Sign in with
Sign up | Sign in
Your question

Slightly disappointed with D70 viewfinder

Last response: in Digital Camera
Share
January 24, 2005 6:42:46 PM

Archived from groups: rec.photo.digital (More info?)

So I went over to my local large camera store with my set of Nikon
lenses (85mm 1.8 & 50mm 1.8) to check out the D70. I had mentioned in a
previous post that since I use almost exclusively my 85 1.8 that I'd be
somewhat happy with my 50 1.8 on the smaller sensor. So I figured I'd
run over to the store with my N80 and two lenses. Well the camera seems
great when you first pick it up - everything is where I'd expect it,
having experience with the N80.

I figured that the image in the viewfinder would be the same size, just
magnified to take into account the crop factor. But when I put it to my
eye I was disappointed to see that the image was the same - just
cropped in the viewfinder. It was definitely a smaller image that
seemed further away from you. Why wouldn't nikon take advantage of the
whole viewfinder? Seems like some simple optics.

This doesn't kill it for me, I'm sure I'd just get used to it. But I
was a bit disappointed. Anyone else have a similar opinion or know why?
-Josh
January 24, 2005 9:46:10 PM

Archived from groups: rec.photo.digital (More info?)

Well right now the status quo is shoot with my N80 using 85 1.8
(mostly) onto Portra VC 400 and get 2000 x 3000 pixel scans on my local
(good) printer's Noritsu printer (the same one that makes the prints I
guess similar to the Frontier but supposedly better). Of course I also
get the standard double prints at 4 x 6. I am thinking more and more
that I should probably just keep doing the same thing and just use the
6 megapixel scans I'm getting (about 2.2 megs each as JPEGs). He only
charges me an additional $2.50 for the CD on top of the $15 for the
prints. Maybe I'll get the next rendition of the D70 when they improve
on a few things.

-Josh
January 24, 2005 11:08:55 PM

Archived from groups: rec.photo.digital (More info?)

Josh wrote:


>
> This doesn't kill it for me, I'm sure I'd just get used to it. But I
> was a bit disappointed. Anyone else have a similar opinion or know why?
>

I found the same thing and is one reason I chose the Olympus Dslr, better
viewfinder. If I already owned some nikon or canon lenses it might have
swayed me but I don't want a viewfinder I have "to get used to" if I can
help it. Makes composing hard if the viewfinder is too small.
--

Stacey
Related resources
Can't find your answer ? Ask !
Anonymous
January 25, 2005 1:10:01 AM

Archived from groups: rec.photo.digital (More info?)

"Josh" <jayelwin@gmail.com> writes:

> So I went over to my local large camera store with my set of Nikon
> lenses (85mm 1.8 & 50mm 1.8) to check out the D70. I had mentioned in a
> previous post that since I use almost exclusively my 85 1.8 that I'd be
> somewhat happy with my 50 1.8 on the smaller sensor. So I figured I'd
> run over to the store with my N80 and two lenses. Well the camera seems
> great when you first pick it up - everything is where I'd expect it,
> having experience with the N80.
>
> I figured that the image in the viewfinder would be the same size, just
> magnified to take into account the crop factor. But when I put it to my
> eye I was disappointed to see that the image was the same - just
> cropped in the viewfinder. It was definitely a smaller image that
> seemed further away from you. Why wouldn't nikon take advantage of the
> whole viewfinder? Seems like some simple optics.

It'd get a lot dimmer if spread over a larger screen.
--
David Dyer-Bennet, <mailto:D d-b@dd-b.net>, <http://www.dd-b.net/dd-b/&gt;
RKBA: <http://noguns-nomoney.com/&gt; <http://www.dd-b.net/carry/&gt;
Pics: <http://dd-b.lighthunters.net/&gt; <http://www.dd-b.net/dd-b/SnapshotAlbum/&gt;
Dragaera/Steven Brust: <http://dragaera.info/&gt;
Anonymous
January 25, 2005 4:25:25 AM

Archived from groups: rec.photo.digital (More info?)

There is alot about this camera that is less than perfect, starting with the
viewfinder.
Picture quality can be very good (I would not call it excellent) providing
you know how to manipulate raw images in Photoshop; default jpeg settings
are no better than many P&S cameras (and worse than some), and I would not
buy a D70 if that is how you plan to use it.
If you think that with the D70 you are going to get the digital equivalent
of shooting color film with your N80 you will be in for the shock of your
life. There is no digital equivalent of the color slide that is an end
product in itself after pressing the shutter button.
For good or bad shooting with hi-end digital cameras requires some
rethinking and it is simply not possible to get quality results unless you
also learn post exposure digital image manipulation.
If you are the type of user who posts on this newsgroup arguing about the
quality differences between Walmart and Costco prints do not get a D70.
Anonymous
January 25, 2005 4:25:26 AM

Archived from groups: rec.photo.digital (More info?)

"bmoag" <apquilts@pacbell.net> wrote in message
news:9ehJd.5618$8Z1.192@newssvr14.news.prodigy.com...
> There is alot about this camera that is less than perfect, starting with
> the viewfinder.
> Picture quality can be very good (I would not call it excellent) providing
> you know how to manipulate raw images in Photoshop; default jpeg settings
> are no better than many P&S cameras (and worse than some), and I would not
> buy a D70 if that is how you plan to use it.
> If you think that with the D70 you are going to get the digital equivalent
> of shooting color film with your N80 you will be in for the shock of your
> life. There is no digital equivalent of the color slide that is an end
> product in itself after pressing the shutter button.
> For good or bad shooting with hi-end digital cameras requires some
> rethinking and it is simply not possible to get quality results unless you
> also learn post exposure digital image manipulation.
> If you are the type of user who posts on this newsgroup arguing about the
> quality differences between Walmart and Costco prints do not get a D70.

