Image problem

[Alan W]

I use 2 D70's and took some photos outside the other day and in some of the images on both cameras I noticed that the background was a bit contrasty and looked like it had been 'scanned'. I had just received two new 1Gb CF cards and immediately presumed it was them as it was the first time that they had been used. However, I now think that it may have been caused by the extremely cold weather on the day. Has anyone experienced this or read that this can happen?

[BarryM]

i wouldnt have thought it was anything to do with the cards, there just storage area for the images, so id assume it has to be camera related. But no idea what.

[tonymidd]

Hi Alan,
Have you tried swoping the cards between the cameras or using other cards in them? Have you had this problem with the cameras before? I never did with the D70 you had off me so it's more than likely the cards. What make are they?

[Stemmy]

Post an image lets have a look. Sounds like a camera issue rather than card.

[tonymidd]

Post an image lets have a look. Sounds like a camera issue rather than card.
On both cameras? From what Alan is saying he's not had a problem before but with these new cards he has had this happen. As it was a cold day that may have had an effect tho' I've shot with both D70 and D200 in very cold conditions without a problem.
Agree sight of an image would maybe help.

[Stemmy]

Sorry hadnt noticed it was two cameras. Aplogies :o

Cant believe its a card problem. Its just numerical dfata isnt it. Its either there or it isnt. The file is either readable or corrupt.

I cant see how a change in temperature could affect contrast within an image.

Maybe it was just one of those situations that seem to fool the cameras metering systems. I still can never 100% predict what the camera is going to do. That is the big difference - with my Bronica I always felt 100% in control.

[Choogster]

could have been the cold weather affecting the cameras but I`ve used Canon in -20 without problems. Not familiar with Nikon in cold weather.

[Alan W]

Okay, heres an example. Just loook at the noise in the zoomed in crop. Shot in RAW and at low iso.

[StagePhotography.co.uk]

I got a D70.... (in at jessops at the moment getting fixed... I broke it it's not a fault...)...

I'll let you know when it comes out, but I never noticed
noise like that...

How low ISO?

[Alan W]

Hi Stage. 320 iso.

[DennisB]

The background looks a little over exposed to me. Was the overall image tone lifted in processing from raw? I have seen similar noise in some of my own images (not same camera) where I have lifted the image tone to recover datail from underexposed shadows. Particularly noticeable at higher ISO. I would not class 320 as low ISO, but I suppose it all depends on what is available on your camera. To me, low ISO would be 100, and by preference, I would select 200 for most outdoor work in reasonable lighting.

Like other posters, I have never noticed any problem shooting in cold weather, except that the batteries always seem to be just coping with things and need a bit of pampering to keep going.

[Choogster]

320 or 3200? if its 3200 then its no wonder. I shoot the majority of stuff in ISO100.

[Stemmy]

I have also seen some noise when bringing back an underexposed image.

I shoot 200 ISO for outside.

[tonymidd]

I wonder if this amount of 'grain' at the degree of enlargement of the part image would have been objectionable at the correct viewing distance if ISO400 colour neg film had been used?

I don't know if Alan has used film in the past but those who have mainly used digital are somewhat paranoid about noise. Now I'm the first to admit there's nice noise that's sharp and film grain like and these's slush. The latter usually found in in-camera processed jpegs. The noise off a Nikon is usually the former, as this is.

Personally I like a bit of texture in my pictures so long as it meets the above parameters.

[Tim]

I've never heard of an ISO rating of 320 (just me?).. It does look like the kind of noise you'd get from ISO3200.. Hopefully that's the case and you have nothing to worry about..

[tonymidd]

I've never heard of an ISO rating of 320 (just me?).. It does look like the kind of noise you'd get from ISO3200.. Hopefully that's the case and you have nothing to worry about..
digi Nikons go in 1/3 stops; the D70 and D200 go ISO200, 250, 320, 400, 500, 640, 800, 1000, 1250, 1600, no 3200 on the D70 tho the D200 has three high ISO settings which I don't use as I understand they are interpolated in some way, it also has lower settings of 100, 125, & 160.
I'm wondering which conversion prog Alan has used as some are much better than others at handling noise.

[Tim]

Ah that explains it (Canon man myself!!) :)

[Alan W]

Sorry guys, Iv'e been a bit tucked up and been unable to join in the thread. Okay, let me confirm.. the cameras are both set at 320 iso. I took some more photos the other day and once again saw this grain ij the shadows. The shots are RAW and I use the RAW converter in CS2 to open the files which are then saved as highest quality jpegs. The images were fairly underexposed so Martin may have a point there. I am going to do some more tests to see what I can come up with.

