EF-S going L?

I'm glad that the EF-S 10-22 has an UD element and Ring USM but
isn't rated as a L, because it would cost 200 $ more for the red
stripe alone.

regards
I'm just saying if Canon wanted to they could have put a red stripe
and a "L" label on the 10-22 lens because it met what they have
always said the criteria for "L" classification is: having one or
more of the the following, UD Glass, Fluorite or Aspherical
elements. The 10-22 has two of the three. Again from the press
release "the EF-S 10-22mm f/3.5-4.5 USM includes a Super UD
(ultra-dispersion) glass lens and two Aspherical elements in its 13
element" I'm not totally sure of the specs for all of Canon's lens
line up; but, can you name a lens in the lineup that has any of
those elements that isn't an "L" ?
The EF 24-85mm f/3.5-4.5 USM has an aspherical element. So does the
28-135 IS, as well as the 28-90mm f/4-5.6 II USM and the 28mm
f/2.8. Also the 28mm f/1.8 USM and the 135mm f/2.8 with Soft focus.
The 28-200mm f/3.5-5.6 USM has 2 such elements. The EF 400mm f/4 DO
IS USM has 2 fluorite elements.
--
Chuck
--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
 
Wm. Bates wrote:
.
I think Canon leaving the L off the 10-22 EF-S was strictly a
marketing decision. Some how some might think that putting the red
racing strip and "L" on a lens that is aimed at consumers and not
pros would lessen the "L" image.
I'd put money down that this point was at least discussed at length. The fully pro cameras don't have 1.6x sensors. L lenses are designed for pros. I have no doubt somebody in marketing argued that no L lens could be in a mount that won't fit the fully pro cameras.

--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
 
Right now, you can't get L quality ultra wide angles with the
smaller format sensors at any price.
What, exactly, does L image quality at 10mm look like? If we
presume, for a moment, that whatever quality these new lenses
provide is "not L," what is the presumption that Canon has the
currrent ability to do it any better?
You may be right about that :-)
The only reason I took issue with your earlier post is that you said:

"Why is there an expectation of doing it at both the highest quality AND the cheapest price?"

I just wanted to clarify that Edward wanted the best quality possible, but seemed willing to pay the higher prices that that would entail. It is of course quite possible that, "L-worthy" or not, the 18-85 EF-S lens will be as fine optically as current technology permits.
--
Chuck
 
Right now, you can't get L quality ultra wide angles with the
smaller format sensors at any price.
What, exactly, does L image quality at 10mm look like? If we
presume, for a moment, that whatever quality these new lenses
provide is "not L," what is the presumption that Canon has the
currrent ability to do it any better?
You may be right about that :-)
The only reason I took issue with your earlier post is that you said:
"Why is there an expectation of doing it at both the highest
quality AND the cheapest price?"
I just wanted to clarify that Edward wanted the best quality
possible, but seemed willing to pay the higher prices that that
would entail. It is of course quite possible that, "L-worthy" or
not, the 18-85 EF-S lens will be as fine optically as current
technology permits.
--
Chuck
I meant to write that "It is of course quite possible that, "L-worthy" or not, the 10-22mm EF-S lens will be as fine optically as current technology permits."
Sorry for the confusion.
--
Chuck
 
Even 8mp is getting close to exceeding lens resolution. From now on
more pixels won't make much difference other than allowing you to
print larger without interpolation (though with no improvement in
quality over a properly res'd up 8mp image.) More resolution will
require different sensor technologies, Foveon-like perhaps, and
breakthroughs in optical design.

But IMO even 6mp with an APS-C sensor is enough resolution. The
vast majority of 10D/300D owners don't come close to using what
they've already got. I'll take a camera with greater dynamic range
over more pixels any day.

-Dave-
Oh I agree that 6mp is enough; heck, I have an old Kodak dcs-330 right now with 3mp and love it. The sensor is about the size of the new 4/3 sensor. The pictures for me have plenty of detail (of course that may have something to do with my removing the AA filter). Don't know how they'll increase dynamic range, guess we'll see this fall whether Fuji's S3 really works
 
So, how can a picture taken with Kodak 25 film resolve alot more then a 6MP 10D with the same lenses.
If they're cramming 7 megapixels now into those itty bitty teeny
tiny 1/1.8" sensors then they should be able to get, what, about
2000 megapixels out of an APS sensor? (well, almost, give or take)
Even 8mp is getting close to exceeding lens resolution. From now on
more pixels won't make much difference other than allowing you to
print larger without interpolation (though with no improvement in
quality over a properly res'd up 8mp image.) More resolution will
require different sensor technologies, Foveon-like perhaps, and
breakthroughs in optical design.

