My un-scientific observation is that comparing the 1.6 crops only, the 50D is winning the sharpness battle when the other two are up-rezzed. I will admit that it's a very small differance between the 450 and the 50D, so it is possible that the 50D is reaching a limit with todays technology. Isn't that what we should be wanting though? The best that they can come up with todays tech?

As far as noise I felt the 50D won and I like the colors a little better and I thought the DR was a little better. Therefore I feel the 50D is the best 1.6 crop on the market.

David Estes wrote:
My un-scientific observation is that comparing the 1.6 crops only, the 50D is winning the sharpness battle when the other two are up-rezzed. I will admit that it's a very small differance between the 450 and the 50D, so it is possible that the 50D is reaching a limit with todays technology. Isn't that what we should be wanting though? The best that they can come up with todays tech?

As far as noise I felt the 50D won and I like the colors a little better and I thought the DR was a little better. Therefore I feel the 50D is the best 1.6 crop on the market.

Knowing Canon I am sure the 50D will be a spectacular camera, even if the IQ isn't way ahead of the 450D.

It's the 1.5 croppers from 'the dark side' I'd be concerned about, they are making some mighty fine gear now after a slow start

David

gml1 wrote:
Here's a 150% crop from the 50D image compared to the uprezed versions of the same image from the 5D (top-right), 450D (bottom-left), and the 40D (bottom-right).



Seems to me like the 50D outresolves the rest. Just try to read the address of the winery or bottle company:
SOLE AGENTS: MERCHANT DU VINCORP. etc etc
seems like the ZIP is 98158. That's from the 50D label, the others are more vague.
Granted, the diff between 12, 12.8 and 15 is small, but compared to the 10 MP of the 40D, there is appreciable increase in detail...

I just compared my new 50D with a 28mm lens and my (1.274x sensor) Mk III with the 35mm L lens (both 45mm FF equivalents). After identical settings and RAW processing workflow the 50D substantially out resolves my Mk III at 200%. This leads me to believe that there is real world benefit from a 1.6x 15Mp crop sensor.

SoundHound wrote:
I just compared my new 50D with a 28mm lens and my (1.274x sensor) Mk III with the 35mm L lens (both 45mm FF equivalents). After identical settings and RAW processing workflow the 50D substantially out resolves my Mk III at 200%. This leads me to believe that there is real world benefit from a 1.6x 15Mp crop sensor.

Well on paper that is what you'd expect, with the 1D3 being equal to about 6MP in a 1.6x camera.

Jonathan Wong wrote:
gml1 wrote:
The reality, though, is that 15mp on a crop camera is beyond the resolving power of most commercial lenses.
So, these 15mp give you nothing more than empty magnification (aka digital zoom).



So the 1DsIII was a scam all along?



Been there done that. Conclusion...

Yeah, kinda.

-m

mauriceramirez wrote:
Jonathan Wong wrote:
gml1 wrote:
The reality, though, is that 15mp on a crop camera is beyond the resolving power of most commercial lenses.
So, these 15mp give you nothing more than empty magnification (aka digital zoom).



So the 1DsIII was a scam all along?



Been there done that. Conclusion...

Yeah, kinda.

-m

In what way?

Use a zoom and all that expensive resolution goes to pushing mushy pixels around. Canon zooms coupled to Canon AF sensors definitely has limits, and those high MP's are now exceeding them.

Using primes is all good though.

-m

mauriceramirez wrote: Canon zooms coupled to Canon AF sensors definitely has limits, and those high MP's are now exceeding them. Using primes is all good though.

Actually, the 70-200/4 IS zoom lens is sharper than a number of primes. It's risky to overgeneralize.

RalphJ wrote:
mauriceramirez wrote: Canon zooms coupled to Canon AF sensors definitely has limits, and those high MP's are now exceeding them. Using primes is all good though.

Actually, the 70-200/4 IS zoom lens is sharper than a number of primes. It's risky to overgeneralize.


