So I received a 50D + 18-200 kit lens last week and based on the sample shots I've been taking (posted some in my prior 18-200 shots thread), I haven't been impressed with the sharpness I've been getting from the camera. But I thought that was just a problem with the lens. However, after mounting my 35 f/1.4, I'm still getting shots that just do not seem sharp to me. Shots that I'm getting seem to be fuzzier than I'm used to coming from the 5D. So I took a quick comparison shot with the 35 on my 5D and 50D and this is what I'm seeing. These are 100% crops. Focus on the DANT lettering of the word antioxidant.
Both photos were taken with the same aperture, ISO and shutter speed (1/125s, f/1.4, ISO400) and I tried to frame the shot similarly to make up for the crop factor. Besides being grainier, the 50D shot to me just looks softer overall. I do not believe it's a focusing issue as I tested focusing and it was fine on both bodies. So am I just expecting too much from the 50D, or am I missing something or is there just something wrong with this particular body?
It would be more representative if you took these shots stopped down a bit instead of wide open. You can see the obvious DOF difference between a FF and 1.6x camera, thus it's difficult to tell which is sharper (or softer).
I already tested for front/back focus and didn't find a problem with it. With the sample shots above I shot them at an angle so if there were front/back focus, you'd be able to see it since either the letters behind the DANT would be more in focus or more of the MANGO word on top would be in focus. As for lighting, I guess I could put on a flash but I shoot a lot in low light w/o flash so if I have to look at results like the 50D constantly I don't think I can take it.
Well, the 5D is pretty tough competition for any camera WRT image quality. I had the same impression last year when I tried the 40D (I ended up returning it). I had to sharpen the 40D images more aggressively to achieve the same apparent sharpness, and this exacerbated the coarser structure of the 40D files. Even the Nikon D3 and D700 suffer (slightly) by comparison with the 5D at lower ISOs.
Keep in mind that the 50D has more pixels, so 100% comparisons do not necessarily tell the entire tale. It may show better sharpness once image sizes between the two cameras are equalized for output. You might try doing some upsample/downsample testing to see.
This isnt a fair test. You are bound to be stretching the lens performance more with the pixel density of the 50D than the 5D. The only way to make a fair comparison between true sharpnesses would be to use a lens which you know outresolves the 50D and shoot at f8 or similar so that everything is as sharp as it could be.
This test proves nothing other than the fact that the 50D given the same lens (soft) lens as the 5D will appear to show worse sharpness, which is afterall what you would expect.
Hi Gary, yup I considered the more pixels thing with the 50D but downsizing the 50D sample to the same size as the 5D shot before taking the crop showed negligible difference unfortunately. Your description of your impression of the 40D definitely sounds similar to what I'm currently going through with the 50D.
I think you're right: the 50D shots are not as sharp as one would hope. Did you turn on sharpening in-camera? I routinely use +4 sharpening on my 40D (I shoot only .jpegs) and get good results. Without that added sharpening in camera I was disappointed.
Also: if you want to isolated sensor issues, stop down to f8, use a tripod and good light for your tests. The 35 1.4 may in fact perform differently on the 50D, plus 1.4 is too large an apeture to rule out focus (as opposed to sensor) issues.
Alex Nail wrote:
This test proves nothing other than the fact that the 50D given the same lens (soft) lens as the 5D will appear to show worse sharpness, which is afterall what you would expect.
Wait, so my 35L is not a sharp lens? Dang, I thought it had pretty decent sharpness wide open prior to this. =) But anyway, so I can expect to get softer photos on a 50D using the same lens wide open than on a 5D? I was expecting worse high ISO low light IQ but not sharpness.
24Peter wrote:
I think you're right: the 50D shots are not as sharp as one would hope. Did you turn on sharpening in-camera? I routinely use +4 sharpening on my 40D (I shoot only .jpgs) and get good results. Without that added sharpening in camera I was disappointed.
Also: if you want to isolated sensor issues, stop down to f8, use a tripod and good light for your tests. The 35 1.4 may in fact perform differently on the 50D, plus 1.4 is too large an apeture to rule out focus (as opposed to sensor) issues.
Both photos were originally shot in RAW and saved as JPEG via ACR 4.6 beta with no sharpening or any other PP applied. Sure, I can add some sharpening PP but I wasn't expecting such a difference between the 2 bodies with no sharpening applied.
I guess what I'm really curious to know from the *0D owners out there is that should I be ok with this low light wide open result from the 50D? I haven't used a *0D class body in awhile so I don't know. I knew the 50D shots would have more grain and noise but this looks to me more than just a difference in grain and noise but obviously I could be wrong.
Photos of bottles, brick walls, cereal boxes, etc are not cause enough to get worried. Please wait till some real-world samples come in - with better lighting, better exposure, etc.
The word "Mango" on the 50D is sharper than that on the 5D, especially the G and the O. I know you checked the AF aspects on both cameras and you did not see any issues. However, on that particular shot, it looks like the 50D was front focusing ever so slightly. Your test is bordering on pixel peeping activities and I believe that you won't see the difference under normal real life shooting conditions. But hey, whatever floats your boat .
kenshin wrote:
Wait, so my 35L is not a sharp lens? Dang, I thought it had pretty decent sharpness wide open prior to this. =) But anyway, so I can expect to get softer photos on a 50D using the same lens wide open than on a 5D? I was expecting worse high ISO low light IQ but not sharpness.
Well its clearly not critically sharp, which it would need to be for this to be a fair comparison between bodies.
Since the 50D has a greater pixel density, a 100% crop is using a smaller area of the lens (in really simple terms) than the 5D so it is going to show its imperfections to a greater degree.
kenshin wrote:
But anyway, so I can expect to get softer photos on a 50D using the same lens wide open than on a 5D? I was expecting worse high ISO low light IQ but not sharpness.
Yes. The 50D pixel density is way too high.
So high, in fact, that with most commercial lenses (especially when they are wide open), you won't see more detail than 12mp cameras - but the photos will look softer, because of the extra (empty) magnification.
I got flamed for these statements but here they are again: https://www.fredmiranda.com/forum/topic/693849
gml1 wrote:
Yes. The 50D pixel density is way too high.
So high, in fact, that with most commercial lenses (especially when they are wide open), you won't see more detail than 12mp cameras - but the photos will look softer, because of the extra (empty) magnification.
Not flaming you but not buying it either.
The examples we are seeing are not lab samples. Just regular Joes like us shooting and posting. If folks are looking for HD quality in 100% crops they will be looking for a long time or had better go to point and shoots.
The new DPP has a default luminance NR setting of 2 so if some of us are not aware it will throw us off. Back it off to 0 and see better results.
I love seeing the submitted shots here. I do keep in mind they come from all kinds of conditions. Gotta love it.
Will, I see you have a 1DIII and are using the 50D for these shots. I am wanting to get something a little faster and better ISO for nature then my 1DsMKII to use on my 500 f/4. Do you have any input on the 50D Vs, 1DIII? It's half the price and I don't shoot much so spending the extra money for the 1DIII requires some thinking.
I'm surprised no one has mentioned the weaker AA filter of the 5D. Out of camera unprocessed shots have always been sharper with the 5D than any 1 series I have owned except maybe the 1D and 1Ds. I certainly don't see the problem with the 50D image and yes it does look like it is front focusing regardless of your test.
and I believe that you won't see the difference under normal real life shooting conditions. But hey, whatever floats your boat