[tonymidd]

. The shots are RAW and I use the RAW converter in CS2 to open the files which are then saved as highest quality jpegs.
Why shoot RAW and save a lossey Jpeg?? Surely a lossless Tiff would be better. Why take an image with all the data and then throw at least 25% of that data away?

http://www.nikonians.org/html/resources/guides/digital/jpeg_tiff_or_raw_1.html

[Alan W]

Hi Tony, the answer to that is it's because my lab requires images as jpegs. Don't nearly all of them?

[tonymidd]

Hi Tony, the answer to that is it's because my lab requires images as jpegs. Don't nearly all of them?
The lab I use, Ist Cameras in Newcastle Staffs, will print from 300ppi tiffs.
http://www.1stcameras.com/printshop/1st%20cameras%20-%20how%20to%20send.htm
I send them a disk as they specify and 3/4 days latter I've the prints, very good quality ones and their prices are very reasonable.
In your situation I'd still save the corrected images as tiffs and convert these to sized jpegs for the lab using either a PS action or the excelent batch converter in Irfanview. The reason labs want jpegs is because they are smaller files which take less time to process, the fact that don't give the best quality is irrelavent to most labs. I stopped using at least two local so called pro labs when they stopped accepting tiffs.

[Alan W]

The trouble I see with TIFFS Tony is the size. A friend of mine does this, he converts his Canon CR2 RAW files using the Canon sotware, then saves them as TIFFS. He prints his own stuff but still has to convert them to jpegs before he does so. To me that seems a lot of messing around and his TIFF files take up a lot of hard drive. If anything, I would be tempted to convert to PSD's instead of TIFF. However, I convert my RAW's in CS2's RAW editor, do all my main tweaking (CS saves these settings as a small sidecar file), then open in CS2 for any final adjustments or actions and then save as a high quality jpeg for printing. The beauty is that I still have the untouched RAW file (my negative), the sidecar file - which allows me to return to the point where I had finished editing in the RAW editor, and the jpeg. I have been very happy with the printing by my Lab of jpegs so I can't really see any reason for mr to create TIFFS unless I was doing stuff for a lithiographic printer. Anyway, back to my image. Here is the image as it came off the camera - albeit converted to jpeg for posting. It is very underexposed - I have this problem all the time when photographing on the Thames - a lot of reflected light bouncing around and the white parts of the boats, all these trick the meter into underexposing and to be fair - it is only trying to preserve the highlights. I tend to go with the meter and sort out the image later. But on this occasion I was obviously asking too much and it brought out this grain. A guy in another forum said that he had seen similar noise when he had badly underexposed an image. But believe me, these are very difficult conditions to get the right exposure in and very inconsistent. Anyway, here's the original image:

[tonymidd]

Hi Alan,
The size of tiffs does not worry me (56mb 16 bit or 28mb 8 bit off the D200) as I never save completed work on the hard drive. As soon as a job is finalised it's burnt on to duplicate disks and an external drive and then removed from the hard drive. I learnt the hard way when a HD went up the spout with 3 large shoots on it, only work in progress; the untouched camera files are also on disks so if any thing does go wrong I have these to fall back on; and programmes are on the HD. Disks are cheap and their life should outlast the majority of modern weddings. http://www.wedding-photography.org/forum/images/smilies/wink.gif

The answer to your river side metering is a good old incident light meter, a Weston with an Invercone but in this digi age autobracketing should also solve the problem. There is a gadget that you put on your lens to get an incident reading, don't know how good they are.

This image proves nothing as it was a jpeg but I just had to have a go.

.

[tonymidd]

To give you an idea just how good the D70 is noise wise this shot is at ISO 1600 and the detail is 500x.

ops something went a bit wrong with the detail I'll re-post it.

that's bettter, it's a section of those dark leaves in the bottom LH corner.

[Stemmy]

Thats impressive Tony - much better than the canon.

[tonymidd]

Thats impressive Tony - much better than the canon.A Canon man saying a Nikon is impressive!!!!!!!!!http://www.wedding-photography.org/forum/images/smilies/smile.gif http://www.wedding-photography.org/forum/images/smilies/smile.gif
If you think those are impressive you should see the D200's results, I'll see if I can find a couple to post.
Actually as with all photography getting the exposure right helps, the briar leaves were auto bracketed +/- 1/3 and 2/3 stop. I used to do this with wedding dresses in bright light knowing full well the - ones would be binned. Why don't the manufacturers give you the option to bracket to one side or the other?

[tonymidd]

D200 with 50mm f1.8 at f2.2 1/15sec hand held. The images were about 1/2 stop underexposed in very yellow available light.
The top two have been pp'd in NC4.4, all I've done is balance the histogram for each colour channel, resized and applied a very light lab sharpening in PS before saving for the web. No noise control what so ever.

just downloaded the trial nija and quickly appied it to the top image with the result shown in the bottom image, what do you think, worth $70? Strange it's altered the exposure and colour, maybe I've applied it incorrectly. It's getting late and I need a glass of red, try again tomorrow.