30D vs 5D

Hmmm,
Spend about $2800 for a 5D and a 17-40L or $650 for a 10-22 to shoot landscapes in ultrawide angle? I'd say if you print larger than 8x12 or unless you like the bigger viewfinder and are willing to pay over $2000 premium for it, no. The biggest difference between the 5D and 20D in landscape will be in the filesize, while pixel peeping this maybe a big deal, on normal sized prints, it won't really show up. According to Practical photography, 87% of people including photogs couldn't tell which print was made with a 1DsMII or a XT, and this is between a 16 v8MP.

jevidon wrote in post #2364777
i began noticing this while shooting with my buddies 5D for my trip to the Grand Canyon, but it's a simple fact of life...and IMO i would much rather have a FF camera and have to deal with vignetting.

Depends on what you shoot. I add it to some of my stuff and I really like it........but I make that choice. For my purpose there is no way it could have worked, I shoot Vitural tours that are stills merged, I got a bad dark line were every frame came together. And there is no way I was shooting raw and processing each frame to make it go away. And sure I could have stopped down but my 580 doesnt have the power to light a 20X30 room at F16......unless I'm at like 1000 ISO........and then I'm running each image though noise software. Anyway you sliced it, it was not good.

SeanH wrote in post #2364858
Depends on what you shoot. I add it to some of my stuff and I really like it........but I make that choice. For my purpose there is no way it could have worked, I shoot Vitural tours that are stills merged, I got a bad dark line were every frame came together. And there is no way I was shooting raw and processing each frame to make it go away. And sure I could have stopped down but my 580 doesnt have the power to light a 20X30 room at F16......unless I'm at like 1000 ISO........and then I'm running each image though noise software. Anyway you sliced it, it was not good.

Wow - this is really a problem ? I see u have the mkIIN - its a 1,3 crop - so i guess its not a problem there? But again its not FF.
Is it a big diff i real photog life between 1,6 and 1,3 ?
I have seen all those " inner square is the 1,6 next 1,3 and so on things" but is there i diff in the viewfinder for example ? Brighter - bigger ? so much u can see it ?

Nilsen wrote in post #2365426
Wow - this is really a problem ? I see u have the mkIIN - its a 1,3 crop - so i guess its not a problem there? But again its not FF.
Is it a big diff i real photog life between 1,6 and 1,3 ?
I have seen all those " inner square is the 1,6 next 1,3 and so on things" but is there i diff in the viewfinder for example ? Brighter - bigger ? so much u can see it ?

Yes. It can be.

When photographing with my 120-300. I lose a field of view of nearly 100 mm. It would have been useful to have that at the USC-UCLA game this past saturday as positioning yourself to get a photo is not too easy.

My 50mm has the FOV of a 65mm lens on my 1D compared to 80mm on the 30D i have. It's DOF is also shallower on the 1D.

Tee Why wrote in post #2364788
Hmmm,
Spend about $2800 for a 5D and a 17-40L or $650 for a 10-22 to shoot landscapes in ultrawide angle? I'd say if you print larger than 8x12 or unless you like the bigger viewfinder and are willing to pay over $2000 premium for it, no. The biggest difference between the 5D and 20D in landscape will be in the filesize, while pixel peeping this maybe a big deal, on normal sized prints, it won't really show up. According to Practical photography, 87% of people including photogs couldn't tell which print was made with a 1DsMII or a XT, and this is between a 16 v8MP.

I completely understand your point of view (no pun intended)...however there is more to a 5D than just the increased MP. For me, the major difference is that one camera is giving you everything you should be able to see (yes including any vignetting, etc.) and the other is simply cutting all of that out before you get your hands on the file. I have been extremely pleased with the performance of my 17-40 and it seems silly to have to purchase a lens that will only fit on a handful of cameras to get the view I should already be getting with my 17-40. And because I eventually plan on getting either the 5D or a 1Dmk2n, why should I spend $650 on a lens that will eventually render itself useless?

And as I said, it isn't vignetting. It's a totally different phenomenon, and try actually looking at comparable f.l. (10 mm) and aperture on the 20D. People tend to look more critically at things they pay a lot of money for. It's basic psychology. Compare the threads on "Is my L backfocussing/sharp?" vs. "Is my kit lens backfocussing/sharp?" and tell me which category you see more of.

