When will electonic viewfinders improve?

RichA wrote:

On Mon, 20 Dec 2004 13:51:04 -0500, Alan Browne
wrote:


RichA wrote:



Interesting looking new camera. The "apochromatic" lens is a
question. True apochromatism means three primary colours infocus
at one point = no colour error, but I'll believe it when I see it
as all these prosumers have noticeable chromatic aberration.


APO does not mean eliminates chromatic aberation, but it does of course reduce
it significantly. There are many APO lenses out there, I have three (1 prime, 2
zooms).

Cheers,
Alan


This guy sells telescope optics that make Zeiss camera lenses look bad
by comparision:

I won't challenge that statement as I'm more ignorant about telescope optics
than Zeiss optics... and I'm pretty ignorant there too...

The fellow below does say enough to show that even if his optics are fantastic
APO's, they cannot possibly correct every wavelength of light coming through the
pipe. There will be some perfect matches, and many-many very-very close, but
not perfect.

Cheers,
Alan



Defining Apochromatism
by Thomas Back

Updated 6-29-03

With the proliferation of apochromatic refractors that are available
to the amateur astronomer, it is time to define the parameters of a
true apochromatic objective lens. The modern definition of
"apochromat" is the following: An objective in which the wave
aberrations do not exceed 1/4 wave optical path difference (OPD) in
the spectral range from C (6563A - red) to F (4861A - blue), while the
g wavelength (4358A - violet) is 1/2 wave OPD or better, has three
widely spaced zero color crossings and is corrected for coma.


--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: there's no such thing as a FreeLunch.

John Doe wrote:
"Alan Browne" wrote in message
...
David J Taylor wrote:

Alan Browne wrote:
[]

The EVF on the Dimage A2 is about 922,000 pixels, so perhaps focus
is possible there to greater accuracy.


Be aware that these are Sigma-Pixels (TM)! Each RGB is counted
separately. It's actually a 640 x 480 VGA display - just 307200 RGB
pixels.


Good point. I've go to read the fine print!

I wonder why the claim 922K pixels then? The Minolta site says:

"The A2’s 922,000-pixel TFT LCD delivers clear images and sharp,
contoured information."

Are the R,G,B's spatially separated? If so, then it is little
different than calling an 8MPix R,G,B sensor 8 Mpix.

Cheers,
Alan

But that isn't the viewfinder that is the LCD screen. The viewfinder
is the thing you put your eye up to and look through. Not all of them
are LCD screens, very few of dSLR are. The ones I have seen (point
and shoot) are pretty sucky. But one should not confuse the
viewfinder with the LCD screen.
John

On that camera, the 922,000 pixels is the EVF (there is no conventional
finder). It's actually not too bad to use.

David

Alan Browne wrote:
David J Taylor wrote:

Alan Browne wrote:
[]

The EVF on the Dimage A2 is about 922,000 pixels, so perhaps focus
is possible there to greater accuracy.


Be aware that these are Sigma-Pixels (TM)! Each RGB is counted
separately. It's actually a 640 x 480 VGA display - just 307200 RGB
pixels.


Good point. I've go to read the fine print!

I wonder why the claim 922K pixels then? The Minolta site says:

"The A2’s 922,000-pixel TFT LCD delivers clear images and sharp,
contoured information."

Are the R,G,B's spatially separated? If so, then it is little
different than calling an 8MPix R,G,B sensor 8 Mpix.

Marketing - why else lie?

Like all LCD/TFTs, there are RGB triplets. To reduce the triplet
visibility, the alternate rows may be staggered by 1.5 pixels
horizontally.

I don't see it in the same light as calling cameras 8MP. On the camera
8MP of RGB are actually delivered, with an interpolation process from the
2M sensor quads. The viewfinders cannot be driven with 0.9M RGB triples,
just 0.3M. Please don't suggest that we should start naming monitors the
same way we name cameras! I have a 3072 x 768 monitor! G

Cheers,
David

David J Taylor wrote:

Alan Browne wrote:

"The A2’s 922,000-pixel TFT LCD delivers clear images and sharp,
contoured information."

Are the R,G,B's spatially separated? If so, then it is little
different than calling an 8MPix R,G,B sensor 8 Mpix.


Marketing - why else lie?

Like all LCD/TFTs, there are RGB triplets. To reduce the triplet
visibility, the alternate rows may be staggered by 1.5 pixels
horizontally.

I don't see it in the same light as calling cameras 8MP. On the camera
8MP of RGB are actually delivered, with an interpolation process from the
2M sensor quads. The viewfinders cannot be driven with 0.9M RGB triples,
just 0.3M. Please don't suggest that we should start naming monitors the
same way we name cameras! I have a 3072 x 768 monitor! G

I suspect you're right. The possibility exists that the pixel components of a
triplet are in fact spatially seperate, in which case the 0.9M is a true claim.
What is most likely is that they follow the VGA spec, so the monitor h/w is
dividing a pixel into its color components, so 0.3M is correct.

