Nikon D800E with 14-24

Started Jul 16, 2012 | Discussions
Marianne Oelund Veteran Member • Posts: 7,788
Oops

Dana wrote:

No autofocus is set up to activate with half press of the shutter or in this case the remote cable release.

That would imply that the photos were taken with the phase-detect AF system ("viewfinder focusing") active, and the Live View focus setting wasn't used. Unless you had the lens AF switch off, or the camera body AF switch off. Could you clarify?

Luke Kaven Veteran Member • Posts: 5,715
Proofish -- here's the code

It looks to me like they're using a Huffman coding. Here's the source code from dcraw.

See the routine "nikon_compressed_load_raw".

http://ninedegreesbelow.com/photography/floating-point-dcraw/dcraw-float.c

Now why in the world would anyone use a huffman code in any way that was not truly lossless?? It's not an approximate method. I don't know how you could use huffman for lossy encoding unless you lossied up the data before doing the huffman coding.

primeshooter
primeshooter Veteran Member • Posts: 5,990
Agree but...

Marianne Oelund wrote:

Jaelkay wrote:

Marianne Oelund wrote:

By the way, there is no reason to use uncompressed RAW in photography; it's > just a waste of space.

Huh?

Use lossless compressed, at least.

You do have a solid and verifiable reason for making these statements? Yes?

About as solid as it gets: I've developed compression/decompression code for use in flight control computers. Lossless means precisely that: There is no difference between the original uncompressed data, and the data that comes back from the decompression process. If there was any difference - even a single bit - Boeing 777 airplanes wouldn't be flying.

I agree, however there is one difference. Because if you select uncompressed your pc doesn't have to work as hard to uncompress the file, it makes editing programs go faster, worth noting.

stoic little New Member • Posts: 21
Re: Agree but...

Not necessarily true. Though the CPU will have to decompress the file first (piece of cake for most CPU), it will have a smaller file to work on which more is significant since reading large file is slower than decompression.

primeshooter wrote:

Marianne Oelund wrote:

Jaelkay wrote:

Marianne Oelund wrote:

By the way, there is no reason to use uncompressed RAW in photography; it's > just a waste of space.

Huh?

Use lossless compressed, at least.

You do have a solid and verifiable reason for making these statements? Yes?

About as solid as it gets: I've developed compression/decompression code for use in flight control computers. Lossless means precisely that: There is no difference between the original uncompressed data, and the data that comes back from the decompression process. If there was any difference - even a single bit - Boeing 777 airplanes wouldn't be flying.

I agree, however there is one difference. Because if you select uncompressed your pc doesn't have to work as hard to uncompress the file, it makes editing programs go faster, worth noting.

PhotoBabble Forum Member • Posts: 87
Re: Oops

As far as I can tell, when in LV mode, a half press of the shutter will engage contrast based AF, as long as the camera is configured with AF on the shutter release. My first thought was that LV focussing required an AF-ON press, regardless of how the cam was configured (I'm probably just thinking this because that's how I have my cameras set up). But the D800 manual suggests otherwise.

After a little more thought, this has to be the case, because in LV the mirror is flipped up and there's no way for the phase detect AF module to "see" the scene.

OP, just to humor us, could you try focussing via AF-ON first, and then take the shot via cable release - just in case LV mode defaults to AF via AF-ON only (unlikely, but it's a simple test and worth ruling in or out).

PhotoBabble Forum Member • Posts: 87
Just Tried It On My D800/14-24

Just tried LV focus via a half press of the shutter release on my D800/14-24, and it does indeed use contrast-based AF.

OP, I assume you've verified AF is being activated in your setup? By racking the focus until the LV image is clearly OOF, and then engaging AF and seeing the LV image come into focus.

Marianne Oelund Veteran Member • Posts: 7,788
OP procedure?

PhotoBabble wrote:

Just tried LV focus via a half press of the shutter release on my D800/14-24, and it does indeed use contrast-based AF.

OP, I assume you've verified AF is being activated in your setup? By racking the focus until the LV image is clearly OOF, and then engaging AF and seeing the LV image come into focus.

Do we know whether the exposure was actually taken during LV, or was LV only used to focus, then disengaged?

Of course, if we could see a sample that's at 100%, we could determine whether AF is an issue.

John M Roberts Senior Member • Posts: 2,649
Re: Nikon D800E with 14-24

anotherMike wrote:

FYI: The sharpest focal length on the 14-24 that I've found in a few rounds of testing is 16mm - where it's quite seriously sharp by f/5.6 - F/7.1 (on my d800E), 14mm isn't quite as nice, but still pretty good, and 20mm is also pretty good. The lens is weakest in the corners and edges at 24mm, but still pretty sharp in the center at that focal length.

Do you find the 24 on the 14-24 perform better than the 24-70's 24?

Raul Veteran Member • Posts: 8,583
The danger here is not lossless losing but we losing

Ms. Olelund. She may get tired of nonsense and depart as most true experts have done. I have all my fingers crossed so this won't happen.

-- hide signature --

"In the days when manufactures laugh at their clients face."

John M Roberts Senior Member • Posts: 2,649
Re: Too small to be able to tell

Also the plane of focus is limited to a very small part of the overall image to judge. Choose a scene with more included all the way to the edges in your point of focus.

PatFahey Regular Member • Posts: 453
Re: It's only lossless using Nikon's software, so no marketing issue....

Luke Kaven wrote:

Most all the programs out there that read raw files, all use the same library, LibRaw, based on the original dc_raw. The code can be freely inspected.

Thanks, Luke.

Is there any evidence to show that Nikon is using these routines for the D800? Is this something that Nikon even discusses?

