These being with high ISO NR off: there's no question in my mind that this beats the D700 at high ISOs, and it's not a marginal victory for the D800 either.
ausemmao wrote:
These being with high ISO NR off: there's no question in my mind that this beats the D700 at high ISOs, and it's not a marginal victory for the D800 either.
That's a bold statement to me since I am aware of the strengths and weaknesses of the D700's high ISO files. There are a lot of chroma noise at the expense of details in the D800's 6400 samples when viewing them at 100% crop on a computer screen. That's unlike anything I've seen with my D700 files at ISO 6400.
What I'm curious to see is how the 12 MP compares to the downsampled 36 MP side by side when both are shot at the same high ISO settings. At this point, I think it's too premature to say that the D800 outperforms the D700 at high ISO.
honorerdieu wrote:
That's a bold statement to me since I am aware of the strengths and weaknesses of the D700's high ISO files. There are a lot of chroma noise at the expense of details in the D800's 6400 samples when viewing them at 100% crop on a computer screen. That's unlike anything I've seen with my D700 files at ISO 6400.
What I'm curious to see is how the 12 MP compares to the downsampled 36 MP side by side when both are shot at the same high ISO settings. At this point, I think it's too premature to say that the D800 outperforms the D700 at high ISO....Show more →
Chroma noise can be removed without destruction of detail. That's why LR3 has a +25 ChNR on by default. Here is a 100% crop of D700 ISO 6400 raw in LR3 with ChNR set to zero. Notice how the noise varies across the exposure scale.
honorerdieu wrote:
That's a bold statement to me since I am aware of the strengths and weaknesses of the D700's high ISO files. There are a lot of chroma noise at the expense of details in the D800's 6400 samples when viewing them at 100% crop on a computer screen. That's unlike anything I've seen with my D700 files at ISO 6400.
What I'm curious to see is how the 12 MP compares to the downsampled 36 MP side by side when both are shot at the same high ISO settings. At this point, I think it's too premature to say that the D800 outperforms the D700 at high ISO....Show more →
There's loads of chroma noise in a D700 file as well.
If the D800 looked the same at 100% view as the D700 did at 100% it would be hands down the best low light camera ever produced.
In other words, you're comparing a 50" image from the D700 to an 86" image for a D800 and requiring the 86" be as good as 50" to deem them equal. That's absurd.
As it is, it's definitely better. The reason I was hesitant before is because I hadn't checked exif and so did not want to assume a lower level of NR than had been used. Now I know it's NR off and therefore worst case, it's not difficult at all to see which is better.
ausemmao wrote:
There's loads of chroma noise in a D700 file as well.
If the D800 looked the same at 100% view as the D700 did at 100% it would be hands down the best low light camera ever produced.
In other words, you're comparing a 50" image from the D700 to an 86" image for a D800 and requiring the 86" be as good as 50" to deem them equal. That's absurd.
As it is, it's definitely better. The reason I was hesitant before is because I hadn't checked exif and so did not want to assume a lower level of NR than had been used. Now I know it's NR off and therefore worst case, it's not difficult at all to see which is better....Show more →
You overlooked the part where I stated it would be interesting to see the comparison of the D700 12 MP image to the 36 MP that has been downsampled. I am aware of the chroma noise in the D700 files, but it wasn't at the expense of detail. There's a difference... and that's not absurdity.
Absurdity is making a bold claim about the D800 outperforming the D700 at high ISO based on one sample image when the D800 hasn't even been released.
Comparing a 12 and 36 MP file both at 100% and being surprised that the 36 file has less detail at high iso at 100%...
It's utterly trivial to see a downsampled image - set it to fit screen size. Or go up in size from there, or download the files and run a downsampling algorithm of your choice on it.
I'm making that 'bold' claim based on the 640, 800, 3200, 6400 and 25600 samples, and using a D7000 to project an FX camera with similar pixel technology...Unless you wish to assert that Nikon is deceiving us and have messed around with the files in a way that reviews will reveal, it's not at all a 'bold' claim.
I've been working in LR with the ISO 25K image of the Fuji on the table. I can pull a nice image off of it that'd I'd be happy delivering to a client if I were shooting an event or theatrical performance. That's impressive to me.
Comparing a 12 and 36 MP file both at 100% and being surprised that the 36 file has less detail at high iso at 100%...
