Reading mode: Light Dark

Canon FF white paper: FF will always cost much more

Go to last post
David Martin said:
Find me a single statement where Canonsaid that they intend to
discontinue the 1.3!
Click to expand...
I think you would be hard pressed to find a statement from an industry leader in any field that says they plan to discontinue a product - while it is still doing the bulk of their professional sales. That kind of statement might be nice for deciding these debates, but thats about it!

--
-CW
 
David Martin said:
David Martin said:
Wayne Larmon said:
David Martin said:
Every statement Canon make indicates that they are quite happy with
three sensor sizes. Here is just the most recent one:
'"We'd like to continue using that size as well," Westfall said.'

http://news.com.com/Digital+SLRs+bring+lens+quandary+-+page+2/2100-1041_3-6068289-2.html?tag=st.num
Find me a single statement where Canonsaid that they intend to
discontinue the 1.3!
Click to expand...
Ah, I hadn't seen that article. I was going by an interview from
several years ago where a Japanese Canon executive (not Chuck) said
something to the effect that FF has the highest quality, but they
couldn't get FPS fast enough for sports shooters and some PJs.
They could get the speed with 1.6x, but it was too noisy at high
ISOs for professional use (this was several years ago--maybe the
10D era?) APS-H has the speed and is quiet enough for pro use.
But APS-H was seen as a stopgap until such time that they can get 8
FPS with FF (and maybe get the cost down a bit) so they could drop
APS-H.

Apparently, Canon has changed their mind since that interview.

Wayne
Click to expand...
Dunno if this is the article you are thinking about, but it is the
usual one when people think that Canon have said that they are
going to drop the APS-H.
The rep said Canon were going to combine the 1 series and have just
one camera, FF and high speed - that should be with us before long,
either at Photokina or in the new year.
However, that is a long way form saying that they will discontinue
the APS-H - it just means they would have to put it somewhere else.
The statement I have quoted is just the latest in a line of Canon
statements that the are happy to carry on with three sensor sizes.
Actually, it is really simple - suppose you were in the boardroom,
and had a company objective of moving all of the company's DSLR's
to FF witht he exception of entry level. as Canon have said they
have - would you try to do it in one go, given the costs as laid
out in the White Paper, or would you move to APS-H first at a
fraction of the cost?
I know which I would do!

I
--
Regards,
DaveMart

'Just a wildebeast on the plain of life'
Please see profile for equipment
Click to expand...
--
Regards,
DaveMart

'Just a wildebeast on the plain of life'
Please see profile for equipment
Click to expand...
 
Joseph S. Wisniewski said:
nickleback said:
bionet said:
I say end the show and give us back 70s or early 80s sized bodies,
and without those weird round shapes.
Click to expand...
Yeah, because as we all know, the easiest shape for our hands to
grip is a rectangular brick. Get real.
Click to expand...
But a rectangular brick achieves ergonomic perfection when you
place the shutter speed control exactly where the adjustment shaft
exits a mechanical shutter, and the aperture adjustment collar
surrounds the aperture mechanism of a lens. Ask any Leica fan,
they'll tell you that the M is the most ergonomic camera on the
planet. And the best part is that they keep a totally straight face
while saying this, as if they actually believe it.
Click to expand...
That must be why Argus C3's were so popular: they're even more ergonomic than Leicas!

--
Leonard Migliore
 
Petteri Sulonen said:
austinphotog said:
Adding a 1.4x converter instead of getting an APS would seem to
just decrease image quality and add cost.
Click to expand...
Add cost, certainly. But decrease image quality? Not necessarily.
Don't forget that the "crop factor" doesn't magically increase the
lens's reach: to get to any given print size, you'll need to
magnify the captured image more. A teleconverter does the
magnification optically. A "neutral" TC is exactly as good as
enlarging the image in printing (which is what you're doing with a
"crop factor" camera.) A purpose-built teleconverter designed to
match a set of specific lenses can actually add correction to the
lens, i.e. it'll do a significantly better job than cropping and
enlarging.

