steve_dunn2

November 19, 2012

I don't have a 30D, but the very first page of the "Image Playback" section of my 7D's manual has the following warning:

<blockquote>About images taken with another camera:

The camera might not be able to properly display images captured with a different camera or edited with a computer or whose file name was changed.</blockquote>

I believe the manual for my 20D said something similar, and while I've never had a 30D, I'd imagine its manual may also say something like this.

My guess, and it's purely a guess, is that the JPEG needs to be at exactly one of the resolutions in which your camera can save a JPEG and/or it's looking for whatever EXIF tags are required in order to display the various pieces of data that it can display about the image (e.g. which AF points were used, what shooting parameters were used). But I don't believe I've ever seen a description of what the camera needs in order to be able to display an image, which isn't all that surprising since the camera's LCD is intended to be used to review pictures you've taken with the camera, not as a general-purpose display device for images from any source.

October 2, 2012

It may help. In essence, that's a form of HDR. You're not getting a range beyond what the sensor can capture in one shot like you would if you shot a bracketed sequence and then did HDR on it. But by doing multiple RAW conversions, you can optimize each one for a different range, as you suggest. That switches the challenge from how to dodge and burn much of the picture into how to blend your multiple renditions into one seamless image. Either approach can work; one approach or the other may be better for some scenes and/or some photographers.

FWIW, when Canon's DPP RAW conversion software recently added basic HDR abilities, I briefly played around with doing HDR from one RAW file. Basically, this is an automated version of what you're talking about: it takes one RAW image, converts it multiple times with different settings, and then applies your selected tone mapping and other settings to produce one final image. It's not a full-blown HDR product and it lacks a lot of the control that you'd want if you were serious about HDR. And most of its presets have the HDR look that you don't like (and, for the most part, neither do I). But for some images, I have to admit that its automated HDR process from a single RAW file, using conservative settings, gives me a good starting point in less time than it would take me, a Photoshop user of limited skill, to produce something similar manually.

So I'd say go ahead and try blending multiple renditions of the same RAW file together and see how it works for you. If you're comfortable with taking multiple layers and blending them into a final image, as I think you probably are, I suspect you'll find it's a workable method for high-contrast scenes.

September 26, 2012

You didn't mention what camera this is, but if it's even remotely recent, it ought to support 8 GB cards. The camera's manual may say (or it may not; some cameras come with manuals that say nothing at all about the maximum card size the camera supports).

The thing that springs to mind for me is whether this is a legit Sandisk product. Did you actually buy it from Amazon, or did you buy it from a third-party vendor selling through Amazon? The former should be reputable; the latter may or may not be. There have been plenty of reports of fake "name-brand" memory cards being sold by unscrupulous vendors on eBay and I believe also through Amazon.

September 20, 2012

Back/front focusing jumps out at me as a viable possibility, as does the necessity of doing controlled testing in order to get to the bottom of this. That means camera and subject both absolutely stationary (i.e. camera on a tripod, shooting an inanimate object), and using a subject with features such that you can be certain that the AF system is locking onto exactly the thing you want it to focus on. On that last point, remember that the actual area covered by the AF sensor is substantially larger than the box in the viewfinder, so the feature on which you're focusing has to be surrounded by an area where there's nothing for the AF system to see. There are a number of articles on the Web about testing back/front focus, and they should cover all of this.

Good luck! Hopefully it turns out to be something simple, such as needing to do AF microadjustment.

September 19, 2012

I can't really help you much with the bodies, other than to suggest that I suspect you'll find one of the 5D models to be a better match for you, since you're coming from a pro body. Not that the 6D isn't a highly capable body, just that it's aimed lower than the 5D.

The lenses? They're all top-quality and will serve you very well on whichever full-frame body you choose. And I agree that a full-frame body makes the most sense for you. You've built a lens collection that meets your needs based on a full-frame body; if you go with a 1.6-crop body (e.g. the 7D), you'll need to adjust your lens collection to make it fit your needs. I went through that when I migrated from film to a 20D, and not wanting to undo it is one of the multiple reasons I upgraded from the 20D to a 7D rather than to something in the 5D family.

