My little lens and body comparison test

[rottie102]

I've met somebody who has 7d, 5d and a 28-70 L lens, all of which I was thinking of buying.
So I went there with my 450d and Sigma 50mm 1.4 prime to do some testing.

100% (roughly) crops below -







My conclusions:

- 450d has a really good IQ
- after playing with 7d 5d mkI feels so dated!!
- 28-70 L or not - it's not even in the same league as good prime lens.

Edited by rottie102 on Tuesday 5th October 00:14

[RobDickinson]

rottie102 said:

- 28-70 L or not - it's not even in the same league as good prime lens.

Yis, much as I love the 17-55 a prime kills it on IQ.

I'm intrested in a sigma 50/1.4 for my 7D but the focus issues make me quite wary.

[theboyfold]

Do you know anything about the history of the 28-70? I know there are plenty that are slightly 'out' as they can't be serviced by Canon any more. I speak of my experience of buying a duff one

[GetCarter]

Is it me or are they all out of focus?

[theboyfold]

GetCarter said:

Is it me or are they all out of focus?

Which is why I especially ask about the 28-70, as the zoom rings can wear and cause some of the elements to misalign

[fathomfive]

GetCarter said:

Is it me or are they all out of focus?

Thought it was just my eyes

[_dobbo_]

GetCarter said:

Is it me or are they all out of focus?

100% crops with no sharpening applied I guess?

[14-7]

_dobbo_ said:

GetCarter said:

Is it me or are they all out of focus?

100% crops with no sharpening applied I guess?

Shouldn't make that much of a difference.

[Killwilly]

Those photographs have almost confirmed my suspicion that the 450D does overexpose very slightly, mine does the same. I have read a number of posts on other forums where other 450D owners have had the same problem.

[LongQ]

Killwilly said:

Those photographs have almost confirmed my suspicion that the 450D does overexpose very slightly, mine does the same. I have read a number of posts on other forums where other 450D owners have had the same problem.

I think you will find that most cameras since the mass market camera with exposrue setting aids were first introduced, will have been assessed as under or over exposing slightly - usually by about a third of a stop which is the smallest 'auto' adjustment made available by the designers.

For digital cameras overexposing would fit with the 'shoot to the right' concept - ie, when you look at a histogram the peak values will tend to the right hand side rather than centre or left. The rationale for that is that, in the main, you can drag more detail back from overly bright images than you can dark images using an editor.

However ....

Unless you are testing this in a lab with completely known values for both the camera and ambient light; and

you know what metering mode was set and precisely how that is intended to work; and

you are working in a controlled ambient light environment with a calibrated and standardised monitor; and


..... a host of other things ....

a few images crushed by some software and presented on a web forum are not really too good a guide for a general assessment on which to base an approach to adjustments to be applied to the exposures. In any case as a matter of artistic interpretation the idea of what is prefereable - slightly light or slightly dark as the base exposure - tends to be a matter of personal preference.

The comparisons originally posted are interesting, certainly, but in all aspects of what they present seem to offer so many undocumented variables that I am not sure what information they really provide.

[_dobbo_]

14-7 said:

_dobbo_ said:

GetCarter said:

Is it me or are they all out of focus?

100% crops with no sharpening applied I guess?

Shouldn't make that much of a difference.

Looking at these again - is the focus on the reflected blinds in the window glass, rather than on the door itself?

[rottie102]

Ok, just found a moment to reply:

a) of course I know this test could be conducted much better in a lab, I only wanted to see what my options are, I've had 10mins with all the gear so the best I could do was to shoot the same thing with the same settings
b) re focussing - 450d and 7d were definitely, 5d I'm 95% sure focussing in the centre which was roughly in the middle of the door. That's what it looked like in the viewfinder. I haven't had any problems with my 50mm 1.4 before. Steve, don't you think that 450d won't be as sharp as your gear? Of course I'm always happy to learn anything new, in case I'm missing something or should check my lens/body.

Edited by rottie102 on Wednesday 6th October 22:02

[LongQ]

rottie102 said:

Ok, just found a moment to reply:

a) of course I know this test could be conducted much better in a lab, I only wanted to see what my options are, I've had 10mins with all the gear so the best I could do was to shoot the same thing with the same settings
b) re focussing - 450d and 7d were definitely, 5d I'm 95% sure focussing in the centre which was roughly in the middle of the door. That's what it looked like in the viewfinder. I haven't had any problems with my 50mm 1.4 before. Steve, don't you think that 450d won't be as sharp as your gear? Of course I'm always happy to learn anything new, in case I'm missing something or should check my lens/body.

Edited by rottie102 on Wednesday 6th October 22:02

When you say they are 100% crops do you mean they are representing the full image frame, but re-sized to be eaual in size, or that they are small portions of the centre of the frame with the focus point in the middle?

If the latter, and given different focusing systems in the bodies, I sort of wonder how wide the focus area might be compared to the door.

If there was any overlap of the door then the curved windows either side may have an effect, especially when shooting with the lens wide open and so inducing a very shallow DoF.