I bought a d70 and am no fanatic. However what's the advice you'd give if
someone who needs to spend less than 1500$, wants digital, wants instant on
and instant off, wants 3 pics a second at least, wants a good bundled lens,
and wants to do 11x14 prints or maybe bigger? Maybe there will be lots of
options in future, but not many now in this price range. Agree there's a
lot of learning and things aren't perfect. I'm always reading about
manipulating photos, optimizing printer settings etc. Comparison to film
isn't really fair since many of us are primarily interested in digital, for
the many advantages it brings over film. Yes, when the camera first arrived
I took some initial pics that seemed, without any work, worse than with my
olympus 3040! But I would have been limitied by the 3040 (still a great P&S
for the money IMH0). It doesn't have some of the features listed above.
You can't get a sb800 flash for it. With the d70 you can change lenses and
you've got to learn a lot about post processing eventually in this new
digital world. Digital rebel is in the right price range but there's
reasons to consider the d70 over it.
January 25, 2005 8:07:45 AM

Archived from groups: rec.photo.digital (More info?)

Good point! I didn't think of that.

-Josh
Anonymous
January 25, 2005 10:35:02 AM

Archived from groups: rec.photo.digital (More info?)

"David J. Littleboy" <davidjl@gol.com> writes:

> Sure, there are going to be lots of sensor improvements, but the physics is
> already clear that we're close to the limits (see Roger Clark's notes), so
> those improvements are going to be minimal. And no matter how much it
> improves, larger is always better.

Now, that paragraph trips all my bullshit detectors. Probably because
I've watched 40 years of writers explaining why integrated circuits
are nearing fundamental physical limits, and won't be able to continue
to develop. They always had a sane-sounding explanation, too (to this
non-physicist). And, of course, they were always wrong.

While there must really be fundamental physical limits, I've grown a
bit gun-shy of non-experts declaring where they are and when we're
going to hit them.
--
David Dyer-Bennet, <mailto:D d-b@dd-b.net>, <http://www.dd-b.net/dd-b/&gt;
RKBA: <http://noguns-nomoney.com/&gt; <http://www.dd-b.net/carry/&gt;
Pics: <http://dd-b.lighthunters.net/&gt; <http://www.dd-b.net/dd-b/SnapshotAlbum/&gt;
Dragaera/Steven Brust: <http://dragaera.info/&gt;
Anonymous
January 25, 2005 3:13:30 PM

Archived from groups: rec.photo.digital (More info?)

"LarryLOOK" <anonlmp@yahoo.com> wrote:

> Digital rebel is in the right price range but there's
> reasons to consider the d70 over it.

Three very good reasons: instant on, spot meter, and grid lines on the
screen make the D70 _very_ attractive. About the only thing the Rebel has is
the slightly lower noise at ISO 100 (which is significant since it makes
more shadow detail rescuable) and the upgrade path to a full-frame sensor.
The slow turn-on time (I've finally figured out to wake it up _before_
removing the lens cap<g>) and lack of a spot meter are really irritating.

So you are quite right.

But the upgrade path to a full-frame sensor is very important to me. At A4,
for the stuff I do, scanned medium format looks a lot better than 6 or 8MP.
The Epson R800 renders detail that holds up under a loupe, and the
smoothness, tonality, and detail that it can render given enough pixels is
amazing.

If, like just about everyone else, you are happy with 6 or 8MP at 13x19, you
can ignore my ranting<g>.

All but one of my lenses will work on a full-frame camera, which I probably
would not be able to afford if I had to buy new glass for it. (I'm expecting
a mid-range full-frame from Canon in 2006 in the US$3000 to $4000 price
range.)

Of course, Nikon might beat Canon to the affordable full-frame punch, but I
doubt it.

David J. Littleboy
Tokyo, Japan
January 25, 2005 3:13:31 PM

Archived from groups: rec.photo.digital (More info?)

David J. Littleboy wrote:

>
> But the upgrade path to a full-frame sensor is very important to me.

Because you assume sensor technology is at it's peak. If you look at any
other electronic field, even digital photography, you'd see this isn't the
case.

I'm betting sensor technology will improve REQUIRING better optics not that
they will make cheap large sensors. If for no other reason they can get
more sensors in each batch with smaller sensors and the low yield rate with
the large sensors will keep them very expencive. I can't imagine they are
just going to give up on sensor technology development and just make them
bigger like film had to. Guess we'll have to wait to see who is right?

--

Stacey
Anonymous
January 25, 2005 5:33:02 PM

Archived from groups: rec.photo.digital (More info?)

"Stacey" <fotocord@yahoo.com> wrote:
> David J. Littleboy wrote:
> >
> > But the upgrade path to a full-frame sensor is very important to me.
>
> Because you assume sensor technology is at it's peak.

It's not an assumption, it's physics. Besides, as you well know, larger
formats are always better.

> I'm betting sensor technology will improve REQUIRING better optics not
that
> they will make cheap large sensors.

There aren't a lot of better optics than Canon's better lenses. Zeiss Contax
and Leica. Maybe. Certainly not Olympus. And the Zeiss Contax lenses can be
used on Canon bodies.

> If for no other reason they can get
> more sensors in each batch with smaller sensors and the low yield rate
with
> the large sensors will keep them very expencive.

I suspect that the yield rate problem isn't as bad as most people think:
sensors are seriously low-tech (large feature size) circuits (by current
standards), and the clean room technologies are getting better. So it's just
not as hard to make every transistor count as it used to be.

Besides, APS-C sensors can't be more than US$100 or so nowadays, and I'd
gladly pay 10 times that (assuming the rest of the camera stays cheap) for
the 1Dsmk2 sensor. Even US$2000 for the sensor plus US$1000 for the rest of
the camera is fine by me.

> I can't imagine they are
> just going to give up on sensor technology development and just make them
> bigger like film had to. Guess we'll have to wait to see who is right?

Sure, there are going to be lots of sensor improvements, but the physics is
already clear that we're close to the limits (see Roger Clark's notes), so
those improvements are going to be minimal. And no matter how much it
improves, larger is always better.

Not only light collection, but also charge storage is limited by pixel size,
so dynamic range (noise) at low ISOs is limited in smaller pixels.

It's pretty amazing that 15x23mm sensors are better than 35mm, but there's
no way that 15x23mm or smaller is going to be better than 645 film: there
just aren't enough photons and there just isn't enough resolution in that
small an area whatever the sensor does.

