From: nospam on
In article <TpidnQ_vZegdHb3UnZ2dnUVZ_ovinZ2d(a)giganews.com>, Alan Browne
<alan.browne(a)Freelunchvideotron.ca> wrote:

> >> This was our great hope 10 years ago, that we could get digital backs
> >> for our somewhat expensive SLR's.
> >>
> >> Didn't happen. But I would still love a FF back for my Maxxum 9.
> >>
> >> Even if it had no display. Just an ISO setting and raw memory. Nothing
> >> else.
> >
> > that would have been great, but unfortunately, it's impossible to make
> > a digital back for just about all film slrs.
>
> Of course it isn't impossible and I never said "just about all slrs".

it may not be 'impossible' but it requires a lot of work and it is no
longer practical.

> I'm talking about SLR's such as the F-5, EOS-1v and Maxxum 9. All of
> these film cameras have user removable film doors.

that helps.

> The volume occupied by film can easily accommodate power source and
> processing. The 'door' can be a bit thicker to accommodate the sensor
> and supporting electronics.

sure, or put the electronics below the camera, as if it had a motor
drive.

> These cameras have sync connections and that can be used to trigger the
> sensor to 'take' the shot. (May defeat the rear-sync, however). Or
> optical sensors in the back could 'read' the mechanics of the film
> transport (and stimulate the film advance reader). Minor changes might
> need to be made to the camera, hopefully not at the expense of being
> able to shoot film as well as digital).

here's where it gets interesting, and it's anything but minor...

the light sensitive part of the sensor is behind an anti-alias filter
and infrared cut filter and perhaps a glass protective layer, whereas
the light sensitive part of film is on the surface. thus, to maintain
focus, the sensor would have to protrude *into* the camera for it to
line up properly. that means that either the sensor will be limited to
a cropped size (so it can fit into the film gate), or the film gate
would need to be cut to accommodate a full frame sensor. complicating
this is that the shutter mechanism is close to where the film plane is,
and to get the proper alignment, it might also require moving the
shutter assembly inward.

if you go with a cropped sensor (to avoid cutting the film gate) and
assuming the shutter is already far enough forward (which eliminates
most slrs), you now have a viewfinder that no longer matches the
sensor. the solution there is to crop off the excess in the viewfinder
(a.k.a. the 'sports finder' that sigma used on the sd-9). this becomes
even more of a headache with interchangeable screens.

then there's metering. you would need to set the iso *twice*, once on
the camera and again on the sensor assembly, as the camera has no way
to tell the sensor the iso (why would it? film never changed). since
the sensor is cropped, the metering bias is no longer accurate. what
the meter sees in the middle of the image (because the camera thinks
it's still full frame film) is really at the edge of the cropped sensor
(and likely less important) and what it sees at the edge isn't even
captured. also, sensors reflect differently than film so any off the
film metering (e.g., flash) may be adversely affected as well.

and then there's the issue of the memory buffer. the camera will
happily shoot many frames per second for as long as you hold down the
shutter button. meanwhile, can the sensor keep up? in the event the
buffer fills, how will it tell the camera to stop?

> Indeed the above is quite possible. The camera makers have done what is
> likely best, however, integrated cameras that function very well for the
> price. But that was not clear in the early days of DSLR's.

it was fairly clear.
From: nospam on
In article <ZKydnVSv2YYL273UnZ2dnUVZ_uOdnZ2d(a)comcast.com>, John
McWilliams <jpmcw(a)comcast.net> wrote:

> >> Nikon is struggling to keep up with Canon, and the D700X is part of that
> >> strategy.
> >
> > struggling? nikon sold more cameras than canon did last year. a
> > bigger issue is the economy.
>
> Agreed on the economy. Where are sales figures available for Canon and
> Nikon? I would have thought only estimates were published by third
> parties. And please don't mention Google!

here's another report -- september 2008 sales, reported by popular
photography:

