Why does D1x perform much worse than D30 on long exposure?

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X. However, after I saw some pix of 30 sec exposure from D1x and D30, I really think D30 performs much better than D1x in that area. I was wondering if someone can tell me it is true or not? Here are two links of the samples of the pix:

D1x (crop)



D30(original)

Both those samples you point to are VERY noisy, just in different ways. Neither is usable without further work. The D1x sample is what we D1x owners have come to expect with long exposures (I call it "the freckles"). Noise is also temperature related (the warmer the CCD, the more noise, all else being equal).

While others ask for more resolution, I'd take better long exposure and high ISO capability first any day.

Thom Hogan
author, Nikon Field Guide
author, Nikon Flash Guide
author, Complete Guide to the Nikon D1, D1h, & D1x
http://www.bythom.com

David Chen said:

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X.
However, after I saw some pix of 30 sec exposure from D1x and D30,
I really think D30 performs much better than D1x in that area. I
was wondering if someone can tell me it is true or not? Here are
two links of the samples of the pix:

D1x (crop)



D30(original)

Click to expand...

Really? In my monitor, the picture from D30 is very clean. However, the D1x sample is unusable. D1x is about twice price as D30. I can't believe D30 is much better in this area. Is D1H any better?

Thom Hogan said:

While others ask for more resolution, I'd take better long exposure
and high ISO capability first any day.

Thom Hogan
author, Nikon Field Guide
author, Nikon Flash Guide
author, Complete Guide to the Nikon D1, D1h, & D1x
http://www.bythom.com

David Chen said:

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X.
However, after I saw some pix of 30 sec exposure from D1x and D30,
I really think D30 performs much better than D1x in that area. I
was wondering if someone can tell me it is true or not? Here are
two links of the samples of the pix:

D1x (crop)



D30(original)

Click to expand...

Click to expand...

David Chen said:

Thom Hogan said:

While others ask for more resolution, I'd take better long exposure
and high ISO capability first any day.

Thom Hogan
author, Nikon Field Guide
author, Nikon Flash Guide
author, Complete Guide to the Nikon D1, D1h, & D1x
http://www.bythom.com

David Chen said:

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X.
However, after I saw some pix of 30 sec exposure from D1x and D30,
I really think D30 performs much better than D1x in that area. I
was wondering if someone can tell me it is true or not? Here are
two links of the samples of the pix:

D1x (crop)



D30(original)

Click to expand...

Click to expand...

Click to expand...

The D1X provides relatively clean images up to 10 secs, in the d30 a noise reduction system set in whch the D1X does not have. But you are not exactly comparing like for like are you? The image you have given looks realtively poor anyway and not really useable, unless this is the standard you are looking for.

Michael

One reason for this might be the D30s automatic dark frame subtraction capability. With long exposures, the noise created in the image is generated by 'dark currents'. 'Dark currents' are electrical currents that charge the imager elements even when it is not exposed to light, and are generally small enough not to pose a problem with normal exposure times. However, with long exposures (like 30s), it can add up to a substantial ammount of noise.

However, unlike normal image noise, this is somewhat predictable and can be removed rather easilly. The dark current is pretty well uniform, given a specific camera at the specific temperature and shutter speed that the image was taken. Therefore, you can remove the noise by taking a second exposure of exactly the same length right after the image, however with the lens cover on (so the frame should be perfectly black). You can then use Photoshop's layers to subtract the dark frame from the actual image, and you will get a considerably cleaner image.

The D30, however, does this automatically for you. On long exposures, the D30 performs automatic DFS. It first takes the image that you want, then it takes a second image at the same shutter speed (like the D1x, the D30s imager uses an electronic shutter to time the integration), but with the shutter curtains not opening for the exposure. Before saving the image to disk, the D30 then subtracts the dark frame from the image and removes the noise. I'm assuming this feature was enabled on the bellow image, however I may be wrong.

