30D sharpness 100% crops - DPChallenge Forums

Canon EF 28-135mm f/3.5-5.6 IS USM

04/19/2006 10:36:30 AM · #1

Hi. I was wondering if anyone was interested in posting some 100% crops to help me understand my camera better.

I have finished reading my manual the first time, but I haven't started playing with the picture styles to decrease sharpness, so I'm still shooting stock standard mode.

I am finding some jaggies and I'm just not sure how sharp sharp really is on the 30D. I would like to know so I can complete my initial learning process.

I am working a lot right now, so don't have time/opportunity to go shooting :(. Need a wider lens too...

I will post a picture myself in a bit, but I am still experimenting and I can't say for sure what is what yet.

I do most of my shooting in 4-5 minute segments before heading out to the next job... Gotta pay for the durned thing :)

Prof_Fate

04/19/2006 10:43:20 AM · #2

As you move up the food chain model wise in teh Canon camera world, the out-of-camera images get softer (on purpose). My 30D is much softer than my 300D - so I took the standard setting and upped the in-camera sharpening one notch.

is this the right thing to do? I don't know (or really care)- i like it sharper than average.

Also, the first step in post processing is to run a USM of 300%, .3, 0 (so says canon on their tutorial site). this does help a LOT, but it depends some on the image subject.

Have you tried the b&w? Shoot it RAW and you can get the color back, but you can tint it(sepis, purple more), add filters (red, green, etc) - just neat as hell IMO.

You show a 80-200 lens - keep the shutter up at 1/250 or more (at 200mm anyway) to make sure camera shake is not the culprit.

here are 2 shots at 100% from my 30D - Tamron SP24-135 lens. First is as captured, the second with the USM i mention above:

Message edited by author 2006-04-19 10:45:10.

ragamuffingirl

Sony Alpha NEX-5N

04/19/2006 10:47:27 AM · #3

Why would the images get softer on purpose? I thought soft was a bad thing? The second picture is how I want it to look off-camera.

Message edited by author 2006-04-19 10:48:39.

Canon RF 24-70mm f/2.8 L IS

04/19/2006 10:51:50 AM · #4

That's actually high on my list of things to try. I wondered what that purple tint thing is all about... Is that a UV light replicant of some sort?

I am more concerned about those jaggies than anything.

Does anyone know what the magnification is at full zoom in the review on the camera LCD? is it 100% or more? If it's like 300% mag, then it would make sense that it would be showing jaggies. Otherwise....

I will try to punch the sharpening up a notch just for fun. I will move it back down later once I get a hang of the focusing thing a bit better. I am trying to get a reasonable handle on the basics to the point of comfort before getting into things like RAW workflow and fiddling with Picture styles... Of course I'm still going to poke around and have some fun though :).

So far, I'm totally blown away by the focusing. AI Servo was BEYOND cool when trying my friend's EOS 1n at that model air show, and it's flippin' nifty that I have the same function on my 30D (or so I think).

Next purchases: Kenko 300 Pro 2x TC to punch my lens up a bit in length. Fun with birds! Swallow season is just getting started here and while I don't like the look of the swallows here compared to the birds that used to fly nose to nose with me when I was biking to work early morning in Canada, they are still a favorite of mine and a MAJOR challenge for shooting. The best I can hope for right now is a nest/sleeping shot...

kirbic

Canon EOS R5

04/19/2006 10:55:01 AM · #5

Your current lens should be fine for playing with sharpness tests. Just use a solid tripod and ideally a remote release (or use the timer) and mirror lockup. Shoot a subject with fine, high-contrast detail. Use f/8 or thereabouts to minimize the effects of the lens. Shoot in RAW, so you have total control of sharpening in conversion. Convert initially using no sharpening at all, and look at your detail at 100%. Now fiddle with sharpening. The 300, 0.3, 0 recommendation is a decent place to start, but try using 3 for threshold instead of 0; you'll avoid "sharpening noise."
The above should give you a good feel for the out-of-camera image sharpness, and how to best post-process to achieve the best final result.
One comment about the Canon lineup; it's not always true that the higher in the lineup you go, the softer the RAW images are. The 5D, for instance, requires MUCH less sharpening in post, due to a less agressive anti-alias filter. The tradeoff is potential for some minor moire under certain conditions.

