May 24, 2013, 01:55:08 PM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Aglet

Pages: 1 ... 26 27 [28] 29 30 ... 39
406
Software & Accessories / Re: Upsizing pictures for large canvas
« on: August 11, 2012, 02:47:25 PM »
For printing on canvas, you don't actually need as high resolution, 150 ppi is fine and many are printed as low as 100 ppi. If you print at 40x27 inches, that will give you over 140 ppi, which is more than adequate for canvas.

Agreed.  I've got a 36" wide canvas I processed from a Rebel XTi, works out to about 140ppi in final form, and it looks just fine from normal viewing distances, maybe a touch soft from close up but the canvas texture complements it and people love looking at it.

When I tried Perfect Resize I did not like the output from it for close up inspection.  I'd use it on a billboard or a huge print where nobody's going to look too closely at it.  It does do some strange things to improve acutance at a distance but it's not a "photographic" look if you're fussy.  It depends on the image, it may be acceptable.  You can download and try a demo for free.

Upscale your image to the requested 240ppi the shop wants, and, as others have stated, using PS bicubic larger, in ONE step, then carefully sharpen the resulting image for an optimal appearance using some of PS's various sharpening tools.  Some 3rd-party plugins from Topaz or Nik Software can also do slightly different sharpening methods which may provide a slight advantage but not likely enough to warrant using them.

And if your clients are wincing in pain while they wait, maybe they'd prefer something a little softer and soothing to look at anyway.  ;)

407
Extra DR is pnly usefull if the image is taken with it. For the average image with only 8-10 DR then the possible extra from the D800 is just not used.

Good pp ensures the printing issues are resolved before printing.

I agree, somewhat...

Having a camera with extra DR means you'll have cleaner shadows, even if your image didn't contain anything as dark as the extremes the camera can record at some given SNR.  Thing is, many common scenes DO contain a wide DR.

Real world example below.
I just did a recent portrait shoot using 5D2 at ISO 100, outdoors (ughh) but at least on a cloudy day.  Using a 580EX2 for flash fill yet! 

Using the RAW tab in DPP:

Processing in DPP displays a raw luminance histogram ranging from -6.5 to +3 EV un-adjusted, pretty decent considering that it shows a full range for the 5D2 of -9 to +3.8-ish.  This looked fine when chimping the shot on the back of the 5D2 and using the RGB histogram.

But back at the shop
* Things are a little dull so +.67 EV added in DPP. 
* People are wearing dark gray and black dress pants with some sheen, + 2 clicks of shadow adjust so you can see some detail in the dark, but not darkest areas.
* -2 clicks of hilite adjust to soften the sky reflections off a receding hairline
* -2 clicks on contrast to slightly soften the whole scene so no blocked shadows or hilites or unpleasant contrasts.

Image is now beautifully balanced tonally, it will print well.  Except for.. what's that I can see?...  plaid pattern noise on the darker areas of the (almost) black pants!  And they've been lit a little extra from the shadows using fill flash!
In an 8x10 size print, we won't notice it, the pattern noise blends into the dark areas. If I print about 12x18" tho, it'll be visible; to ME anyway.

These are common adjustments to present a nicely balanced, professional-looking image that's not too contrasty. I've seen far too many "pros" frequently provide grossly contrasty images to clueless customers.  They can't complain about shadow noise they can't see because they've clipped it to black, along with some real shadow detail.  These photogs may as well use a cheap point'n'shoot but that's a rant for another time.

Summarizing, many common scenarios, even using some fill light, can provide a wider DR than expected, 9.5EV in this case on a cloudy day using fill light.  Properly toning an image can require lifting the shadow regions enough that the pattern noise can show up in dark areas that are large enough not to obscure them from our perception.  Printing larger will also make the pattern noise more apparent.

If you really light your scenes well, expose the right a little, without clipping, and bring it back down in post, you won't see this issue.  If you work in the real world with minimal lighting control, even a simple cloudy day portrait shoot can show up the weakness in Canon's shadow noise.  If you're shooting landscapes on a sunny day and have a wider DR and you need to compress it to make the image look how you want, you will run into this problem as well.