September 4, 2012

According to Canon's Chuck Westfall:

<blockquote><a href="http://digitaljournalist.org/issue0905/tech-tips.html" target="_blank">H mode on the EOS 5D is basically an ISO 1600 image that's been intentionally underexposed by one stop and then processed internally by the camera to simulate ISO 3200. It's primarily convenient for in-camera JPEGs, but you could achieve the same effect in RAW mode by shooting at ISO 1600 and then applying exposure compensation in your RAW conversion software.</a></blockquote>

So I suspect what I wrote may be incorrect in that I said the RAW converter is responsible for boosting everything one stop; it may indeed be the camera doing it before writing the RAW file. Good catch.

September 3, 2012

<cite>Do Auto Lighting Optimizer and Highlight Tone Priority work if you shoot in RAW? I always ignored them since I figured they would be turned off.</cite>

ALO makes no difference to the actual RAW image data. It presumably sets a flag in the file to tell a RAW converter that ALO was enabled, so that the RAW converter could (if you and/or it wanted) use that as a default for conversion. And I'd imagine it also applies it to the embedded preview image.

I've yet to see conclusive proof one way or the other on HTP. I've seen posts saying it does, in fact, adjust the "RAW" data to apply an appropriate tone curve, and I've seen posts saying that they've done a binary comparison of RAW files of the same thing shot with and without HTP and the only difference in the image data is on the level that you might expect as shot-to-shot variation.

<cite>It is my understanding that the extended high range ISO, as well as the intermediate 1/3 stop ISO's (anything between 200-400, etc) work the same way</cite>

Definitely, the H settings work the same way: they're shot at the highest actual ISO the camera supports (i.e. the highest full-stop increment that shows up as a number), but underexposed by the appropriate number of stops. So on a camera whose highest real ISO is (say) 3200, H or H1 is metered at 6400 but exposed at 3200 (i.e. underexposed one stop), and then the RAW converter (or in-camera JPEG engine) is responsible for boosting everything one stop.

I know that at least some cameras implement the 1/3-stop ISOs similarly: they're metered at the indicated ISO, but exposed at the nearest full-stop value, and again the RAW converter (or in-camera JPEG engine) is responsible for adjusting accordingly. I don't know if this applies to all EOS bodies with 1/3-stop increments or if some of them might actually implement 1/3-stop settings in the sensor itself. I don't believe Canon is in the habit of publishing details on this.

September 3, 2012

<cite>Otherwise, PSE is only 8-bits. You can open a 16-bit image in PSE, but you can't do anything to it until you convert it to 8-bit.</cite>

This is oversimplified to the point of being wrong. Even PSE's predecessor in the 1990s, Photoshop LE, could do one thing at 16 bits (Levels), which is more than "can't do anything." The last several versions of PSE have been able to do a number of things with 16-bit images, including levels, highlights/shadows, saturation, unsharp mask, and many (but certainly far from all) filters. Yes, there are many things you can't do at 16 bits, including anything to do with Layers (and that's a pretty serious limitation for a lot of people).

I don't know why you're seeing it saying 8 bits when you open your files. I use PSE 8 and it shows 16 bits for 16-bit files, regardless of whether they're coming from ACR or as 16-bit TIFFs from a non-Adobe RAW converter. I can't help you with how ACR handles DNGs created from NEFs, but I'd be surprised if they were limited to 8 bits, since ACR can certainly convert at 16 bits from the RAW files from the two cameras I've used (Canon 20D, which produces 12-bit RAW files, and 7D, which produces 14-bit RAW files). One of the things you can't do in PSE is to convert from 8 bits back up to 16, so if you're doing your work and saving a file, and you can verify that the file you've saved is definitely a 16-bit file, then despite what PSE was displaying, you were indeed working at 16 bits.

If the Transformations you mention are the ones I'm thinking of (adjusting the geometry of the image), then they're layer-based, and will only work at 8 bits. But that's an example of something where bit depth shouldn't be all that critical. Tonal adjustments (levels, highlights/shadows, saturation, local contrast enhancement, etc.) really ought to be done at the highest bit depth possible, and PSE supports those at 16 bits if you're doing them directly to the image rather than via Layers (there are pros and cons to both approaches). As well, I've long used the PTLens plug-in to do most geometric corrections (as well as fixing chromatic aberration), and it handles 16-bit images. So the 8-bit limits on Transformations aren't a problem for me.

