Upgrading 6 mp to 10 mp, A big Deal?

[fmw]

That is exactly the question I had as I unpacked the new Nikon D80 and charged the battery. Just how big a deal was the 10 mp sensor compared to the 6 mp sensor in my D50?

I put one of my lenses in the light tent and made an exposure with each camera, recording the image in NEF or the Nikon RAW format. I used the same lens - a 17-55mm f2.8 Nikkor zoom - and the same memory card. I then made a significant crop of each image and enlarged the results to an 8" wide JPEG for viewing on a computer screen. I used a little compression to get the file size reasonably small but not much. I did exactly the same thing to both images. Then I made an extreme crop to show just the 12-24 numbers. I enlarged those to an 8" width so I could see what was going on. Here is the image with the results. At the top is the original shot made with the D50. Below that is the first crop and the extreme crop from the D50 image and then the first crop and the extreme crop from the D80 image.

I can just begin to see a tiny bit of pixelization on the extreme D50 enlargement but none in the D80. You can tell me what you think but I think the images are comparable enough that the increase in sensor resolution was basically insignificant. In truth, I expected more difference but I really don't see it. My conclusion is that once you are at 5 or 6 mp you are pretty much good to go. Beyond that, the improvements are subtle at best. Comments?

 

[Alex_B]

i can see more noise on the lower extreme crop (D80)
did you use different iso?

what was your jpeg compression?

overall to mee it seems that the lens (or the focus) you used is the limiting factor in both cases, not the sensor.

do both cameras use the same in camera sharpenig for the raw iamge? some cameras use no in camera sharpening for RAW, others do.

RAW images from a Bayer sensor are always a bit blurry by principle.


just some thougts from someone not knowing much about your lens or any of the two cameras ... maybe some nikon expert here can help more

 

[fmw]

The compression was 60% in Photoshop to make the JPEG's. The cameras don't use any sharpening for RAW. RAW is Raw. Sharpening would show up in a JPEG but not in a raw, wouldn't it? My processing in both cases was identical. Auto levels to bring up the darkish RAW images to something more normal and light unsharp mask when I resized - used the same sharpening for both images. The focus is right on in both cases as you can see.

My point was like yours, I guess. The sensors aren't a limiting factor. Things need to become pretty extreme to even see any meaningful difference at all between the two. I view 6 and 10 mp as basically the same thing with the 10 mp image just taking up more hard drive space. Perhaps I need to test things differently but I was pretty unimpressed with 10 mp. It is nice to have two control wheels on the camera, however, and nice to be able to do multiple exposures.

 

[Alex_B]

The compression was 60% in Photoshop to make the JPEG's. The cameras don't use any sharpening for RAW. RAW is Raw. Sharpening would show up in a JPEG but not in a raw, wouldn't it? My processing in both cases was identical. Auto levels to bring up the darkish RAW images to something more normal and light unsharp mask when I resized - used the same sharpening for both images. The focus is right on in both cases as you can see.

My point was like yours, I guess. The sensors aren't a limiting factor. Things need to become pretty extreme to even see any meaningful difference at all between the two. I view 6 and 10 mp as basically the same thing with the 10 mp image just taking up more hard drive space. Perhaps I need to test things differently but I was pretty unimpressed with 10 mp. It is nice to have two control wheels on the camera, however, and nice to be able to do multiple exposures.

well, i know cameras from a major brand which do sharpen RAW (raw is not always the pure sensor data in that sense) .. you can switch the RAW sharpening off in tose cases though. but maybe with nikon raw means totally unsharpened, don'T know.

in any case you should take some more complicated images ... maybe with thin lines in them ... see the moiree effects (if any) look at the details there.

in that image you presented here, the details are rather bulky ... go try some trees with leaves from a distance, or some things containing strong contrasts and thin lines ...

 

[Alex_B]

just some add on .. I use an almost 13 MP camera, and even with my best glass for wide apertures the glass is definitely the limiting factor and I would guess 8 MP would not make a difference... for f/8 however, and if everything is perfect (light, focus, etc.) ... 13 MP seems to make a difference.

 

[fmw]

just some add on .. I use an almost 13 MP camera, and even with my best glass for wide apertures the glass is definitely the limiting factor and I would guess 8 MP would not make a difference... for f/8 however, and if everything is perfect (light, focus, etc.) ... 13 MP seems to make a difference.

Seems to? I guess you would admit the differences are subtle as well. I wasn't really trying to get down to image quality. I was just looking for pixelization. I wanted to see if it started sooner in the enlargement process with the the lower resolution sensor. I guess it does but the difference is really subtle. I've had a chance to set up the D80. I set the sharpening to max. I'll make some shots to see if it affects the RAW images. Then I'll quit this nonsense and just go do some photography with the confidence that I have enough resolution for virtually anything - with either camera.

 

[Alex_B]

Seems to? I guess you would admit the differences are subtle as well.