SeanH wrote in post #2364672
Really?.........huh

Call it what you like, but if you like your edges and corners darker with any fast or wide lenses then the 5D is your camera.....LOL. Or you can always PP the dark edges (vignetting) out.........in every shot......unless your shooting JPG's. But what's a couple min to take it out......LOL.........n​o thanks, not for me.

Hey Sean... correcting something like this is so basically simple it's not even funny. Even if you are shooting JPEG.

If you shoot RAW, you just need to take a reference frame... either one of these will do just fine and correct the vignetting in ACR. Then save the profile. Then in the future, when you do an entire day's shooting with the lens on, open Bridge, and go to your folder where all your RAW files are and select all and apply this profile to it. Instantly a sidecar file is created for each RAW file with the dark corners corrected in all of them.

If you shoot JPEG, all you have to do is open the file in Photoshop and apply the lens correction filter to correct the dark corners.. The trick is to record it as an action while you're doing it and then save it. Then in the future when you do an entire day's shooting with the lens on, open Bridge, and go to your folder where all the JPEG's are. Do select all and run the image processor and select the action you recorded. It will go through each file while you drink your coffee or eat your lunch and correct them all.

random question. I own the smaller xti and i really like the camera. Eventually i will be upgrading to 30d (well this all depends on my wife ) The one thing I am getting confused on is why would a full frame camera vignette at wider angle. My understanding is with a crop camera, the width is the same of sensor, just not as tall? I would think that a full frame camera even with wide angle would see more informatoin as a lens can see wider angle in all 360 degrees of its viewing angle, not just side to side? If somebody could just give quick explenation of why this could happen?

thanks

forkball wrote in post #2366594
Hey Sean... correcting something like this is so basically simple it's not even funny. Even if you are shooting JPEG.

If you shoot RAW, you just need to take a reference frame... either one of these will do just fine and correct the vignetting in ACR. Then save the profile. Then in the future, when you do an entire day's shooting with the lens on, open Bridge, and go to your folder where all your RAW files are and select all and apply this profile to it. Instantly a sidecar file is created for each RAW file with the dark corners corrected in all of them.

If you shoot JPEG, all you have to do is open the file in Photoshop and apply the lens correction filter to correct the dark corners.. The trick is to record it as an action while you're doing it and then save it. Then in the future when you do an entire day's shooting with the lens on, open Bridge, and go to your folder where all the JPEG's are. Do select all and run the image processor and select the action you recorded. It will go through each file while you drink your coffee or eat your lunch and correct them all.

Hey thanks for the info, but 2 questions-

#1. Since the "effect" (since it's not vignetting...lol) varies with F-stop and focal length (since we were talking a 17-40 zoom) how do you create one action that knows how much to take out........considering the amount will vary?

#2...........why bother? .........really, I have looked at images side by side with my mk2 and honestly I don't see enough of a difference to even bother with the issues of the FF.

Here's two shots with a Mk2n that have no PP'ing other than sharpening, at what works out to be 15.5mm on a full frame camera.........see any dark corners?...(if anything that would be the flash can't cover that wide)......like I said, why bother with FF

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

the FF sensor is larger both in width and length. As a result it is seeing more of what is passing through the lens, most notably the outer fringes of the photo that are lost with crop sensors (including the XTi). You really need to experience the difference first-hand before making a judgement.

bphillips330 wrote in post #2366622
random question. I own the smaller xti and i really like the camera. Eventually i will be upgrading to 30d (well this all depends on my wife ) The one thing I am getting confused on is why would a full frame camera vignette at wider angle. My understanding is with a crop camera, the width is the same of sensor, just not as tall? I would think that a full frame camera even with wide angle would see more informatoin as a lens can see wider angle in all 360 degrees of its viewing angle, not just side to side? If somebody could just give quick explenation of why this could happen?

thanks

The FF sensor is bigger in all directions than the 350D's sensor. A FF sensor is 24 mm x 36 mm while your 350D has one that's 15x22.5 mm. An EF-S lens is designed to just cover that smaller sensor; at the wide end your kit lens covers about a 60 degree field of view. On a FF camera an 18 mm lens (that was designed to cover the whole FF frame) would cover about 90 deg. But your 18-55 isn't designed to put light across the whole 24x36 mm field of a FF camera; if you were to put it on one you'd see vignetting - the lens housing getting in the way of the light, leaving you with black corners (really black corners, with absolutely no light getting in).

To get an example, try this: put your 70-300's lens hood in front of your kit lens, and zoom to a wide angle. You'll start to see the edges of the lens hood in the picture. That's real vignetting. Something physically obstructs the angle of view of the lens.