So you could probably find some people who do what you need to do in PSE and are satisfied, and others who aren't, in part depending on their personal preferences and the way they organize their workflow. Make sure it doesn't actually meet your needs before spending money on something else.

August 17, 2012

I bought a used 300/4 IS several years ago for use with a film body and was completely pleased with it. Completely. I ended up selling it a little while after going digital, since on a 1.6-crop body, it was now too long a fair bit of the time (and I kept needing something between it and my next-longest lens).

If a 300mm prime is what you need, and the 2.8 is not an option for whatever reason (price, size, weight, etc.), this lens is a great choice. It's very sharp, even wide open (and I found it works well with the 1.4x II as well, though it's a bit soft wide open with the TC; stopping down one stop makes a big difference). IS is very handy, even though it's an outdated IS version (only good for about two stops). With IS and the relatively modest weight of this lens, it's quite handholdable, and IS works well on a monopod, too.

August 9, 2012

Yes, it's been mentioned here before, <a href="http://www.photo.net/canon-eos-digital-camera-forum/00ah7J" target="_blank">on Monday</a> and <a href="http://www.photo.net/canon-eos-digital-camera-forum/00ahFZ" target="_blank">on Tuesday</a>. There are some initial reports from users there.

Yes, previous EOS DSLR firmware updates were free. But none of them expanded a camera's features as extensively as this one; to get that much improvement in the past, you'd have to buy a new model of camera. I suspect that's why the initial poster pointed out that it's free.

August 8, 2012

This is an interesting approach, but it's a bit handicapped by the limits you mention. There's another approach that works just like a regular hard drive but faster: a hybrid drive. It's an SSD and a traditional hard drive in one package, and not only does it physically look and feel like a hard drive, it also acts like one from the perspective of the rest of your computer - so no special hardware or OS support is required. If your computer can use a SATA drive (and SATA drives have been the standard for quite a few years now), it can use a hybrid drive.

<a href="http://www.anandtech.com/show/5160/seagate-2nd-generation-momentus-xt-750gb-hybrid-hdd-review/" target="_blank">Here's a review of one such drive</a> (comparing it to the previous-generation hybrid drive from the same vendor, as well as to pure hard drives and pure SSDs).

August 8, 2012

Puppy, I agree that writing firmware for hardware you control is much less complex than, say, writing an update version of Windows that has to run on a gazillion different computers. But hey, as noted above, Apple has messed up some of their OS versions for hardware they control. And while I don't recall the details, I remember that Canon released a firmware update for the 20D that turned out not to work very well in the real world, and they had to release another version to fix whatever problem that was. (I didn't suffer from that one due to my habit of waiting for a while to see if a new firmware update is stable - which I apply to pretty much all firmware updates, too, not just point-zero ones.)

I'm happy to hear that the initial reports from a few users are positive. Maybe I'll get my bigger buffer and auto ISO limits and in-camera aberration control for JPEGs and whatnot sooner rather than later!

August 7, 2012

<cite>The "Features" I was asking about are issues that people have experienced after the upgrade that would require backing it out, i.e., like the Iphone v5 software upgrade that caused all sorts of problems.</cite>

Then don't be among the first to install it, unless there's something in it that you truly need. I've been an IT guy for the last 20+ years and my general recommendation is to avoid point-zero releases if at all possible, at least until enough others have taken that leap and found them to be stable. Far too much software gets released that really ought to have been tested more than it was, and I have better things to do with my time and energy than dealing with bugs in version *.0 that the vendor will probably fix in version *.01 in the next little while.

Personally, there's nothing in there that I need, though there are a few things that might be nice to have. So I'm not installing it for now. I'll keep my eye out here and elsewhere to see if others' experiences are like Puppy's or if there are reports of problems. If it proves to be solid, I'll upgrade. If 2.01 comes out in the next little while and fixes a bunch of problems that showed up in 2.0, then I'll upgrade directly to 2.01 and skip the problems.

