May 19, 2009
Hello all. I just upgraded to the 50D from the original 300D digital rebel. I absolutely loved the 300D but was approaching 10k pics and happened to be thinking about upgrading when I saw Canon had come out with the 50D. First let me beg that someone with more experience than myself will tell me that I did not make a mistake! LOL This camera cost lotta $$$. The reason I ask is that I took her out this weekend playing around shooting anything from birds, flowers to my little boy crawling around some pine trees gathering pine cones. I have the 28-135 IS USM lens. Out of all the pics I took in various lighting conditions I did not get any that were tack sharp when I blow them up to 100% in Photoshop. I have taken far better pics with my old faithful 300D. Now that I have rambled on I guess my question is this. Are there any settings with this model that I should check that could be causing this or is it just operator error needing more practice with it? The 28 - 135 IS USM has been a good lens on my 300D so I have discounted it. Any advice or discussion tips would be greatly appreciated.
I don't have a 50D, but I can tell you a couple of things that might help.
First, if you're comparing 100% views of a 50D image against a 300D image, remember that you're viewing the 50D image, zoomed in 1.6X as much. The 15 megapixels of the 50D are squeezed into the same space as the 6 megapixels of the 300D.
But when you zoom in to 100%, you're viewing the same number of pixels on your screen with images from either camera. So you're looking at the images "blown up" farther when you view a 50D image than when you view a 300D image.
So that's not a fair comparison. Assuming the lens was the same and the focus was the same, etc., then the 50D image should look softer than the 300D image because you're scrutinizing it more closely. Any blur from the lens, or missed focus, or camera shake, etc., should look 1.6X as blurry when viewed at 100% on a 50D image than a 300D image, actually. It'll look blurrier simply because you're zoomed in on it quite a bit more.
So you can't fairly compare the two cameras viewing at 100%. You need to compare equal sized prints or equal sized views on your computer.
Next, you could have focus issues. Have you played with the 50D's Autofocus micro adjustment? That could help. It may be that that particular lens, when used on your particular 50D is not dead-on for calibration.
It's also possible that because you're viewing the images at a higher "magnification" with the 50D, it's revealing faults in the lens that you simply could not see with the old camera. Make sure you compare equal sized images from the two cameras.
Finally, are you shooting in JPG mode or RAW? Sharpening is something that has to be fine-tuned for different cameras. It's very possible that the default sharpening used by the JPG "engine" in the 300D employs fairly aggressive sharpening.
And the 50D probably sharpens a lot less by default because they figure you'll be more likely to want to process your own images when shooting with a 50D than with a 300D. So the 300D is set up to please people right out of the camera while the 50D probably sharpens less aggressively because that makes the images more flexible and better suited for further processing.
Sharpening should almost always be the very last step in image processing. With the higher-end cameras, they tend to use less sharpening in the camera itself in order to allow you greater flexibility later when you process the images.
To really see the best from the 50D, I'd recommend shooting in RAW and doing no sharpening until after you've resampled the images to their final pixel size. It's amazing how much of a difference that makes.
Thanks for the great info. I will make sure I am comparing similar size prints and see if the difference is still the same. I have been researching other posts and notice there are several others talking about the Auto Focus microadjustments. I have not played with that at all yet. This very well could be the culprit. I will have a chance to play with that tomorrow and let you know the results.
...for maximum sharpness at 100% view with the 50D with any lens, keep your ƒ-stop at or under ƒ 8. ƒ 7.1 or ƒ 5.6, for instance. As a general rule, this prevents diffraction from adding to per-pixel softness.
Your lens has to do a better job -- resolve tighter -- with the 50D because the sensor elements are smaller than your other camera. This can also lead to softness at 100%. See if you can buy or borrow an 85mm ƒ 1.8 Canon lens; at about $400 retail, they're one of the sharpest lenses Canon makes anywhere near this price range, and will show off the 50D's ability to resolve detail very well.
Also... have fun.
Check out the EF-S 60mm f/2.8 Macro. I doubt that Canon makes a sharper lens. And it's relatively inexpensive.
So you get macro capability, a great "portrait length" lens, and have what might well be the sharpest lens you'll ever own.
So if you've got any interest in macro, and if you're looking for a great prime in the 60mm focal length (equivalent to 100mm on a full-frame body), I heartily recommend the EF-S 60.
That's one lens that will show you what any camera is capable of.
The 50D has the finest pitch sensor of any Canon DSLR. If a full frame sensor using the 50D technology would be over 36 mega pixels.
Early on with my 50D is lost a lot of images to motion blur. Imagine motion blur first occurs when pixel level detail in an image moves from one sensor pixel to the adjacent pixel during an exposure. This phenomenon can only be seen at 100% crop but it changes the rules for taking a picture if you want to avoid it.
Basically the extreme fine pitch senor imposes a new hand held exposure rule of thumb. Instead of shooting shutter speeds of no less than 1/focal length the 50D shooters need to use a rule of thumb 1/(focal length x 2).
