Which file format is best for printing?

<p>Which file format would be most suitable if I were to burn images to a disc and then have the images printed from the disc? The pictures were shot in RAW and I've worked on them in Lightroom. Now, as far as putting them on a disc, I'm not quite sure which would give the best resolution...or does it matter? Thanks, Jeff</p>

<p>Yep, it matters. If you're having the prints done by a lab, they will specify the file format. If your doing your own printing, TIFF files give the best results. The differences between TIFF and JPEG may not be apparent with smaller print sizes, but the larger the prints the more critical it becomes....</p>

<p>Stick with TIFF whenever possible. </p>

<p>I gotta disagree on this. I'd bet heavily that I can take your TIFF, save it as a high-quality JPEG, get prints made from each file, and none of you would be able to tell any difference. Now don't get me wrong, a heavily compressed JPEG can lose a lot of quality. But if you use the very highest quality JPEG settings, the compression is either lossless or very, very close to lossless, and you still reduce the file size for average pictorial content by about 60% versus an 8-bit TIFF (and 80% versus a 16-bit TIFF). Again, the key is just to use high quality settings for saving as a JPEG (for the most critical images, I use 1x1,1x1,1x1 subsampling and the lowest compression).</p>

<p>As to anyone who says that TIFF's can store 16-bit images while JPEG's can only retain 8 bits, that's true, but most printers will just toss out the extra bits, and even a printer that theoretically uses them is highly unlikely to deliver a <em>visible</em> improvement from the extra bits. For image editing where you're making major adjustments with curves or levels or color correction, those extra bits can be nice. But once you've got a final, ready-to-print image, they really don't matter.</p>

 

<p>Dave is correct. Are there 16 bit printers anywhere ?<br>

You edit and process using TIFF, or PSD, or whatever your processing tool allows, but for final print you need to create best quality JPG files.</p>

<p>I was in the TIFF camp until I started experimenting with high quality jpegs-up to 11x14 I can't see a difference on prints. On some 16x20's, I have noticed a very slight shift in overall contrast and saturation vs smaller sizes, but it is minor and my images are usually cropped. If you're using close to the full image, it's not likely you will see a discernible difference even @ 16x20. As Dave R. says, go HQ and only use 1st generation JPEGS (JPEGS created directly from your finalized conversion file) with no additional JPEG level edit changes or "saves" so as to avoid compression loses.</p>

<p>First off, a lot depends on what is going to happen to the JPEG after you present it for print. A conversion to an output space? You decide you’ll edit it further? Yes, JPEG is visually lossless after you save it the first time. But unless file size (transportation) is a major factor, it provides no other benefit and actually is more fragile than an 8-bit TIFF and a 16-bit TIFF is going to provide more data in case of the above possible edits. </p>

<p>As for printers only sending 8-bits per color to the printer, it depends. For a lab, probably the case. For those of us printing on the Mac to an Epson pro printer (which many labs use), or any Canon export through Photoshop with their plug-in, it sends the full high bit data to the printer. Or for some using RIPs (ImagePrint comes to mind), again, high bit data is being sent to the driver. So it depends. Many outside labs use such printers and drivers. One could ask. </p>

<p>Bottom line. If a lab demands JPEG, you send them JPEG. If you have a huge file you have to upload and that’s a huge bottleneck, send JPEG. Otherwise, high bit TIFF is the way to go. The extra data may be used, it may not be used but its moot as with a JPEG, its not necessarily moot. </p>

<blockquote>

<p>Dave is correct. Are there 16 bit printers anywhere ?</p>

</blockquote>

<p>In addition to the Epson drivers Andy mentioned, current high end Canon printers also come with 16 bit direct printing Photoshop plugins, to reduce banding that is sometimes visible passing through the OS 8 bit printing path.</p>

<p>All Gimp-print (aka Gutenprint) print drivers are 16 bit. The Mac's UNIX derived print system is also 16 bit.</p>

<p>All in all, I think that means that most serious printers are 16 bit.</p>

<p>In terms of printers that produce 16-bit output, if that makes a visual difference etc, see this discussion on Luminous Landscape:<br /> http://www.luminous-landscape.com/forum/index.php?topic=53914.0<br /> As mentioned there, output using the Canon export module makes a visual difference on the prints.</p>

 

<blockquote>

<p>2. Printed a very difficult synthetic image in both 8-bit and 16-bit: Bill Atkinson’s 28 balls test image. The color space was Adobe RGB (1998) and again, the driver was set for highest quality for both documents. Note also that Dither was turned OFF in Color Settings. <br /><br />Results: I can see a difference, the 16-bit output has smoother graduations in many areas of the various balls. You need to look closely but its quite visible. This is a greater difference than I’ve seen in the past using the Epson driver set both ways. If you have a high bit file, you’d certainly want to pass it to the Canon export module and set it for highest quality (16-bit).</p>

</blockquote>

<p>I guess I should qualify / address what appeared to be an assumption in the original question: presumably the only reason anyone would do a save-as and make a file purely for printing is to send that file to a lab. If you're simply printing to an inkjet attached to your own computer, I can't see any reason not to just print from Photoshop or whatever, with the original .psd file or whatever open.</p>

<p>Now if you are printing to a high-end inkjet, does sending a more-than-8-bit image improve the quality? Andrew cites "a very difficult synthetic image [and] Dither was turned OFF in Color Settings". It may be that in such circumstances some visible benefit might be created, but if you use the settings that don't exaggerate the differences (turning dither off might well exaggerate the difference) and then print actual pictures (as opposed to synthetic images intentionally made very difficult), I strongly believe it that very rarely if ever will you notice a real difference (that is, one you could reliably detect in a double-blind test--if you can't do that, then it's not a truly visible difference).</p>

 

<p>First, dither was turned off for the subtract routine to work correctly (proved the correct deltaE values showing actual difference when converting from 16-bit to 8-bit, the time in which PS would introduce the dithering noise).<br>

When printing through the Canon driver, Dither is not, the export module doesn’t use ACE, its using the OS CMM so the dither is moot here. The output with and without sending 16-bit data is visible on the test image and as others in the Lula posts have written, on other images where smooth gradients are present (skies etc). <br>

As stated in the posts, yes, the differences are subtle, often not visible in many images. But the images are starting out in high bit. The drivers and output device can use the high bit data. Why flip a coin when you print and not use this high bit data when in some cases, its beneficial? I’ll add that there are many outside labs (even Costco) that provide print services using such printers. </p>

<p><em><strong>Which file format would be most suitable if I were to burn images to a disc and then have the images printed from the disc?</strong></em></p>

<p>I suppose you also want to give this CD to customer so they can print the image at Costco or similar? then Jpeg is the way to go in most lab. A TIF or JPEG when print have no visual difference. So no it doestn matter as per your question.</p>

<p>For all the technical around TIFF vs JPEG etc... the other post have plenty of good extra info.</p>

<p> </p>