As an aside in relation to your iOS 5 comment: I have a friend who reliably bricks at least one of her assorted iThings on the day a new OS is released. (This is not an anti-Apple comment; other vendors also have a spotty history with brand new OS versions. It's just that she's an iPerson, so every time she bricks something, it's an iThing because that's all she uses.)

July 31, 2012

This is very far from a complete answer, but here's one option to try. Your camera included a copy of Canon's DPP software. The latest version, downloadable* from Canon's Web site, includes an HDR module. That module can work with either RAW or JPEG (I'd suggest RAW if possible), and while it has several controls to adjust parameters, it also has a handful of presets ranging from subtle to garish. So if it's easy you want, just flip through the presets until you find one you like; do this for a few images and you'll probably settle on one or two of the presets that you like best. And if you want to start with a preset and then tweak the settings on an image until it's more to your liking, you can do that: the presets are just pre-packaged combinations of the settings for the various controls, so you can start there, adjust things, and if you find it's messed up, just pick the preset again and it will reset the controls for you.

Like I said, it's only one option, but since you already have an older version of the software and can download an update for free, and the presets let you do some basic HDR by picking something off a menu, it's worth a shot.

*: Any update you download from Canon will refuse to install unless it finds evidence of a previous version of itself, or related Canon software, on your computer. But you already have the CD that came with your camera, so install the old DPP, download the new one, and the installer will be happy.

July 20, 2012

Before I go into a bit more detail, let me first say that I agree with the first two responses. You ask if it's necessarily a problem, and the answer is no, it's not necessarily a problem. Certainly, in some cases, it's a problem. But in others, it's not. The bottom line in most cases* is this: do you like the way the image looks? If so, then whether there's clipping is pretty much a non-issue. Nobody except weenies and the editors at stock photography sites (and the two groups probably overlap somewhat :-) is going to check the histogram of your image before deciding if they like it.

On a technical level, if a channel clips, it's going to throw off the colour balance of those pixels where clipping occurs. For instance, imagine that part of your scene is an area that's all the same colour but some parts are lit more brightly than others. Let's say the colour is such that its blue component is exactly twice its red and green components (i.e. a moderately saturated blue). In areas where there's no clipping, every pixel has this relationship, which is correct for the item you're photographing. In areas that are brightly lit, though, maybe the blue channel in your image starts clipping: it's stuck at 255 (on the traditional 0-255 scale), yet the green and blue components keep rising. As they rise, it fades from blue towards white, even though the actual item you're photographing is all blue, not white.

An example of this type of effect in the real world is the sky. Ever taken a picture in which the the sky, which was various shades of blue in real life when you took the picture, ends up as much lighter shades of blue and heads towards white in some areas? Yeah, we all have at one time or another. That's because the blue channel is clipping before the others.

But what if it's only slightly clipped? If it's stuck at 255 when the real number should have been 260, then for almost all practical purposes, it's not a problem. You'll end up with a very slightly different shade of that colour, but it's close enough that nobody's going to notice.

And since you ask if the image will be "printable," that brings up another point: your monitor, whatever device you're thinking of using to print the images, and the colour space you're working in probably all have different gamuts (the range of all colours the device/space can represent) anyway. Just because a channel isn't clipped in your image file doesn't mean your monitor and/or printing device won't clip it. Again, backing off from a technical answer into a more practical one, if it looks good, then it is good, regardless of whether there's some clipping going on somewhere in there.

*: There are some technical types of photography in which maintaining precise colour accuracy is crucial, and clipping is going to be a no-no there. But these are the exception rather than the rule.

July 17, 2012

<cite>I was wondering though, shouldn't the Canon-supplied corrections via DPP be better, or at least, easier to apply?</cite>

I have been happy with PTLens for several years, and have recently been experimenting with DPP (since the Digital Lens Optimizer feature appeared). The following are my unscientific comments on what I've found with the two of them; please note that these are just my thoughts and not the result of any formal testing. Also note that I use Photoshop Elements, which uses a crippled version of ACR that doesn't include whatever lens corrections are available in ACR in Photoshop (and presumably also in Lightroom).