It's utterly trivial to see a downsampled image - set it to fit screen size. Or go up in size from there, or download the files and run a downsampling algorithm of your choice on it.
I'm making that 'bold' claim based on the 640, 800, 3200, 6400 and 25600 samples, and using a D7000 to project an FX camera with similar pixel technology...Unless you wish to assert that Nikon is deceiving us and have messed around with the files in a way that reviews will reveal, it's not at all a 'bold' claim. ...Show more →
You are appearing to be more ridiculous as the discussion progresses.
Again, I have said it is premature make any claim about the high ISO performance of the D800 before it has been released. I have stated that I am "curious" and "interested" about the findings.
That means I'm being open minded about whatever the findings are. But it seems you have a problem with me not agreeing with your claims.
Tyl3n0L wrote:
Yes it's sure one hell of a monster at 36mp, but that's about it isn't it? I don't really care about the video feature. I mean the D700 ISO seems to be quite a bit better and the D700 is also faster. Even my older D90 is 4.5fps... This is two specs I was mostly looking for, High ISO & Speed.
I think the main problem here is that Nikon are selling the D800 as the replacement for the much loved D700, which was, still is and will continue to be an awesome camera. People got used to the speed and high ISO capabilities and reasonably expected its replacement to build on those capabilities, improving things as they went.
If the D800 had instead been launched as a replacement for the D3x, which in my mind is what it really is, then perhaps everyone would be looking at things a little differently. After all, the D3x was no speed demon and nor was it a high ISO demon.
Here we are expecting a 36MP camera that is less than half the price of the D3x, at significantly increased resolution to be a sports camera (high fps) and low light monster (D700-D3s territory).
Folks, you really are expecting too much. Sorry.
To me, a true D700 replacement would have been 16MP-18MP with better CR and great high ISO, like a mini D4
Tyl3n0L wrote:
Yes it's sure one hell of a monster at 36mp, but that's about it isn't it? I don't really care about the video feature. I mean the D700 ISO seems to be quite a bit better and the D700 is also faster. Even my older D90 is 4.5fps... This is two specs I was mostly looking for, High ISO & Speed.
If you were comfortable with the cost of the D700 and FX lenses, you'd be very comfortable with the cost of the D800. Your "problem" is that you're comparing apples and oranges, because the costs of DX and FX are nowhere near comparable. Neither are the various performance metrics: this camera is not at all designed for "high ISO and speed". Other cameras are aimed at your needs, but not this one.
Your next camera is likely the D400, but it might also be a used D700, D3, or D7000. This is a year when Nikon renews a whole bunch of things. There is no way the D800 is the likely successor or alternate to your D90... too big a difference in a lot of ways.
Seems par for the course to me. It is after all a 36MP therefore the pixel pitch will be quite a bit smaller and therefore more prone to noise. Stop comparing a tiny 12mp file to high res file. We have seen the same with the 5d2 files too. It is comparing apples to oranges. You can hide so many imperfections with smaller resolution files. Heck that's why people still ogle over the 4MP 1d/d1h files.
When you downsize a 36MP file to 12MP you get results with better definition than if it were shot in the 12MP camera with an AA filter to begin with. To properly compare it you need to downsize it past 12, to perhaps 8MP then upsize it to 12MP and apply noise reduction.
Personally this is a total non-issue to me. I want very high quality images at exposure times of 1-30 seconds at base ISO on a tripod. Or I want an okay image at up to ISO 400-800 handheld. Or I shoot at high ISO and need a 1-2MP image for newsprint, web, etc - it will be so downsized it doesn't matter how noisy it is and the reader wouldn't care anyway. The great thing about the D800E is that it can wear all these hats. One camera (system) is better than three. Consolidation has inherent advantages! (For one, you're more likely to have the right tool for the job on hand at any time.)
pr4photos wrote:
When I say noisy, I meant I downloaded the images, downsized them to a size that I would be sending to clients, and compared it to existing images from the D700 that I have sent to clients. Thats my real world test. I know they will be good as prints, but I have to compare at sizes that my client will receive. And at that size they are noisier than the D700quote]
Did you download the FullSize JPEGs?
You should have also downloaded the better downsized ones.
They have no in camera noise reduction and are better at 25,600 than my D3 images were.
The D700 has that same Sensor. ...Show more →