The 1Ds3 + 70-200 f/4 IS make an excellent combo indeed. However, I find my 24-70L a bit lacking on my 1Ds3 (especially wide open, at the borders and at the longer end of the zoom).

RalphJ wrote:
Is "bugger all improvement" an Aussie idiom?


http://en.wikipedia.org/wiki/Bugger

There you go! Very common slang in the UK as well.

Greg, I said equal subject coverage and equal (200%) magnification-10Mp vs 15 Mp. The point was that there was extra center of the (stopped down) lens detail to be found with the extra sensors of the 50D.

The 50D needs good lenses and focus needs to be accurate to really benefit from the increased MP. The micro focus adjust is quite useful, so go use it. Prints look quite good despite the relatively high noise level. But... I can tell you the IQ is not in the class of the 1Ds MK III or even the 1Ds MK II with a fine lens.

EBH

thw2 wrote:

That's only because the 450D has a lighter AA filter than the 50D.



Is that a guess or something you can give us a link to?

Thx

Andy

RalphJ wrote:
mauriceramirez wrote: Canon zooms coupled to Canon AF sensors definitely has limits, and those high MP's are now exceeding them. Using primes is all good though.

Actually, the 70-200/4 IS zoom lens is sharper than a number of primes. It's risky to overgeneralize.


Another internet myth. I have a sharp 70-200 f4L IS and my 50mm 1.4 is so much sharper than it that it's laughable. Oh and my 50mm comes nowhere near my 85mm 1.8. Same story with my 24-70L, the primes are just so much better, it's the reason I've switched to primes for everything important these days.

Beni wrote:
RalphJ wrote:
mauriceramirez wrote: Canon zooms coupled to Canon AF sensors definitely has limits, and those high MP's are now exceeding them. Using primes is all good though.

Actually, the 70-200/4 IS zoom lens is sharper than a number of primes. It's risky to overgeneralize.


Another internet myth. I have a sharp 70-200 f4L IS and my 50mm 1.4 is so much sharper than it that it's laughable. Oh and my 50mm comes nowhere near my 85mm 1.8. Same story with my 24-70L, the primes are just so much better, it's the reason I've switched to primes for everything important these days.

I don't know about your lenses... but my 70-200 f/4 IS @ 135mm is at least as sharp as my 135L at comparing apertures. My 24-70L @ 35mm is pretty comparable to my 35L from f/5.6 and smaller

wow- this is as much fun as watching 3 color blind people argue about what color the stop light is :lol:

People, there is all this computing power sitting right in front of you.
NOT ONE PERSON has run a Ronchi test on these camera / lens combos?
(http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productID=2343)
Then run a line profile test on them? Geeze!
Until someone gives me hard data I am not going to beleive "looking @ 150% crops of scaled image" to determine a "25%" increase.
You know how hard it is for a human to determine 25%?
If I had a 50D Ill get a ronchi from my work, run a test with the 35L, 24-70 and 70-200 and put them through a line profile test and tell you numerically how many lines they each resolve.
You know all this bickering is so similar - theres a new 5 MP Sony chip on a 2/3" (8.5x7.1mm) sensor. (Equiv of a 27MP 1.6X sensor)
Its supposed to be out resolving all the machine vision lenses in the industry. But the larger canon and nikon lenses are being used one it without problems.
It has 3.4 micron pixels. Thats 144 line pair / mm, the 50D is only 106lp/mm
I dont think our lowly 106 lp/mm will be out resolving any L lenses if they work on this 5mp camera with 145LP requirements

Daan B wrote:
Beni wrote:
RalphJ wrote:
mauriceramirez wrote: Canon zooms coupled to Canon AF sensors definitely has limits, and those high MP's are now exceeding them. Using primes is all good though.

Actually, the 70-200/4 IS zoom lens is sharper than a number of primes. It's risky to overgeneralize.