In fact, in general pre-capture optical magnification works better
than post-capture enlargement. I.e., the full-frame camera with TC
would be expected to perform at least as well, possibly somewhat
better, than the APS-C camera without TC. Assuming, of course, the
purpose-built, high-quality TC's are used.
Petteri Sulonen said:
But the point should be to choose the camera and lenses based on
your needs not the chorus of the forums.
Click to expand...
Precisely.
Petteri Sulonen said:
The bulk of users are happy with zooms for WA, and the camera
manufacturers have made zooms for the APS sized sensors. But if you
want wide angle primes, the choices and focal length have to go to
the bigger sensor. The normal length primes are anouther reason to
go for a 36x24 sensor sized camera.
Click to expand...
Again, precisely. At this time, APS-C is a no-brainer for almost
any type of shooting. Full-frame is a no-brainer for certain
specific types of shooting, and somewhat preferable than APS-C for
a slightly wider range of shooting styles.

Petteri
--
[ http://www.prime-junta.net/ ]
[ http://p-on-p.blogspot.com/ ]
Click to expand...
 
We agree on most of it.

I'm not sure how much enlargement is really needed. We are talking about a 10 megapixel 1.6 sensor. There really isn't much enlargement on an 8x10. A 1.4x converter does intensify lens inperfections. I don't own canon right now, but am thinking of adding the 400D. On my current b&w setup with minolta film the converter seems fine on primes, but degrades my zooms. I would assume this would happen on canon also. For a 16x20 with a tc and 13 or 16 megapixels, I honestly don't know, but my gut says I would be happier with a telephoto zoom and a higher density smaller sensor. I would like to hear your experience on canon lenses and tc combination.
Petteri Sulonen said:
austinphotog said:
Adding a 1.4x converter instead of getting an APS would seem to
just decrease image quality and add cost.
Click to expand...
Add cost, certainly. But decrease image quality? Not necessarily.
Don't forget that the "crop factor" doesn't magically increase the
lens's reach: to get to any given print size, you'll need to
magnify the captured image more. A teleconverter does the
magnification optically. A "neutral" TC is exactly as good as
enlarging the image in printing (which is what you're doing with a
"crop factor" camera.) A purpose-built teleconverter designed to
match a set of specific lenses can actually add correction to the
lens, i.e. it'll do a significantly better job than cropping and
enlarging.

In fact, in general pre-capture optical magnification works better
than post-capture enlargement. I.e., the full-frame camera with TC
would be expected to perform at least as well, possibly somewhat
better, than the APS-C camera without TC. Assuming, of course, the
purpose-built, high-quality TC's are used.
Petteri Sulonen said:
But the point should be to choose the camera and lenses based on
your needs not the chorus of the forums.
Click to expand...
Precisely.
Petteri Sulonen said:
The bulk of users are happy with zooms for WA, and the camera
manufacturers have made zooms for the APS sized sensors. But if you
want wide angle primes, the choices and focal length have to go to
the bigger sensor. The normal length primes are anouther reason to
go for a 36x24 sensor sized camera.
Click to expand...
Again, precisely. At this time, APS-C is a no-brainer for almost
any type of shooting. Full-frame is a no-brainer for certain
specific types of shooting, and somewhat preferable than APS-C for
a slightly wider range of shooting styles.

Petteri
--
[ http://www.prime-junta.net/ ]
[ http://p-on-p.blogspot.com/ ]
Click to expand...
 
I completely agree that a faster processor should not add much to the cost, and many would like the value of the APS-H. I am assuming that canon is going to improve the 5D. I used the OP estimate of APS-H with 12 megapixels. The question is how good can canon make an APS-H camera without hurting the profits of their other cameras. If there are volumes in a $1800 APS-H 16 megapixel with 5fps, and it doesn't kill too many FF sales, it would be wise for canon to pursue it. I just don't think the volume of sales are there. At least we are in agreement on the technical facts, just not the market potential.
David Martin said:
austinphotog said:
You now can sell for a similar profit per unit and similar body a