The D1X can do exactly the same thing, it just takes a little more work

David Chen said:

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X.
However, after I saw some pix of 30 sec exposure from D1x and D30,
I really think D30 performs much better than D1x in that area. I
was wondering if someone can tell me it is true or not? Here are
two links of the samples of the pix:

D1x (crop)



D30(original)

Click to expand...

Michael Oersigh said:

The D1X provides relatively clean images up to 10 secs, in the d30
a noise reduction system set in whch the D1X does not have. But you
are not exactly comparing like for like are you? The image you have
given looks realtively poor anyway and not really useable, unless
this is the standard you are looking for.

Michael

Click to expand...

I'm sorry that I haven't had any DSLR. The pix are actually for testing the noise leve in long exposure. To be honest, the pic of long exposure from D30 is the best I've ever seen. Before I saw the pix, I thought D1x is better than D30 in any aspect. Even now, I still think it may be due to the strange CCD on D1x and not because of Nikon. So, maybe someone can tell me D1H is much better than D1x in terms of long exposure. Or, is there any in camera adjustment in D1x to reduce the noise level to D30's standard?

Regards,
David Chen

I own both the Nikon and Canon systems (utilizing each of their strengths for different purposes) and you are right, the noise situation on the D30 is significantly better. Part of it is the noise reduction system. But that system only kicks in with exposures in excess of one second. The D30 has less offensive noise to begin with. (Some say not necessariy "less noise", but almost universally agree it's less "offensive" noise.) D30 shots made at 800 ISO are extremely usable, and, depending on conditions, even 1600 shots will get you by in a pinch. When I was testing the camera during a family trip to Disney World (a wonderful place for testing out a camera, with virtually limitless shooting conditions), I was amazed at the quality of some 1600 ISO shots handheld at 1/8 second in very dark environments. The low shutter speed was, of course, possible with Canon's amazing 28-135 Image Stabilization lens.

Shortly after the Disney trip, I was asked to shoot some chicken hatchery equipment (don't laugh) shots for Internation Hatchery Magazine (again, don't laugh). I'm typically a social photographer, but met someone at a wedding I shot who wanted me to consider the job. Forced to shoot in the manufacuring facility itself, there was absolutely no room for external lights. I experimented with long exposures and multiple flash gun bursts, but ultimately, I stopped down to about f11, set the ISO on 100, and let the shutter just stay open for about 10 seconds. It was very dark, but the light was somewhat even. I cloned the hell out of a little bit of red backdrop they had in the place, and they were very pleased with the results. I was pleased with such low noise images. In fact, working with the full-size images, I would not have known they were long exposures had I not taken them myself.

The original shots had no background to speak of, just tons of equipment and parts laying around, and people actually walking around. (Well, blur trails.) Here's the three they bought. And by the way, the angles had to be very specific so as to show certain parts of the equipment. Apparantly, chick hatchery pros really know their stuff.

http://michaelphoto.net/systeng/

Would the shots have been possible with Nikon. Sure. But mind you, I didn't have to do ANY processing work for noise. And these were JPGs, NOT raw images. (That's how much I trust the D30, although in all honesty, I had very little experience with the CRW format at the time.) Just profile conversion, levels, and then the masking/cloning work. (Thank god for the polygon mask tool.)

The D30, for me, is the clear winner when it comes to digital noise?

M

Thomas Sapiano said:

However, unlike normal image noise, this is somewhat predictable
and can be removed rather easilly. The dark current is pretty well
uniform, given a specific camera at the specific temperature and
shutter speed that the image was taken. Therefore, you can remove
the noise by taking a second exposure of exactly the same length
right after the image, however with the lens cover on (so the frame
should be perfectly black). You can then use Photoshop's layers to
subtract the dark frame from the actual image, and you will get a
considerably cleaner image.

The D30, however, does this automatically for you. On long
exposures, the D30 performs automatic DFS. It first takes the
image that you want, then it takes a second image at the same
shutter speed (like the D1x, the D30s imager uses an electronic
shutter to time the integration), but with the shutter curtains not
opening for the exposure. Before saving the image to disk, the D30
then subtracts the dark frame from the image and removes the noise.
I'm assuming this feature was enabled on the bellow image, however
I may be wrong.

