What's new
DPR Forum

Welcome to the Friendly Aisles!
DPRF is a photography forum with people from all over the world freely sharing their knowledge and love of photography. Everybody is welcome, from beginners to the experienced professional. Whether it is Medium Format, fullframe, APS-C, MFT or smaller formats. Digital or film. DPRF is a forum for everybody and for every format.
Enjoy this modern, easy to use software. Look also at our Reviews & Gallery!

Digital v Film again

Lynn and Michael, interesting conversation.

Michael, machine gun techniques, nor any automation, will produce many, many great photographs IMO. No more than using a Motor drive assures capturing a decisive moment. Skills, instinct and the eye developed by photographers like Lynn, cannot be replaced by electronics. Knowing how to meter becomes an instinct based on experience. I cannot tell you how often I have turned off all the auto-aids and went totally manual. AE is terrific, as is bracketing. Yet, they are time hungry alternatives often inapproprate for fast breaking situations(per Lynn's contact sheet).
 
> In the UK CDR that can write at 900 mb in three minutes now cost £32 Blank CDs now costs
 
I've found this thread to be extremely interesting. Marc, I'm going to have to do some research about Raw files and the results you get after processing them. I had not realized that you could get such large files, considering they came from a 2000 x 3000 pixel sensor. This was a major reason that I have not taken digital more seriously - obviously I didn't know as much as I thought I did. There must be some kind of sophisticated interpolation going on. I worked with some early DSLR files a few years ago and was not impressed at all with the quality - the images were often pretty nasty, some almost like video captures - but it sounds like the technology has made great strides.

I have been surprised and impressed by some enlargments of DSLR files that I've seen recently, which look much better than I would have ever expected, and didn't have any obvious digital artifacts in them (which I'm extremely sensitive to).

Is there working software available to process Raw files from the ND, or did I misunderstand and it's only currently available for other cameras?

I'll have to get some good-quality current DSLR images to look at and play with.

I've read about some cameras reading luminance information (with the aide of software processing) from every pixel in the sensor, instead of reading straight RGB values only. This sounds similar to the way a Lab file works - look at an image in Lab mode and compare the L channel (which carries all the luminance information) to the a and b channels, which carry all the color information. The a and b are generally mush - it's surprising how little picture information is there compared to what you'd see in individual RGB or CMYK channels. Depending on the image, even a severe blur to the a and b channels can have very little effect on the final image, and can even clear up some noise and nasty digital artifacts. The green channel in an RGB file usually has the most important luminance information, and I'd guess that this why some sensor arrays have 50% green sensors and 25% each red and blue. For instance, if you convert an RGB file to grayscale (or select 'desaturate') the result is made from about 60% green, some red and only a little blue.

What is the current status of the ND? Is it still on the market? B&H doesn't list it.

Yo, Clive - I think I understand where you're coming from with your idea that a physically-smaller 4Mp sensor is somehow superior to a bigger 6Mp sensor, but your logic is seriously flawed. By your reasoning, an 8x10 made from a sharper film in a 110 camera will be superior to a less-sharp film in a 6x9 camera, just because the 110 film can resolve more lines per mm. The 6x9 film will record a vastly greater number of lines across the entire image, and its prints will require much less enlargement, so the 6x9's FINAL PRINT will be much sharper. It's the resolution of the digital file in relation to the final print size that's important, not the physical size of the sensor.

Because it records fewer pixels, the 4Mp camera's file needs to be enlarged more than the ND's 6Mp file would to get the same print size. The ND's sensor is roughly 3000 x 2000, so a full-frame, landscape-format print 10 inches wide would be 300dpi. A full-frame 4Mp file would give you a slightly-taller 10-inch wide print that would be somewhere around 240dpi - I'm guessing because I don't know the 4Mp's exact proportions, but it would be somewhere around 240.

It's the resolution of the FINAL PRODUCT that is important.

- Paul
 
Hi Paul, concerning RAW files. Each digital camera has it's own RAW developer software that comes with the camera. A Kodak MF ProBack that has a 4X4, 16 meg sensor, produces an 18.6 meg RAW file once downloaded to your desktop. You cannot open this file until it is processed through the Kodak RAW developer ( or one of the 3rd party RAW processors such as Adobe RAW or Phase One's RAW program). During the RAW development process you can select either 8 bit which opens as a 47 meg, 12"X12" image, or 16 bit which opens as a 94 meg, 12"X12" image. So, some sort of ups&ling is going on. But it is extremely difficult to detect if at all in the final 100% print.

These files can also be further increased in both size and meg count with programs such as Genuine Fractals, which creates a resolution independant STN file that can be opened in any number of size/meg combinations. Like with film, the quality of the original image is very important as flaws will not only be magnified, but can also cause posterization and artifacts.

