JPEG 9.1

[gerald_wallace]

<p>I noticed that JPEG version 9.1 was available and wondered if any one has tried using it?<br>

Are there any benefits for using version 9.1? (I never use JPEG when taking picture for myself).<br>

Just curious.</p>

[mbkissel]

<p>Well, if what I've read is correct, it will be a welcome addition to the graphical file formats. Lossless compression? Great! And it would simplify upgrades on new camera models by using the new jpeg format rather than proprietary RAW or other methods which means no more waiting for your favorite software to "catch up". </p>

[digitaldog]

<blockquote>

<p>And it would simplify upgrades on new camera models by using the new jpeg format rather than proprietary RAW or other methods which means no more waiting for your favorite software to "catch up".</p>

</blockquote>

<p>How so? We still need (desire) raw, unrendered data. <br>

The proprietary raw issue is solely caused by the camera manufacturers. There's zero reason they have to do this. Raw is based on TIFF anyway, and they just add tiny differences for each release with no justification.</p>

[JDMvW]

<p>more details at http://www.digitaltrends.com/photography/new-jpeg-standard-complete-lossless-option/#!Lbrmw<br>

(wouldn't you know it, it's those Ossies again!)</p>

<p>I agree that it is unlikely to replace the various RAW formats for many reasons, including those pointed to by Andrew.</p>

[mbkissel]

<p>Andrew, with the increase in color depth, there may not be that much difference. However, I was more referring to the fact that you could continue to use your old software without having to convert images to a readable format, like TIFF.</p>

[digitaldog]

I don't understand. What's not much different from what?

[dave_redmann]

<p><em>Lossless compression? Great! And it would simplify upgrades on new camera models by using the new jpeg format rather than proprietary RAW ....</em></p>

<p>At a minimum you'd need not only lossless compression but also 16-bit-per-channel to approach what raw (not RAW--it is not an acronym) gives you. There are other issues too.</p>

<p><em>We still need (desire) raw, unrendered data.... </em></p>

<p>Agreed, which is why even a 16 BPC lossless JPEG is still not a truly satisfactory replacement.</p>

<p><em>Raw is based on TIFF anyway ....</em></p>

<p>Huh? TIFF and JPEG are both rendered by de-mosaicing the raw data (and typically other operations, like applying white-balance), so I don't see how one can assert that raw is based on TIFF?!</p>

<p>Anyway, I don't see much point. Raw is what it is, and is something I want, at least for the foreseeable future. TIFF, PSD, etc. serve purposes, mostly as intermediaries. Regular old 8 BPC JPEG's are just fine for almost any end use. But that is IMOPO, I guess some disagree. And what remotely-modern software can't use pretty much any JPEG?!</p>

[mbkissel]

<p>Dave, I'm not advocating this format as a replacement for RAW (by the way, most of the time I see it written, it <strong>is</strong> in all caps; even Canon uses it in this manner on their website. But thanks for the clarification and you may continue to correct me as you see fit). With lossless compression and 12 bpc color depth, JPEG shooters may see less IQ difference than previously possible with JPEGs or TIFFs converted from RAW files.</p>

[Wouter Willemse]

<p>Either way, JPEG 9.1 is not something you can just try unless you program something to use the new library (or codec), which was made available. Most software at present probably isn't updated yet, and hence cannot read or write JPEGs that use the new functionality. It'll take time before this trickles down to the normal software packages as well as cameras and smartphones.</p>

[digitaldog]

<p>This new JPEG implementation like the last will <strong>never</strong> be a substitute for raw capture. It's baked, it's rendered (by the camera). This defeats the purpose of capturing raw data for most of us*. Doesn't matter what the bit depth is, or if it's lossy, it's rendered and worse, camera JPEG (currently) provides two RGB working space encodings that are insufficient in gamut to contain colors we can capture from raw data. So while it might be nice to see the camera manufacturers implement it, I'll never use it and it's no substitute for raw. I further doubt this will be implemented if you look at the history of these camera manufacturers and how slow and unwilling they are to update anything other than their proprietary raws which is totally unnecessary.</p>

<p>Yes, many raw files are based on TIFF:<br />http://en.wikipedia.org/wiki/Raw_image_format</p>