I agree .. subtle. So they only really come out in some special situations, but in 90% of all images the difference is probably negligible. And for those 10% you might still be happy with the smaller MP number for most purposes.

then I'll quit this nonsense and just go do some photography with the confidence that I have enough resolution for virtually anything - with either camera.

yes, that is always the best idea ...!! go shooting

sometimes all the worries about technical details distract us from what it is all about in the end ...!

 

[fmw]

Here goes. After shooting again, I would say that Raw is Raw. Sharpening doesn't appear to have an effect on raw images. It may be possible, however, that the higher resolution needs more sharpening and that applying the same sharpening to both doesn't work. I redid the D80 shot and sharpened it to what seemed to make it look like the D50. I didn't compare them. It's here. We can compare on line.

And since we're worried about image quality, here it is again this time shot with a tack sharp 60mm Micro Nikkor - a fixed focal length macro lens.

The zoom isn't too bad is it?

That's enough for me, I guess.

 

[forceflow1049]

I think the D50 produces an overall better image, once noise, etc are taken into account.

 

[Big Mike]

As far as I'm aware...you can set the sharpening level when shooting RAW...but that sharpening is not applied to the image until the conversion stage...so you can choose to sharpen the image or not.

When you are comparing the cropped images (in the first post)...I think I see a difference. You are zooming in more, on the D80 image...becuase it's a bigger image to start with. If you zoomed in say, 200% on both images...the 10MP image should be wider and have more info....shouldn't it? :scratch:

That's the big difference...the actual size of the image (and resultantly, the file size). You have more pixels to work with.

Maybe my logic is fuzzy...

 

[bitteraspects]

nikon produces pixelated images when blown up either way.

 

[fmw]

As far as I'm aware...you can set the sharpening level when shooting RAW...but that sharpening is not applied to the image until the conversion stage...so you can choose to sharpen the image or not.

When you are comparing the cropped images (in the first post)...I think I see a difference. You are zooming in more, on the D80 image...becuase it's a bigger image to start with. If you zoomed in say, 200% on both images...the 10MP image should be wider and have more info....shouldn't it? :scratch:

That's the big difference...the actual size of the image (and resultantly, the file size). You have more pixels to work with.

Maybe my logic is fuzzy...

No, the images were taken at the same focal length, the same distance and cropped identically. The top part of the image, showing the full frame, was shot with the D50 but I shot one just like it with the D80 to get to the cropped and enlarged images. I just didn't include it to keep the JPEG size under control. I might have been off a tiny bit but certainly not much. All the images, including the one with the macro lens, were made as exactly equal as I could.

I tend to agree with ForceFlow. I think the D50 does make a slightly better image in this case. The reason might be that it is easier technically to make a 6 mp sensor than a 10 mp sensor on a given equal surface area and the 6 mp just works a little better. People have told me that the same resolution on a larger area makes a better image. This might be showing that sort of thing.

In the end I still conclude that the difference between 6 and 10 mp with these cameras is meaningless. I also conclude that the 100% difference in their selling prices makes the D80 a poor value in comparison unless you simply must have some of the features it has that the D50 doesn't have and there aren't many of them. Perhaps we should be talking about going from 6 mp to 10 mp as a downgrade.

At any rate, I needed a backup DSLR and now I have it. I just need to decide which is the backup.

 

[benhasajeep]

I was asked if I wanted to get the new D200 or D80 and I said no. I will wait till the next set of new cameras. I believe the only advantage to 10mp is being able to make a slightly larger print without upsizing the file. And really do you need a larger mpix camera if you are only going to print 8x10's or even if you don't even print and just do computer or web pics???

I do print to 13x19 on occasion but the cameras I have now cover that just fine.

 

[fmw]

Either Nikon is going to have to move to a 35mm size sensor or there will have to be a big improvement in technology to get concerned about resolution. At the moment I would say resolution is probably not something that should be at the top of the list of issues for choosing a camera.

To be fair, the D80 does have some nice features all packed into the same basic body as the D50. Most of them are the differences between the D50 and the D70. There are also a lot of in camera editing features that would be of interest to someone who doesn't use Photoshop. But the price difference sure doesn't make any sense to me. I guess you could view it as a higher resolution version of the D70 in the body of the D50.

If you compare it to the D50 it looks like a poor value. But if you compare it to the D200, it starts looking better. The only meaningful thing the D200 has is the metal body which would be important for a pro who is hard on equipment but not very important for me.

Those of you who bought the D50, I think, found the best value available in the Nikon DSLR line.

 

[benhasajeep]

Those of you who bought the D50, I think, found the best value available in the Nikon DSLR line.

My wife enjoys the d-50. Hard for me to get my hands on it. Not that I am stuck with the Canon 20D. Just I have better glass for the Nikon. And somehow inbetween dating and marriage the D-50 I bought and paid for before we were married and all the 2.8 lenses have somehow now become hers???