$1300 APS-C 10 megapixel 5fps
$1500 APS-H 12 megapixel 4fps
$2000 FF 16.7 megapixel 3fps
Click to expand...
Your math here is about right, although in fact processing power is
so cheap now that you would not actually have to decrease the frame
rate for the APS-H - it would add very little to the cost.
But what you are omitting is that there is a bell-curve on cost, ad
$2000 is well on the high side of it, so sales are a lot lower.
It also tends to understate the advantage of the APS-H sensor, as
the densties you suggest whilst reasonable would also allow lower
density, giving better high ISO and so on.
But to coampare like with like, yoou are looking at a factor of
around 1.6, so IOW you sould do just as well with 16MP in the APS-H
as 10 in the APS-C
so for anothe couple of hundred you have 1.6 times the real estate
to play with.
Member said:
If you raise the price of the FF you probably won't sell that many
less, so its likely that you maximize profit by selling it for more
money. But it does cost more to sell 3 cameras than 2, and canon
is committed to FF. Would the APS-H camera do anything other than
take sales away from the APS-C or FF? I doubt anyone would switch
to canon from anouther brand for that camera.

If my numbers are way off, than that would change things.
Click to expand...
Click to expand...
 
I don't think they are going to hurt their FF market whatever they put in a 1.3.

Talk to 5D owners - once they have gone FF you would have to prize the camera from their cold, dead fingers.

Anyway, the point really is that Canon have said that they are going FF for everything except entry level, and surely with the costs they outline in their White Paper the only sensible way of doing that is to migrate via a 1.3.

If what you really are saying is that you can't see how Canon work out that FF will ever be viable at the xxD level, I couldn't agree more, I can't either!
But I can see how they would get to the half way stage.
So the line-up might go:
400D 10MP 1.6
40D 12MP 1.3
6D 16MP FF
1D 22MP FF

Now although Canon have told us a lot about how going to larger sensor sizes increases their costs, they have been pretty quiet about how it may reduce them

Not trying to cram lots of pixels in must increase their count of good sensors - and perhaps means less stages in the processing, reducing costs further.

Assuming that Nikon and Sony have to stay broadly in touch on pixel count, perhaps it is even tougher to keep increasing the density without being killed by diffraction and so on.

When the D2x comes up for upgrade going to 16MP on that size chip is not going to be pleasant or easy, and perhaps more costly than increasing the size to APS-H.
austinphotog said:
David Martin said:
austinphotog said:
You now can sell for a similar profit per unit and similar body a

$1300 APS-C 10 megapixel 5fps
$1500 APS-H 12 megapixel 4fps
$2000 FF 16.7 megapixel 3fps
Click to expand...
Your math here is about right, although in fact processing power is
so cheap now that you would not actually have to decrease the frame
rate for the APS-H - it would add very little to the cost.
But what you are omitting is that there is a bell-curve on cost, ad
$2000 is well on the high side of it, so sales are a lot lower.
It also tends to understate the advantage of the APS-H sensor, as
the densties you suggest whilst reasonable would also allow lower
density, giving better high ISO and so on.
But to coampare like with like, yoou are looking at a factor of
around 1.6, so IOW you sould do just as well with 16MP in the APS-H
as 10 in the APS-C
so for anothe couple of hundred you have 1.6 times the real estate
to play with.
austinphotog said:
If you raise the price of the FF you probably won't sell that many
less, so its likely that you maximize profit by selling it for more
money. But it does cost more to sell 3 cameras than 2, and canon
is committed to FF. Would the APS-H camera do anything other than
take sales away from the APS-C or FF? I doubt anyone would switch
to canon from anouther brand for that camera.

If my numbers are way off, than that would change things.
Click to expand...
Click to expand...
Click to expand...
--
Regards,
DaveMart

'Just a wildebeast on the plain of life'
Please see profile for equipment
 
austinphotog said:
I'm not sure how much enlargement is really needed. We are talking
about a 10 megapixel 1.6 sensor. There really isn't much
enlargement on an 8x10. A 1.4x converter does intensify lens
inperfections.
Click to expand...
So does imaging with the same pixel count on a sensor that is 1.6 x smaller. A 1 pixel imperfection on the 24mm x 36mm sensor becomes a 1.6 pixel imperfection

