More 5D impression vs 30D/40D, 2nd week.

Continue from here. I got a hold of the 24-105L. My main interest is to have a quick comparison between 5D+24-105L vs 5D+24-70L vs 30D+17-55IS. I also have a 40D with me, so I will also make references to it. Again, nothing scientific here, no test chart shooting, no signal-to-noise calculation from the pictures. This is subjective, I went out with 2 camera combos at the time, and I came up with my own impressions.

Lens Build: 24-70L(70L) > 24-105L (105L) > 17-55IS (55IS)
The 2nd ">" is given. After using 70L for a week, I expected the same from 105L, but it isn't quite there. 105L has more plastics, actually it is all plastic although very high quality platics. The 70L's inner barrel is anodized almuninum. The whole lens feel more solid and tough.

AF in good light: 40D(55IS) >= 30D(55IS) >= 5D(105L) > 5D(70L)
70L on 5D is noticeable slower than others in achieving initial lock. 105L on 5D is almost as fast as 55IS on 30D/40D. But 30D combo still feels more responsive. 40D combo feels even faster. First 3 combo are all very good, I cannot ask for more. I wish 70L can be a little faster

AF in low light: 40D(55IS) > 30D(55IS) > 5D(105L) > 5D(70L)
40D > 30D. 40D can lock where 30D cannot (in very servere condition). 40D also does it quicker. The same apply to 30D(55IS) vs 5D(105L).

Both 40D and 5D quite a few times, gave out AF lock comfirmation while it is actually NOT in focus. Not just slightly off, it is way off. 30D in the similar condition will simply not lock. It never give out fail confirmation, not once in 2 years. 70L is again noticeably slower than others.

Noise at PIXEL LEVEL: 5D > 30D >= 40D
At ISO1600, 30D and 40D look similarly noisy. If I have to pick a winner, it would be 30D. 30D noise pattern is in smaller piece and sharper looking, 40D is larger pieces but softer looking. This is true even with HIGH ISO NR enable. 5D is better than 30D, but NOT one stop better. 30D@800 is better than 5D@1600. 30D@1600 is better than 5D@3200.

Noise when view on 24" monitor: 5D>=40D>=30D
It is Dell 2407, 1920x1200. They look all very good at 1600, you really need to pixel peep or print large to see difference. Have that said, 5D maintains most detail, then 40D, the 30D. I observed 2 key things that makes 5D look better. 1) 5D capture significantly more details. If you apply NR so that the same detail is maintained or resize to same output size, 5D pictures will look quite a bit better. Probably more than one stop. 2) 5D meters "hotter", so help expose more to the right and help on noise.

Sharpness at pixel level: 5D+70L > 5D+105L > x0D+55IS
I thought 55IS is very sharp, but 5D+Ls are better. At least for center. 70L give the best out of camera pictures, best color, most POP. But overall, all are very good. If your shoot RAW, their output can be adjusted to the same easily. Again, I am happy with all of them.

AA filter: 5D has the weakest
People talked about it, but until now, I see hard proof myself. I shot a few pictures inside my car, which by chance captured the LCD screen of te NAV unit. I saw a lot of moire with 5D, not nearly as much on 30D or 40D. But this is the only place I noticed it, so in real world shooting, I doubt it is a problem.

Vignetting: 105L on 5D is BAD
I am not a picky guys about it. People complain about it on 55IS, which I think it is totally acceptable. But with 105L on 5D, it is quite bad at 24mm f/4 even with no filter no hood. You can even see this in the viewfinder under the right condition. And to make it worse, vignetting with 105L is not gradual. The image doesn't get darker until near the corner, then suddenly get darker. Vignetting in 55IS or 50f1.8 is more gradual and less offensive. Again, in real life situation, this is probably not a issue. You only notice it if the background is in bright flat color.

Quality of bokeh: 5D(70L) >> x0D(55IS) > 5D(105L)
70L's bokeh is very smooth and creamy, quite a bit better than the other 2. 105L give the most distracting bokeh especailly if background is foilage type.

