Upgrade from D300 for a poor college student?

Started Apr 7, 2013 | Discussions thread
ultimitsu
Senior MemberPosts: 5,422
Like?
Re: Upgrade from D300 for a poor college student?
In reply to Ray Ritchie, Apr 9, 2013

Ray Ritchie wrote:

Yes, but as I said in my reply to your first post when you introduced the DxO pmp measure, I don't think the analogy holds as well once we start comparing different lenses on different bodies.

The analogy merely explains what pmp suppose to represent.

I note on DxO's comparison tool, for example, that if I compare D300 and D700 with the same lens, the difference in pmp rating changes depending on what lens I select.

Yes, this is because each lens's imperfection is different. DXO unfortunately uses whole numbers, so one lens may result 9.8pmp on D700 and 7.4 on D300, pmp will round them and you would get 10 vs 7. while another lens may be 10.4 on D700 and 8.5 on D300, after rounding you would get 10 vs 9.

And I don't think the pmp difference between D300 and D700 is the same as the difference between, say, a D7100 and a D600, even though in both cases we're comparing cameras with the same size sensors.

You are probably right. There is law of diminishing returns. the impact of lens imperfection would be greater going from D300 to D7100 than it is from D700 to D600.

There are a lot of factor at work determining the differences once we start changing both the lens and the camera, and I'm not yet confident that DxO's pmp metric is a consistent basis for these complex comparisons. Not to mention the fact that there are a number of other factors that go into IQ besides "simple" resolution (vignetting, distortion, chromatic aberration,...)

That is true. But pmp is a good "rough" guide.

I also don't understand what DxO is really doing when they say they "correlate" the rating with human visual perception.

I think it is just bad choice of words.

There are so many parameters other than pixel count that determine the IQ of a camera/ lens combination that I can't quite wrap my brain around what that specific number may mean.

Pmp does not attempt to represent other IQ metric - such as Dr or color depth, it only concerns with resolution.

I think it does take into account other factors than just sensor resolution, though - for example, the nature of the anti-aliasing filter used in each body. That is, two different bodies may have different maximum pmp ratings with ideal lenses, even though they are both 12 megapixel sensors. And DxO says differences of 10% or less are not significant, but how "significant" is a difference of, say, 20%?

That may be so too. We will only know for sure when DXO publishes some pmp results from AA-less cameras such as D800E, D7100, and K5IIS.

But for present purpose, I think DXO's figures can be relied on to establish that with the same lens you are like to get better results from D700 than D300 (except extreme corner and vignetting on some lenses).

I understand 17-55 is a better built lens with more exortic elements than 24-85. it has slightly less distortion and less CA. But I do not think these will be significant in real life. On the other hand 24-85 has a 2 stop VR, that is very significant when photographing static subjects in less than optimal light.

I have seen the advantages of the 24-85VR's stabilization in some shooting situations on my D800. But the 17-55 is more massive, and thus, less subject to vibration from things like mirror slap,

VR (or Canon IS) is never meant to fight against mirror slap. they are intended to mitigate basic level of hand shake.

What I have to concede is I find D600's mirror slap is hard, quite a bit harder than my Canon 60D. I get more sharp 1/50s shots from 50mm lens on 60D than I get from D600 (even though 60D has significant higher pixel density and should in theory require faster shutter speed)

and it is about a stop faster - so I don't think the 24-85 always realizes 2 stops of advantage vs. the more expensive lens on DX.

Yes, But as I said before D700 is over 3 times better than D300 in high iso. D600 is almost 5 times better. That more than negates that 1 stop advantage in 17-55's F2.8.

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow