Fred Greissing wrote: Looks like the D800 is better than the 5D Mark III at ISO 12,800 when comparing scaled RAW files.
The inset is the d800 sample scaled to match the 5D sample.
LMAO.. and to think I thought you knew something about camera's... LOL... now I KNOW your just a troll.
The fact that the OP is suggesting that the D800 is better at ISO 12,800 than the 5DIII reveals what little and limited knowledge he has despite his assertions. I have not seen a single reputable review (or any review for that matter) or comparison between the two that would suggest that the D800 is better at ISO 12,800. This should be a red flag for anyone reading the OP's "analysis" of this or any equipment in the future.
Next we will be hearing that the D800 has faster FPS... lol..
They are both excellent camera's that offer their own advantages. Do your homework by educating yourself with reputable reviewers and take absurd claims such as this, as nothing more than an attempt to garner attention.
I don't think my has a greenish hue but I do swear my focus point is moving by itself or the selector is more sensitive and I'm inadvertently moving it by accident...
David Poata
Posts: 421
Hamilton, Waikato, New Zealand
ACPhotography wrote:
I don't think my has a greenish hue but I do swear my focus point is moving by itself or the selector is more sensitive and I'm inadvertently moving it by accident...
MC Grain wrote: Can you explain the scaling process? For whatever reason, I don't totally get it.
My real question is, doesn't scaling always make for a better image?
If you have two identical sensors, but one is a crop, won't the appropriate scaling method make the FF look better?
In that case, it's kind of hard to say which is the better sensor - no problem to sy which is the better results.
What I did is quite simple. I took the sample of the Nikon D800 on the dpreview page and scaled it down so that the scale matched the scale of the watch in the canon sample. I then made a gif that animates between the two to shot the difference.
So far all ISO comparisons I have seen are looking at the image scaled to 100%, but the true difference should be show with both images reproducing the subject at the same size. the Nikon is 36MP and the Canon is 22MP. So to compare the two in the real world so to speak you need to scale down the Nikon.
That said they are both pretty darn good, but from what I have seen the Nikon does a better job especially in the shadows and above all skin tones in and approaching the shadows.
I'm also referring to high ISO, not ultra high ISO.
I don't need anything above 12,800.
Mike Haftel wrote: Me, too. But I'm actually waiting to see if Nikon fixes the LCD screen issue.
I don't know that there is a screen issue. The only site talking about it as an issue is Nikon Rumors. As far as I can tell, those who are complaining have their LCD screens set wrong.
Fred Greissing wrote: Looks like the D800 is better than the 5D Mark III at ISO 12,800 when comparing scaled RAW files.
The inset is the d800 sample scaled to match the 5D sample.
You think the inset looks better? Really?
It took me a while to figure out which one was which, and I assumed you meant the mushy, low-contrast inset was the Mk III. Only after looking at the animated GIF in another browser (I keep all animations turned off in my primary browser) did I come back to see which one you said was which.
London Fog wrote: Good luck with this one Fred, even though you are 100% right!
These Canon boyz are so touchy, so easily offended and end up getting you brigged over nothing - beware!
I'm a bit of a Canon fan... about 8 years with digital canons. Only one repair required.
That was done nice and quickly, been just fine since. I also have a slew of Canon lenses. I was ready to snap up a 5D mark III, but I was disapointed to see no improvement in the shadows. One of the main differences image quality wise that I see in my Phase One cameras and my Canon's is the dynamic range and shadow quality. From what I am seeing the Nikon appears to be significantly better in that area. Good detail into the blacks is like the foundation of good image quality.
David Poata
Posts: 421
Hamilton, Waikato, New Zealand
Doobie the destroyer wrote: 36mp is for photo quality A2 prints. That's commercial advertising sizes. I highly doubt most people, including pros, will utilize that.
Although, why not go medium format? I imagine for prints of those sizes you want the best editing versatility you can extract, which the 16-bit files are a godsend for.
I definitely think the D800 would be better for field work, but I still don't think the D800 is an alternative for MF, more so a supplement. Like how a compact can be a supplement for a DSLR.
Doobie the destroyer wrote: Than it's probably a good choice for you.
Although, why not go medium format? I imagine for prints of those sizes you want the best editing versatility you can extract, which the 16-bit files are a godsend for.
I definitely think the D800 would be better for field work, but I still don't think the D800 is an alternative for MF, more so a supplement. Like how a compact can be a supplement for a DSLR.
