S100 Image Quality Comparison

First -- This isn't really a fair test for the Canon S100. Considering its I.Q. will be compared to a DSLR camera and EF lens costing much much more.......

One would expect the Canon 550D coupled with the Canon EF 50mm F/1.4 lens wouldn't be a contest, but in these first tests it appears the S100 is doing a good job of holding its own.

The first setup isn't scientific, and isn't down to the milli-meter, but it can give us an idea as to how much image quality we can expect from a static shot, indoors, under normal lighting conditions. The EXIF data has been omitted to keep the peakers from magically guessing which picture goes to which camera .

First Shot: Fireplace Mantel

Shot A ----- ISO 100 White Balance set to Cloudy

HTTP response: NOT FOUND | MIME changed to 'text/html'

HTTP response: NOT FOUND | MIME changed to 'text/html'

HTTP response: NOT FOUND | MIME changed to 'text/html'

HTTP response: NOT FOUND | MIME changed to 'text/html'

Might be a tad of oversharpening in both of them, from the look of the edges of the rose in the crop. And it looks like their "Cloudy" WB settings don't exactly match; the first looks to be slightly more aggressive (bluer). But based on the relative noise levels and the slight OOF in the crop of Shot B, I'd say that B was the 550D.

More details....

The S100 was shot at approximately 80mm f/5 (wide open)
The 550D was shot at 50mm f/4 (stopped down to f/4)

In case you're wondering why the S100 was shot at 80mm and the 550D at 50mm. This is because the S100 @ 80mm is 80mm on a 35mm format. This is in stark contrast to the 550D which has a 1.6 crop factor in comparison to 35mm. So, to compensate or get the same framming, one must divide 80 by 1.6 to get 50mm. By doing the 1.6 conversion this helps to ensure we have the same framming of the Fireplace Mantel photo.

David

That would be 80 mm equiv. on the S100; that's really closer to 17 mm actual f. l.

B definitely looks smoother and more contrasty to me, so I would guess that is the 550. I *hope* I am wrong though because I own an S100 and would love it if it was the better of the two here.

Jon wrote in post #13667809
Might be a tad of oversharpening in both of them, from the look of the edges of the rose in the crop. And it looks like their "Cloudy" WB settings don't exactly match; the first looks to be slightly more aggressive (bluer). But based on the relative noise levels and the slight OOF in the crop of Shot B, I'd say that B was the 550D.

Yes, Crop B is out of focus. I didn't realize this when I first set up, but the tripod and camera are not completely perpendicular to the fireplace. It seems objects to the right hand side are more in focus where extreme left hand side of the frame is not. That's what we get when we aren't paying attention... doh!

Not sure why we have jaggies.... considering the sharpening was fairly minimal in PSP.

Jon wrote in post #13667855
That would be 80 mm equiv. on the S100; that's really closer to 17 mm actual f. l.

That would be correct... Having said that though... I was going by the read-out on the camera LCD which seems to automatically convert 5.2mm - 26mm to a readable 24mm - 120mm respectively.

Appreciate your comments Jon.

David

michaelnel wrote in post #13667912
B definitely looks smoother and more contrasty to me, so I would guess that is the 550. I *hope* I am wrong though because I own an S100 and would love it if it was the better of the two here.

"B definitely looks smoother"

You have touched on a reason why sensor size makes a difference. I'll leave it at that for now.

David

davidc502 wrote in post #13668036
"B definitely looks smoother"

You have touched on a reason why sensor size makes a difference. I'll leave it at that for now.

David

Yes... I know that. B is smoother, and that is the main reason I think it is from the camera with the larger sensor.

If you had included images from a FF or 4x5 digital back, they'd be smoother yet.

I think the increased contrast is due more to the wonderful 50 f1.4 lens than the sensor though.

That said, in the full size images, most people wouldn't see much difference, especially if they didn't see them side by side. Either one would be considered acceptable to most.

michaelnel wrote in post #13668144
Yes... I know that. B is smoother, and that is the main reason I think it is from the camera with the larger sensor.

If you had included images from a FF or 4x5 digital back, they'd be smoother yet.

I think the increased contrast is due more to the wonderful 50 f1.4 lens than the sensor though.

That said, in the full size images, most people wouldn't see much difference, especially if they didn't see them side by side. Either one would be considered acceptable to most.

I think you bring up some very interesting points! I agree in that most people are not going to see a difference. This has been my argument between the 550D and the 5Dmk2 as well as a few other more expensive DSLR's.

Last comparison here, but this time it is lanscape. Both of these photos looked sharp on my screen, but after I downsized them to 1024X623 lost sharpness. It's been a while since I've done landscape, so it's probably operator error on my side.

Both photos were taken at F/5.6 ISO-100 WB = Cloudy and 24mm on the 550D and 38.4mm on the S100. I don't know what the exact/proper focal lenght would be on the S100, but that's what the camera is telling me.

Other than a light sharpening in PSP, no other adjustments made to the photos.

Photo A

HTTP response: NOT FOUND | MIME changed to 'text/html'

HTTP response: NOT FOUND | MIME changed to 'text/html'

michaelnel wrote in post #13668144
Yes... I know that. B is smoother.

Sorry, if I'm coming off like mr. know it all. Hope no one is taking it that way. Just trying to show some side by side photos between two different cameras.

David

In those two, I can't pick. I do see they have slightly different tonal qualities, with the second one being warmer. But I don't see IQ differences that would allow me to select which camera.

However, based only on the warmth, I would choose the B as being the 550, since the previous example had a warmer rendition for the 550.

This one is MUCH harder to choose though, and it's more like the pictures most casual shooters would shoot, so it shows there isn't a lot to be gained by humping around that DSLR if you are going to make snapshots like these.

davidc502 wrote in post #13668320
Sorry, if I'm coming off like mr. know it all. Hope no one is taking it that way. Just trying to show some side by side photos between two different cameras.

David

Naw, I didn't take it that way. I am enjoying this thread.

Photo B also seems to have more even illumination, which would fit with the 50 1.4 on the crop body.

However you really need to remember that unless you're only going to look at your photos on a computer, cell phone, or other relatively low resolution device, a 1024x768 pixel image isn't a fair test. Most commercial printing operations will do around 240-300 dpi, so even at the low end a typical 4x6" print will have on the order of 960x1440 pixels; and a 5x7 that people will put on the wall or a desk will be between 1200x1680 and 1500x2100 pixels. There you'll start seeing things like sensor noise, CA, and other weaknesses showing up, regardless of the camera/lens combo.

If your 50mm f1.4 is like mine, the AF isn't always accurate. So for the first group I'd guess B is the T2i/550D.

For the 2nd set, the trees that are further away seem sharper in B. I'm guessing B is the S100.

100% crops out of the raw files with no PP would be interesting to look at.

But the DSLR with that 50 f1.4 will likely win every time.