We are talking motion blur so there will be times when an image taken at a slower shutter speed just happened to come in sharp but on average you might conclude the 50D is soft if you’re looking a 100% crop and don’t reign in motion blur to a new level. I took a long sequence of HH 200mm shots at a 1/60 sec last month 6.3 fps and several were tack on sharp while others wouldn’t stand a close look.
50D - Full equipment list in profile.
In my opinion, you really won't get the best that the camera can offer.
You're locked into the processing settings that you have dialed into the camera at the time of the shot. If you shoot in RAW, you can choose how to process the image later.
And one of the great benefits of that is being able to avoid any sharpening until the proper point in the processing (which is at the very end).
Shooting in JPG, the camera will sharpen the image (according to what you've got the in-camera sharpening set for). So that means that further sharpening done by you to the shot will mean that you're "double sharpening" the images.
And sharpening an already sharpened image can lead to nasty artifacts.
So if you must shoot in JPG mode, I'd recommend setting the in-camera sharpening to its lowest setting (some kind of negative number, I think).
Then, after you've resampled your images to their final size, apply some "smart sharpen" in Photoshop while viewing at 100%. You'll be amazed at the results.
The camera can only sharpen at its native image resolution because that's all it has to work with (unless you shoot in one of the reduced-resolution modes).
Don't be afraid of trying RAW.
Try setting the camera to shoot in RAW + Large-Fine JPG. That way, you've got the out-of-the-camera JPGs just like always, but you've also got the RAWs for all of your shots.
Then, when you've got some time to play a little, install Digital Photo Professional. It's a free RAW converter program that comes with the camera.
You can then play with converting your RAW files.
You'll notice that you've got the same processing settings as the camera has, plus some additional ones. And the beauty is that now, with the RAW data, you can play with all of those settings while viewing the result on the screen.
So you can play with the same image, seeing what it looks like with different settings for white balance, picture style, contrast, color tone, color saturation, and even sharpness.
That beats the heck out of needing to predict, before you even take the shot, what settings you'll want for each and every shot. In my opinion, that makes shooting a LOT easier because I don't need to worry about any of those image processing decisions while shooting.
I can concentrate on composition, exposure, focus, and capturing the moment while shooting, and leave all of the image processing for later.
Shooting in JPG, you have to make the image processing decisions before you take the shot. That's backwards, and makes shooting so much harder IMO.
And most importantly for this discussion: Shooting in JPG means that you're stuck with the sharpening you had set up in the camera before the shot was taken. Further sharpening can then cause problems. So you're locked in.
If you've allowed the camera to save the RAW data, then you'll have it forever, and you can process the image exactly how you want to get the effect you want for that particular use of that particular image.
The RAW data is never destroyed, so you can actually make multiple conversions of that same RAW image data, each one using a different processing "recipe".
If you're making a final image for web posting or emailing, you can reduce the size of the image and then sharpen at that reduced size to get the best possible look on your computer screen. Or, from that same RAW image, you can process the image to optimize it for a huge print. Different processing to get the best results for different end applications of that very same shot! Cool
When I first got my 20D, I was initially disappointed that I could not get images as sharp as what I got from my $200 6 megapixel point-and-shoot camera. It turns out that my P&S did a LOT of in-camera sharpening, and the 20D was far "lighter" in its sharpening.
Once I got more used to processing my 20D images, I realized that I could do so much more with it, and the images were, indeed, better than what I could get with the P&S. But initially, I was disappointed in the out-of-the-camera sharpness.
The more consumer-friendly the camera, the more they tend to sharpen. That's great for times when you don't want to do any image processing, but it's bad for times when you do want to do your own processing. The higher-end DSLRs tend to cater more to the case where you'll be wanting to do your own processing, so to that end, they do less aggressive in-camera processing.
That's a good thing, but it means that you are somewhat forced to process your own images from these cameras to get the best results. Fortunately, it's fun and relatively easy
You could also try just cranking the in-camera sharpening up to a higher level for the sake of your out-of-camera JPGs, but be advised that the resulting JPGs, while looking sharper when viewing at 100% on your computer screen, may suffer from more artifacts if and when you process them further. Sharpening is something that can't be "undone", so while those images might look great upon initial 100% inspection, they may cause problems down the line if you do additional processing.
But here's one thing to remember:
The in-camera processing settings do NOT affect the RAW image data. They do affect the JPGs from the camera and those settings are used by DPP as its "as shot" processing settings. But you can change those settings to anything you want when you process the RAWs.
So you can shoot in RAW + Large-Fine JPG, play with the in-camera settings all you want to see how they affect the JPGs, and never have to worry about spoiling the RAW data. Thus, you can experiment with the in-camera sharpening but still have the RAW data untainted by those experimental settings.
RAW is so handy!
At the "image level" (comparing equal sized prints from a 6 megapixel camera to a 15 megapixel camera, for example) the images would be identical for motion blur.