-Pat

anotherMike Forum Pro • Posts: 10,105
Re: Nikon D800E with 14-24

Actually, a good question. The 24-70 is quite weak at 24 with every previous body I've tested it on, so I never even bothered to test it on the d800E at 24mm - just didn't figure it was worth the time at that specific focal length (the level of testing I do takes significant time). At some point it might be worth looking into, although I'd be really surprised if the 24-70 was suddenly good at 24mm (Note, I don't knock the 24-70: it's REAL good at 35mm; I've proven this to myself beyond any shadow of a doubt on the D800E).

Not gonna happen for a while though, due to time constraints.

-m

Luke Kaven Veteran Member • Posts: 5,715
Re: It's only lossless using Nikon's software, so no marketing issue....

PatFahey wrote:

Luke Kaven wrote:

Most all the programs out there that read raw files, all use the same library, LibRaw, based on the original dc_raw. The code can be freely inspected.

Thanks, Luke.

Is there any evidence to show that Nikon is using these routines for the D800? Is this something that Nikon even discusses?

Well, I think everyone who ever wrote a routine to demosaic a bayer sensor has studied dc_raw to start. It's been enhanced by different developers for inclusion in their own products.

But one thing seems clear, and that is that Nikon uses a Huffman coding for their lossless compression. Bill Claff also notes the same. The code is pretty clear about that.

Huffman codes are not something you would ever use for lossy encoding.

Some others, e.g., DNG, use "lossless JPG" encoding. Yes, oddly enough, there is a lossless version of the JPG spec.

Robin Casady Forum Pro • Posts: 12,898
Re: Agree but...

primeshooter wrote:

I agree, however there is one difference. Because if you select uncompressed your pc doesn't have to work as hard to uncompress the file, it makes editing programs go faster, worth noting.

Depending on the computer, it often takes longer to read the uncompressed file than to read and decode the compressed version.
--
Robin Casady
http://www.robincasady.com/Photo/index.html

OP Dana Junior Member • Posts: 48
Re: Just Tried It On My D800/14-24

Yes. It is definitely focusing with a half push of the remote cable button. At this point, I don't know whether it is a lens issue or a camera issue. The only other lens I have to test is the older 24-85 and when I check focus through LV with both set at 24mm and enlarge the image to max size and alternate between AF and manual focus, they both look about as sharp. This surprises me as I would think the 14-24 should be have better sharpness. Am I wrong?

Luke Kaven Veteran Member • Posts: 5,715
Re: Just Tried It On My D800/14-24

Hi Dana,

The near side of the wooden bench looks like the focal point. There is abundant detail in the twig sticking out over the bench, and in nearby foliage. And I have to say that this looks correct to me as far as the expected behavior of the lens.

Unlike the UWA lenses of old, this lens requires more in the way of planning. For some reason, slight errors in focus are much more noticeable on the 14-24 than they are on historic lenses.

Perhaps it is in contrast to its native sharpness that out of focus regions look more out of focus. But when things are slightly out, they look out on this lens. In this case, it might be the fact that you are focusing rather close to the lens, and the depth-of-field at that distance isn't that much.

Dana wrote:

Yes. It is definitely focusing with a half push of the remote cable button. At this point, I don't know whether it is a lens issue or a camera issue. The only other lens I have to test is the older 24-85 and when I check focus through LV with both set at 24mm and enlarge the image to max size and alternate between AF and manual focus, they both look about as sharp. This surprises me as I would think the 14-24 should be have better sharpness. Am I wrong?

John M Roberts Senior Member • Posts: 2,649
Re: Nikon D800E with 14-24

Thanks Mike. I do have the impressions from other posters with similar observation. I try to stay no less than 28mm on mine and if I'm going to be wider, lug the bulbous 14-24 out. If you have experience using a grad ND with yours I have another thread in nikon lenses that I can't seem to generate any knowledge to it.

OP Dana Junior Member • Posts: 48
Re: Just Tried It On My D800/14-24

Luke,

I was using a focus point in lower right corner. From your experience is that a weaker part of the lens? I was also shooting at 5.6 and 1/8 sec. Even though it was on a tripod, and was triggered with a remote cable, the ground was not particularly hard. Do you think 1/8 sec could pick up slight vibration, enough to affect the focus? I will definitely work on some photos in a more controlled environment. I am not excited about the thought of sending in the camera and lens to Nikon. Thanks.

Robin Casady Forum Pro • Posts: 12,898
+1: The danger here is not lossless losing but we losing

Raul wrote:

Ms. Olelund. She may get tired of nonsense and depart as most true experts have done. I have all my fingers crossed so this won't happen.

I am amazed by the idiots arguing with Ms. Olelund in this thread. I hope she will ignore them and continue as a valued contributor to this forum.
--
Robin Casady
http://www.robincasady.com/Photo/index.html

Robin Casady Forum Pro • Posts: 12,898
Re: Just Tried It On My D800/14-24

Dana wrote:

Do you think 1/8 sec could pick up slight vibration, enough to affect the focus? I will definitely work on some photos in a more controlled environment. I am not excited about the thought of sending in the camera and lens to Nikon. Thanks.

This is a shutter speed that is very likely to be affected by mirror flap. Were you shooting in Live View, or did you turn LV off before releasing the shutter?

For more controlled shooting, disable AF activation by the shuttter release button. It is Custom Setting a4. Active focus with the AF-ON button only. If you are going to use LV to check focus, you do not want it changed when you press the release button.

The extreme AF points tend to be the least accurate.
--
Robin Casady
http://www.robincasady.com/Photo/index.html

Keyboard shortcuts:
FForum MMy threads