Interesting Lens Hood observation:
Hood for 55IS is designed for 27mm FOV. It is 1/2 inch longer than hood for 105L, but I found that I can use hood from 55IS on 105L on 5D without introducing any more vignetting. This tells me that the OEM hood for 105L is under designed.

Flare: 55IS flare the most
This I already know, and I don't think it is an issue. This lens does so many things so beatifully.

Overall IQ: 5D(70L) > 5D(105L) > x0D(55IS)
Having said this, I have absolutely no complain about x0D(55IS). The additonal sharpness and IQ 5D body brings is welcomed. The differences however is very small. Again, need pixel peep or print large to see.

Overall usage: 5D(105L) >= x0D(55IS) > 5D(70L)
By overall usage, I mean normal day to day usage. NOT profession on the job type of usage. I picked 5D(105L) first because mainly it is wider than x0D(55IS). Other than that these 2 combo are very similar. They give about same level of background blur control, similar AF speed, samilar weight and dimension. The 5D(105L) gives a more feeling that is solid, secure, "it won't fail". I picked 5D(70L) last because of its weight, size, lack of IS and slow AF

Low Light usage: 30D(55IS) >= 5D(105L) > 5D(70L)
In very low light, say ISO1600, f2.8, 1/20s, IS lens has advantage. 1 stop of light or 1 stop of ISO cannot make up for it. Plus slow AF with the 70L, I pick 5D(70L) last. For 30D(55IS) vs. 5D(105L), it is tough pick. 30D(55IS) has one stop of light and AF advantage. However, I do think that with 5D(105L) at ISO3200, f4, 1/20s, resize 12MP file to 8MP, apply NR to match same detail, the final output will be match or better than 30D(55IS) at ISO1600, f2.8, 1/20s. But this is a lot of works that I am not sure I want to do it often. I wish I can have more time with all 5D and 105L for realy life shooting comparison. Unfortunity, 5D body, 40D body, 105L, 70L need to go back to their owners soon.

My concludsion:

Before going into this comparison, I didn't consider upgrading to 40D from 30D. And this didn't change after this side by side comparison. For my usage, 40D doesn't offer any meaningful advantages. Or I should say 40D doesn't make enough difference for the final output. I have never had issue with 30D AF. I hardly use live view for the time I owned 40D. 40D noise at 1600 isn't better. I don't see enough reason to spend near $500 for this upgrade.

As for 5D, it make difference in the the ways that I like, such as:
1)shallower DOF. There is no easy way it can be done with crop. For my usage, this is a welcome change.

2)ability to go wider. This is specially true for standard zoom choice. With 30D and 17-55IS, I found I need to go wider sometimes, not too often. So I also have 10-22. On 5D, 24mm is very wide, I don't think I will need a ultra wide like 17-40.

3)access to wide angle prime lens. I shoot a lot of indoor low light. I always wish I can go beyond f2.8 with wide angle. With crop, there isn't any choice. Even 24L and 35L aren't that wide on crop. With 5D there are cheap alternative like 28f1.8, 35f2. If I go crazy, there are 24L and 35L

4)more useable high ISO, mainly 1600 and 3200. I think we should not argue this. 5D is better at these ISO. 1 stop advantage for the same size output (NOT pixel level) is NOT an exageration.

I do wish 5D has onboard flash build in. It can come in handy although no big deal here. My plan now is to find good deal to get 5D+105L kit. Then I will sell either 17-55IS or 24-105L, depend on which one I like better (30D+55IS vs 5D+105L). As for 30D, I think I will just keep it as a backup or use it when I want more reach.

Thanks for analysis.

For me, most of my shots are indoors of daughter running around.

I'm using a friend's 40D while my 30D is @ Canon Service. I am amazed by increased ability of 40D to AF in low (next to no) light. Not worth the $1000 for me to buy a 40D, but if I were trying to decide between the two, for me it is worth the extra $300+ to buy 40D over 30D.

I totally agree. If I start from new and choosing between 30D and 40D, no contest here, definitely a 40D. It is simple a better body. But to sell my 2 years old, trusty 30D and upgrade to 40D, I just don't see it. Now worth the effort and money. And I don't see enough difference in the final output image. But this is just me.

eel

Mystwalker wrote in post #5923719
Thanks for analysis.