But anyway, the point behind my teleconverter comment wasn't to suggest that people just add teleconverters to existing lenses. The point is that from a design standpoint, that is all you'd have to do and doing so wouldn't add a whole lot to the lens weight and size. In other words, as a general rule, there is no need for 36x24mm lenses to be much bigger than those made for APS-C, or 4/3 and so forth - so long as you try to keep the effective performance (DoF and FoV) similar. You see significant size and weight differences when bot of those equivalencies are not maintained.
austinphotog said:
For a 16x20 with a tc and
13 or 16 megapixels, I honestly don't know, but my gut says I would
be happier with a telephoto zoom and a higher density smaller
sensor.
Click to expand...
I agree. The higher density sensor is probably a better "teleconverter" if we are talking separate non-matched components. But that is a tangent to the point I was making about size.

--
Jay Turberville
http://www.jayandwanda.com
 
Chato said:
This covered by Galileo in the spotting scope forum? Isn't he the
guy who took those great pictures of some leaning tower of Pizza?
:)
Click to expand...
Yes. His invention of the very narrow field Galilean telescope that lead to the reverse Galilean "peep" viewfinders on inexpensive cameras lead to his discredit and his best research on the accelleration of DSLRs due to the force of gravity went largely unnoticed.

--
Jay Turberville
http://www.jayandwanda.com
 
I guess I was taking you too literally. Yes, slower lenses should be straight forward to design for FF that are around the same size and weight of efs lenses, and provide similar light over the sensor area. The only problem then becomes if someone manufactures the lens you want, and if the lens is fast enough for the focus system to work properely. The first problem can be solved partially by converters.
Jay Turberville said:
Jay Turberville said:
I'm not sure how much enlargement is really needed. We are talking
about a 10 megapixel 1.6 sensor. There really isn't much
enlargement on an 8x10. A 1.4x converter does intensify lens
inperfections.
Click to expand...
So does imaging with the same pixel count on a sensor that is 1.6 x
smaller. A 1 pixel imperfection on the 24mm x 36mm sensor becomes
a 1.6 pixel imperfection

But anyway, the point behind my teleconverter comment wasn't to
suggest that people just add teleconverters to existing lenses.
The point is that from a design standpoint, that is all you'd have
to do and doing so wouldn't add a whole lot to the lens weight and
size. In other words, as a general rule, there is no need for
36x24mm lenses to be much bigger than those made for APS-C, or 4/3
and so forth - so long as you try to keep the effective performance
(DoF and FoV) similar. You see significant size and weight
differences when bot of those equivalencies are not maintained.
Click to expand...
 
While it's true that these older mixed signal processes foundries will have been fully amortized, and in fact they already are, that's not good enough to keep an inefficient process running if it's not anywhere near it's capacity. The building and the skilled workers can be much more effectively leveraged on competitive processes that are running at or near capacity.

The market for large FF 35mm imagers and other ultra-specialized imagers which need these older process geometries by itself, is nowhere NEAR sufficient to keep an obsolete fab in production, when it would otherwise be phased over to much newer, smaller geometries. Even with all of the up front costs fully amortized away, the variable costs of keeping an old fab running at 25% or less of capacity will be fatal for the process, unless the price premium for those chips is so great that the $ per wafer is high enough to approach to what can be done on a newer process making devices other than FF imaging sensors. That's why, to keep the old, big geometry process needed to make a Full Frame imager in existence, the price per chip can't come down much. As a result, I don't think we will ever see these sensors drop much below $300-400 per good device, meaning that these cameras won't ever get below the ~ $2K level for the camera. Which insures that it will be a low volume niche, relative to $400-500 DSLRs built with $20-30 APS or 4/3 sensor chips. And that's not ever going to change.
 