Just because the files are 16 bit doesn't mean they have more information. The D800 almost certainly has more dynamic range.
David Poata
Posts: 421
Hamilton, Waikato, New Zealand
Doobie the destroyer wrote:
Than it's probably a good choice for you.
Although, why not go medium format? I imagine for prints of those sizes you want the best editing versatility you can extract, which the 16-bit files are a godsend for.
I definitely think the D800 would be better for field work, but I still don't think the D800 is an alternative for MF, more so a supplement. Like how a compact can be a supplement for a DSLR.
MF entreats a concise shooting technique, slower work flow and at times chained to a tethered station. Not my ideal choice but I will concede to MF depending on client requirements.
I much prefer the faster AF, higher ISO, smaller form factor/portability of the D800. So for me the extra resolution is a God send.
Can someone correct me if I'm wrong when the 5d MKII and the D700 came out everybody did the scale down process from the 5d to the d700 and showed similar results as these? Meaning the bigger MP file had better noise performance.
I pretty much don't care about the leapfrog dilema, I work with two old rebels and the results I'm seeing from the 5d and the d800 are flawless compared to my rebels lol so What ever anybody gets (be it nikon or canon) they're pretty much golden.
Looking at those source images, the Nikon's noise is more bothersome to me.
You made a JPEG from posted JPEGs, right?
People used to compare two tracks on a CD, one was an analog recording and one was digital. They comments were always "This one sounds digital, this one sounds analog." What never made sense to me is once they're on a CD, shouldn't they both sound digital?
In photography terms, don't we really need to compare the RAWs side by side? And even then, do you compare them in the same RAW converter or the manufacturer's RAW converter?
And, if you're going to scale to have a results based comparison, shouldn't you be applying NR? I don't know if there's such thing as one taking NR better, but if the idea is to compare likely results shouldn't they be converted images with contrast and color adjustments too?
No one besides the photographer is ever going to see a straight RAW image.
Mike Haftel wrote: Me, too. But I'm actually waiting to see if Nikon fixes the LCD screen issue.
Suggestions for affected LCD's are the green cast could be fixed by adjusting the LCD monitor hue - see page 52 of the D800 manual
ei Total Productions wrote:
I don't know that there is a screen issue. The only site talking about it as an issue is Nikon Rumors. As far as I can tell, those who are complaining have their LCD screens set wrong.
Nope. I've done my research. The LCD issue is very real and has nothing to do with the Live View Screen Hue adjustment, and there are people reporting about it all over the place, not just on NR. Resetting LCD Hue also resets every time you turn the camera off.
Either way, I'm waiting a month or so (by then the camera is hopefully available in stores).
Just because the files are 16 bit doesn't mean they have more information. The D800 almost certainly has more dynamic range.
The difference from a 14-bit file to 16-bit is pretty is pretty big. It's 16,384 units of information vs 65,536. Also, I don't think Nefs are true 14 bit files. You have a much bigger margin of error for shooting and a much better file to work with in post. A 3fr can recover like 2-3 stops either way with no visible quality loss.
A retoucher, designer, and graphic artist would have much better file to play with from something like a 3fr rather than a Nef.
Also, a MF cameras will still render better clarity and bokeh as well, just based on the size of the sensor and lenses.
I definitely agree with that the D800 will be faster and easier to shoot with, though. However, I also imagine shooting speed isn't so much of a factor when your doing commercial work that requires A2s, and it is more about the highest quality output at every step of the workflow.
David Poata
Posts: 421
Hamilton, Waikato, New Zealand
Doobie the destroyer wrote: MF cameras will still render better clarity and bokeh as well, just based on the size of the sensor and lenses.
I definitely agree with that the D800 will be faster and easier to shoot with, though. However, I also imagine shooting speed isn't so much of a factor when your doing commercial work that requires A2s, and it is more about the highest quality output at every step of the workflow.
As an ergonomic compromise the Leica S2 looks to be the best of both worlds in terms of MF quality in a 35mm form factor
Doobie the destroyer wrote: The difference from a 14-bit file to 16-bit is pretty is pretty big. It's 16,384 units of information vs 65,536.
14 bit or 16 bit ... that is the "container". It does not mean that it is filled using all the 16,384 steps per channel.
From my experience with both 14 bit and 16 bit files the difference is not huge, but it is there. However a very good film scan at 16 bit better than the best 16 bit file from a camera CCD.
| 
| 
| edu