So to those few I often spar with here on this issue, the above is a real world scenario that's not even an extreme example and it'll STILL show up the weakness of Canon's sensor tech some of us complain about.
Hence whey I ask if they shoot (in real world lighting), process, and print. Because if they do, and strive for a nice looking final image, then they will run into this. It's just a fact. 8)

408
The problem I see with this is that, if the 12 stops of the 5DIII isn't enough, then the 14 stops of the D800 is unlikely to be sufficient, either...and, in any case, you'll get far superior results with even a two-exposure HDR rendering.

Now, consider that the 5DIII has an in-camera auto-HDR mode that will do a 6 frames per second bracket and even give you a not-miserable JPEG preview of the results...and the dynamic range advantage of the D800 vanishes in a flash.

That leaves the only remaining real-world advantage to the D800's dynamic range being for high-contrast scenes not suited to tripod work where you'll still have the luxury of being able to do significant amounts of post-processing...and, for the life of me, I can't think of anybody who actually shoots like that. Outdoor daytime sports published in high-end magazines? If so, good luck getting the shot in the first place, between the D800's framerate, autofocus, and Nikon's heavyweight supertelephotos....

Cheers,

b&

Do you shoot?
Do you print?
Do you do ANY post-processing to make your print or screen image look BETTER?!?

If so, then you'd appreciate the value of the extra DR.  After all, especially when printing, you have to compress the DR of the captured scene to very limited effective DR of print media otherwise you have a very contrasty image with blocked up shadows and blown hilites.

Of course, if all you do is get your snapshots printed at the local drugstore then you'll likely never realize the benefit of a superior DR image if you don't know the value of it in post-processing.

And how is a high-contrast scene not suited to tripod work?  Do tripods melt in the summer sun?

I still like my Canon gear, especially the modern glass, but I'm ecstatic about the results I get with my modern Nikon bodies, even using 30 year old lenses on them, because the noise levels are so much cleaner at low ISO.  The more I use and compare the two camera systems, the more I see the weaknesses in Canon's noisy low ISO images, even when controlling the lighting to some extent.

Sure wish I COULD mount my Canon glass on a Nikon body tho. And I DO actually prefer Canon's ergonomics and user interface but the Nikons get me the file that make the prints that makes me smile.

409
HDR - High Dynamic Range / Re: Post your HDR images:
« on: August 08, 2012, 12:46:10 PM »
My first attempts at HDR... I processed in Photoshop CS5. It is very difficult to make anything that looks "real" with Photoshop's HDR processing.

They have potential. I'd reduce exposure a bit and maybe increase contrast or combo of both. One thing that makes photographs look real is some black clipping needed... HDR's expose everthing, that seems unnatural to some people.

I did a 3 minute PP on it... not sure if it looks better to you... but the idea was to make it a bit more natural.

yup, off to a pretty good start.
PS is not quite as easy to use for HDR as specialized SW but it's flexible, even when combining multiple bracketed images.
Messing with the tone-mapping you CAN create a fairly photographically realistic looking image by keeping the DR compression to reasonable amounts; there are a few ways to do that.  All depends on how "cooked" you like your HDR final image to look.

I still prefer to expose to retain hilite detail then push all the shadows in post using one good raw file.. but that requires using a very noise-free camera system like, dare I repeat myself yet again, a D800 or D5100.  In the same area as the shot above I took another one of a charred black tree, with the charcoal parts in the shade.  I shot it with both the 60D and a D800.  I can make a "realistic" HDR from either shot but with the D800 there is far more crisp textural detail in the dark areas because no noise reduction is required and I can still lift the darkest areas much more if desired without seeing pattern noise like the 60D produces.