<ul>

<li>Curvilinear distortion: each product has a lens database it uses to correct this automatically, based on shooting information such as lens used (and zoom position in the case of a zoom). Someone who does something like architectural photography, where "close enough" isn't good enough, may find one product does it better than the other; for my purposes, either one works just fine.

<li>Chromatic aberration: DPP presumably has a database for this, too, as it automatically gets it right without generally needing any fine-tuning. PTLens makes you manually adjust the correction parameters for each shot. I'm happier with DPP on this one.

<li>Vignetting: Again, DPP generally pretty much gets it right without fine-tuning, so I suspect it has a database for this, while, PTLens makes you adjust it manually. (But as I have a 1.6-crop camera and all but one of my lenses are full-frame, I rarely have any significant amount of vignetting to fix.)

<li>De-fishing: both can do it, but as I don't have a fisheye lens, I haven't played with it.

<li>Perspective correction: as far as I know, DPP lets you rotate and crop the image but won't fix your picture if you were standing on the ground looking up at a tall building. PTLens will. But there are other tools that will let you fix it, too, so the main advantage PTLens might have here is that you can roll this into one process along with the other corrections. It's left to the reader to determine whether that is of any benefit to them and their workflow.

<li>Digital Lens Optimizer: DPP has it for selected lenses, and in some cases, it's magic. PTLens doesn't have it. If you're working with an image that can benefit from this, DPP is the only way to go. (But, as mentioned in a previous response, you can always use just the bits of DPP you find useful, convert the image to a 16-bit TIFF, and then use any other tool that can read 16-bit TIFFs to do the rest.)

</ul>

So, after years of using ACR as my converter and PTLens to fix the above defects, I'm now starting to figure out which tool would work best on each image, and using a mixture of ACR and DPP.

July 16, 2012

There are some add-ins for some browsers that allow you to read at least some EXIF data, such as <a href="https://addons.mozilla.org/en-US/firefox/addon/fxif/">FxIF</a>. There are also standalone programs like <a href="http://opanda.com/en/iexif/">IExif</a>, some of which may have a browser plug-in available. And, of course, there are full-blown image viewer/editor programs that also let you view EXIF data, though typically you'd have to save the image using your browser and then open it in the viewer/editor.

As for losing EXIF data when editing, it depends on what editor you're using and what you're doing. Some editors may do a better job than others of preserving EXIF data during editing. I use Photoshop Elements and as far as I know it preserves whatever EXIF data is there (but I haven't don a before-and-after comparison to see if anything's gone missing). I think <a href="http://www.irfanview.ca">Irfanview</a> also preserves EXIF. But that's assuming you choose to save the EXIF data when saving. There may be an option in the dialog box when saving that says whether or not to save the EXIF data; many programs also have a "Save for Web" feature which may strip the EXIF data to reduce the file size.

And focus distance? Not all cameras record it in the first place (my old EOS 20D did not; my new EOS 7D does), and even with a camera that does record it, not all lenses report it to the camera.

July 10, 2012

Also see if the manufacturer of your video card has a movie player available, and (if so) whether it can read the .MOV files your camera produces. Most graphics chips from the last several years include hardware acceleration for at least part, and often all, of MPEG4 decoding, and since they have dedicated hardware for it, they have no trouble keeping up with even 1080p video. The video card maker's player, if there is one, would know this and would take advantage of it; other video players may or may not (you'd have to check the documentation) and may force the CPU to do the decoding instead.

Just a thought ... I don't do video so I can't give you a list off the top of my head of which players can utilize hardware acceleration for video decoding and what drivers/GPUs they support this on.

July 10, 2012

There's already a JPEG embedded within the .CR2 file (though I'm not sure what size JPEG it is). <a href="http://www.google.ca/search?hl=en&as_q=extract+jpeg+from+cr2&as_epq=&as_oq=&as_eq=&as_nlo=&as_nhi=&lr=&cr=&as_qdr=all&as_sitesearch=&as_occt=any&safe=off&tbs=&as_filetype=&as_rights=" target="_blank">Google tells me there are tools available to extract this</a>, so if you need both but don't want to use up the space on your cards*, you could just shoot RAW and then use one of those tools to extract the embedded JPEGs once you download the images to your computer.

