I just saw 5dmk3 and Nikon D800 samples on their official sites and I'm...

Started Mar 6, 2012 | Discussions thread
Skip M Veteran Member • Posts: 7,174
Re: I just saw 5dmk3 and Nikon D800 samples on their official sites and I'm...

bobn2 wrote:

Skip M wrote:

bobn2 wrote:

Skip M wrote:

bobn2 wrote:

That depends on how much DR you have. You are used to a camera with around 11 stops of Dr, not one with around 14 stops.

So, Bob, you're expecting the D800 to have a two stop improvement in DR over the D3s??

A bit more than 2 stops, in fact.

Wow, that's some technological improvement.

Not so much, it's about 0.6 stop over the APS-C version of this sensor, and not so much a 'technological improvement' as an area increase.

Yes, I'm used to a camera with slightly over 11 stops of DR, and I'm seeing maybe one stop improvement over that in the D800, possibly a little more, but not two stops.

You wouldn't be seeing anything if the exposure and ISO setting has not been judged to take advantage of that DR.

The 5D mkII is rated by DxO as 11.9 stops, the D3s at 12, so 14 stops would be a two stop improvement over those two cameras. Do you have any actual, concrete proof that we can anticipate that much of a jump?

Absolutely. We have DxO's DR for the APS-C version of this sensor, 13.9 stops, we have the increase that will come form the larger size/pixel count, 0.6 stops to give 14.5 stops. there is not really any room for argument over the DxO rating, as DxO itself says:

" the Nikon D800 will inevitably achieve the highest DxO Mark sensor scores ever measured "

But then DxO knows how their measurements work.

There looks to be, going by DxO's scores, something of a compromise, though. The D3s scores 3253 on low light ISO, the D7000 comes in at a relatively paltry 1167. And this for a DR improvement over the D3X of 1/5 of a stop, a camera that rates somewhat higher in ISO rating (1992).

To comment one needs to understand what the DxO measurements mean. The 'high ISO rating' is the ISO exposure at which the sensor achieves a normalised 30dB SNR. This SNR is a 'total light' thing, so depends not just on the exposure (which should be fixed by the ISO rating) but also the sensor size. Since the D3s has a sensor 2.25 times the area of the D7000, the smaller camera is actually performing at the same level as the a FF camera at 2625 ISO. not quite up to the D3s, but less than 1/3 stop behind, exactly where the D800 was expected to be. As it happens the D800 is turning out a bit better than this.

I'm not sure which I prefer, having less of a chance of blowing all the highlights of a bride's dress but preserving the details in the groom's tux (or blocking up the shadow, but getting all the detail in her dress!) or being able to get dang good, clean images in a dimly lit church without flash.

Again, you don't have to choose because the D800 gives you the full 2+ extra stops DR while only being 1/5 stop behind in low light capability. Of course to get the full 14.6 stops DR you need to set it at base ISO, but its low light capability at that setting is nearlt as good as it is at higher settings.

But it's 2 stops behind the 5D mkIII in ISO capability. I have to decide which is more important to me. And, frankly, ISO capability is more important. Also, if the D800's sensor behaves like that of the D7000, you lose all of your DR addvantage by ISO400, according to DxO. It seems to drop precipitously from ISO100 to ISO 325 or so, and actually drops below that of the 5D mkII and D700 from there on, if only by a little bit. I spend most of my time shooting at ISO200, so it would still have an advantage over those two cameras, but as soon as I moved indoors, it would lose that advantage.

By the way, if, as you often assert, smaller pixels result in lower noise levels, why is it that the SNR of the D700 is lower than either the D7000 or the 5D mkII?

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