David J. Littleboy
Tokyo, Japan
January 25, 2005 5:33:03 PM

Archived from groups: rec.photo.digital (More info?)

David J. Littleboy wrote:

>
> "Stacey" <fotocord@yahoo.com> wrote:
>> David J. Littleboy wrote:
>> >
>> > But the upgrade path to a full-frame sensor is very important to me.
>>
>> Because you assume sensor technology is at it's peak.
>
> It's not an assumption, it's physics. Besides, as you well know, larger
> formats are always better.

Higher resolution also is better. 35mm today is better than medium format
was using film from the 80's and that was a developed technology.
How long have they been making hi rez sensors, 10 years at best? IMHO they
are just starting to get really good.

>
>> I'm betting sensor technology will improve REQUIRING better optics not
> that
>> they will make cheap large sensors.
>
> There aren't a lot of better optics than Canon's better lenses. Zeiss
> Contax and Leica. Maybe. Certainly not Olympus.

Smelling like a canon troll.

How many of the new ZD zuiko lenses have you used? Or any OM lenses for that
matter... And how many people actually buy "canon's better lenses" to take
advantage of what's avalible now? Given most are optimized for full frame
sized capture, you don't think lenses optimized for smaller sensors might
be capable of higher resolution? Given 35mm optics resolve more than
medformat ones, I don't see why that wouldn't be the case here as well.


>
> Besides, APS-C sensors can't be more than US$100 or so nowadays, and I'd
> gladly pay 10 times that (assuming the rest of the camera stays cheap) for
> the 1Dsmk2 sensor. Even US$2000 for the sensor plus US$1000 for the rest
> of the camera is fine by me.

Most people aren't going to pay >$3000 for a camera body, which will make
them even more expencive. If there is a small market, they can't sell
enough to get the economy of scale down where it has to be. You don't think
they'd already be marketing a camera like you suggest if they could?

>
>> I can't imagine they are
>> just going to give up on sensor technology development and just make them
>> bigger like film had to. Guess we'll have to wait to see who is right?
>
> Sure, there are going to be lots of sensor improvements, but the physics
> is already clear that we're close to the limits

That's what they said about HD speeds and size, memory densities, network
transmission rates etc etc. Yea I'm sure they have given up making any big
improvements...

>
> Not only light collection, but also charge storage is limited by pixel
> size, so dynamic range (noise) at low ISOs is limited in smaller pixels.

But IMHO the MP count is already high enough for most uses. You don't think
they will work on better image quality at the same MP count?

>
> there
> just aren't enough photons and there just isn't enough resolution in that
> small an area whatever the sensor does.
>

Maybe not using the old 35mm designed canon lenses? :-)

I could easily see them going to either some sort of 3chip system like video
cams use to eliminate the bayer filter or just better, more sensitive chip
technology. Guessing the chips or this technology has already reached a
dead end seem silly given what has happened on almost every other digital
front so far. Like I said, time will tell but I'm not going to believe they
are going to stop development on the sensors and they are "close to the
limits".
--

Stacey
Anonymous
January 25, 2005 8:17:07 PM

Archived from groups: rec.photo.digital (More info?)

"Stacey" <fotocord@yahoo.com> wrote:
> David J. Littleboy wrote:
> > "Stacey" <fotocord@yahoo.com> wrote:
> >> David J. Littleboy wrote:
> >> >
> >> > But the upgrade path to a full-frame sensor is very important to me.
> >>
> >> Because you assume sensor technology is at it's peak.
> >
> > It's not an assumption, it's physics. Besides, as you well know, larger
> > formats are always better.
>
> Higher resolution also is better. 35mm today is better than medium format
> was using film from the 80's and that was a developed technology.

You are simply wrong on that. Plus-X pan in 120 made lovely 11x14s (and that
was in 1970!) that 35mm still can't dream of. And things like Tech Pan and
Panatomic X that were available then aren't now. The better slide films
today are nice, but they're not very sharp (MTF50 is only 45 lp/mm or so)
and K25 is ancient history.

Ektar 25 and Konica Impressa 50 are also history.

Basically, film has gotten _worse_ in the last 10 years.

> How long have they been making hi rez sensors, 10 years at best? IMHO
they
> are just starting to get really good.

Like film, sensors are getting worse: more pixels at the cost of more noise.
The only sensor that represents a real improvement at this point is the
1Dsmk2.

Cheap shot: I don't own NeatImage. The 300D's noise is simply not a problem.
Yet I've seen you discussing noise reduction tricks in gory detail here.

> >> I'm betting sensor technology will improve REQUIRING better optics not
> > that they will make cheap large sensors.
> >
> > There aren't a lot of better optics than Canon's better lenses. Zeiss
> > Contax and Leica. Maybe. Certainly not Olympus.
>
> Smelling like a canon troll.

Better than an Olympus troll. At least canon trolls have their facts right.

> How many of the new ZD zuiko lenses have you used? Or any OM lenses for
that
> matter...

OM was my 35mm system, but when I looked for a wide angle camera, it quickly
became very clear that Olympus was a cult camera, and that it was very much
second tier mfr in the Japanese market. Sure, they were nice. But nowhere
close to Nikon or Canon in optics.

> And how many people actually buy "canon's better lenses" to take
> advantage of what's avalible now?

I certainly do. And if people don't want to, that's their choice. And they
have a _lot_ of choices.

> Given most are optimized for full frame
> sized capture, you don't think lenses optimized for smaller sensors might
> be capable of higher resolution?

The lp/mm will be higher, but the lines per frame height will be lower. As
is always the case between different formats: the larger format captures
more detail, the smaller format is cheaper and more convenient. Pick your
weapon, but don't tell people that the smaller format is better. That's
seriously silly.

The ridiculous thing here is that you know all that...

> > Besides, APS-C sensors can't be more than US$100 or so nowadays, and I'd
> > gladly pay 10 times that (assuming the rest of the camera stays cheap)
for
> > the 1Dsmk2 sensor. Even US$2000 for the sensor plus US$1000 for the rest
> > of the camera is fine by me.
>
> Most people aren't going to pay >$3000 for a camera body, which will make
> them even more expencive.