One of the most serious flaws of the ND is its RAW feature. In the camera, the small buffer slows you way down in terms of any fast sequence of shooting, and there is no LCD preview when shooting with RAW. Plus, the camera only captures in sRGB color space. The Contax RAW developer is the second half of the weakness. I have been unable to produce a clean file using this RAW developer. Forum contributer Irakly has a technique to overcome this. A giant disappointment to all ND owners was that Adobe did not include the ND in it's list of RAW files that their new RAW developer would recognize. However, with the aid of other forum members such as DJ, I was able to discovered that just shooting in J-peg large gave me very nice images, completely solved the buffer problem, and allows review on the LCD screen while shooting. So, it is how I now use the ND almost exclusively.

The chief difference between the Contax 6 meg sensor in the ND, and say for ex&le the Canon 6 meg sensor in the 10D, is the size of the sensor itself. While each have about the same number of effective pixels, each pixel in the ND is larger. The result isn't in the area of resolution as much as it is in tonal gradations. Tonal gradation in prints from the ND are smoother, especially noticeable in transitions from middle tones to light tones and highlights. Blown whites with the ND is less of problem, where it is an issue with the 10D that you must stay aware of. This superior tonal range is probably a contributing factor to why B&W conversions with the ND files are some of the best from any digital camera I use except for the Kodak ProBack.
 
Paul,
Thank you for that explanation. It makes perfect sense when put like that. Now I just need to figure out whether the ND's and other full size sensors have larger than normal pixels or gaps between them and whether Sigma's claim that its 3Mp sensor does actually equate to conventional DSLR's of three times the CCD resolution.
 
Hi Lynn,

Having re-read your previous posting I can see that I got the wrong end of the stick. Note to self: read things properly before commenting!

Paul, Mark and Clive, re your discussion on sensor size, I thought bigger pixels were better in terms of not only dynamic range but also noise control.
 
Mike, I'm not sure about the noise control. The Canon 1Ds is 11 meg, full frame and exhibits some noise in the blacks. But that could be due to any number of different factors.

Clive, to my knowledge there are no gaps between pixels in any CCD or CMOS sensor.
Most detailed spec's for digital camera sensors will quote the actual pixel size. I don't recall the actual numbers, but I do remember that the NDs' are larger than the 10D's with both being approx. 6 meg.
 
Instead of sending individual posts and contribute to the mass frey of them...I'll comment on a few things in one...

Hi David,

> 1. One commentator recently mentioned that digital is not so good in > low light applications.

Interline sensors that are used in Video cameras are pretty good at low light applications. The larger sensors, called "full frame" (not meaning that they are the size of the 35mm, but that they take the image all at once) may have noise problems in low light situations. Some CCDs get color fringing, and the Foveon is in fact terrible at low light. Some of the low light issues with full frame sensors is due to the Bayer pattern processing algorithm, not the sensor... So, as a general rule, it's wrong to say digital is not as good in low light, but certainly implementtions don't work as well as others.

-------------------------

Clive,

> I am intrigued however at whether the 6Mp spread over the larger CCD > is detrimental to the image compared with a small chip 6Mp camera...

No. Typically, the larger the sensor element is, the lower the noise, hence the larger Canon sensors give better noise characteristics than the smaller Nikon ones. Of course, this is very dependant on implementation...so it's far more than just the sensor that contributes to image degredation.

> and > what the conventional sensors performance is compared to the Sigma. > Those questions are not only valid to myself, but it just might > interest others.

But, I don't agree that this is the forum for this discussion. These issues/questions have far better venues elsewhere. These issues have been discussed ad-nauseum elsewhere, and I'd suggest looking elsewhere to satisfy your curiosity.

> Now I just need to figure out whether the ND's and other full > size sensors have larger than normal pixels or gaps between them

They have larger sensor elements (they are NOT pixels, unless we are talking monochrome). The gap is pretty much dictated by the manufacturing processes technology.

> and > whether Sigma's claim that its 3Mp sensor does actually equate to > conventional DSLR's of three times the CCD resolution.

In a nutshell, no...but again, this is a question better discussed/asked elsewhere, and with a little research, you'd find it's been discussed ad-nauseum.

---------------------------

Hi Mike,

> Paul, Mark and Clive, re your discussion on sensor size, I thought > bigger pixels were better in terms of not only dynamic range but also > noise control.

Correct ;-)

Regards,

Austin
 
No worries, Mike (Nunan). It gets confusing keeping up with all the posts in here at times!

Marc, I'm glad to hear your reply, thank you. I have become very nostalgic about my old days of shooting horses now that I've dug up my old stuff. I certainly do miss it in retrospect. Shooting that sort of stuff was extremely satisfying and exciting, kept me on my toes with plenty of challenge (and physically fit running the course to catch the riders). -Lynn

(PS: To Dirk, I can no longer get messages to the list by email, only by direct post in the forum. 4 attempts by email in 2 days have failed. Thought you'd want to know)
 
Film or Digital

To me they are different tools, eg. B&W, Color, Kodachrome, Ektachrome etc.

I agree with Lynn on disciplined study, but since I like to shoot slides and the news just reported that Kodak is moving away from film. I wish everyone would bracket and shoot 10,000 rolls of film a year. I like film and my 50 year old technology cameras.

I hope there is a place for both film and digital and not just one or the other.

Regards

Gilbert
 
You must log in or register to reply here.
Back
Top