The Cos^4 light falloff is because, especially at wide angles and with fast apertures, the average distance a light ray has to travel to reach the corners of the sensor is greater than one going to the center of the lens. This breaks down into 2 parts.

1) With a very wide angle lens, it's further from the rear nodal point (essentially the point that "looks like" the back of the lens as far as image-making goes) to the corners of the sensor than it is to the center. Since light spreads out as the square of the distance, if the corner's 1.4x as far away as the center, it gets half the light (that'd be your 18 mm lens on FF, or 12 mm on a crop). And the distance changes as the cosine of the angle from the lens axis (through the nodal point) to the sensor site where the measure's taken.

2) A lens focusses all the rays from any given point in the subject to the same point in the image. That means the incoming light rays for a point will be striking all over the front of the lens (essentially the aperture), and being passed through the nodal points of the lens. With a large aperture, the light hitting the edges of the lens has to travel further, so, again, it falls off a bit. And again, the falloff is directly related to how far off the lens axis the particular spot on the sensor is.

Now, in reality you don't see anything like the theoretical "thin lens" light falloff this says you should get. Lens designers spend a great deal of time correcting (and it is possible to compensate) for this light falloff at the same time they're dealing with spherical and chromatic aberrations and pincushion/barrel distortion. So light falloff from cos^4

SeanH wrote in post #2366748
Hey thanks for the info, but 2 questions-

#1. Since the "effect" (since it's not vignetting...lol) varies with F-stop and focal length (since we were talking a 17-40 zoom) how do you create one action that knows how much to take out........considering the amount will vary?

#2...........why bother? .........really, I have looked at images side by side with my mk2 and honestly I don't see enough of a difference to even bother with the issues of the FF.

Here's two shots with a Mk2n that have no PP'ing other than sharpening, at what works out to be 15.5mm on a full frame camera.........see any dark corners?...(if anything that would be the flash can't cover that wide)......like I said, why bother with FF

Answers:

#1 you can't, but you can create an action for each focal length and aperature setting, then write a script that will take from the EXIF data that information and apply the correct action. OR... two products, PTLens and Dx0 work very well in correcting this and other lens corrective issues and can integrate into an automated workflow very easily.

#2 From everything I've heard... image quality is pretty comparable between the two. The pixel density should be close to the same if not exactly the same. My post wasn't to say that the 5D is a better choice than the mkIIn or any other camera... it was just to point out that vignetting is easy to fix.

jevidon wrote in post #2366760
the FF sensor is larger both in width and length. As a result it is seeing more of what is passing through the lens, most notably the outer fringes of the photo that are lost with crop sensors (including the XTi). You really need to experience the difference first-hand before making a judgement.

That is why you choose different lenses to do the same thing with different cameras. If you need a super wide angle you get a 10-22 for your XTi or a 16-35 for your 5D and take exactly the same picture in most circumstances. (You have to use a slightly smaller aperture with the 5D to get the same depth of field as the XTi and the 10-22 does not open wide enough to get as shallow depth of field as f/2.8 on the 5D.)

The argument that a 16-35 mm lens is supposed to give a certain view on every camera just because it gives that angle of view on an old movie camera is nonsense. People have just become used to that angle of view and are to lazy or obstinate to change.

Yes, you are wasting image circle when you use an EF lens on an EF-S camera. The people who should be complaining are the ones who use EF-S cameras because they are forced to buy lenses that provide a wider field of view than they need. They would be much better off if Canon provided more lenses that were just engineered to provide a better image in the size of image circle they need. For example, if someone needs a 300 f/4 lens for their XTi, they have to buy the EF L version that provides a much larger image circle than they need. They would be much better off if they could buy a lens that had the same image quality and cost less because it only provided a smaller image circle or a lens that cost the same, but provided a sharper image because the engineers could concentrate on image quality for a smaller area instead of having to compromise centre quality for corners beyond the sensor.

http://www.imaging-resource.com/IMCOMP/CO​MPS01.HTM

This link is from another thread comparing the 1D to the 5D.
When comparing the 5D to the 30D I find the 5D to be inferior in some areas.

Well, I've owned a 20D for a year and now a 5D.

There is no comparison - the image quality produced by the 5D is superior. The camera is better built than the 20D and it handles better.

I sold my 20D.

Yes, there is some light falloff with 17-40 and 24-105 lenses, but only under certain lighting conditions and easiliy corrected in PS if needs be.

My 5D is a keeper!