People jumped on the D30 in a big way. That much money for a 3MP camera
seemed pretty silly in those days. The same money for a 16MP camera seems a
much better deal.

> If there is a small market, they can't sell
> enough to get the economy of scale down where it has to be. You don't
think
> they'd already be marketing a camera like you suggest if they could?

My bet is that when the market for $8,000 cameras is saturated and sales
slow down to a trickle, Canon will suddenly discover that there really was a
need for a camera between the 20D and the 1Dmk2 after all. And a lot of
people will find they need a 16MP camera.

> >> I can't imagine they are
> >> just going to give up on sensor technology development and just make
them
> >> bigger like film had to. Guess we'll have to wait to see who is right?
> >
> > Sure, there are going to be lots of sensor improvements, but the physics
> > is already clear that we're close to the limits
>
> That's what they said about HD speeds and size, memory densities, network
> transmission rates etc etc. Yea I'm sure they have given up making any big
> improvements...

No, they didn't say that at all about memory devices. Ever. The limits have
always been seen as 5 to 10 years off with the expectation that the problems
will be solved or another technology will come along. It's been that way
since the 70s.

> > Not only light collection, but also charge storage is limited by pixel
> > size, so dynamic range (noise) at low ISOs is limited in smaller pixels.
>
> But IMHO the MP count is already high enough for most uses. You don't
think
> they will work on better image quality at the same MP count?

As I said, it's not high enough for the printers that already exist. And the
only way to significantly improve performance at the same pixel count is to
make bigger sensors. (Note the "significantly" in there: incremental
improvements are, of course, possible.) But as I keep pointing out, we're
already at the limits.

> > there
> > just aren't enough photons and there just isn't enough resolution in
that
> > small an area whatever the sensor does.
>
> Maybe not using the old 35mm designed canon lenses? :-)

Cute cheap shot. But wrong.

> I could easily see them going to either some sort of 3chip system like
video
> cams use to eliminate the bayer filter or just better, more sensitive chip
> technology.

Again, the sensors are already photon noise limited. See Roger Clark's notes
here. And between the (mathematical) need for a low-pass filter and the
human eye's lack of color resolution, 3-chip systems make no sense.

> Guessing the chips or this technology has already reached a
> dead end seem silly given what has happened on almost every other digital
> front so far.

Sensors are radically different from other digital technologies. They have
to count photons and store a proportional charge or signals of some sort.
That takes space. Logic circuits have no intrinsic physical limitation, and
so can be made smaller down to quantum mechanical levels.

But again: in photogrtaphic sensors larger is always better. Always. Any
technology that makes 4/3 better makes full-frame that much better as well.

David J. Littleboy
Tokyo, Japan
Anonymous
January 26, 2005 1:21:29 AM

Archived from groups: rec.photo.digital (More info?)

David J. Littleboy wrote:
>
> "Stacey" <fotocord@yahoo.com> wrote:
> > David J. Littleboy wrote:
> > > "Stacey" <fotocord@yahoo.com> wrote:
> > >> David J. Littleboy wrote:
> > >> >
> > >> > But the upgrade path to a full-frame sensor is very important to me.
> > >>
> > >> Because you assume sensor technology is at it's peak.
> > >
> > > It's not an assumption, it's physics. Besides, as you well know, larger
> > > formats are always better.
> >
> > Higher resolution also is better. 35mm today is better than medium format
> > was using film from the 80's and that was a developed technology.
>
> You are simply wrong on that. Plus-X pan in 120 made lovely 11x14s

and 20x30, and 40x40 and 60x60 inch prints.