blakey mcgill said:
blakey mcgill said:
Really? How about with fast lenses above 35mm. That 50mm f/1.4 in > the DPReview test vignetted rather severly. Its much better at f/2.0, > but then you just lost a chunk of the larger sensor's inherent advantage.
Click to expand...
But did the 50 1.4 vignette like that on film?
Click to expand...
I don't know. The point is performance on the two different sensor sizes. So even if the vignetting is entirely unrleated to microlens and incident ray angle issues, the bottom line is that the 50mm f/1.4 performance is coming at a pretty noticable cost in quality. Stopping down to f/2.0 controls the vignetting to what would be pretty acceptable for most images. So the final question is how well the 50mm f/1.4 performs at f/2.0 on the 5D vs. how a 30mm f/1.4 Sigma performs on a a D2X. I don't know the answer to how that test would turn out. But I can imagine that it might be pretty close. And if so, then the larger format has lost some of its advantage - whatever the reason. You just can't know where the advantage is and how large it is until you do some testing - or unless you have a pretty complete technical description.
blakey mcgill said:
Also, there is no 50
1.4 in 4/3rds, so are you sure it would be alot better?
Click to expand...
No I'm not. You can certainly design a 4/3s lens so that it will vignette. B
blakey mcgill said:
I have seen tests on the Canon forum, comparing that lens on film
and FF digital. The vignetting was virtually identical. Now, that
was only one test, done by someone I do not know, so that's why I
say it's debatable.
Click to expand...
The guy who did these tests seems to agree with you. But he also seems to see things differently than me. One of images are quite clearly shows vignetting on the 24x36mm DSLRs but virtually no vignetting on the film camera.
http://www.photo.net/bboard/big-image?bboard_upload_id=30518584

Others show a quite different character of vignetting. The film camera has moderate vignetting that eventually goes off the cliff, while the digitals have a smooth ramping of the vignetting that starts almost immediately.
http://www.photo.net/bboard/big-image?bboard_upload_id=30518184
http://www.photo.net/bboard/big-image?bboard_upload_id=30518384

Yes, vignetting can be largely a function of lens design. But it can also be partially a funtion of cosine falloff that is a bigger factor on a larger sensor or film format given the same backfocus and microlens acceptance angle can also play a part.

--
Jay Turberville
http://www.jayandwanda.com
 
austinphotog said:
I guess I was taking you too literally.
Click to expand...
Sure. But that's understandable. I probably could have been clearer.
austinphotog said:
Yes, slower lenses should
be straight forward to design for FF that are around the same size
and weight of efs lenses, and provide similar light over the sensor
area. The only problem then becomes if someone manufactures the
lens you want, and if the lens is fast enough for the focus system
to work properely.
Click to expand...
My main point is that when you compare truly equivalent lenses, the differences tend to be minimal.

The secondary point is that the particular gear/capability you want is almost always more important of a consideration than a generalization such as Canon's assertion that full frame always leads to higher image quality.

--
Jay Turberville
http://www.jayandwanda.com
 
bionet said:
The grip makes the camera unnecessary large, and it's not needed
anyway.
Click to expand...
The grip just sticks out in front a bit, where the lens woud anyway, and makes placement of the shutter button and exposure controls much better.
bionet said:
The old SLRs weren't exactly bricks, the edges were rounded, and
they're certainly more ergonomic than the 350D.
Click to expand...
I'll have to agree with you on the 350D. Probably the least ergonomic of current DSLRs, unless you have tiny hands.
bionet said:
If people need a thick "pacifier for the hand" to have a steady
grip, then why do most rifles have one that isn't much thicker than
an old SLR body. You have the strongest grip on fairly small shapes
and not on something you can't get your fingers around.
Click to expand...
You are contradicting yourself. The 350D is uncomfortable because the grip is too small.

--
Seen in a fortune cookie:
Fear is the darkroom where negatives are developed
 
Chato said:
Named Galileo proved this wrong. He proved, if you believe this
kind thing, that an SLR and a compact point and shoot will hit the
ground at the same time.
Click to expand...
In a vacuum. In the real world there's this strange stuff called "air" that changes that. Dropping a camera is one thing, it's too short a distance and the objects are too similar for the effect to matter.

But try a feather and a ping pong ball, or a loaded logging truck and a compact car on top of a big downhill. I'll ride in the truck...

--
Seen in a fortune cookie:
Fear is the darkroom where negatives are developed
 
Quick reply as I'm in a rush.