04/19/2006 10:57:35 AM · #6

Rag, I'm no expert on these things obviously, but two things jump to mind immediately as to why things would get softer. First, there is an 'anti-alias filter' that can be helpful as well as the fact that most people with more expensive cameras have a greater interest in PP.

The bottom line is that the camera can do a nice job, but generally PP can do it better.

Lower end camera users also often do not shoot in RAW. Shooting in JPG makes more sense when using the in-camera processing because a fair bit of it is done to the image BEFORE compression, so at least it works with as much information as possible before it is thrown away. If you shoot in medium compressions/image sizes, it's not uncommon for a number of camera features to become more sensible (like digital zoom, sharpening etc). 30D users aren't usually after this type of shooting.

04/19/2006 11:01:25 AM · #7

Thanks Dr. Kirbic :). I was thinking about what to do to do image testing. I will get on the f/8, 135mm ISO 100, 2 second timer, mirror lockup test ASAP. I am embarrassed at how few of these things I had actually thought of myself...

My version of that test would have been f/5.6, 200mm, ISO 100, 2 second timer w/o mirror lockup. I might actually use the 85mm f/1.8.

Thanks for covering my bases for me!

Inexperience is a dangerous tool when wielded....

Message edited by author 2006-04-19 11:02:44.

Prof_Fate

04/19/2006 11:10:10 AM · #8

Some (quickie) ISO tests...to better learn the camera

And the 2X TC will soften things a bit. I did some tripod tests and it's not bad, but I went to the Zoo and actually use the 2X in the real world, as well as a 1.4 and none at all - i can see the softness of the 2X when i look at all the pics.

04/19/2006 11:16:39 AM · #9

Odd, in your ISO test shots, the 3200 ISO seems MUCH sharper and higher contrast than the other shots... camera shake eliminated? handheld? what happened there?

You are referring to the Canon 2x TC or the Kenko? I've used the Kenko a bit with a friend and although I haven't had a chance to look toooo closely, I haven't been able to see any serious softness issues. I have been getting the impression lately that the Kenko 2x is better than the Canon...

It's certainly more versatile. It seems to have less of an aperture drop too possibly? My friend uses his with a Canon 75-300 f?? - ?? and I can't imagine that at full telephoto there would be any use of AF left. He seems to indicate that the camera doesn't have any problems or issues at all with using it... He tends to be a bit loose with his facts though... I intend to borrow and use before I buy...

Prof_Fate

04/19/2006 12:33:55 PM · #10

The canon TC has a lip or protrusion that limits what lenses you can use it with. I have the regular tamron converters.

Here are some shots i took with my 300D and 70-210 2.8 Tamron for converter testing.
BOTH converters (1.4 and 2.0!)

the 2.0 alone

the 1.4 alone

none at all

and my rebel did AF will all of it, even with the converters stacked!

On the ISO tests - all handheld, nuthin fancy or scientific. The 3200 one seems sharper to me too, but when you run it through neat iamge it'll soften up ;P
3200 is supposedly a contrived ISO setting, that's why it's listed as H on the LCD but it shows in the exif as 3200. The 5D has a setting below 100 called L on the LCD, or 50 everywhere else.

Here i did some PP on the ISO800 image.
Look at the distant arm and near side of the face to see things sharpen up with USM but get noisy, So i did a NI bit and it cleaner, but softer, so then i did a USM on the NI version. I just did the 300%, .3, 0. These were resized for the web. Normally i do one more sharpening for web display (often 45%, .9, 1) but didn't on this image.

Message edited by author 2006-04-19 12:35:10.

Mr_Pants

SiPix Blink

04/19/2006 01:12:33 PM · #11

Originally posted by eschelar:

It seems to have less of an aperture drop too possibly? My friend uses his with a Canon 75-300 f?? - ?? and I can't imagine that at full telephoto there would be any use of AF left. He seems to indicate that the camera doesn't have any problems or issues at all with using it... He tends to be a bit loose with his facts though... I intend to borrow and use before I buy...

You would still lose two stops, no matter what. The focal length has doubled and the aperture has remained the same size. With a 75-300, it's possible that the effective focal length is not reported to the camera, so it tries to autofocus anyway (much like the old 'tape the pins' trick). In good enough light, it should be possible.