Cheers,
Alan


--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: there's no such thing as a FreeLunch.

Alan Browne wrote:
David J Taylor wrote:

Alan Browne wrote:

"The A2’s 922,000-pixel TFT LCD delivers clear images and sharp,
contoured information."

Are the R,G,B's spatially separated? If so, then it is little
different than calling an 8MPix R,G,B sensor 8 Mpix.


Marketing - why else lie?

Like all LCD/TFTs, there are RGB triplets. To reduce the triplet
visibility, the alternate rows may be staggered by 1.5 pixels
horizontally.

I don't see it in the same light as calling cameras 8MP. On the
camera 8MP of RGB are actually delivered, with an interpolation
process from the 2M sensor quads. The viewfinders cannot be driven
with 0.9M RGB triples, just 0.3M. Please don't suggest that we
should start naming monitors the same way we name cameras! I have a
3072 x 768 monitor! G

I suspect you're right. The possibility exists that the pixel
components of a triplet are in fact spatially seperate, in which case
the 0.9M is a true claim. What is most likely is that they follow
the VGA spec, so the monitor h/w is dividing a pixel into its color
components, so 0.3M is correct.
Cheers,
Alan

Alan, in both CRTs and LCDs the R, G and B elements are spatially
separated. Take a loupe to them and check for yourself.

Are you suggesting that my LCD monitor would be more correctly described
as 3072 x 768 pixels or 1024 x 768 pixels? I would take issue if you
were! I view it (no pun intended) as 1024 x 768 true-colour pixels, not
as 3072 x 1024 single-colour pixels.

Cheers,
David

David J Taylor wrote:

Alan Browne wrote:

I suspect you're right. The possibility exists that the pixel
components of a triplet are in fact spatially seperate, in which case
the 0.9M is a true claim. What is most likely is that they follow
the VGA spec, so the monitor h/w is dividing a pixel into its color
components, so 0.3M is correct.
Cheers,
Alan


Alan, in both CRTs and LCDs the R, G and B elements are spatially
separated. Take a loupe to them and check for yourself.

I realize that. What I meant was if the source pixel (with RGB info in it) is
decomposed and presented as spatially seperated R,G,B pixles, then the display
is truly .3M.

Are you suggesting that my LCD monitor would be more correctly described
as 3072 x 768 pixels or 1024 x 768 pixels? I would take issue if you
were! I view it (no pun intended) as 1024 x 768 true-colour pixels, not
as 3072 x 1024 single-colour pixels.

I'm sure that is the case, for the reason I describe above.

But if the sptially seperated R,G,B's in the display represented spatially
seperated information from the sensor, then the spatially seperated information
in the display would be 0.9M.

In the end, they are likely driving a VGA spec'd device, and so it is 0.3M.

Cheers,
Alan
--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: there's no such thing as a FreeLunch.

Alan Browne wrote:
[]
But if the sptially seperated R,G,B's in the display represented
spatially seperated information from the sensor, then the spatially
seperated information in the display would be 0.9M.

OK, I see what you mean. I'm sure it's not the case as you would have to
"bend over backwards" to dematrix from the sensor quad and re-matix to the
RGB triples!

In the end, they are likely driving a VGA spec'd device, and so it is
0.3M.
Cheers,
Alan

Agreed.
David

David J Taylor wrote:

Alan Browne wrote:
[]

But if the sptially seperated R,G,B's in the display represented
spatially seperated information from the sensor, then the spatially
seperated information in the display would be 0.9M.


OK, I see what you mean. I'm sure it's not the case as you would have to
"bend over backwards" to dematrix from the sensor quad and re-matix to the
RGB triples!

Never question what a software weenie will do with a pile of twinkies and a
bootle of Coke at his side will do at 2 am.

(To think of the weekends I passed in the lab 'just to see if...')

Cheers,
Alan


--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: there's no such thing as a FreeLunch.

Alan Browne wrote:
[]
Never question what a software weenie will do with a pile of twinkies
and a bootle of Coke at his side will do at 2 am.

(To think of the weekends I passed in the lab 'just to see if...')

Cheers,
Alan

I'm afraid I'm not familiar with "twinkies".
I sounds slightly rude, I'm sure it isn't!

What weekends you wasted (as did I!).

Cheers,
David

David J Taylor wrote:


I'm afraid I'm not familiar with "twinkies".
I sounds slightly rude, I'm sure it isn't!

Crappy chocalate cakes from a vending machine. US actually. Here it would be
"May West" or "Joe Louis"... crap just the same. I never ate many myself, more
coke or coffee ... not much coke actually, just too much coffee and chocolate bars.

Cheers,
Alan

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: there's no such thing as a FreeLunch.