<http://flash.popphoto.com/blog/2008/11/top-selling-cameras-for-septembe
r-2008.html>

although canon has the #1 spot, nikon has the #2, #3 *and* #4 spot.
From: Alan Browne on
nospam wrote:
> In article <TpidnQ_vZegdHb3UnZ2dnUVZ_ovinZ2d(a)giganews.com>, Alan Browne
> <alan.browne(a)Freelunchvideotron.ca> wrote:
>
>>>> This was our great hope 10 years ago, that we could get digital backs
>>>> for our somewhat expensive SLR's.
>>>>
>>>> Didn't happen. But I would still love a FF back for my Maxxum 9.
>>>>
>>>> Even if it had no display. Just an ISO setting and raw memory. Nothing
>>>> else.
>>> that would have been great, but unfortunately, it's impossible to make
>>> a digital back for just about all film slrs.
>> Of course it isn't impossible and I never said "just about all slrs".
>
> it may not be 'impossible' but it requires a lot of work and it is no
> longer practical.
>
>> I'm talking about SLR's such as the F-5, EOS-1v and Maxxum 9. All of
>> these film cameras have user removable film doors.
>
> that helps.
>
>> The volume occupied by film can easily accommodate power source and
>> processing. The 'door' can be a bit thicker to accommodate the sensor
>> and supporting electronics.
>
> sure, or put the electronics below the camera, as if it had a motor
> drive.
>
>> These cameras have sync connections and that can be used to trigger the
>> sensor to 'take' the shot. (May defeat the rear-sync, however). Or
>> optical sensors in the back could 'read' the mechanics of the film
>> transport (and stimulate the film advance reader). Minor changes might
>> need to be made to the camera, hopefully not at the expense of being
>> able to shoot film as well as digital).
>
> here's where it gets interesting, and it's anything but minor...
>
> the light sensitive part of the sensor is behind an anti-alias filter
> and infrared cut filter and perhaps a glass protective layer, whereas
> the light sensitive part of film is on the surface. thus, to maintain
> focus, the sensor would have to protrude *into* the camera for it to
> line up properly. that means that either the sensor will be limited to
> a cropped size (so it can fit into the film gate), or the film gate
> would need to be cut to accommodate a full frame sensor. complicating
> this is that the shutter mechanism is close to where the film plane is,
> and to get the proper alignment, it might also require moving the
> shutter assembly inward.

See my reply to "Me".

>
> if you go with a cropped sensor (to avoid cutting the film gate) and
> assuming the shutter is already far enough forward (which eliminates
> most slrs), you now have a viewfinder that no longer matches the
> sensor. the solution there is to crop off the excess in the viewfinder
> (a.k.a. the 'sports finder' that sigma used on the sd-9). this becomes
> even more of a headache with interchangeable screens.

Just change the screen once. The new screen has prominent crop marks.

> then there's metering. you would need to set the iso *twice*, once on
> the camera and again on the sensor assembly, as the camera has no way
> to tell the sensor the iso (why would it? film never changed). since
> the sensor is cropped, the metering bias is no longer accurate. what
> the meter sees in the middle of the image (because the camera thinks
> it's still full frame film) is really at the edge of the cropped sensor
> (and likely less important) and what it sees at the edge isn't even
> captured. also, sensors reflect differently than film so any off the
> film metering (e.g., flash) may be adversely affected as well.

In the Maxxum system there is a honeycomb matrix meter sensor.

Spot: no difference
Weighted: Horizontal matrix over full matrix
Scene: all matrix points.

So cropping the exposed area based on those meters would have little if
any effect.

See my other reply to "Me". There is a film door to camera data path to
support data recording option on the Maxxum 9. (Would likely require a
firmware change in the body ... but there it is).

Later Maxxum cameras (Maxxum 9 and 7, maybe more) use pre-flash as well
as OTF flash metering.

> and then there's the issue of the memory buffer. the camera will
> happily shoot many frames per second for as long as you hold down the
> shutter button. meanwhile, can the sensor keep up? in the event the
> buffer fills, how will it tell the camera to stop?

There are always compromises. For this an audible beep to tell the
shooter to stop as nothing is getting recorded...

>> Indeed the above is quite possible. The camera makers have done what is
>> likely best, however, integrated cameras that function very well for the
>> price. But that was not clear in the early days of DSLR's.
>
> it was fairly clear.

Not to everyone discussing it. There was also the notion of maintaining
film capability, although that desire has completely waned in the great
majority of shooters.