> (and that was in 1970!) that 35mm still can't dream of. And things like Tech Pan and
> Panatomic X that were available then aren't now. The better slide films
> today are nice, but they're not very sharp (MTF50 is only 45 lp/mm or so)
> and K25 is ancient history.
>
> Ektar 25 and Konica Impressa 50 are also history.
>
> Basically, film has gotten _worse_ in the last 10 years.
>
> > How long have they been making hi rez sensors, 10 years at best? IMHO
> they
> > are just starting to get really good.
>
> Like film, sensors are getting worse: more pixels at the cost of more noise.
> The only sensor that represents a real improvement at this point is the
> 1Dsmk2.
>
> Cheap shot: I don't own NeatImage. The 300D's noise is simply not a problem.
> Yet I've seen you discussing noise reduction tricks in gory detail here.
>
> > >> I'm betting sensor technology will improve REQUIRING better optics not
> > > that they will make cheap large sensors.
> > >
> > > There aren't a lot of better optics than Canon's better lenses. Zeiss
> > > Contax and Leica. Maybe. Certainly not Olympus.
> >
> > Smelling like a canon troll.
>
> Better than an Olympus troll. At least canon trolls have their facts right.
>
> > How many of the new ZD zuiko lenses have you used? Or any OM lenses for
> that
> > matter...
>
> OM was my 35mm system, but when I looked for a wide angle camera, it quickly
> became very clear that Olympus was a cult camera, and that it was very much
> second tier mfr in the Japanese market. Sure, they were nice. But nowhere
> close to Nikon or Canon in optics.
>
> > And how many people actually buy "canon's better lenses" to take
> > advantage of what's avalible now?
>
> I certainly do. And if people don't want to, that's their choice. And they
> have a _lot_ of choices.
>
> > Given most are optimized for full frame
> > sized capture, you don't think lenses optimized for smaller sensors might
> > be capable of higher resolution?
>
> The lp/mm will be higher, but the lines per frame height will be lower. As
> is always the case between different formats: the larger format captures
> more detail, the smaller format is cheaper and more convenient. Pick your
> weapon, but don't tell people that the smaller format is better. That's
> seriously silly.
>
> The ridiculous thing here is that you know all that...
>
> > > Besides, APS-C sensors can't be more than US$100 or so nowadays, and I'd
> > > gladly pay 10 times that (assuming the rest of the camera stays cheap)
> for
> > > the 1Dsmk2 sensor. Even US$2000 for the sensor plus US$1000 for the rest
> > > of the camera is fine by me.
> >
> > Most people aren't going to pay >$3000 for a camera body, which will make
> > them even more expencive.
>
> People jumped on the D30 in a big way. That much money for a 3MP camera
> seemed pretty silly in those days. The same money for a 16MP camera seems a
> much better deal.
>
> > If there is a small market, they can't sell
> > enough to get the economy of scale down where it has to be. You don't
> think
> > they'd already be marketing a camera like you suggest if they could?
>
> My bet is that when the market for $8,000 cameras is saturated and sales
> slow down to a trickle, Canon will suddenly discover that there really was a
> need for a camera between the 20D and the 1Dmk2 after all. And a lot of
> people will find they need a 16MP camera.
>
> > >> I can't imagine they are
> > >> just going to give up on sensor technology development and just make
> them
> > >> bigger like film had to. Guess we'll have to wait to see who is right?
> > >
> > > Sure, there are going to be lots of sensor improvements, but the physics
> > > is already clear that we're close to the limits
> >
> > That's what they said about HD speeds and size, memory densities, network
> > transmission rates etc etc. Yea I'm sure they have given up making any big
> > improvements...
>
> No, they didn't say that at all about memory devices. Ever. The limits have
> always been seen as 5 to 10 years off with the expectation that the problems
> will be solved or another technology will come along. It's been that way
> since the 70s.
>
> > > Not only light collection, but also charge storage is limited by pixel
> > > size, so dynamic range (noise) at low ISOs is limited in smaller pixels.
> >
> > But IMHO the MP count is already high enough for most uses. You don't
> think
> > they will work on better image quality at the same MP count?
>
> As I said, it's not high enough for the printers that already exist. And the
> only way to significantly improve performance at the same pixel count is to
> make bigger sensors. (Note the "significantly" in there: incremental
> improvements are, of course, possible.) But as I keep pointing out, we're
> already at the limits.
>
> > > there
> > > just aren't enough photons and there just isn't enough resolution in
> that
> > > small an area whatever the sensor does.
> >
> > Maybe not using the old 35mm designed canon lenses? :-)
>
> Cute cheap shot. But wrong.
>
> > I could easily see them going to either some sort of 3chip system like
> video
> > cams use to eliminate the bayer filter or just better, more sensitive chip
> > technology.
>
> Again, the sensors are already photon noise limited. See Roger Clark's notes
> here. And between the (mathematical) need for a low-pass filter and the
> human eye's lack of color resolution, 3-chip systems make no sense.
>
> > Guessing the chips or this technology has already reached a
> > dead end seem silly given what has happened on almost every other digital
> > front so far.
>
> Sensors are radically different from other digital technologies. They have
> to count photons and store a proportional charge or signals of some sort.
> That takes space. Logic circuits have no intrinsic physical limitation, and
> so can be made smaller down to quantum mechanical levels.
>
> But again: in photogrtaphic sensors larger is always better. Always. Any
> technology that makes 4/3 better makes full-frame that much better as well.
>
> David J. Littleboy
> Tokyo, Japan
January 26, 2005 5:03:18 AM

Archived from groups: rec.photo.digital (More info?)

David Dyer-Bennet wrote:

> "David J. Littleboy" <davidjl@gol.com> writes:
>
>> Sure, there are going to be lots of sensor improvements, but the physics
>> is already clear that we're close to the limits
>
> While there must really be fundamental physical limits, I've grown a
> bit gun-shy of non-experts declaring where they are and when we're
> going to hit them.

Especially with a technology this young, film is how old and it's still
being improved?
--

Stacey
January 26, 2005 5:24:16 AM

Archived from groups: rec.photo.digital (More info?)

David J. Littleboy wrote:

>
> "Stacey" <fotocord@yahoo.com> wrote:
>> David J. Littleboy wrote:

>>
>> Higher resolution also is better. 35mm today is better than medium format
>> was using film from the 80's and that was a developed technology.
>
> The better slide
> films today are nice, but they're not very sharp (MTF50 is only 45 lp/mm
> or so) and K25 is ancient history.
>
> Ektar 25 and Konica Impressa 50 are also history.
>


Seems ALL you ever care about is measuable things like lpmm etc. On one hand
all you want is high ISO performance from a Dslr, then compare modern films
to 25 ASA ones?

>
>> How long have they been making hi rez sensors, 10 years at best? IMHO
> they
>> are just starting to get really good.
>
> Like film, sensors are getting worse: more pixels at the cost of more
> noise. The only sensor that represents a real improvement at this point is
> the 1Dsmk2.
>
> Cheap shot: I don't own NeatImage. The 300D's noise is simply not a
> problem. Yet I've seen you discussing noise reduction tricks in gory
> detail here.


Because I'm looking for medium format quality from a digicam. How do your
11X14's+ look from 6mp? Like 645 images?

>> How many of the new ZD zuiko lenses have you used? Or any OM lenses for
> that
>> matter...
>
> OM was my 35mm system, but when I looked for a wide angle camera, it
> quickly became very clear that Olympus was a cult camera,

So again how many of the ZD lenses have you used to post this "fact"? Zero?
Have you checked out the MTF on the new 7-14? It blows away the wide angle
lenses made by anyone else.



>
>> Given most are optimized for full frame
>> sized capture, you don't think lenses optimized for smaller sensors might
>> be capable of higher resolution?
>
> The lp/mm will be higher, but the lines per frame height will be lower. As
> is always the case between different formats: the larger format captures
> more detail, the smaller format is cheaper and more convenient. Pick your
> weapon, but don't tell people that the smaller format is better. That's
> seriously silly.
>
> The ridiculous thing here is that you know all that...

If the smaller sensor is higher rez and the lens can resolve more, it's not
a problem. What's nice is then the lenses for the same effective focal
length are smaller. The 50-200 ZD is half the weight and size of the
100-400L, is sharper and is 2 stops faster.


>
> People jumped on the D30 in a big way.

I didn't. Less than 35mm quality for $3000?

>
>And a lot of
> people will find they need a 16MP camera.