<blockquote>

<p>Many raw file formats, including IIQ (<a href="http://en.wikipedia.org/wiki/Phase_One_(company)">Phase One</a>), 3FR (<a href="http://en.wikipedia.org/wiki/Hasselblad">Hasselblad</a>), DCR, K25, KDC (<a href="http://en.wikipedia.org/wiki/Kodak">Kodak</a>), CR2 (<a href="http://en.wikipedia.org/wiki/Canon_(company)">Canon</a>), ERF (<a href="http://en.wikipedia.org/wiki/Epson">Epson</a>), MEF (<a href="http://en.wikipedia.org/wiki/Mamiya">Mamiya</a>), MOS (<a href="http://en.wikipedia.org/wiki/Leaf_Medium_Format_Photography">Leaf</a>), NEF (<a href="http://en.wikipedia.org/wiki/Nikon">Nikon</a>), <a href="http://en.wikipedia.org/wiki/Orf_format">ORF</a> (<a href="http://en.wikipedia.org/wiki/Olympus_Corporation">Olympus</a>), PEF (<a href="http://en.wikipedia.org/wiki/Pentax">Pentax</a>), RW2 (<a href="http://en.wikipedia.org/wiki/Panasonic">Panasonic</a>) and ARW, SRF, SR2 (<a href="http://en.wikipedia.org/wiki/Sony">Sony</a>),<strong> are based on the TIFF file format.[4] These files may deviate from the TIFF standard</strong> in a number of ways, including the use of a non-standard file header, the inclusion of additional image tags and the encryption of some of the tagged data.</p>

</blockquote>

<p>http://www.digitalpreservation.gov/formats/fdd/fdd000073.shtml</p>

<blockquote>

<p>TIFF/EP_1 was intended as suitable as a raw camera image format and to be as compatible as possible with TIFF 6.0. A TIFF/EP_1 file is a valid TIFF file that contains the TIFF/EP format identifier and conforms to certain restrictions described in ISO 12234-2:2001.</p>

</blockquote>

<p>Adobe's DNG is based on TIFF. Adobe owns and controls TIFF!</p>

<p>I don't care how many companies or users write <em>RAW</em> instead of <em>raw</em>, it's wrong at least in the context of talking about digital camera files. <strong>RAW isn't an acronym</strong> like TIFF or JPEG or TWAIN and as such should not be used this way. Google image search WWE RAW and now you have a proper usage of the acronym! If you use RAW when discussing a digital camera file, a number of your audience will look down at you as someone who doesn't understand fully what you're writing on the subject. As the Chinese proverb says: The first step towards genius is calling things by their proper name.<br /> *http://www.lumita.com/site_media/work/whitepapers/files/pscs3_rendering_image.pdf</p>

[mbkissel]

<p>Okay, Andrew, thank you. You are entitled to look down on me anytime and as often as you like. I don't plan to become a genius, so that first step won't help me. Oh, and I can't even read Chinese either. So I'll continue to use it in all caps from way down here at the bottom of the intellect barrel.</p>

[digitaldog]

<p>Mark, if you want to use terminology incorrectly by all means do so. I'm not looking down at you specifically, I too, many years ago, used <em>RAW</em> instead of <em>raw</em> until someone pointed out it wasn't the correct usage so I've changed my methods of communication. It's called learning!<br>

You can continue to use the language incorrectly if you want, but at least you now know that there's no acronym for raw image data and no reason to write RAW. The continued incorrect use of terminology and how people now perceive your writing is in <strong>your</strong> hands. </p>

[mbkissel]

<blockquote>

<p>It's called learning!</p>

 

</blockquote>

<p>Thank you. I've certainly learned a lot here today.</p>

[picpocket]

<p>Andrew, you are clearly speaking for more than yourself when you say "how people now perceive your writing ...". Will you agree that is way too generic statement attributed to a crowd which you may have no clear way of knowing the correctness of. still you don't hesitate to throw it out there. Perhaps you meant "some touchy people who have low tolerance on grammatical mistakes or common usage language deviations"?<br>

<br>

Having said that, raw is correct usage - i stick to that. RAW is widely prevalent usage though, and with camera makers using it themselves, i (applies to me only) would just stop short of perceiving them as anything lower than my intellect in this field. <br>

<br>

It was all nice till the point when Mark was made aware of the folly, and he explained he understood but gave his reasons on why he may continue using it. I think the message was conveyed and choices made then. Everything after that was just ego brunch IMO. Someone once said - what's in a name - and people love that notion, or they don't. If we are going to debate it still after hundreds of years, so be it </p>

 

[joelclarke]

<p>As photographers, we use precision instruments, software and methods all the time, which we then update - debating with others before we do so as whether those updates are an improvement - because presumably we are seeking to perfect what we already have.<br>

Language is also a precision instrument and should be treated with the same respect.</p>

[q.g._de_bakker]

Saying that an image file format is 'based' on the TIFF format says very little. Just that it contains bitmap data and instructions how to interpret those.<br>The entire idea behind the TIFF format was that it is not a specific format, but allows a multitude of different formats to be wrapped up inside it, plus the ability to expand the number of formats that can be packaged inside a TIFF file.

