Reinaldokool Loc: San Rafael, CA

:thumbup: :thumbup:

Photoninny Loc: Monterey Bay Area

The Microsoft codec link listed will not work with Windows 8. If you are using Win8.1 then use this link: http://www.microsoft.com/en-us/download/confirmation.aspx?id=41337 for the proper codec.

pecohen Loc: Central Maine

I think we're splitting hairs here and that really is what I was trying to say.

If you have a JPEG image, you can't actually view it until appropriate hardware processes that image to convert all of those zeros and ones into a light pattern that you can view. In fact software comes into play usually as well to decode the JPEG compression and compute the actual pixel values.

On the other hand, if you have a RAW image then you still have a pixel by
pixel numeric representation of the image that may not even be compressed; but if it is compressed it is compressed by the much more straightforward LZW scheme. The LZW decoding probably will be done by software though there are ways to do it strictly in hardware. At the microprocessor level there is a very fuzzy line between hardware and software in any event.

In principle at least, that decompressed RAW image can be displayed on a monitor (at least one that can handle 16-bit color channels) without further processing by software, but for a nice image the image has to be processed further and there is a header in the RAW file that suggests how that processing might be done.

In summary, the processing for putting a JPEG image onto a monitor is different than for a RAW image. It is different processing and for a RAW image the software processing to get an acceptable image may be significantly greater, but in both instances both software and hardware have to process the data to get it onto the monitor so we can see it.

By the way, it is true you can't see a RAW file. But neither can you see a JPEG or a TIFF file - that is why I used the word representation. In each case what you are talking about is a computer file, strings of bits that might have any number of different forms including electric charges and magnetic fields. None of us are equipped to see these bits directly or to make much sense of them if we could, we can only hope to see representations of these files after they have been processed so that we can understand them as images.

amehta Loc: Boston

The jpeg and tiff files contain RGB data which is the same format that monitors display and printers print. Raw sensor data is not RGB data, each pixel contains only a single luminosity value. The data about the specific sensor, and ideally the lens used, is needed to convert that to RGB image data. So your statement above (in bold) is not true for a raw file, while it is true for a jpeg or tiff file.

This is not splitting hairs, this is a qualitative difference between raw sensor data and processed image data.

pecohen Loc: Central Maine

However the data from a sensor is formatted, the data word for a pixel is not a single intensity value. With a single intensity value per pixel you can only get a monochrome image.

If you have three 16-bit intensity values you can certainly pack them into a 48-bit word and it is conceivable that you might choose to do that in some very obscure and complicated way; in cryptology, they do this kind of thing all the time. But in a camera it is hard to imagine why the designers would want to make the design more complicated or obscure than necessary.

I admit ignorance about what format might the data from a sensor might be in, but I find it hard to imagine that it cannot be unpacked in a fairly rudimentary way into three, or perhaps four data values, each representing a coordinate in some color space. Most likely that would be RGB but it could be some other coordinate system. It doesn't matter, there are known formulas for converting between different systems.

Without question, the raw data out of the sensor will be deficient in a variety of ways and it should be corrected. To the extent that the lens introduces further distortion that needs to be factored in. Certainly before displaying an image you would want to at least make a reasonable effort at doing some of this processing.

But this is all just processing. It takes processing to display a JPEG image and it takes processing to display a RAW image but that image, in one form or another is in the file. The processing in each case is conceptually different but so is the processing conceptually different for a JPEG than for a TIF file.

It seems fair to conclude that we have different opinions of what constitutes hair-splitting and we should probably leave it at that; its not so important. We're clearly looking at the question from different viewpoints. The discussion was a good one though and I thank you for that.

amehta Loc: Boston

You're right, this is not completely relevant to the original topic, and I would drop it but I don't want anyone else to read what you wrote and think it is correct.

For a typical digital sensor, each pixel has a colored filter in front of it. Each pixel records a single intensity level, not three values for RGB. The raw converter takes the information about the filter color for each pixel and combines it with the intensity value at each pixel to create RGB values. This means the 16-bit tiff file from the raw conversion has more information than the raw file, and it is several times larger. One raw converter will not produce exactly the same RGB image as another one, even if no additional processing is done, because each will apply lens corrections differently. Displaying a jpg, on the other hand, follows a standard algorithm, so it will generally look the same regardless of which tool is used.

The second part of the raw conversion is the "basic" post-processing which is done. This includes applying white balance adjustments, other color corrections, exposure adjustments, noise reduction, and sharpening. All of this is done by the camera when it produces a jpg. When the OP said the "raw images look wicked grainy", it meant that the noise reduction being applied by the post processing software was not as effective as the noise reduction the camera did when producing jpgs. As others have said, though, there are quite a few ways to do better noise reduction in post-production, which should help the OP with this issue.

Pablo8 Loc: Nottingham UK.

*******************************************
That could have been eliminated at the taking stage, by NOT using a built-in flash, or one very close to the axis of the lens.