Diffraction effects, camera shake, and lens focus/quality do NOT cause images from a 50D to be worse than images from, say, a 300D for the same shooting conditions.
It's only that the 50D is able to resolve finer details than the 300D. So in order to get the best image that the 50D is capable of, you do have to be more careful.
But if you were to use the same lens, shoot the same scene, and use the same settings, the images, from the 50D would make as good or better prints of the same size as the images from the 300D shot under identical conditions.
In no case will the extra resolution of the 50D cause you to get a worse image than what you'd have gotten with the 300D with the same settings, lens, and shooting situation.
It's only because we tend to view our images at 100%, and thus, view the images from the higher resolution camera at a higher magnification, that we're fooled into thinking that the higher resolution camera is somehow, more "sensitive" to these problems.
The real truth is that those problems are present to the very same degree in our images from the lower resolution camera. It's just that we are not scrutinizing the images as closely because 100% viewing does NOT represent a constant magnification level.
And further, because of its higher resolution, the 50D could make better images than the 300D if everything is perfect. So the 50D has the potential to get better images. But it won't get worse images given the same shooting situation.
Comparing equal-sized prints or downsampling the 50D images to six megapixels for the comparison is the only fair way to judge this.
So again, it's only important to use a new rule of thumb for shutter speed IF you're now going to actually make prints that are 1.6 times as large from the 50D as the prints you were making from the 300D shots. That's not likely or a fair comparison if you see what I mean.
Imagine how large a print would be if you made it to match the image size you're viewing when you look at your shots at 100% on your monitor. Set your view to 100%, then pan around over the image while imagining how large that print would be!
Do that for both a 300D image and a 50D image. See how much larger the 50D image would be?
You need to set your view to 63% for the 50D image if you're going to compare it to a 100% view of a 300D image.
A camera can't really be "soft." A lens can. The new camera is simply showing you the shortcomings of your old lens. Try printing out an A/B comparison at the same size.
Either that, or your new lens/camera combo has a focusing issue that didn't show up with your old camera due to tolerances and stuff. You can now play with liveview and it's AF to see if the sharpness improves significantly.
The 50d with the appropriate lens is a killer camera. You can see at 100% crops details no other camera will provide at the moment.
Example and 100% crop taken with the EF70-200 f4L IS at 200mm, ISO160, f8, 1/500.
You get details like individual branches on the trees on the glass reflection, etc. I think even a 200% version is usable.
There is no softness to the 50D. period. Set it up right and get better taking the pictures with the camera. I've got 6000 shots since October and am just getting better and better. And that's sharp at 100%. Almost all of the shots are perfect at 13x19 printing let alone smaller prints.
Also be aware of the strengths and weaknesses of the camera. I find the 50D creates tack sharp images without any problems when there are clear contrast edges for the AF to pickup.
With the above settings mentioned (micro af adjustment is a MUST), I feel that the sharpness setting in the profiles is more of a "softness setting" in reverse. I like it better with the setting at 7.
here's a photo with no sharpening, blown up to 100%.
50D Canon 24-105 L IS Canon 70-200 L IS Canon 50 1.8 Canon 2x converter
Basically the extreme fine pitch senor imposes a new hand held
exposure rule of thumb. Instead of shooting shutter speeds of no less
than 1/focal length the 50D shooters need to use a rule of thumb
1/(focal length x 2).
Using my 50D with a 100-400mm IS L (hand held) will give me sharp images at around 1/800th second. And that is with IS on. So the "focal length x 2" bit might be a bit optimistic at least for long lenses photographing things a bit away from the camera.
Thanks for the info and it just so happened that I had been reading earlier about RAW so I set my camera to RAW + JPG L. Ok the first thing I did was gasp and laugh when my 2GB Sandisk went from 365 pictures to 60. LOL However off my daugher and I went to her soccer practice where I shot everything from the kids to Halls cough drop wrapper on the ground. I am playing now. I did play a lot today and have some much better shots but am looking forward to playing with RAW and learning more of the capabilities. Again thanks for the information.
Those are great shots. Nice to have L glass huh. Not in my budget right now since I just purchased camera but I am saving my pennies!!!
It was a good lens but not a wow lens. I found it best on the long end. Even the budget 50 1.8 prime will kick the snot out of your current lens for sharpness.
When it comes to a much better resolving sensor like the 50D, glass and shooting skills are everything. Next is your ability to do top notch PP work.
Google Canon Micro adjustment and you will get all the advice you need to attempt it. You may find that it helped or you may realize that your OEM adjustment is perfect as it is now.
Cat's eye at 100%, 70-200 f4L at 135mm and lighting via a single 430EX and Cactus V4 on a pod.
bettagl wrote:-- hide signature --
Those are great shots. Nice to have L glass huh. Not in my budget
right now since I just purchased camera but I am saving my pennies!!!
...try calibrating your lens to your 50D with the micro-adjustment feature for focusing through this link (scroll down just over half way through the web page):
5% lighting, 5% composition. 90% location. Get there.