But IMO even 6mp with an APS-C sensor is enough resolution. The
vast majority of 10D/300D owners don't come close to using what
they've already got. I'll take a camera with greater dynamic range
over more pixels any day.

-Dave-
--
Henrik
 
So, how can a picture taken with Kodak 25 film resolve alot more
then a 6MP 10D with the same lenses.
The answer.... decent lenses resolve WAY more than 8MP bayer. My primes would laugh at 8MP. I connect up a 2x TC to the back of them, and they resolve gobs more detail that wasn't there before the TC was added. In other words my 8MP sensor didn't have enough resolution to extract all the detail from the lens so I had to add the TC as effectively a magnifying glass.

Jason
 
Oh I agree that 6mp is enough; heck, I have an old Kodak dcs-330
right now with 3mp and love it. The sensor is about the size of the
new 4/3 sensor. The pictures for me have plenty of detail (of
course that may have something to do with my removing the AA
filter). Don't know how they'll increase dynamic range, guess we'll
see this fall whether Fuji's S3 really works
Do you see increased moire effects with the AA filter gone? I suppose resolution is lens-limited, though, with that camera.

If Bayer mosaic technology continues to dominate over the next half-decade or so we may see higher resolution cameras with true color decoding as an option. That is, each output pixel would be made up of four photosites on the sensor. No de-mosaicing needed. A 24mp sensor would thus give you a 6mp true color image. This could be done already, and I'm kinda surprised no-one has written a RAW converter that does it. I'd love to see what a 1.5mp true color 10D image looks like. Or a 2.75mp 1Ds image. :-)

-Dave-
 
Do you see increased moire effects with the AA filter gone?
Oh yeah! Not a problem to get rid of in Photoshop though.

I suppose resolution is lens-limited, though, with that camera.

Mm, not really, the pixel size is actually slightly larger than cameras like the 10D et al.
If Bayer mosaic technology continues to dominate over the next
half-decade or so we may see higher resolution cameras with true
color decoding as an option. That is, each output pixel would be
made up of four photosites on the sensor. No de-mosaicing needed. A
24mp sensor would thus give you a 6mp true color image. This could
be done already, and I'm kinda surprised no-one has written a RAW
converter that does it. I'd love to see what a 1.5mp true color 10D
image looks like.
Mm, interesting idea, but I'm not sure it would provide any benefit over taking the regular data and just downsizing. I would think you would might end up with more "jaggies".
 
I think the EF-S as a temporary thing until the cost of producing
full frame sensors or 1.3x crop in pro-sumer cameras at cheaper
price arrives in the future.
BUT how do we not know that the EF-S lenses are the new standard as Full frame lenses go by their 35mm wayside. Not within the next few years - but soon, film is dead.

Travis
http://www.pbase.com/fototravis
Canon Digital Rebel and Powershot A80 User
 
I meant to write that "It is of course quite possible that,
"L-worthy" or not, the 10-22mm EF-S lens will be as fine optically
as current technology permits."
Sorry for the confusion.
The optimist will say, "This is the best a 10-22mm zoom can be!" The pessimist will say, "This is the best a 10-22mm zoom can be."

I'm afraid the pessimist may be right. There must be some reason none of the major manufacturers has come out with a lens that short for full frame. The 16-35mm L is long in the tooth--and for that matter, so is the 1Ds. If we don't see either of them replaced by April, I'll wonder about the longevity of Canon's full-frame digital interests.

--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
 
I suppose resolution is lens-limited, though, with that camera.
Mm, not really, the pixel size is actually slightly larger than
cameras like the 10D et al.
Oops, It didn't occur to me 'til after I posted that the DCS-330 might be a D-SLR. :-) I was thinking older compact digicam.

-Dave-
 
So, how can a picture taken with Kodak 25 film resolve alot more
then a 6MP 10D with the same lenses.
The answer.... decent lenses resolve WAY more than 8MP bayer. My
primes would laugh at 8MP. I connect up a 2x TC to the back of
them, and they resolve gobs more detail that wasn't there before
the TC was added. In other words my 8MP sensor didn't have enough
resolution to extract all the detail from the lens so I had to add
the TC as effectively a magnifying glass.
Exactly my point.