Another internet myth. I have a sharp 70-200 f4L IS and my 50mm 1.4 is so much sharper than it that it's laughable. Oh and my 50mm comes nowhere near my 85mm 1.8. Same story with my 24-70L, the primes are just so much better, it's the reason I've switched to primes for everything important these days.

I don't know about your lenses... but my 70-200 f/4 IS @ 135mm is at least as sharp as my 135L at comparing apertures. My 24-70L @ 35mm is pretty comparable to my 35L from f/5.6 and smaller

If your 70-200 is as sharp as your 135L at f4 then that lens is over hyped! My 50mm beat the pants off it for sharpness and contrast at f4. Neither is this comparison nonsense, this simple test was what drove me in a decision to sell it and buy a Gitzo 2542L (finally!) and a 100mm f2 and just rent a 70-200 for the few occasions that I need it as a zoom.

I've all but stopped using zooms for anything but action these days, I've owned 3 copies of the 24-70L and none have come close to that little 50mm at f5.6 (certainly not at f2.8) nevermind the 85mm which matches the zoom at f4 when the prime is wide open (f1.8) for resolution and sharpness, CA is a different story! .

crops are a stupid way of confirming sensor sharpness. Too many variables to account for. It's not the lens that it is outresolving, it's your displays.

These new Canon sensors in high pixel count make awesome large prints like nothing seen before in DSLR's.

You seen them? The 5D mkII isn't even out yet! Not that I'm denying it, just interested as to the basis of your statement.

Beni wrote: If your 70-200 is as sharp as your 135L at f4 then that lens is over hyped! My 50mm beat the pants off the 70-200 for sharpness and contrast at f4. I've all but stopped using zooms for anything but action these days.

Here's the 70-200/4 zoom wide open compared to the 50mm prime stopped down to f4 (mouse over for the latter):

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=404&Camera=9&Sample=0&FLI=0&API=0&LensComp=115&CameraComp=9&SampleComp=0&FLIComp=0&APIComp=3

That chart reinforces what I said above (and what Beni ridiculed as "an internet myth"): the 70-200/4 IS zoom is sharper than many primes.

Just because you've had bad experiences with zooms doesn't mean everybody has.

As the EXIF also says that the focal length is 70mm I take it that the lens is a Sigma 70/2.8 as the Canon 50/2.5 Macro is, well, a 50mm!
As you can't use the Sigma on a 5D - I know it fits but..............you can't compare directly so the test is not a fair test as I understand it.

I've spoken to a couple of Pros using the 20+ megapixel 1D MKIII and they say there's no lose of quality or depreciation in resolving power. Now, granted when Canon and Nikon begin to make 60 Megapixel cameras (which I heard might start as soon as late 2010 *rumor?*) then Canon L and Nikkor lenses will probably need to be reinvented and will cost as much as Hasselblad and Leica glass.

But you never know, what if they find a way to make the sensor capture larger images using the current glass? No one can predict 3 to 4 years ahead so we'll just wait and see. As far as 21.1 megapixels I have full confidence in my L glass and prime lenses. EF-S and cheaper EF lenses...no so confident.

But look at the sample from canon of the young girl. That was taken with today's lenses, and the picture is perfectly sharp.

minhnestrone wrote:
I've spoken to a couple of Pros using the 20+ megapixel 1D MKIII and they say there's no lose of quality or depreciation in resolving power.

You must mean the 1Ds MK III, not 1D MK III. My 50D clearly has better detail in the center 38% than my 1Ds MK III. However, the 1Ds MK III has quite good imaging characterisctics, both overall and pixel for pixel.

EBH

minhnestrone wrote: You never know, what if they find a way to make the sensor capture larger images using the current glass? No one can predict 3 to 4 years ahead so we'll just wait and see.

I think it's fairly safe to say that it would be pointless to make a sensor for the current glass that is larger than the image circle cast by the current glass.