Secondly the most sensitive focus point, the one in the centre, is working with horizontal lines. Anything less than horizontal, as seen by the focus system shot to shot, may compromise things. The blinds in the window are almost horizontal but maybe to the point where a slight misalignment could throw things off rather than improve focus decisions made by the camera. If you were hand holding as well (and in a hurry too) there may be enough compromises introduced to make comparison of the results marginal. What shutter speed were you seeing/using?

Please don't think I'm trying to be purposely negative here. In recent weeks I have been comparing (over extended shoots) a 7D with my 400D and more recently a 1D MkIII to see where the value proposition lies. It's not easy and things are not as full proof as one might expect once you get to that level though one common factor - the use of good glass - offers benefits in all cases though whether the benefits are enough justification for the probably costs is another matter.

[rottie102]

LongQ said:

rottie102 said:

Ok, just found a moment to reply:

a) of course I know this test could be conducted much better in a lab, I only wanted to see what my options are, I've had 10mins with all the gear so the best I could do was to shoot the same thing with the same settings
b) re focussing - 450d and 7d were definitely, 5d I'm 95% sure focussing in the centre which was roughly in the middle of the door. That's what it looked like in the viewfinder. I haven't had any problems with my 50mm 1.4 before. Steve, don't you think that 450d won't be as sharp as your gear? Of course I'm always happy to learn anything new, in case I'm missing something or should check my lens/body.

Edited by rottie102 on Wednesday 6th October 22:02

When you say they are 100% crops do you mean they are representing the full image frame, but re-sized to be eaual in size, or that they are small portions of the centre of the frame with the focus point in the middle?

Small portions of the centre of the frame. The whole picture looked like this:



Shutter speed in this case 1/80

[GetCarter]

Ah ok... 1/80 is pretty slow. It'd need to be rock solid (tripod) to keep distant text in focus.

[rottie102]

LongQ said:

rottie102 said:

Ok, just found a moment to reply:

a) of course I know this test could be conducted much better in a lab, I only wanted to see what my options are, I've had 10mins with all the gear so the best I could do was to shoot the same thing with the same settings
b) re focussing - 450d and 7d were definitely, 5d I'm 95% sure focussing in the centre which was roughly in the middle of the door. That's what it looked like in the viewfinder. I haven't had any problems with my 50mm 1.4 before. Steve, don't you think that 450d won't be as sharp as your gear? Of course I'm always happy to learn anything new, in case I'm missing something or should check my lens/body.

Edited by rottie102 on Wednesday 6th October 22:02

If you were hand holding as well (and in a hurry too) there may be enough compromises introduced to make comparison of the results marginal.

One more thing: but you see, that's how I photograph most of the time. Handheld, very often in a hurry. So for me these results are useful for comparison purposes.

[LongQ]

rottie102 said:

LongQ said:

rottie102 said:

Ok, just found a moment to reply:

a) of course I know this test could be conducted much better in a lab, I only wanted to see what my options are, I've had 10mins with all the gear so the best I could do was to shoot the same thing with the same settings
b) re focussing - 450d and 7d were definitely, 5d I'm 95% sure focussing in the centre which was roughly in the middle of the door. That's what it looked like in the viewfinder. I haven't had any problems with my 50mm 1.4 before. Steve, don't you think that 450d won't be as sharp as your gear? Of course I'm always happy to learn anything new, in case I'm missing something or should check my lens/body.

Edited by rottie102 on Wednesday 6th October 22:02

If you were hand holding as well (and in a hurry too) there may be enough compromises introduced to make comparison of the results marginal.

One more thing: but you see, that's how I photograph most of the time. Handheld, very often in a hurry. So for me these results are useful for comparison purposes.

That context may be important for others looking at the comparison.

At a personal level it is quite valid if you are assessing whether an upgrade and its associated costs are value for money for you and your requirements. I have looked at things from the same angle many times and it's fair to say that almost any dSLR body and lens can deliver something satisfactory for personal use at a lowish cost compared to what one needs to spend to get incremental improvements consistently. But at the other end of the scale - Carteresque landscapes for example - technique PLUS equipment can move things on a long way. Whether the investment is justifiable is a personal decision.

I strongly suspect that as the higher levels of equipment the results for the money invested may be somewhat disappointing without also adjusting technique. It's not that one does not get good shots, just that they may not seem that much better (or may not be achieved more consistently as one might hope for and expect) unless technique is also adapted to suit.

[rottie102]

Completely off topic - I've made up my mind and I'm buying a 60D. Just for the articulated display.

[LongQ]

rottie102 said:

Completely off topic - I've made up my mind and I'm buying a 60D. Just for the articulated display.

Probably a good call. I have a couple of Canon bodies with articulated displays and whilst the bodies are somewhat out of date the articulated displays can be so useful I don't plan on retiring them just yet.

However I also have a Phottix Hero that allows me to observe what a dSLR is doing via an attachment to the viewfinder that then transmits a wireless video signal to a hand held screen viewer from which I can activate the shutter release. Quite useful in some situations.