First, I've used the Canon 50 1.4 on a 5D at f1.4, and I don't think 1.4 is unusable from a vignetting perspective. It's easily corrected in post. I've seen quite a few examples of 1.4 working quite well.

Second. If the situation were reversed, would you expect Olympus to mention that noise is 2/3 stop worse than 1.5 crop?
Jay Turberville said:
Jay Turberville said:
Jay Turberville said:
Really? How about with fast lenses above 35mm. That 50mm f/1.4 in > the DPReview test vignetted rather severly. Its much better at f/2.0, > but then you just lost a chunk of the larger sensor's inherent advantage.
Click to expand...
But did the 50 1.4 vignette like that on film?
Click to expand...
I don't know. The point is performance on the two different sensor
sizes. So even if the vignetting is entirely unrleated to
microlens and incident ray angle issues, the bottom line is that
the 50mm f/1.4 performance is coming at a pretty noticable cost in
quality. Stopping down to f/2.0 controls the vignetting to what
would be pretty acceptable for most images. So the final question
is how well the 50mm f/1.4 performs at f/2.0 on the 5D vs. how a
30mm f/1.4 Sigma performs on a a D2X. I don't know the answer to
how that test would turn out. But I can imagine that it might be
pretty close. And if so, then the larger format has lost some of
its advantage - whatever the reason. You just can't know where the
advantage is and how large it is until you do some testing - or
unless you have a pretty complete technical description.
Jay Turberville said:
Also, there is no 50
1.4 in 4/3rds, so are you sure it would be alot better?
Click to expand...
No I'm not. You can certainly design a 4/3s lens so that it will
vignette. B
Jay Turberville said:
I have seen tests on the Canon forum, comparing that lens on film
and FF digital. The vignetting was virtually identical. Now, that
was only one test, done by someone I do not know, so that's why I
say it's debatable.
Click to expand...
The guy who did these tests seems to agree with you. But he also
seems to see things differently than me. One of images are quite
clearly shows vignetting on the 24x36mm DSLRs but virtually no
vignetting on the film camera.
http://www.photo.net/bboard/big-image?bboard_upload_id=30518584

Others show a quite different character of vignetting. The film
camera has moderate vignetting that eventually goes off the cliff,
while the digitals have a smooth ramping of the vignetting that
starts almost immediately.
http://www.photo.net/bboard/big-image?bboard_upload_id=30518184
http://www.photo.net/bboard/big-image?bboard_upload_id=30518384

Yes, vignetting can be largely a function of lens design. But it
can also be partially a funtion of cosine falloff that is a bigger
factor on a larger sensor or film format given the same backfocus
and microlens acceptance angle can also play a part.

--
Jay Turberville
http://www.jayandwanda.com
Click to expand...
 
Huh? I just took the figures you gave, of 93 by 94, and that works out pretty close to Joe's 87% coverage

I was not commenting on the 5D coverage - just on the figures you yourself provided.
bionet said:
David Martin said:
bionet said:
Joseph S. Wisniewski said:
The A1 was a 0.83x magnification, 87% coverage viewfinder.
Click to expand...
The specs said 93 vertical and 94 horizontal or something..
Click to expand...
Click to expand...
Your math is off. The 5D's VF is horizontally and vertically 95%,
the A1's is 93/94 H/V which isn't much less.
Click to expand...
--
Regards,
DaveMart

'Just a wildebeast on the plain of life'
Please see profile for equipment
 
You must log in or register to reply here.

Similar threads

Jeroen1969
Newbee to the Medium Format Universe
2 3
Replies
49
Views
11K
SeeingSomeThingsMissingMany
SeeingSomeThingsMissingMany
2
Looking for system on a budget - am I asking too much?
Replies
19
Views
901
MarkInSF
M
R
APS-C, full frame, medium format, how much a difference
4 5 6
Replies
115
Views
3K
GarysInSoCal
GarysInSoCal
JimKasson
Should I buy a medium format digital camera?
6 7 8
Replies
150
Views
22K
JimKasson
JimKasson
A
What compact (FF) camera should I get?
Replies
7
Views
207
gganon
G

Keyboard shortcuts

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