen is all over the place.

> And if you want to have the whole group in focus using manual focus you
> first focus on something in the middle then you check if the rest of the
> group is in focus by pointing the camera to the other members of the
> group, and if not everybody is in focus you choose a smaller aperture.


Why not measure far and near, set the focus ring to the middle
between them and then stop down the appropriate amount?

>> In the real world CHAF measures the targets at discrete steps of
>> the lens. Each step is as large as possible to speed up the AF.
>> Especially in cameras that don't have any other AF.
>> Ponder that.


> That is pure speculation by you.


Just as it is pure speculation that an EVF displays images from
the main sensor. It could after all be a secondary sensor!

> It's a piece of information you don't
> have. This information is not published anywhere and only the camera
> manufacturers have it.


And *you* speculate wildly on the future of EVFs and OVFs.
By your own standards, you should never utter a thing about them.
Because you don't have information about the future.

On the other hand, I have a sound theory of how CHAF works.
This is based on observation and knowledge of basic physical
phenomena and practical limits.

If you don't agree, please lay out your theory how CHAF works,
and why using as large steps as possible while hunting will not
speed up CHAF.

>> So PDAF is at least as accurate as CHAF --- and likely more
>> accurate, since you don't want a slow AF.


> People often report back and front focus issues.


People often report flat batteries as well.

> Some cameras even allow
> micro AF adjustments to address this issue. The problem exists.


All EVF cameras allow recharging batteries or even changing them
to address this isue. The problem is very wide spread.

See where your logic leads to?

On the other hand, everyone knows and measures that CHAF is slower
than PDAF and even the fastest CHAF isn't up to fast action.
Unlike front or back focus you cannot send in your camera or even
*gasp* correct it yourself.

BTW, the Sony A55 will also suffer from front or back focus.
Even with a EVF.

>> This is not always possible, unless you only photograph static scenes.


>> > Or
>> > you can set the focus first, then go back to full view and frame again.


>> This is only possible if you have enough time. Sometimes you don't.


> Obviously you use manual focus either only on static scenes, or when you
> can predict where the subject will be when you press the shutter and you
> can focus on something at the same distance.


How did people manage before AF, I wonder?

> Note that this doesn't depend on whether you use an OVF or EVF.


No, but at least today you still need a mirror for fast PDAF.
When you use a mirror, you might just as well add an OVF.

>> Ah, yes, compared to, say, Canon 1D Mark whatever or their Nikon
>> equivalent, they sell well. Compared to camera phones the sale
>> numbers are a joke.


> Every phone may have a camera, but very few people use camera phones to
> take photos.


And yet you see them showing their pictures on their mobiles
(and shot with their mobiles) everywhere. Maybe snapshots aren't
photos to you.

> Walk to a touristic place and you will notice that.


Maybe only touristic places are where photos are taken, according
to you.

> Only in east Asia I've seen people use phones to take photos.


And East Asia is a very small country.

> Given that digital cameras with a optical zoom start at 50 Euro,


Even camera phones can have optical zooms. And many better
models have.

A 50 EUR camera is likely a complete waste of cash, so you'd
better pay twice that.

> it is
> understandable why people might prefer to use a real camera to take
> shots.


Most people carry their mobile phone every day. Most people don't
carry 'real cameras' every day. When they carry real cameras,
it is understandable that more and more people carry DSLRs,
which after all, start at 300 EUR, and are vastly better than
the extra cost over a 50 EUR point'n'shoot. After all, too many
people are put off by experiencing 50 EUR cameras.

>> >> So, yes, accurate manual focus with the OVF is possible. Let
>> >> me reiterate for you:
>> >> 1. Use the screen and enlarge OR


>> > But then you are not using the OVF to focus. You are using the LCD
>> > screen instead.


>> And that is forbidden or cheating, because it delivers what
>> EVFs promise?


> It's not forbidden, but you posted this as an example of "manual focus
> with the OVF" which it clearly is not.


Ooh! I need to rephrase that for you: "manual focus despite
having an OVF, not a super-duper can-enlarge-100-times EVF".

No reason you *must* use an OVF, if you don't want to, not on
today's DSLRs.

>> >> 2. use the focus confirmation OR


>> > This is not manual focus. You are essentially relying on the camera's AF
>> > (which might fail).


>> Unfortunately for you, it is manual focussing using a focussing
>> aid. Like enlargement. Which might fail as well.


> It's not manual focus, because you are using the camera AF. You are
> simply replacing the lens motor with your hands.


> It's manual focus when you determine yourself (not the camera) if
> something is in focus.


Ah, but *you* determine it. Just as you determine the exposure
in M mode, despite an over/underexposure needle of the camera,
coupled to a very simple or very sophisticated metering system.

>> >> 3. use a manual focussing screen.


>> > Crop DSLRs do not have focussing split screens.


>> Do you need URLs to Katzeye and Haoda, or will you be able to
>> google yourself?


> No crop DLSRs are sold with these split screen focusing aids.