For what? And you think the current lenses will support a sensor this good?
The 100-400L sure doesn't. The curent full fame sensors are more than this
lens can deal with. I'm sure very few of the canon "line up" can resolve
well enough for this sensor. It's one reason the ZD lenses aren't cheap.


>>
>> That's what they said about HD speeds and size, memory densities, network
>> transmission rates etc etc. Yea I'm sure they have given up making any
>> big improvements...
>
> No, they didn't say that at all about memory devices. Ever.

I'm sure someone has, just like you have here. No one designing the sensors
is claiming what you are, just people who are defending Canon's market
position do!

>
> You don't
> think
>> they will work on better image quality at the same MP count?
>
> As I said, it's not high enough for the printers that already exist.

Yes for pixel peepers and people looking at images with a microscope they
might not be. I've read you do this, most people never would.


>> > there
>> > just aren't enough photons and there just isn't enough resolution in
> that
>> > small an area whatever the sensor does.
>>
>> Maybe not using the old 35mm designed canon lenses? :-)
>
> Cute cheap shot. But wrong.

Then why doesn't the 100-400L lens perform up to the standards of a 1D?

>
>> I could easily see them going to either some sort of 3chip system like
> video
>> cams use to eliminate the bayer filter or just better, more sensitive
>> chip technology.
>
> Again, the sensors are already photon noise limited. See Roger Clark's
> notes here. And between the (mathematical) need for a low-pass filter and
> the human eye's lack of color resolution, 3-chip systems make no sense.

Because these filters affect the total resolution. Look how shitty the
resolution is on a TV yet this makes a HUGE difference in video cam
performance and how clean a low light signal they can generate. You claim
high ISO noise is such a big deal, that's what 3CCD cameras help.


> But again: in photogrtaphic sensors larger is always better. Always. Any
> technology that makes 4/3 better makes full-frame that much better as
> well.
>
>

But you assume these large sensors can be made at reasonable prices. Larger
chips of any sort go up cubically in price by their size.
--

Stacey
Anonymous
January 26, 2005 6:20:52 AM

Archived from groups: rec.photo.digital (More info?)

<<Higher resolution also is better. 35mm today is better than medium format
was using film from the 80's and that was a developed technology.>>

You forget that MF puts more area into the same photo, so you have that many
more pixels or grains devoted to the same detail in MF as in 35mm... 2.7x as
many pixels or grains for the same amount of subject on a 645 frame (the
smallest MF frame). So MF is inherently going to produce a nicer tonal
rendition. Then you have to magnify the MF neg less, so you see less grain in
the same final print size...16x makes 16x20 from 35mm, but only 9.7x makes the
same print from 645. The MF lens only has to deliver 60% of the ll/mm as the
35mm lens onto the negative as a result of this advantage, too. The 35mm DSLR
with very high pixel count comes very close to the quality of MF print scanned
in digitally, but MF still has some advantage in image quality, as a couple of
recent test reports have shown.
January 26, 2005 6:20:53 AM

Archived from groups: rec.photo.digital (More info?)

Wilt W wrote:

> <<Higher resolution also is better. 35mm today is better than medium
> format was using film from the 80's and that was a developed technology.>>
>
> You forget that MF puts more area into the same photo, so you have that
> many more pixels or grains devoted to the same detail in MF as in 35mm...

My point was using say 400asa film from the 80's on medium format vs 400asa
film on 35mm today the 35mm stuff will look better.

> The 35mm
> DSLR with very high pixel count comes very close to the quality of MF
> print scanned in digitally, but MF still has some advantage in image
> quality, as a couple of recent test reports have shown.

Of course.

--

Stacey
Anonymous
January 26, 2005 7:59:05 PM

Archived from groups: rec.photo.digital (More info?)

"Stacey" <fotocord@yahoo.com> wrote:
> Wilt W wrote:
>
> > <<Higher resolution also is better. 35mm today is better than medium
> > format was using film from the 80's and that was a developed
technology.>>
> >
> > You forget that MF puts more area into the same photo, so you have that
> > many more pixels or grains devoted to the same detail in MF as in
35mm...
>
> My point was using say 400asa film from the 80's on medium format vs
400asa
> film on 35mm today the 35mm stuff will look better.

You must be on another planet. I've tried the modern ISO 400 films, and they
are really really gross compared to, say, Reala. Basically, they are
unacceptable for quality imaging, even in 645. Sheesh, grain is visible in
A4 prints from 645.

If you had been using Tri-X in the '80s (or '70s or '60s or 50s) in medium
format, you'd have better prints than if you use ISO 400 color films in 35mm
today.

(The largest changes have been in the Fuji ISO 100 slide films, which are
quite wonderful. But again, not wonderful enough for 35mm to compete with
the MF B&W ISO 100 films from 1960 (or whenever Plus-X was first released).)

David J. Littleboy
Tokyo, Japan
Anonymous
January 26, 2005 8:54:41 PM

Archived from groups: rec.photo.digital (More info?)

"Stacey" <fotocord@yahoo.com> wrote:
> David J. Littleboy wrote:
> > "Stacey" <fotocord@yahoo.com> wrote:
> >> David J. Littleboy wrote:
> >>
> >> Higher resolution also is better. 35mm today is better than medium
format
> >> was using film from the 80's and that was a developed technology.
> >
> > The better slide
> > films today are nice, but they're not very sharp (MTF50 is only 45 lp/mm
> > or so) and K25 is ancient history.
> >
> > Ektar 25 and Konica Impressa 50 are also history.
>
> Seems ALL you ever care about is measuable things like lpmm etc.

ROFL. You've completely missed the point of my rants about the lp/mm
counters over on the MF list.

But yes, I care that my image look good when peope look closely.

> On one hand
> all you want is high ISO performance from a Dslr, then compare modern
films
> to 25 ASA ones?

But of course. Slow MF film for quality imaging, DSLRs for low light work.

At ISO 100, even the 8MP dSLRs are only just begining to edge out 35mm;
they're a factor of two to five away from MF. And the inkjet printers are
good enough to actually render that difference, even at A4.