--
Henrik
 
So, how can a picture taken with Kodak 25 film resolve alot more
then a 6MP 10D with the same lenses.
The answer.... decent lenses resolve WAY more than 8MP bayer. My
primes would laugh at 8MP. I connect up a 2x TC to the back of
them, and they resolve gobs more detail that wasn't there before
the TC was added. In other words my 8MP sensor didn't have enough
resolution to extract all the detail from the lens so I had to add
the TC as effectively a magnifying glass.
Exactly my point.
Your point misses the fact that I was referring to the 1.6x sensor. I'd be the last person to deny that larger sensors offer greater potential resolution. I've used the Phase One P25 22mp medium format back. It's in another league entirely from anything Canon makes. It also uses a 49x37mm sensor with about half as many photosites per sq. mm as the 20D's sensor. The larger photosites combined with the greater sensing area make for a superior image (in technical terms) despite the lower density. USM is an option, not a necessity, with P25 images.

Maybe with higher quality photosites chip designers will be able to eek out a little more detail in the 1.6x format from existing lenses compared to the 20D's sensor. I wouldn't expect much more, though.

-Dave-
 
Maybe with higher quality photosites chip designers will be able to
eek out a little more detail in the 1.6x format from existing
lenses compared to the 20D's sensor. I wouldn't expect much more,
though.
With good lenses, LOTS more detail. First, there is a lot of detail bayer misses. Secondly, a good lens will still resolve well right down to the pixel level with a 2x TC mounted. That implies that is 4 times the resolution in the 1.6 crop up from grabs if the resolution doesn't fall off much at the edge of the 1.6x crop. For the lenses I am talking about, in fact the quality doesn't drop much at all to the edge fo the 1.6x crop.

Jason
 
With good lenses, LOTS more detail.
Agree, As a limited proof you can just refere to the Nikon D70, the lack of AA-filter (or a week one). At the places the D70 show 'color moire' the lens have extracted extra detail that the sensor can not resolve correctly.
First, there is a lot of detail bayer misses.
Secondly, a good lens will still resolve well
right down to the pixel level with a 2x TC mounted. That implies
that is 4 times the resolution in the 1.6 crop up from grabs if the
resolution doesn't fall off much at the edge of the 1.6x crop. For
the lenses I am talking about, in fact the quality doesn't drop
much at all to the edge fo the 1.6x crop.
What sensor are you refering to ? 6MP, 1.6x ?

6*4=24MP bayer pattern pixels on 1.6x. Sounds correct.
(60MP bayer on 36x24mm :-)

With color film, there might not be a use of higher resolution 35mm optics.

But we are no were near to see the limit of electrical sensors yet, maybe we need new optics in the future :-) ... distant future :-(

--
Henrik
 
Does anyone think Canon will make L lens for the EF-S series?
What I see happening is Canon is testing the waters. If Canon sells
a lot of 10-22s at good fraction of its MSRP this will indicate back to
Canon that there is demand for medium dollar EF-S lenses. Then
Cannon can respond with any product Canon wishes to lob at that
market segment....

However consider:

32mm F/1.4 EF-S based on the 50mm F/1.4 USM
22mm F/1.4 EF-S based on the 35mm F/1.4 USM
17mm F/1.8 EF-S based on the 28mm F/1.8 USM

Anything larger (FL > 50mm) is well served by the current primes.

One could argue that the 32mm is not needed, however comparing
the MTF charts of the 50mm F/1.4 and the 35mm F/1.4 we see
that the 50mm optical prescription has a distinct advantage in the
astigmatism department. And that's what we would get with a
1.6X shrink of the regular EF focal length prescriptions.
--
Mitch
 
Since most likely based on non L f2 35mm, not L1.4
Does anyone think Canon will make L lens for the EF-S series?
What I see happening is Canon is testing the waters. If Canon sells
a lot of 10-22s at good fraction of its MSRP this will indicate
back to
Canon that there is demand for medium dollar EF-S lenses. Then
Cannon can respond with any product Canon wishes to lob at that
market segment....

However consider:

32mm F/1.4 EF-S based on the 50mm F/1.4 USM
22mm F/1.4 EF-S based on the 35mm F/1.4 USM
17mm F/1.8 EF-S based on the 28mm F/1.8 USM

Anything larger (FL > 50mm) is well served by the current primes.

One could argue that the 32mm is not needed, however comparing
the MTF charts of the 50mm F/1.4 and the 35mm F/1.4 we see
that the 50mm optical prescription has a distinct advantage in the
astigmatism department. And that's what we would get with a
1.6X shrink of the regular EF focal length prescriptions.
--
Mitch
--
http://public.fotki.com/wibble/public_display/

 

Keyboard shortcuts

Back
Top