No crop DSLRs are sold with:
- ultra wide lenses
- macro lenses
- fast ~f/1.4 lenses
- high end zooms or fixed focal length lenses
- teleconverters/extenders
- polarizers
- ND filters
- memory cards
- aftermarket batteries
- etc.

According to your logic, crop DSLRs can't have any of the above.

>> > The A55 can shoot at 10 fps


>> 9.09 fps.
>> http://www.imaging-resource.com/PRODS/AA55/AA55A.HTM


> It's indeed 10fps. This has been tested by pointing the A55 on a running
> stopwatch and taking a series of shots. The A55 will take 20 frames in 2
> seconds. Not always camera reviews are accurate.


Fence post error. That's at best 9.5 fps. Think it through.

Hint: A farmer wants to have a 200 meter fence. He needs a fence
post every 10 meters. How many fence posts does he need?

PS: Why should a stopwatch test be more accurate than a camera review?

>> > No DSLR with a moving mirror can do that,
>> > not even the most expensive models. It's an inherent problem of the
>> > moving mirror design.


>> Sure. The D3 can't do 11 fps. Must be imagination.
>> And the Canon 1D Mark IV can't do 10 fps. All lies.


> From what I hear these cameras do not *measure* focus while they are
> taking the shots. They just measure focus at the beginning of the the
> sequence, then use predictive AF to adjust the focus between frames.
> This focus prediction might be right or wrong.


You shouldn't believe everything you hear. Or, pray tell, why
do people complain at times that their camera (like the above)
in a series of dozens of shots (of action, where no guessing from
seconds ago will ever match) started to track some foreign object,
like a fence post that got in the way while tracking the action?

Of course these cameras use predictive AF: They measure the speed
towards/away from the camera (needing multiple measurements)
and then adjust the focus to where the object will be by the time
the shutter opens.

>> > I don't have exact numbers for both cameras. I hear the Sony A55 can do
>> > up to 600-900 shots on one battery charge.


>> One google for 'sony A55 CIPA' would have told you: 380 life view,
>> 330 with EVF. Compared to e.g. 1500 with the Canon 1D Mark IV.
>> Just over 4.5 times more.


> People who actually have *used* the A55 have reported these 600-900
> shots figures.


Sure. CIPA is not "manual AF, depress the shutter until the
battery goes empty", but emulates a rather different usage
pattern, with time between shots etc. You can google it and read
it yourself.

BTW: The photo industry developed that method for fair
measurements.

On the same note, shooting as these people do, the 1DIV will have
a much larger battery lifetime as well.

>> Of course, if you don't wait and don't refocus and don't turn off
>> the camera every so often and so on you can get many more shots
>> from one battery.


>> >> How much can a DSLR do?


>> > Highly irrelevant.


>> Because it lets your EVF cameras look bad.


> No, because very few people shoot over 1000 shots/day.


Of course. Just as very few people use EVFs.


>> > You actually get more than 300 shots on a battery charge (see above).


>> Yes, 330. 10% more.


> No, up to 600-900.


Not according to CIPA. Which is what the makers consider fair.


>> > But again, almost nobody shoots 2000 images in a day.


>> And of course, you can always recharge at night.


> Unless you are on an expedition with no access to A/C,


It's sooo trivial to find other reasons, I don't even need to try.

> in which case you
> would need to bring additional batteries with you.


And there *is* a difference between having to bring 4 batteries
and 17 (taking optimistic battery life).

>> Almost nobody uses EVFs, either.


> Wrong.


Let's see:
- DSLRs? No (and a minority of cameras)
- Compact cameras? Name one with EVF, most don't have any VF.
- Phone cameras? No
- No camera? Still a rather large part of the world population.
- Bridge cameras? A small minority of cameras.
- EVIL? Also a rather small minortity of cameras.
- Video cameras? Some have an EVF. Rarer than compact cameras.

>> > http://www.dphotoexpert.com/2007/09/...-the-cheating-
>> > dslr-viewfinder/


>> Opiniated, and while correct in some observations, I think it misapplies
>> them. Doesn't explain why OVFs *must* *always* be wrong about DOF. See
>> the comments it got.


> Did you notice those chessboard images? It's very obvious that OVFs show
> more DOF than there really is.


Did you notice that the A55 can't work? It's very obvious that
the majority of light has to go to the light hungry AF system.

Did you notice that you cite a *single* camera in a *single*
situation (very shallow DOF) with a *single* focussing screen not
made for manual focussing or DOF preview as a 'proof' that *all*
OVFs will *always* misrepresent DOF?

>> > in fact significant enough. Sometime you need
>> > to frame very accurately.


>> Post processing.


> Which means throwing away parts of the image.


Yes. Sometimes even a percent or two. Because you didn't chimp
after the shot, or used the nice big screen on the back, as you
microadjusted your tripod. After all, post processing is common
even in the analog world.

>> I find a big difference between good and usable.