But film become really gross at ISO 400 and up, so it doesn't make sense to
use it, especially since the dSLRs are so much better.

The low-noise dSLRs provide interesting imaging possibilities that film
didn't.

So MF film for quality imaging, and dSLRs for low-light fun makes perfect
sense.

> > Cheap shot: I don't own NeatImage. The 300D's noise is simply not a
> > problem. Yet I've seen you discussing noise reduction tricks in gory
> > detail here.
>
> Because I'm looking for medium format quality from a digicam.

MF is in the 16MP and up range. (If you can get the information from the
film to a print. I'm beginning to suspect that your lab that made your MF
prints was problematic.)

> How do your 11X14's+ look from 6mp? Like 645 images?

Of course not. 6 and 8MP are no where close to MF.

But, FWIW, noise is simply not a problem with the 300D/10D/20D at ISO 100.
That you are finding NeatImage critical says something about the camera you
bought.

> >> Given most are optimized for full frame
> >> sized capture, you don't think lenses optimized for smaller sensors
might
> >> be capable of higher resolution?
> >
> > The lp/mm will be higher, but the lines per frame height will be lower.
As
> > is always the case between different formats: the larger format captures
> > more detail, the smaller format is cheaper and more convenient. Pick
your
> > weapon, but don't tell people that the smaller format is better. That's
> > seriously silly.
> >
> > The ridiculous thing here is that you know all that...
>
> If the smaller sensor is higher rez and the lens can resolve more, it's
not
> a problem.

You've missed the point: the lp/mm may be higher for smaller lenses, but the
lines per frame height is higher for the larger format. The pixels in the
1Dsmk2 are a lot larger (requiring less lens resolution) than the E300, even
though there are twice as many.

I don't expect four times the resolution from the 1Dsmk2 as from the E300, I
expect twice.

> What's nice is then the lenses for the same effective focal
> length are smaller. The 50-200 ZD is half the weight and size of the
> 100-400L, is sharper and is 2 stops faster.

Smaller formats are always more convenient than larger formats. You just
have to decide if the loss of quality is acceptable. You are trying to have
it both ways.

It would be interesting to see your claim tested: I wonder if the 100-400L
(a lens I'm not familiar with) is really so gross that, say, 11x14s (300
dpi) with it would be anything other than a lot better than 11x14s (200 dpi
or so) from the E300 + 50-200.

> > People jumped on the D30 in a big way.
>
> I didn't. Less than 35mm quality for $3000?

Agreed. What was interesting, though, was that it blew the 5MP consumer cams
out of the water. Much better color rendition, in particular.

> >And a lot of people will find they need a 16MP camera.
>
> For what?

Quality prints. 16MP isn't quite 645, but it's a heck of a lot more than
35mm.

>And you think the current lenses will support a sensor this good?

Yes.

> The 100-400L sure doesn't. The curent full fame sensors are more than this
> lens can deal with. I'm sure very few of the canon "line up" can resolve
> well enough for this sensor. It's one reason the ZD lenses aren't cheap.

The Canon prime telephotos have no problem far exceeding the resolution of a
16MP sensor. All the way out to the corners.

> Yes for pixel peepers and people looking at images with a microscope they
> might not be. I've read you do this, most people never would.

People I hand A4 prints to put their noses on them. Interesting images
demand a closer look. And people at galleries walk up to prints; not
grain-sniffing distances, but 12 to 18" is pretty typical. So I'd like to be
able to make prints that hold up at those distances.

> >> > there
> >> > just aren't enough photons and there just isn't enough resolution in
> > that
> >> > small an area whatever the sensor does.
> >>
> >> Maybe not using the old 35mm designed canon lenses? :-)
> >
> > Cute cheap shot. But wrong.
>
> Then why doesn't the 100-400L lens perform up to the standards of a 1D?

There are lots of lenses that do, so it really doesn't matter if there are
some that don't.

> >> I could easily see them going to either some sort of 3chip system like
> > video
> >> cams use to eliminate the bayer filter or just better, more sensitive
> >> chip technology.
> >
> > Again, the sensors are already photon noise limited. See Roger Clark's
> > notes here. And between the (mathematical) need for a low-pass filter
and
> > the human eye's lack of color resolution, 3-chip systems make no sense.
>
> Because these filters affect the total resolution. Look how shitty the
> resolution is on a TV yet this makes a HUGE difference in video cam
> performance and how clean a low light signal they can generate. You claim
> high ISO noise is such a big deal, that's what 3CCD cameras help.

Comparing low-res video to high-res still photography isn't going to result
in sensible results.

> > But again: in photogrtaphic sensors larger is always better. Always. Any
> > technology that makes 4/3 better makes full-frame that much better as
> > well.
>
> But you assume these large sensors can be made at reasonable prices.

No, I don't. But if they're cheaper than my guitar, then they're
"affordable".

> Larger chips of any sort go up cubically in price by their size.

MF digital is $10,000, APS-C digital is $900. The geometric mean is $3,000.
Quite a bit cheaper than a Gibson L5.

But so? MF is more expensive, heavier, less convenient than 35mm. But it
looks a lot better. If you don't care about the quality, use the more
convenient system. But don't try to tell us that the absolute quality is
better, because it's not.

David J. Littleboy
Tokyo, Japan
January 27, 2005 6:05:29 AM

Archived from groups: rec.photo.digital (More info?)

David J. Littleboy wrote:

>
> "Stacey" <fotocord@yahoo.com> wrote:

>> >
>> > You forget that MF puts more area into the same photo, so you have that
>> > many more pixels or grains devoted to the same detail in MF as in
> 35mm...
>>
>> My point was using say 400asa film from the 80's on medium format vs
> 400asa
>> film on 35mm today the 35mm stuff will look better.
>
> You must be on another planet. I've tried the modern ISO 400 films, and
> they are really really gross compared to, say, Reala.

Have you compared them to 400asa film from the 70's? Try READING the post.

Of course reala is better than ASA 400, where did I say anything about
that?


> Basically, they are
> unacceptable for quality imaging, even in 645. Sheesh, grain is visible in
> A4 prints from 645.

