Beginner D3200 v D5100 image quality

G

guess&press

Messages
310
Name
Dave
Edit My Images
Yes
Hello all,

First post so please excuse any 'doh!' moments...

I have a Nikon D3200 with a Sigma 150-500mm lens, and while it's OK I recently had the chance to attach a D5100 to the lens. It seems to me the 5100 captures a lot more detail and generally produces a better image. I've tried to show this in a comparison shot.

https://dl.dropboxusercontent.com/u/5159779/D3200.JPG
https://dl.dropboxusercontent.com/u/5159779/D5100.JPG

Obviously not very scientific, not least because the 5100 is set to vivid and the 3200 to standard (argh!) but the same thing seems to happen quite generally. Do people agree there is more detail in the 5100, and if so would this be because of the settings (I know vivid increases the sharpness a bit, and maybe different brightness and contrast might bring out more detail I supose) or because the 5100 has better processing? Or am I seeing things that aren't there? :)

The 5100 clearly works better in low light and takes longer to ramp up ISOs as shown here, and that by itself may be worth upgrading for. I use the camera mainly for birding so this is important. So is croppability, but the pixel count seems much less significant to the result. Would people agree?

Thanks for any comments.
 
Last edited:
What you are seeing is that both sensors are out resolving the lens. For that reason the lower MP sensor looks sharper at a pixel level but in reality its the lens that is the limiting factor in sharpness, not the body.
 
Thanks. That's really helpful. It explains a lot, actually, such as that this Sigma lens is supposed to be OK at 400mm but soft at 500, but there seems to be little/no difference on the 5100 while there is on the 3200--so the higher MP sensor shows up the limits of the lens more clearly. Does that sound right? And this is why some shots at 150-200 mm look really good at 100%--because the lens is no longer a limiting factor?
 
Last edited:
You've got it :)

I like to think of it in terms of magnification, the higher you try and magnify things the more precise the equipment needs to be. At 500mm the magnification is very high and therefore, to get pixel sharp shots you need VERY good glass, at 150mm (i.e. lower magnification) then the quality of the lens is less important so pictures appear sharper.

You can also magnify by using a smaller sensor (Full Frame vs. Cropped discussion) or by increasing the number of megapixels.

Ultimately though, if you are printing at sensible sizes then it makes little difference, it's only when you pixel peep (which we can't help but do) then the differences become obvious.

The other differences I see in your pictures above are probably due to the different 'engines' (i.e. EXPEED 2 vs. 3 - they meter differently in my experience), different in camera settings (i.e. vivid vs. normal) and also technique - at 24mp you need to be very accurate with your focus because if you look at pixel levels it will shot up a missed focus easily. Of course, as said, this isn't visible when printing, even at A4.
 
Thanks again Ned, this is tremendously informative for a beginner like me, and reassuring in its way.

It does raise some other questions, but it's a bit late for that so I'll regroup for another go soon!
 
Last edited:
OK, I've decided I'm going to upgrade. The 3200 seems to choose higher ISOs a lot more quickly than the 5100, and I'm finding not being able to see the ISO in the viewfinder a real bugbear now. Getting control of noise reduction, a choice of RAW+JPEG settings, 14 bit RAW and so on now have an appeal I didn't know about at the start.

But...I have another question. Which one to go for--5100 or 5200? The 5200 seems to have (even) better low light performance. But the main thing is, I'm pretty sure I'll end up getting a 300mm f4D with a 1.4 tc. Would the 5200 allow that combo to shine more than the 5100? Is there any downside to the 5200 (apart from needing a steadier hand!)?
 
Sounds like a bit of G.A.S. to me ;)

Not saying you shouldn't upgrade, but don't expect an increase in image quality. With the Sigma at the long end, you'll be lens-limited on any camera. Also, long focal lengths like that need very careful technique to get the best from them.

The differences you're seeing in the two images are simply down to in-camera image settings and slightly different exposures. It's also possible that shooting into the light like that, even a tiny difference in framing could induce a bit of flare - maybe light bouncing off a lily pad just out of shot.
 
@HoppyUK: Thanks for the pointers, especially on the lighting, and again on technique. Never thought about off-screen light. I do find that the 3200 seems to select higher ISOs more quickly than the 5100 though.

I had to look G.A.S. up. :) There may be a bit of that, but I've only been at this for about six months and I think I'm getting a feel for what I want out of a camera, so if possible I want the next set-up to last a long time as I work on technique and no doubt a lot of other things as well. It looks like the Nikkor 300mm F4 + TC is a very effective combo (90% of my images are of birds) without starting to pay eyewatering amounts (all relative, I know). I did read somewhere that a 5200 would gain more benefit from good quality long lenses than the 5100. Does this sound right? Should I be looking at a different series completely given the subject matter? :eek:

Hmm. Can you go off topic on your own thread? :)
 
Last edited:
The reason that is said about the D5200 and long lenses is that it is a high resolution (I.e. Magnification) sensor and so you need good lenses to make the most of it.

If you want to make a step up to last you then I wouldn't bother with the D5X00 series as it isn't really a step up, I would jump up to a D7X00 series camera - the D7000 is a real bargain nowadays.
 
minnnt said:
+1 for the D7k. :) The extra controls and step up in useability is a real plus for me.
Click to expand...

Just picking up on this, and the principle really. Very long lenses are hard to use well, and in practise the limiting factors are often too long shutter speeds (though OS helps with the Sigma of course) and particularly focusing. Doesn't matter how good the lens or camera is if you can't nail good focus reliably and that's a know difficultly with slower lenses, ie f/6.3 at the long end.

So a camera upgrade might be worth looking at that way.
 
Thanks all for comments, explanations and suggestions. I've got some thinking to do and some things to work on. But I'm a lot clearer on why the images look the way they do, and also how to approach pixel peeping a bit more sanely! I've seen minnnt's gannet pics and they are absolutely superb IMHO--I'd love to get images like that--yet at 100% they look a little strange to me, almost as if they'd been painted. So treating the camera like a microscope is right out. :)
 
@David: Well, entirely a good thing given the result! Perhaps 'painterly' would have been a better word, but what I was trying to say was that 100% views don't really help you see the image as a whole, just as standing too close to a painting more or less forces you to lose the overall effect and start looking at brushstrokes. Which may be great for art critics trying to assert their expertise but entirely misses the point when it comes to photography, IMHO. It's the whole picture that counts, not expecting to see an almost cellular level of detail when zooming in. So pixel peeping has its place but it's a limited place. And that's an understanding which I'd started to form but which has been crystallised because of this thread, so again thanks all round.
 
Last edited:
Thankyou :)

I'm happy with the images and I only view at 100% if I'm making local adjustments. These images have been pushed quite a bit in PP as the conditions on the day were very harsh so bright sunshine and dark shadows were plenty and I've had to make up for my lack of skill with the camera in PP so what you're seeing at 1:1 is probably increased noise from pulling out shadows etc.
 
You must log in or register to reply here.
Back
Top