> The JPEG output of many DLSRs is good, if you set up the camera
> properly. Not always of course, but for a percentage of JPEGs RAW
> processing does not deliver better results.


And yet some photographers think that RAW is giving them better
results than the inbuild JPEG engine. Especially as they can
tinker with sharpening (including defining regions to be specially
treated) etc. etc.

But if your JPEGs always look as good as RAW, more power to you.
Mine don't.

-Wolfgang
 
Reply With Quote
 
Wolfgang Weisselberg
Guest
Posts: n/a
 
      03-05-2011
Alfred Molon <(E-Mail Removed)> wrote:
> In article <(E-Mail Removed)>, Wolfgang
> Weisselberg says...


Ah --- interesting snippage. So you agree with all you snipped?

>> Just as it is pure speculation that an EVF displays images from
>> the main sensor. It could after all be a secondary sensor!


> Make an example of one camera with *EVF* from a secondary sensor.


Make one example of a DSLR not being able to focus while
doing 10+ fps. That was an expicit claim of yours!

And you agree with my theory how CHAF works, since you snipped it and
didn't provide an alternative theory.

And you agree that the problem of flat batteries is not designed to be
fixed by users, unlike AF microadjustments?

And you agree that camera phones rule the earth?

And you argee that your 'no crop DSLR is sold with' logic was spurious?

>> >> > The A55 can shoot at 10 fps


>> >> 9.09 fps.
>> >> http://www.imaging-resource.com/PRODS/AA55/AA55A.HTM


>> > It's indeed 10fps. This has been tested by pointing the A55 on a running
>> > stopwatch and taking a series of shots. The A55 will take 20 frames in 2
>> > seconds. Not always camera reviews are accurate.


>> Fence post error. That's at best 9.5 fps. Think it through.


>> Hint: A farmer wants to have a 200 meter fence. He needs a fence
>> post every 10 meters. How many fence posts does he need?


>> PS: Why should a stopwatch test be more accurate than a camera review?


> If the A55 takes one shot every 100ms,


If the moon was made of cheese, ...
Claiming the thing you intent to prove leads to circle logic.

> Point the A55 on a running stop watch and the individual images the A55
> takes show 1.33s, 1.43s, 1.53s, 1.63s, 1.73s and so on.


Point one at a stopwatch and see 1.33s, 1.44s, 1.55s, 1.66s,
1.77s, 1.88s, 1.99s, 2.10s, 2.21s and 2.31s. 9.09 fps.

Feel free to provide an URL to your mythical stopwatch test.
Prove that your stopwatch test is accurate, but that a review
by people who do that for a living isn't accurate.

Again: you are just speculating wildly.


And you agree that your hearsay about 'predictive focus' was way wrong?


>> On the same note, shooting as these people do, the 1DIV will have
>> a much larger battery lifetime as well.


> The battery of the 1DIV has a capacity of *25.5 Wh*, while the battery
> of the A55 has a capacity of *7.7 Wh*.


So what? The battery of an electrocar is also much larger
than one for a diesel powered one.

> http://www.amazon.de/Akku-Canon-EOS-.../dp/B0034CUPA6


> With 25.5Wh energy the A55 would take 330*25.5/7.7= 1093 CIPA shots.


So you can fit a LP-E4 into an A55? No? Well, then it's
irrelevant. Neither can you fit an electrocar battery into the 1D4.

> And the A55 has in-built GPS and in-built flash, which the 1DIV has not.


The A55 can't take Canon EF lenses. Which the 1D4 can.
Guess what is more important to the users of the 1D4.

> CIPA mandates that "Full illumination flash shall be used for one of
> every two pictures taken"
> http://www.cipa.jp/english/hyoujunka...f/DC-002_e.pdf


True, but not applicable to the 1D4. Which is covered by the CIPA.

> With no flash used and GPS disabled, the A55 (with a 25.5 Wh battery)
> would even exceed 1500 CIPA shots.


With wings sewn on and gravity disabled, pigs would fly.
Unfortunately, while you might sew on wings, you cannot disable gravity.


>> Did you notice that you cite a *single* camera in a *single*
>> situation (very shallow DOF) with a *single* focussing screen not
>> made for manual focussing or DOF preview as a 'proof' that *all*
>> OVFs will *always* misrepresent DOF?


> Even if the test were repeated with 25 different cameras, you would
> still dispute the results.


The test wasn't repeated, so your point is moot.
And pure speculation.
Again.

-Wolfgang
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Optical Viewfinders availability almost extinct aaronep@pacbell.net Digital Photography 41 07-20-2010 07:06 AM
a techie sort of question about p&s cameras and optical viewfinders albert Digital Photography 9 12-15-2008 07:03 PM
Why new cameras does not have optical viewfinders? zalek Digital Photography 45 02-06-2008 03:44 PM
Bell beginning to toll for reflexive optical viewfinders? RichA Digital Photography 26 08-20-2007 01:40 PM
P&S optical viewfinders Colin Brace Digital Photography 15 04-03-2006 10:06 PM



Advertisments