The D1X can do exactly the same thing, it just takes a little more
work

David Chen said:

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X.
However, after I saw some pix of 30 sec exposure from D1x and D30,
I really think D30 performs much better than D1x in that area. I
was wondering if someone can tell me it is true or not? Here are
two links of the samples of the pix:

D1x (crop)



D30(original)

Click to expand...

Click to expand...

The idea of a 30" exposure on a digital thats clean is a few years away I believe. Even if you had a totally noise free image, in a 30" exposure you are going to get lots of gama radiation noise. This has no effect on film, but will effect a CCD or CMOS sensor.

The D30 in being CMOS has a major advantage on the D1x or any CCD camera, the CMOS chip has a lower energy requirement and most likely has considerable less emissions. Both of these have an effect on noise in a time exposure.

The higher the resolution of the CCD or Cmos chip will also effect the noise as the higher res. CCD currently made are very dense and the pixels are not as light sensitive as the larger D1 or D30 pixels.

Some of the noise in a 30" will also be stuck pixels. The longer you keep the CCD on, the greater chance of a stuck pixel. I am seeing more stuck pixels on my D1x at 1" or 2" exposures than I ever did on my D1. May just be my camera.

The best solution is the already mentioned dark frame subtraction in photoshop.

Paul Caldwell
[email protected]

Michael Oersigh said:

David Chen said:

Thom Hogan said:

While others ask for more resolution, I'd take better long exposure
and high ISO capability first any day.

Thom Hogan
author, Nikon Field Guide
author, Nikon Flash Guide
author, Complete Guide to the Nikon D1, D1h, & D1x
http://www.bythom.com

David Chen said:

Hi,

I'm about to buy my first DSLR. I'm truely like Nikon D1H/X.
However, after I saw some pix of 30 sec exposure from D1x and D30,
I really think D30 performs much better than D1x in that area. I
was wondering if someone can tell me it is true or not? Here are
two links of the samples of the pix:

D1x (crop)



D30(original)

Click to expand...

Click to expand...

Click to expand...

The D1X provides relatively clean images up to 10 secs, in the d30
a noise reduction system set in whch the D1X does not have. But you
are not exactly comparing like for like are you? The image you have
given looks realtively poor anyway and not really useable, unless
this is the standard you are looking for.

Michael

Click to expand...

I have my D1X set to 800 Asa permanently and get perfectly clean usable pictures from it, although I am only an amateur. The exception is some noise in darker areas, which disturbs me although it does not seem to disturb non-specialists. I think someone ought ot write some special software that filters this out. When it gets really ugly eg. dark velvet cloth, I have to photoshop the texture out brutally.

Edmund

Quantum Mechanic does a great job cleaning up digital camera NOISE.

Edmund Ronald said:

I have my D1X set to 800 Asa permanently and get perfectly clean
usable pictures from it, although I am only an amateur. The
exception is some noise in darker areas, which disturbs me although
it does not seem to disturb non-specialists. I think someone ought
ot write some special software that filters this out. When it gets
really ugly eg. dark velvet cloth, I have to photoshop the texture
out brutally.

Edmund

Click to expand...

Qimage Pro has a quite sophisticated dark frame subtraction function...

you take sample shots at the same speed+aperture with the lens cap on and store these in a directory, then you simply tell Qimage where they are and it automatically locates the closest match to any given image when you apply the filter.

More details in the "learning by example" section of the extensive built in help to Qimage Pro.

Cheers

Bri

Edmund Ronald said:

I have my D1X set to 800 Asa permanently and get perfectly clean
usable pictures from it, although I am only an amateur. The
exception is some noise in darker areas, which disturbs me although
it does not seem to disturb non-specialists. I think someone ought
ot write some special software that filters this out. When it gets
really ugly eg. dark velvet cloth, I have to photoshop the texture
out brutally.

Edmund

Click to expand...