The Imaging Resource have just published their (first) 50D samples.
Here's a comparison between the 50D, 450D, and the 5D:

50D ( 15mp CROP)
http://www.imaging-resource.com/PRODS/E50D/FULLRES/E50DhSLI00100.HTM

450D ( 12mp CROP)
http://www.imaging-resource.com/PRODS/XSI/FULLRES/XSIhSLI0100.HTM

5D ( 12mp FF)
http://www.imaging-resource.com/PRODS/E5D/FULLRES/E5DhSLI0100.HTM

Observations:
-) the 50D image (15mp) does not have more detail than the 450D image (12mp)
-) the 50D image is softer than the 450D image
-) the 50D image (15mp) has less detail than the 5D image (12mp)

Unfortunately, this is totally expected.
People see 15mp and think big prints and room for cropping.

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) over 12mp.
Also take a look at the samles for the K20D (14mp) and A350 (14mp) on the Imaging Resourse web site. Both cameras do not have more detail than the 12mp 450D.

Disclaimer: these observation is based on CROP camera resolution of 15mp.
A FF equivalent would be ~30mp.

Which lens?

EBH

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?

What lens was used in the comparison? There is in fact no point comparing sensor performance with a lens out resolved by the sensor in the first place now is there?

I guess there is no reason to get the 5D MKII either.

Before making some sweeping statement about the resolving power of the 50D vs the XSi and 5D you need to know what lens and settings were used and how was the file processed.

If the 50D is a jpg out of the camera with the factory settings for NR then there is your softness and less detail. Pull a raw file into PS without any in-camera NR and then compare. I have done it with a MKIII and 40D and there is a noticeable difference.

Jim

Much as I like Imaging Resource this test proves absolutely nothing and most certainly does not support your conclusions. Who started this nonsense about the current lens being outresolved by a 15mp 1.6 sensor. There simply is no such thing in reality, please.

davenfl wrote:
Much as I like Imaging Resource this test proves absolutely nothing and most certainly does not support your conclusions. Who started this nonsense about the current lens being outresolved by a 15mp 1.6 sensor. There simply is no such thing in reality, please.

Agreed 100%

I spent a couple of hours comparing a 40D, 50D and 5D, taking identical shots. 3 people found the same differences between camera bodies independently of each other, so we weren't inventing them. And yes, the difference between 40D and 50D does exist (L glass, JPEGs straight out of the cameras, A4 prints).
And if the 50D is softer than the 40D, then I've got an extra-soft 40D, 'cause it sure doesn't look that way, and I've got the opinions of 2 pros plus mine and the prints to prove it.
Though I'll admit readily that (1) my testing was limited, and (2) my setup did not exhaust all possibilities, far from it. So others are free to differ.
In any case, 50D has hit the shelves, so it is easy for any really interested party to make up his/her own mind. So much better than taking other people's word for it.

Jim Victory wrote:
I guess there is no reason to get the 5D MKII either.

5D MK II should be a good backup to the 1Ds MK III when size/weight is a consideration. It should also be good for landscapes and other purposes where AF versatility is not critical.

EBH

philber wrote:

In any case, 50D has hit the shelves, so it is easy for any really interested party to make up his/her own mind. .

Since when? I must be out of the loop.

EBH

haha wow, what a helluva blanket statement.

Yeah, it's definitely out. I held one and shot a few shots with one at my local store today. I love the new grip rubber...it's a lot softer and very nice and grippy. The new LCD is beautiful, and while I didn't take a bunch of shots, the AF seemed very snappy. They had the 18-200 mounted on it, and that is a much better built lens than it appears at first glance. Very nice, though still a bit overpriced.

