Drawbacks of 21MP's

Ok, so I guess the 1Ds crowd can help me here.

What are the drawbacks of 21MP? I am asking only with respect to image quality? Not for workflow, I understand the issues there.

I have read that as megapixels go up the image detail becomes softer? I've read this online over and over. But have no way to know what the truth is.

Can someone dispel this myth, or prove it?

I don't think anyone can really get a handle on this until we see production-body, RAW images that have been properly processed.

21Mp Sensor + Digic IV = New Animal...

However, that won't stop the conjecture so, Game On!

umm, he's asking the 1Ds people, how can they not answer since that body is a year old??

as for the OP, I'm also wondering this same thing with the 5Dii looming. But one thing to make you feel a bit better? As you downsample images from ~20mp to ~10mp, they will have similar detail to something shot on 10mp... So it's a lot easier to go down, than uprez a 10~12mp image up to 20mp. Not that you can't go up, just that if you want 10 and are using 21, it's pretty easy to go back down to 10.

jacobsen1 wrote in post #6357885
As you downsample images from ~20mp to ~10mp, they will have similar detail to something shot on 10mp...

Agreed, but this is the paradox. See if I use 10mp (less than 5D), then I do not get the gained detail. High end wedding photographers use the 1Ds because they can get the detail in the veil. down sampling will loose that detail.

Oh wait! You mean downsize out of camera. Ok, but still want to know about the soft issue.

jacobsen1 wrote in post #6357885
umm, he's asking the 1Ds people, how can they not answer since that body is a year old??

Because even the 1Ds people don't have the same hardware that's in the 5DMkII...

It may be close (Dual Digic III Vs Digic IV.) Hell it may turn out to be exactly the same.

But so far, the only images available are jpgs so I stand by my original comment: No one knows yet how 5DMkII images (from RAW) will ultimately turn out.

he's asking if details get soft as the MP go up. How is that not a perfect question for 1DsmIII owners??

anyway, yeah, personally I see this as being a high ISO killer that you might need to use at 10mp (sRAW) to get great high ISO files, but at ISO100, I doubt you're going to have issues with details. Look at the samples that are out at 100% then look at the 5D files you have at 100%. The 5D is awesome, don't get me wrong, but once you own it you tend to stop checking your files at 100% quite as much. So when a new option comes out and you chimp it's images at 100% it's not as good as you remember until you go back.

I've been doing some close checking of my 5D files at 100% and thinking about this myself. OBVIOUSLY we won't know 100% until it's in consumers hands, but I feel very comfortable it won't be an issue for me.

drawbacks / sideffects? what is happening.....

iseecanonpeople wrote in post #6358015
drawbacks / sideffects? what is happening.....

People are freaking out over vaporware...

FlyingPhotog wrote in post #6358043
People are freaking out over vaporware...

how exactly people freaking out over vaporware or are you just trying to be difficult? the other thread is talking about a 21mp file size. it can be from a Nikon, Canon, Leica or even a p&s. this thread is talking about the affect of a more MP. again, it can be from anything not just the 5DmkII.

if you are against all of this "vaporware" speculation, why do you keep responding to these threads?

If people want to compare detail between the 1Ds/5DII and the 5D, they'd need to upsample the 5D files to 21 MP, to see how fine the 1Ds/5DII detail is, or downsample the 1Ds/5DII files to 12 MP to see if they match the 5Ds richness/cleanness of detail. Viewing the original files at 100% won't give a match. Among other things, the 21 MP file will show more effects of things like camera/lens movement and lens softness when viewed at 100%.

Another interesting comparison would be to compare a 1Ds/5DII file at 100% with a 30D file shot with the same distance and focal length (matching the field of view).

Since I don't have a 1Ds or a 5DII I can't run any comparisons -- sorry!

Maybe I wasn't clear.

What I am trying to understand is......Does high pixel count lead to soft images.

1. could be more sensitive to hand shake.
2. could be its reaching the resolving power of a lens (you can see mine in my sig)
3. could it be that there is an inherent problem similar to diffraction

I don't know, that's why I'm asking.

The 1Ds owners have had this high pixel count for some time. They would know the most about this subject.

21.1Mp will bring in more detail than 12.8Mp on the same FF format. It's by no means softer. It's sharper.
Try uprezing a 5D image to 21.1Mp and you'll see the real softness of that.

please read this:

http://www.fredmiranda​.com/forum/topic/68897​6/0

main point: a 21mp will show the lens's defects more readily than a 12mp sensor, but once you downsize the 21mp to 12mp, you can expect at the very least the same results as a 12mp camera, nothing worse. if you want to compare details, you need to upsize the 12mp camera to 21mp, or downsize the 21mp to 12mp.

I would think that the resolving power of the lens used will be the biggest factor. You can't use a cheap lens, or your output will be garbage.

davidfig wrote in post #6358452
Maybe I wasn't clear.

What I am trying to understand is......Does high pixel count lead to soft images.

1. could be more sensitive to hand shake.

Certainly, at 100%, any shake is more evident at higher resolutions.

As per my earlier post, it will vary with the lens, but will not create a problem -- as above, if viewed at 100% you will see more lens softness over a lesser-resolved image if viewed at 100%

I don't know about the 1Ds, but the 5DMII has takens steps to keep edge softness down. Again, there should not be a loss at higher resolutions as long as you do the "real world" comparison -- in other words, softness will show up more at the 100% level but overall softness and detail should never be worse than the lower resolution image, unless the design is worse.

I have not heard of a 1DsMIII owner complainging about a net loss of sharpness (that is, compared to a full frame image from, say, a 5D when viewed at the same size). On a 100% per-pixel level, I would expect any shake or lens softness to show up more, but even then it would be about the same as the 8 MP 20D/30D, both of which handle detail quite well.