For me, most of my shots are indoors of daughter running around.

I'm using a friend's 40D while my 30D is @ Canon Service. I am amazed by increased ability of 40D to AF in low (next to no) light. Not worth the $1000 for me to buy a 40D, but if I were trying to decide between the two, for me it is worth the extra $300+ to buy 40D over 30D.

Compare AF performance with the same lens mounted on all 3 bodies.

I still think 40D > 5D > 30D...not 40D > 30D > 5D.

There is no way the AF on the 30D is better than the 5D.

I am pretty sure the inner barrel of the 24-105 is the same as the 24-70. the 24-70 weighs more and perhaps that is swaying your impression a little bit.

eelnoraa wrote in post #5923514
My concludsion:

Before going into this comparison, I didn't consider upgrading to 40D from 30D. And this didn't change after this side by side comparison. For my usage, 40D doesn't offer any meaningful advantages. Or I should say 40D doesn't make enough difference for the final output. I have never had issue with 30D AF. I hardly use live view for the time I owned 40D. 40D noise at 1600 isn't better. I don't see enough reason to spend near $500 for this upgrade.

As for 5D, it make difference in the the ways that I like, such as:
1)shallower DOF. There is no easy way it can be done with crop. For my usage, this is a welcome change.

2)ability to go wider. This is specially true for standard zoom choice. With 30D and 17-55IS, I found I need to go wider sometimes, not too often. So I also have 10-22. On 5D, 24mm is very wide, I don't think I will need a ultra wide like 17-40.

3)access to wide angle prime lens. I shoot a lot of indoor low light. I always wish I can go beyond f2.8 with wide angle. With crop, there isn't any choice. Even 24L and 35L aren't that wide on crop. With 5D there are cheap alternative like 28f1.8, 35f2. If I go crazy, there are 24L and 35L

4)more useable high ISO, mainly 1600 and 3200. I think we should not argue this. 5D is better at these ISO. 1 stop advantage for the same size output (NOT pixel level) is NOT an exageration.

I do wish 5D has onboard flash build in. It can come in handy although no big deal here. My plan now is to find good deal to get 5D+105L kit. Then I will sell either 17-55IS or 24-105L, depend on which one I like better (30D+55IS vs 5D+105L). As for 30D, I think I will just keep it as a backup or use it when I want more reach.

I completely agree. It appears you went through the same situation I did... I have a 30D and needed another body to relegate my 30D as a backup... The 40D didn't do enough when compared to the 30D while the 5D (despite all the hootin' and hollaring about "old technology") DID prove a meaningful result for my shooting needs... Also, I too have never really had a problem with the focus system on the 30D / 5D...

eelnoraa wrote in post #5923514
4)more useable high ISO, mainly 1600 and 3200. I think we should not argue this. 5D is better at these ISO. 1 stop advantage for the same size output (NOT pixel level) is NOT an exageration.

that is consistent with what DPreview came to while reviewing the 40D.

nicksan wrote in post #5923945
Compare AF performance with the same lens mounted on all 3 bodies.

I still think 40D > 5D > 30D...not 40D > 30D > 5D.

There is no way the AF on the 30D is better than the 5D.

This maybe true if the same lens is used. But with 17-55IS on 30D, it is definitely faster than both L on 5D. It is the lens. You need to use one to realize how fast this lens AF.

nicksan wrote in post #5923945
I am pretty sure the inner barrel of the 24-105 is the same as the 24-70. the 24-70 weighs more and perhaps that is swaying your impression a little bit.

The 70L has a one piece inner barrel to the front thread. When I screw the filter onto the 70L, the sound is metal touch metal. If the 105L also has a one piece design, then the barrel is definitely not metal. When screwing filter onto 105L, the sound is metal on plastic, same as other canon lens like 17-55IS. Further more, 70L's inner barrel feels cool when you touch it. Not the same for 105L. Of course, i cannot verify the material, maybe one one who has taken them apart can let us know the true.

eel

Great analysis, thanks for sharing!
Pretty much confirms my own views..