*: Also, the RAW+JPEG takes a bit longer to write to the card since there's more data, and on at least some models, it may also reduce the number of shots you can fit in the buffer, which could be a problem if you're shooting bursts.

July 5, 2012

Check the manual and/or specs to find out if your printer's print heads are part of the printer or part of the ink cartridges. (I don't know about current Epson printers, but at least in the Olden Days, Epson's print heads were part of the printer.)

The residual ink that's left in the print head after you've printed tends to dry up if it sits there unused for a prolonged period, and it can get to the point where even the printer's head cleaning function can no longer unclog the print heads. For printers that have the print heads as part of the ink cartridges, buying a new ink cartridge cures the problem quite easily (at the cost of having to buy a new print head every time you buy a new cartridge, even if your current print head is working properly). For printers that have the print heads as part of the printer, though, you're looking at replacing the print head, which is not necessarily designed to be cheap or easy to do.

So if your printer has built-in print heads, you may want to make sure you use every ink occasionally to avoid having print heads get clogged.

June 28, 2012

I'm pretty sure the range of new and improved features in this firmware update far exceeds anything they've done before in a firmware update for any EOS body. Two things that are sort of missing from the list above (they'd be lumped under the new in-camera RAW editing feature) are correction of distortion and chromatic aberration; not only did the camera not have in-camera RAW editing before, but it didn't have either of these corrections at all! The only item it had to correct optical flaws in the lens was peripheral illumination correction. Oddly, I haven't seen mention anywhere (yet) as to whether these can be set to apply automatically when shooting JPEGs, as peripheral illumination correction can, but since the firmware includes the code needed to do this, it would make sense to do it.

Now, most of this isn't significant to me. I shoot stills, almost exclusively RAW, process them on my computer, and even when I had a 20D with its 6-shot (for RAW) buffer, I rarely shot a burst that ran into that limitation. Still, I can see that I might use some of these features from time to time, and I will be upgrading my camera eventually. Just not immediately; it's almost always a Bad Idea to be among the first to use a major new release of any software, and chances are pretty good there'll be at least one minor point release coming along to fix bugs. So I'll wait either for that, or until it becomes clear that the new firmware is solid.

<cite>Also curious about how they did that w/ a simple firmware update</cite>

The press release ascribes this to "powerful memory management algorithms taken from Canon's flagship EOS-1 series".

June 20, 2012

Yakim, thanks for thinking of asking Chuck! If anyone can give us whatever amount of technical information Canon is prepared to release, it would be him.

June 20, 2012

Thank you to those who repsonded to my query and confirmed that 3.11.31 can use lens data you've previously downloaded with an earlier version!

<cite>The download time was extremely slow.</cite>

As with Rob Bernhard's comment earlier today, I found the same thing when I downloaded the 90 MB of data required to support my four current lenses. I didn't time it or measure the actual download rate, but it took way longer than I'd have expected (and, as an IT person, I'm quite accustomed to downloading large installers, patches, etc.).

<cite>I was expecting a list closer to the one you get for adding Peripheral Illumination Correction to a camera in EOS Utility.</cite>

The lens correction info is far more complex than the peripheral illumination correction and I'd expect it would take Canon a lot more effort to produce these files. I won't be surprised to see the list of supported lenses grow gradually over time, but I also won't be surprised if some lenses never make it onto the list. But that's just my guess.

June 19, 2012

<cite>they artificially handicapped 3rd party lenses by programming in a 'cap' for them at Group C</cite>

That's not what the article says:

<blockquote>We have no way to tell with third party lenses if the central points are acting as single or dual-cross. (Other than knowing the dual-cross sensors are NOT active if the maximum aperture of the lens is smaller than f/2.8, because Canon has said that.) Anyway, for purposes of this post, we’ll call them all Group C, because that’s all we can tell.</blockquote>

June 19, 2012

Interesting. If you already have the lens correction data for your lenses loaded and then update DPP to the new version, which according to the release notes fixes a problem that I (and doubtless others as well) had reported with the HDR tool, will it happily use the existing correction data, or does it manage to lose access to what you'd already downloaded?

Sign In

Profiles

Forums

Blogs

Events

Downloads

Gallery

Store

Posts posted by steve_dunn2