[digitaldog]

<blockquote>

<p>Language is also a precision instrument and should be treated with the same respect.</p>

</blockquote>

<p>Agreed and as someone who's been in the business of educating photographers about digital imaging since the early 90's, but who didn't use the terminology correctly (until having the pleasure of working with Adobe and Thomas Knoll when both began processing it in ACR at the beginning), I'm simply attempting to <strong>stop</strong> the incorrect use of the language: RAW. <br>

From John Nack the product manager of Photoshop for years:<br>

http://blogs.adobe.com/jnack/2005/10/more_dng_momentum.html<br /></p>

<blockquote>

<p>PS–I’ve always preferred the nice, simple “raw” as the term for this sort of format. Saying “RAW” seems a little aggro (“RAW is WAR!”), like you need to make the little devil-horns with your hand while saying it. <strong>The term is neither an acronym (RAW) nor a proper name (Raw),</strong> but rather a generic descriptor for a whole class of formats. Therefore Adobe just says “raw.”</p>

</blockquote>

<p>Again, I don't care how many companies or people here or elsewhere use RAW instead of raw, it's an <strong>incorrect use of the terminology</strong> we find in the imaging field. You all know the facts now. If anyone here can find a reputable source (peer reviewed) that explains what RAW stands for in terms of imaging and camera capture, let's see the references please. Otherwise stick to RAW and look silly now that you know it's not an acronym or get on board using the language correctly. There's no excuse otherwise IMHO and that of many of my peers and anyone working at (for one example) Adobe Systems. </p>

<blockquote>

<p>Saying that an image file format is 'based' on the TIFF format says very little.<br /></p>

</blockquote>

<p><em>Many (</em>not all<em>)</em> raw file formats are based on TIFF/EP. You must have missed the link above I provided along with it's text I pasted! What isn't clear about: Many raw file formats, including (examples) <strong>are based on the TIFF file format? </strong><br>

http://en.wikipedia.org/wiki/TIFF/EP: <strong>One of the uses of TIFF/EP is as a </strong><strong><a title="Raw image format" href="http://en.wikipedia.org/wiki/Raw_image_format">raw image format</a>.</strong><br>

One example is Nikon's NEF raw file format, which uses the tag TIFF/EPStandardID (with value 1.0.0.0).[1] Adobe's DNG (Digital Negative) raw file format was based on TIFF/EP, and the DNG specification states "DNG ... is compatible with the TIFF-EP standard".[2] Several cameras use DNG as their raw file format, so in that limited sense they use TIFF/EP too.[3]<br>

http://www.barrypearson.co.uk/articles/dng/raw.htm#contents: A typical raw file is highly structured internally, often (not always) using a tagged structure <strong>based on TIFF.</strong><br>

<strong> </strong>Sorry if the <strong>facts</strong> of both the correct usage of the term used to define a camera raw capture and what much of that data is based upon gets in the way of your <strong>opinions</strong>. Everyone is entitled to be wrong despite the facts. </p>

<blockquote>

<p>Someone once said - what's in a name</p>

</blockquote>

<p>And long before that, Confucius wrote <em>Rectification of names</em>, and unlike some here, you don't need to read Chinese to understand why calling things by their right name is the right thing to do.</p>

<blockquote>

<p><em>A superior man, in regard to what he does not know, shows a cautious reserve. If names be not correct, language is not in accordance with the truth of things. If language be not in accordance with the truth of things, affairs cannot be carried on to success. When affairs cannot be carried on to success, proprieties and music do not flourish. When proprieties and music do not flourish, punishments will not be properly awarded. When punishments are not properly awarded, the people do not know how to move hand or foot. Therefore a superior man considers it necessary that the names he uses may be spoken appropriately, and also that what he speaks may be carried out appropriately. What the superior man requires is just that in his words there may be nothing incorrect. </em></p>

</blockquote>

[q.g._de_bakker]