You know we have a new camera in the 50D with a host of new controls and settings and without exhaustive testing, we don't have a clue what they all do and what the setting really provide at default values or otherwise. This is all software driven stuff and it is going to take weeks of experience to realize what setting on a 40D are equivalent to what setting on a 50D. Also we have a whole bunch of people running around publishing photos manipulated in CS3, ACR, etc, etc, and none of them are equipped fully at this time to handle 50D raw images, at the moment only DPP can be trusted. I wish everyone would just take a deep breath until we get some experienced experts to run the new camera and DPP software through it paces so we can see what we have. From what little I have seen to date it look very promising. Canon by design has a history of providing less than interesting images right out of the camera.

I've seen sharp 100% crops from various 50D users. If anything, I think the high resolution of the 50D and the resulting IQ is exposing the users who don't know how to handle their cameras correctly.

davenfl wrote:
You know we have a new camera in the 50D with a host of new controls and settings and without exhaustive testing, we don't have a clue what they all do and what the setting really provide at default values or otherwise. This is all software driven stuff and it is going to take weeks of experience to realize what setting on a 40D are equivalent to what setting on a 50D. Also we have a whole bunch of people running around publishing photos manipulated in CS3, ACR, etc, etc, and none of them are equipped fully at this time to handle 50D raw images, at the moment only DPP can be trusted. I wish everyone would just take a deep breath until we get some experienced experts to run the new camera and DPP software through it paces so we can see what we have. From what little I have seen to date it look very promising. Canon by design has a history of providing less than interesting images right out of the camera.

I agree 100% with the exception of your ACR comment. ACR does have a release candidate out that includes 50D support. It works as well as DPP.

gml1, if you really wanna pixelpeep, you should do that with RAW files. Try to compare 40D to 50D RAW files in DPP with identical settings for both. There is NO lens-sensor limitation as far as Imaging Resource tests are concerned. It's all in your head.

what would be nice is a comparison of a 30d,40d,50d a real world one.
shoot the same picture with each use a 35L. Shoot in RAW, sharpen to 2
no other PP then convert to a tiff and print a 13x19 of each no cropping

Here is the EXIF data from the 50D image the OP linked to:

http://www.imaging-resource.com/PRODS/E50D/EXIF/E50DhSLI00100.HTM

It say the lens is: Canon EF 50mm f/2.5 Macro or Sigma Lens

After looking at the small writing under "Proportional Scale" on the round dial on the right hand side my feeling is that there is actually a little more detail in the 50D image. Around the scale of the dial there are clearly lines on the 50D image around the 40" mark where the 5D image is just solid grey.

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?

The 1DsIII is not a crop camera. The lens is of course important.

gml1 wrote:

Disclaimer: these observation is based on CROP camera resolution of 15mp.
A FF equivalent would be ~30mp.

You don't have the numbers quite right. If a full frame sensor had the same pixel density as the 50D it would be a 38.4 megapixel sensor. If a full frame sensor had the pixel density of the 450D it would be 32 megapixels.

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).

This is how the differences will look like on 75x50 inch print @ 96ppi:
http://i.pbase.com/o6/17/716817/1/104037501.th7Ri0Bo.1.bmp

davenfl wrote:
Much as I like Imaging Resource this test proves absolutely nothing and most certainly does not support your conclusions. Who started this nonsense about the current lens being outresolved by a 15mp 1.6 sensor. There simply is no such thing in reality, please.


Thank you. Anyone know the wavelength of visible light? Think about it.

Ok dcains, I have to admit I don't get that last one... what does 625nm of light waves have to do with the 50D test? not that comparing cameras makes any sense at all at this point...shoot with whatever the hell you want and leave it be.

carl

I'm just commenting on the notion that sensors and/or lenses may "outresolve" one another. Sure, it may happen in some cases, as the weak part will always determine the limit, but under optimal circumstances, there's a long way to go before a sensor will routinely outresolve a lens or vice-versa.

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).

This is how the differences will look like on 75x50 inch print @ 96ppi:
http://i.pbase.com/o6/17/716817/1/104037501.th7Ri0Bo.1.bmp


Shouldn't we be viewing these 150% crops from like 15-20 (or pick a number) feet away?

Tough room. We're comparing 150% crops now.