So what? SOme people are more interested in the image than obcessing about
grain. Then again maybe this is why you are so anal about noise?

>
> If you had been using Tri-X in the '80s (or '70s or '60s or 50s) in medium
> format, you'd have better prints than if you use ISO 400 color films in
> 35mm today.

Lets see, you're comparing B&W film to color?

>
> (The largest changes have been in the Fuji ISO 100 slide films, which are
> quite wonderful. But again, not wonderful enough for 35mm to compete with
> the MF B&W ISO 100 films from 1960 (or whenever Plus-X was first
> released).)
>

Again try comparing apples to apples.
--

Stacey
Anonymous
January 27, 2005 6:31:47 PM

Archived from groups: rec.photo.digital (More info?)

David Dyer-Bennet <dd-b@dd-b.net> wrote:
> "David J. Littleboy" <davidjl@gol.com> writes:

>> Sure, there are going to be lots of sensor improvements, but the physics is
>> already clear that we're close to the limits (see Roger Clark's notes), so
>> those improvements are going to be minimal. And no matter how much it
>> improves, larger is always better.

> Now, that paragraph trips all my bullshit detectors. Probably because
> I've watched 40 years of writers explaining why integrated circuits
> are nearing fundamental physical limits, and won't be able to continue
> to develop. They always had a sane-sounding explanation, too (to this
> non-physicist). And, of course, they were always wrong.

There are a few areas that could be improved without breaking any of
the laws of physics. Firstly, the well capacity could be increased
without increasing the sensor pitch. Also, readout and amplifier
noise could be reduced. Also, the RGB filters could be improved. But
the fundamental limits won't go away.

Andrew.
January 27, 2005 9:52:36 PM

Archived from groups: rec.photo.digital (More info?)

On Mon, 24 Jan 2005 20:08:55 -0500, Stacey <fotocord@yahoo.com> wrote:

>Josh wrote:
>
>
>>
>> This doesn't kill it for me, I'm sure I'd just get used to it. But I
>> was a bit disappointed. Anyone else have a similar opinion or know why?
>>
>
>I found the same thing and is one reason I chose the Olympus Dslr, better
>viewfinder. If I already owned some nikon or canon lenses it might have
>swayed me but I don't want a viewfinder I have "to get used to" if I can
>help it. Makes composing hard if the viewfinder is too small.

I've been using an 8008 and an F4S for years. I never noticed the
image was smaller in the viewfinder until it was mentioned in this
thread. I guess it just wasn't one of those things I pay attention
to.

Roger Halstead (K8RI & ARRL life member)
(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com
Anonymous
January 28, 2005 12:39:45 PM

Archived from groups: rec.photo.digital (More info?)

<andrew29@littlepinkcloud.invalid> wrote in message
news:10vi2b3dcrkkk15@news.supernews.com...
> David Dyer-Bennet <dd-b@dd-b.net> wrote:
> > "David J. Littleboy" <davidjl@gol.com> writes:
>
> >> Sure, there are going to be lots of sensor improvements, but the
physics is
> >> already clear that we're close to the limits (see Roger Clark's notes),
so
> >> those improvements are going to be minimal. And no matter how much it
> >> improves, larger is always better.
>
> > Now, that paragraph trips all my bullshit detectors. Probably because
> > I've watched 40 years of writers explaining why integrated circuits
> > are nearing fundamental physical limits, and won't be able to continue
> > to develop. They always had a sane-sounding explanation, too (to this
> > non-physicist). And, of course, they were always wrong.
>
> There are a few areas that could be improved without breaking any of
> the laws of physics. Firstly, the well capacity could be increased
> without increasing the sensor pitch. Also, readout and amplifier
> noise could be reduced. Also, the RGB filters could be improved. But
> the fundamental limits won't go away.

If you read between the lines in Sony's blurbs for their sensors, this is
exactly what you see. They bust their butts to tweak the fabrication
processes, microlens designs, device characteristics. But a 3 dB improvement
in readout noise won't sell cameras, so each generation of chips has better
_per square mm_ characteristics and worse _per pixel_ characteristics.
Really depressing: the Sony F717 is a better 5MP camera than any 5MP camera
you can buy today, even though current sensor technologies could make a
better 5MP 2/3" sensor than the F717's. Sigh.

Look through the Sony CX-NEWS back numbers here:

http://www.sony.net/Products/SC-HP/cx_news/backnumber.h...

David J. Littleboy
Tokyo, Japan
Anonymous
January 28, 2005 12:39:46 PM

Archived from groups: rec.photo.digital (More info?)

David J. Littleboy <davidjl@gol.com> wrote:

> <andrew29@littlepinkcloud.invalid> wrote in message
> news:10vi2b3dcrkkk15@news.supernews.com...
>>
>> There are a few areas that could be improved without breaking any of
>> the laws of physics. Firstly, the well capacity could be increased
>> without increasing the sensor pitch. Also, readout and amplifier
>> noise could be reduced. Also, the RGB filters could be improved. But
>> the fundamental limits won't go away.

> If you read between the lines in Sony's blurbs for their sensors,
> this is exactly what you see. They bust their butts to tweak the
> fabrication processes, microlens designs, device
> characteristics. But a 3 dB improvement in readout noise won't sell
> cameras, so each generation of chips has better _per square mm_
> characteristics and worse _per pixel_ characteristics.

But that's fine, and quite to be expected. Let's say that you could
split each pixel into four: the readout noise of these four pixels
would sum as RMS, while the signal would sum linearly because nearby
pixels are highly correlated. So, you'd double the noise and also
double the linear resolution. If you combine that with process
improvements to reduce the nose floor a little and perhaps increase
the signal a little, it doesn't sound like such a bad deal.

> Really depressing: the Sony F717 is a better 5MP camera than any 5MP camera
> you can buy today, even though current sensor technologies could make a
> better 5MP 2/3" sensor than the F717's. Sigh.

> Look through the Sony CX-NEWS back numbers here:

> http://www.sony.net/Products/SC-HP/cx_news/backnumber.h...

What, all of them?

Andrew.
!