410
HDR - High Dynamic Range / Re: Post your HDR images:
« on: August 08, 2012, 04:56:48 AM »
Here's a single shot HDR from one raw file, the only way I like to do them. This is a bit more processed than I'd normally do, I prefer a more realistic look.

Done primarily with ACR 7.x in Photoshop and a little NIK software tweak to provide more of the typical HDR look by adjusting local contrast.
Original is included, straight from DPP, no adjustments.
Shot with 60D in manual mode.

411
A lot of user's first response is: Will this improve the low ISO noise? LOL!
Haha!. If a new camera body won't improve low ISO noise (the 5D3), then people should not expect it from a mere firmware update. :D

Sure we can EXPECT it, since banding on some 7Ds seems to be a result of the dual-channel readout process.  It is possible, depending on hardware design, to better calibrate the dual readout channels for a better match and thus, reduce banding. 

Will have to put mine to the upgrade test after I hear how v2 firmware works for more critical functions... like AF.

412
Agreed.  Honestly, you can't go wrong with any of the top brands of DSLR's.  They all have the capacity to make magnificent images.  It's nice when the tools have all the bells and whistles... but specs don't take great images...

some more than others
specs define the limits of what you can do with the captured image and some of us can tell you, after shooting other systems, Canon's images are a bit more limiting in post.  We're not talkin' snapshots here. And you can't always control the light.



413
EOS Bodies / Re: No DXO results on 1DX until at least September...?
« on: August 03, 2012, 12:51:49 PM »
Lend me your new 1DX, I'll take some shots with the lens cap on for you and tell you what it can't do. :D

414
5D MK III Sample Images / Re: 5D3 for landscapes
« on: July 29, 2012, 05:13:05 AM »
I set ACR and PS to use ProphotoRGB 16bit (LR automaticall gives you Melissa which is ProphotoRGB 16bit with a different internal TRC). No sense in clipping anything you don't have too. Even AdobeRGB will clip stuff, especially flowers and sometimes sunsets/fall foliage/etc.

I've seen that as well.
And Adobe RGB can be a delivery color space as some printer ink sets are now covering a wider gamut than sRGB.  Best to learn when to use it, sRGB is adequate for most images.

I sometimes wish I had one of the big Canon pro printers so I could try some 16bit/channel prints directly from DPP to maximize tonality and possibly gamut.  Anyone have experience with this method for extreme color like sunsets and highly saturated flower colors?


415
Lenses / Re: Your 70-200 f/2.8L IS II...
« on: July 29, 2012, 04:47:49 AM »
I previously owned the 70-200 f/2.8L IS mark I. I wasn't very happy with it; it seemed to lack in sharpness. The mark 2 is a world of difference!

+1
I've had the previous vI IS and the non-IS versions of this lens.
I have a slight preference for the bokeh in both the older versions but this v2 is much sharper on a 5D2 from end to end and right to the corners (almost), so much so I can forgive some loss of bokeh sweetness.  There's a narrow range of focal length, aperture, and non-focus subject distance tho where the bokeh is horrid.  You may rarely see it but if you do, back up or move forward a few feet and recompose to get past the icky aspherical element aberrations.

Great IS, AF is nice and fast and my first copy was as close to perfect as I could hope for
$2200 cdn Dec. 2011.

I would have kept my v1 IS but it was always giving me very soft images at the long end where I most needed the performance, I had to replace it. 

And it's way nicer to use, IMO, than the v2 Nikon equivalent for those who also use that system.
The Canon lens is a joy to use MF, the Nikon has way too much sticktion in the MF ring so it's nearly impossible to do a precise MF with it.  It's otherwise fast and very sharp corner to corner like the Canon.

3rd party note, I also have the Tamron 70-200 f/2.8.  Got one new for $700!  No IS, not fast or precise AF, but very sharp at both ends wide open, weaker in the mid range until you stop down a couple.  Hard to MF with short focus throw but I don't mind it for use on a tripod when I'm not in a hurry.  I hope Tamron will release a stabilized version of this lens that performs as well or better.  Why do I bother with this one?.. Good bokeh in most situations, very smooth, sharp and cheap.  It's worth considering for some users because of the IQ/price.