The link to a Wikipedia article, linking to a page on the net that lists formats, mentioning that some are based on TIFF files, is not really convincing proof. That's the New Science of today: internet utterances referring to other internet utterances to prove they are correct. Wikipedia is the worst thing that hit internet.<br><br>As demonstrated here. Look at what the things are, not at what someone somewhere on the internet says that might agree with what you say. Facts and opinions, right?<br>The TIFF format is a container format, providing a way to package different bitmap formats inside a file. The thing (back then) peculiar to TIFF files is that, as a container format, they allow many different formats to be wrapped inside them. Not that they contain bitmap data plus metadata that helps decoding the bitmap.<br>The description of what can be put inside a TIFF file is up to whoever devises the formatting of whatever it is. The TIFF standards prescribes adhering to one definition or a set of versions of whatever it is that is going to be wrapped up inside a TIFF file.<br>The JPEG definition, for instance, defines how to encode and decode data. The TIFF standard tells us that it is to be expected to encounter a JPEG coded dataset inside a TIFF file, and (possibly, when applicable) what style of JPEG is allowed. The TIFF file itself contains a header, identifying it as a TIFF file, and a map telling us where to find the essential bits for understanding what it is that we find inside that TIFF file. Those essential bits are what you also find in the different data file formats when not wrapped inside a TIFF file: a header that identifies the data format and whatever info we need to decode the data, plus the data themselves. A humble BMP format file follows the same scheme, yet is not based on the TIFF format.<br>Raw formats are defined by however made them up. They may or may not follow the same scheme, depending on whether the owner wants to give clues to anyone, and how many, that help decode the file. The fact that they may provide such a 'road map' as also found in TIFF files does not make them TIFF based. Nor BMP based.<br><br>Saying that a image file format is based on TIFF format (supposing it would be true) says very, very little. So little that it doesn't make sense. I's like saying that JPEG 9.1 is based on BMP, because both allow compressed and uncompressed image data.

[digitaldog]

<blockquote>

<p>The link to a Wikipedia article, linking to a page on the net that lists formats, mentioning that some are based on TIFF files, is not really convincing proof.</p>

</blockquote>

<p>Fair enough, disprove it.</p>

<blockquote>

<p>The TIFF format is a container format, providing a way to package different bitmap formats inside a file.</p>

</blockquote>

<p>Agreed. As such, why can't or isn't the raw formats Wikipedia (and other links) listed above being based on TIFF <strong>not</strong> correct or possible?</p>

<blockquote>

<p>Raw formats are defined by however made them up.</p>

</blockquote>

<p>Partially yes, I'd agree. There's a lot of proprietary tags and bits that make it necessary to 'hack' the format each time a new one comes along so Adobe and all other raw converters can access that data. But again, you're suggesting this is proof that the format <strong><em>isn't</em></strong> based on TIFF? Fine, <strong>back it up with proof the references provided are wrong! </strong></p>

[digitaldog]

<p>Since Wikipedia is being dismissed for whatever reason, See page 4: http://www.color.org/documents/Camera_raw-the_basics.pdf<br>

This was written by Jack Holms of HP whom I've worked with, co- authoring a paper for the ICC of which we both belong. <br>

Get ISO:12234-2<br>

<img src="http://www.digitaldog.net/files/TiffEP.jpg" alt="" /><br>

Further: </p>

<blockquote>

<p>The structure of raw files, including the ISO standard raw image format ISO 12234-2 follows a common pattern - a short header, camera sensor metadata, image metadata such as exposure settings, camera and lens model, date and sometimes the location. A thumbnail, a smaller JPEG file for quick viewing and the sensor image data. NEF is based on the TIFF format.</p>

</blockquote>

<p>http://www.zamzar.com/convert/nef-to-tiff/</p>

[q.g._de_bakker]

Again: look at the thing discussed, not at what references you can find on the web of lies, half truths, and maybe truths.<br>The internet is very much based on TIFF files, since it is full of pointers that direct us to 'the next bit' but leaves it to those bits to explain what they are and how to interpret them.<br><br>There is no question, no answer to whether the list saying that some data formats are based on TIFF is correct or not. It makes as little sense as "saying that JPEG 9.1 is based on BMP, because both allow compressed and uncompressed image data".<br><br>TIFF is a file format, not a data format.<br><br>So what proof do you expect there to be?

[digitaldog]

Since you are totally unable to dismiss that raw is based on TIFF despite ISO and all other references provided, best

move is to ignore you!

[q.g._de_bakker]

Better move still would be to educate me (and everyone else who might read this) how they are based on TIFF. Point out the TIFF in them.<br>So please do.<br><br>By the way: you do know that some 20 years ago, Adobe decreed that TIFF is not an acronym? ;-)

[digitaldog]

That would be akin to attempting to educate someone who believes the Earth is 6000 years old what carbon dating is: a

waste of time. We're done!

[q.g._de_bakker]

Meaning you can't. Which is of course correct: it can't be done. All that can be done is point out how more and more things are stuffed inside a TIFF file.<br>These things themselves come with their own definition and description. Those are added to the TIFF standard so that people who write software that is supposed to be able to open TIFF files know how to deal with that. Or not, as is the case with proprietary raw formats<br>Again: TIFF is a file format. Not a data format. It provides a header that points to where data can be found. And that's pretty much the end of the TIFF part. That's why it is so successfull: it will accomodate whatever anyone wants.<br>But the fact that you can stuff all sorts of things inside something else only makes that container just that: the container of those things. it does not make the things it contains a derivative of the container.