--
-- 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: Remove FreeLunch.
-- usenet posts from gmail.com and googlemail.com are filtered out.
From: nospam on
In article <3sydnfW44rjxPb3UnZ2dnUVZ_vjinZ2d(a)giganews.com>, Alan Browne
<alan.browne(a)Freelunchvideotron.ca> wrote:

> > the light sensitive part of the sensor is behind an anti-alias filter
> > and infrared cut filter and perhaps a glass protective layer, whereas
> > the light sensitive part of film is on the surface. thus, to maintain
> > focus, the sensor would have to protrude *into* the camera for it to
> > line up properly. that means that either the sensor will be limited to
> > a cropped size (so it can fit into the film gate), or the film gate
> > would need to be cut to accommodate a full frame sensor. complicating
> > this is that the shutter mechanism is close to where the film plane is,
> > and to get the proper alignment, it might also require moving the
> > shutter assembly inward.
>
> See my reply to "Me".

in that post, you mentioned a 3mm clearance. i don't know offhand if
that's sufficient or not but it's certainly cutting it *very* close if
it is. most cameras will require moving the shutter, which basically
means it's no longer practical to bother adapting it.

> > if you go with a cropped sensor (to avoid cutting the film gate) and
> > assuming the shutter is already far enough forward (which eliminates
> > most slrs), you now have a viewfinder that no longer matches the
> > sensor. the solution there is to crop off the excess in the viewfinder
> > (a.k.a. the 'sports finder' that sigma used on the sd-9). this becomes
> > even more of a headache with interchangeable screens.
>
> Just change the screen once. The new screen has prominent crop marks.

so you lose the functionality of interchangeable screens, although i
suppose a whole new set of screens could be released.

> > then there's metering. you would need to set the iso *twice*, once on
> > the camera and again on the sensor assembly, as the camera has no way
> > to tell the sensor the iso (why would it? film never changed). since
> > the sensor is cropped, the metering bias is no longer accurate. what
> > the meter sees in the middle of the image (because the camera thinks
> > it's still full frame film) is really at the edge of the cropped sensor
> > (and likely less important) and what it sees at the edge isn't even
> > captured. also, sensors reflect differently than film so any off the
> > film metering (e.g., flash) may be adversely affected as well.
>
> In the Maxxum system there is a honeycomb matrix meter sensor.
>
> Spot: no difference

it's now effectively a larger spot.

> Weighted: Horizontal matrix over full matrix
> Scene: all matrix points.

does it look at the edges of the frame? those are no longer in the
photo, but the meter still analyzes them.

> So cropping the exposed area based on those meters would have little if
> any effect.

maybe, but you still have to set iso twice.

> See my other reply to "Me". There is a film door to camera data path to
> support data recording option on the Maxxum 9. (Would likely require a
> firmware change in the body ... but there it is).

and a firmware change means you no longer can have a module that the
user can drop in. they'd have to send the camera out for updating.
and since new functionality is being added to the camera's firmware
(namely, sending shooting information to the sensor), there may not be
enough room in the existing rom chip to hold it all, which would
require a larger capacity chip and quite possibly, changing quite a bit
more.

> Later Maxxum cameras (Maxxum 9 and 7, maybe more) use pre-flash as well
> as OTF flash metering.

preflash would be required.

> > and then there's the issue of the memory buffer. the camera will
> > happily shoot many frames per second for as long as you hold down the
> > shutter button. meanwhile, can the sensor keep up? in the event the
> > buffer fills, how will it tell the camera to stop?
>
> There are always compromises. For this an audible beep to tell the
> shooter to stop as nothing is getting recorded...

that's quite a compromise!

> >> Indeed the above is quite possible. The camera makers have done what is
> >> likely best, however, integrated cameras that function very well for the
> >> price. But that was not clear in the early days of DSLR's.
> >
> > it was fairly clear.
>
> Not to everyone discussing it. There was also the notion of maintaining
> film capability, although that desire has completely waned in the great
> majority of shooters.

it's not practical.
From: Ray Fischer on
nospam <nospam(a)nospam.invalid> wrote:
>In article <ZKydnVSv2YYL273UnZ2dnUVZ_uOdnZ2d(a)comcast.com>, John
>McWilliams <jpmcw(a)comcast.net> wrote:
>
>> >> Nikon is struggling to keep up with Canon, and the D700X is part of that
>> >> strategy.
>> >
>> > struggling? nikon sold more cameras than canon did last year. a
>> > bigger issue is the economy.
>>
>> Agreed on the economy. Where are sales figures available for Canon and
>> Nikon? I would have thought only estimates were published by third
>> parties. And please don't mention Google!
>
>here's another report -- september 2008 sales, reported by popular
>photography:

Get real. One month's data means almost nothing. Pick another month
and Canon could hold the top spots, depending on what cameras were
recently released.

--
Ray Fischer
rfischer(a)sonic.net