416
..spending money on a 10-22mm when you've already got a 16-35mm f/2.8 seems pointless if you ever have aspirations of going full frame because the cost of that lens alone makes up a good portion of the difference between a high end crop and a full frame.

unless you opt to buy a used 10-22mm, which you can always sell next year for nearly the same you'll pay for it now. 
Great thing about good glass, if you can afford to buy it, it'll cost you almost nothing after you sell it if you keep it in nice shape.  Even revised new versions rarely depress the price much, if any at times.
Don't be too afraid to do a little "horse-trading."  I've done plenty, rarely cost me more than a few hours of my time over the years.

417
There really is no such thing as highlight recovery with a standard digital sensor. It's a hard cut-off and they have linear capture. You simply find the saturation point and then you measure the black point noise.

I suspect RGB channels will saturate at different levels tho, so that's about the only "recovery" available but then color accuracy is lost.
Some cameras tone curves vary a fair bit as they approach cutoff.  Canon seems to take it a bit more gently than Nikon from what little I've checked but I'm careful not to clip any color channel when taking a shot and that can be really tricky with some subjects.

It is. The Canon sensors themselves have a LOT more dynamic range captured than the result that gets collected after all the reading and converting of the sensor. I forget the details but I think the 5D2 sensor itself grabs well over 13 stops at the screen level, closer to 14 while the RAW file, after all the electronics damage the signal, has like only 11 stops left at the screen level.

I'd be very interested to know how Canon's sensors would perform if they used the same type of on-sensor processing as SoNi-kon's.

418
Yeah but it's awfully tricky if you want exact measurements out of that since you need a super stable light and power source and high-end equipment to measure output or some high end device to regulate the light and a detailed table of how much it outputs per various fed currents I think. I don't think many people have the setup to be able to do that very well.

i was thinking of single shot analysis, step wedge would be good but not necessary, just some varying reflectance targets from black to a few EV above.

The kind of analysis I'm thinking of would be to analyze the noise (variance) and try to determine the banding effect that way rather than averaging it or however DxO does it.  in fact, I'd be most interested in Peak-to-peak variance at those various levels because that's what becomes the visible distraction.

not that this is even necessary!
We can see from a typical shot that the raw files contains this to an annoying degree.
I don't need to measure HOW annoying it really is.  ;D

.. I don't feel like bothering with the time that would take..

+3!

I barely have time to go shoot for fun, let alone monkey in the studio testing stuff beyond what I've already done to decide I needed to get a better body for some of my work.


419
I agree w previous poster, try get a competent cropper like the 7D or even 60D and keep saving for a FF, hopefully the next one to be announced will be worth it.  If not, used 5D2's are out there for about $1600 and there's nothing wrong with its AF, especially compared to the 1000D, that's something I mount on fence-posts to get picked at by buzzards while waiting for a lighting strike.  ;D

420
I run an i7 iMac and use nearly everything BUT Aperture.  Whenever I try demos it chokes and crashes too often so I haven't even been able to get a sense of how well it works.

I do most of my work in Canon's free Digital Photo Pro.
When i need to do something DPP can't then I use DxO 6 & 7, Lightroom 3, Photoshop Ext. with Nik Suite and Topaz suite of add-ons. 

DxO doesn't have as many i-p & o-p file formats so a bit of a kink in possible workflow.  Free demo is worth trying.
I tried early LR 4 demo (pre-release) and haven't tried it since, performance was very sluggish but it did work well.

On One Software's suite I've demo'd a couple times with poor experiences on Mac as well.
Not sure if PhotoMechanic is available for Mac but my event-shooting buddy relies on it with his wintel laptop.

DPP, LR, DxO and PS.
If I had to (just) pick one that I had to pay for, it'd be Lightroom, does wonders for the price.

Pages: 1 ... 26 27 [28] 29 30 ... 39