5D3 + DPP = lack of banding.

ron purdy said:

ron purdy wrote:
This is not a contest in who can wreck a RAW file.

Click to expand...

Apparently for some it is .

All kidding aside Horshack is one of the more reasonable about the issue on this forum.

BTW, the problem is not with ISO 400 or 800 files but ISO 100. It's most likely still there with DPP if you beat the file up enough but the real question is does it matter and for me it doesn't.

Bob

--
http://www.pbase.com/rwbaron

Without wanting to drive my security software to extremes I have to agree with Ron with what I can glean from the clip and the thumbnail. Appears to be driven to extremes! Lets see some more normal, somewhat underexposed raws!

Horshack said:

I guess I misunderstood the point of your OP.

Click to expand...

When your underexposed file is properly exposed using DPP adjustments, it looks great to me. Even at 100%. Sharp, good colors, no banding.



--

ron purdy dot com

marshwader said:

I'm sure you are right. You get no bells and whistles in Dpp. It can be akward and clunky but the banding seems much dimished (and it's free). What happens if you convert in Dpp and move on to do any further processing in PS? I just think Dpp knows more about how Canon cook their "raws" then any reengineered converter.

Click to expand...

My experience with my 7D (don't have a 5D3 yet) is that Adobe RAW conversions seem to boost the reds unnaturally compared to DPP. I did a comparison on a file and the ACR version was noisy with a lot of false color in the boosted shadows compared to converting with DPP and applying the same shadow boost on the 16 bit Tiff in PS.

Bob

--
http://www.pbase.com/rwbaron

You've probably done this. So, apologies, if this is irrelevant. But if you select camera standard as against adobe standard things look much more similar, although there are still marked differences in the shadows and reds. Ron's version of the "banded" file looks great to me. Maybe everyone has extreme expectations.

ron purdy said:

ron purdy said:

I guess I misunderstood the point of your OP.

Click to expand...

When your underexposed file is properly exposed using DPP adjustments, it looks great to me. Even at 100%. Sharp, good colors, no banding.

Click to expand...

I can see banding in my image with only a 1EV push. Your OP contained an image pushed +2EV, bringing it to the point of overexposure. I did the same +2EV push with a different image yet somehow I was trying to "wreck a RAW file" and my "exposure was too high". I'm not following your reasoning.

Horshack said:

I can see banding in my image with only a 1EV push. Your OP contained an image pushed +2EV, bringing it to the point of overexposure. I did the same +2EV push with a different image yet somehow I was trying to "wreck a RAW file" and my "exposure was too high". I'm not following your reasoning.

Click to expand...

Pardon me, I didn't mean that you in particular were trying to wreck sh!t. I was referring to the many threads that can be found on this site. Sorry about that.

Now, my point is that when either the samples I posted, or your sample is properly exposed, they look great IMO.

Also, did you turn the Auto Light Optimizer Off in DPP? I would do that, make a clean TIFF, and continue with adjustments in Photoshop. And next time, try to get your exposure right Just kidding.

--

ron purdy dot com

I agree and IMO this ISO 100 shadow lift thing has been taken to the extreme by a few on this forum.

Certainly, for some extreme applications like shooting early morning or late afternoon sun over terrain with a single shot, exposing for the sun and then lifting the dark terrain 4 or 5 stops in post will show pattern noise with a Canon and some Nikon's but not with a Sony Exmor. The question is why would you do that if you didn't need to. With a 4 or 5 stop lift you'll still have ISO 1600 or 3200 noise in the shadows whether banding is there or not and color fidelity will most likely suffer too. If the situation allows just use an ND grad or blend 3 exposures in post to get a really pristine result and this applies to any camera with or without an Exmor sensor. Now someone from the cult will say "well, what if the ND grad doesn't follow the shadow line or your subject is moving"? OK, those are some instances where this post process recovery method may save a shot but how many times does that happen for most of those shooting with these cameras?

Some of us learned to recognize and control good light from years of shooting slide film and 11 or 12 stops still seem like a gift. Sure, at times under very specific circumstances 14 is better but it isn't a perfect 14 either.

Bob

ron purdy said:

ron purdy said:

I guess I misunderstood the point of your OP.

Click to expand...

When your underexposed file is properly exposed using DPP adjustments, it looks great to me. Even at 100%. Sharp, good colors, no banding.



--

ron purdy dot com

Click to expand...

--
http://www.pbase.com/rwbaron

Is this a 5D2 or 3 file?

Horshack said:

ron purdy said:

kevindar, if you have 5D3 files which have banding when jacking up the shadows, I encourage you to process them in DPP and post the results. That would make more sense than just talking about it IMO ;-)

Click to expand...

Pushed 2 stops:



Full sized: http://horshack.smugmug.com/photos/i-RFZ5K7g/0/O/i-RFZ5K7g-O.jpg

Raw: https://www.transferbigfiles.com/d205b582-d04b-431e-9bdd-a7f7bc4640eb?rid=5qHADZEemHMAjl0MwGqKEg2

Click to expand...

--
http://www.pbase.com/rwbaron

rwbaron said:

Is this a 5D2 or 3 file?

Click to expand...

III

Here again is the file, the same area which Horshack posted at 100%, at +1.3 in DPP, and NR (and ALE) turned off. IMO this is the correct exposure and DPP converter settings for this file.



--

ron purdy dot com

Hey Ron,

I'm with you 100%.

Here's something I wrote a couple of months back and have referred others to a few times.

http://forums.dpreview.com/forums/read.asp?forum=1032&message=39742232

Keep up the great work -- I love your photography.

Regards,
Humayun

ron purdy said:

I am trying to reproduce the banding which I am seeing on the web.

The ligher shots is exposure at +2 and shadows at +5 in DPP – which is already way more than I ever have to adjust my files.

I can not see any banding in the shadows/black areas – even at this extreme adjustment level.

(tried with several RAW files from imaging-resource.com.)

HOWEVER, I AM USING DIGITAL PHOTO PRO, which no one else seems to be using for their tests.

DPP has always been the best converter IMO for the 5D2 in terms of noise, color accuracy, sharpness, etc., so this is what I use.

I would like someone who is seeing banding to try DPP and get back to us.

Following is one example…



-ron

--

ron purdy dot com

Click to expand...

ron purdy said:

ron purdy said:

I can see banding in my image with only a 1EV push. Your OP contained an image pushed +2EV, bringing it to the point of overexposure. I did the same +2EV push with a different image yet somehow I was trying to "wreck a RAW file" and my "exposure was too high". I'm not following your reasoning.

Click to expand...

Pardon me, I didn't mean that you in particular were trying to wreck sh!t. I was referring to the many threads that can be found on this site. Sorry about that.

Now, my point is that when either the samples I posted, or your sample is properly exposed, they look great IMO.

Also, did you turn the Auto Light Optimizer Off in DPP? I would do that, make a clean TIFF, and continue with adjustments in Photoshop. And next time, try to get your exposure right Just kidding.

Click to expand...

The original image had ALO on (didn't think to check it, I guess it was shot that way). When I turn it off it adjusts the exposure lower, which makes the banding less obvious but it's still visible. Btw, this wasn't my photo and I agree, I wouldn't do the adjustment I applied here for this particular scene. It was only for demonstration purposes. When you wrote an open invitation for DPP banding I figured this photo was a good example, because I thought you were looking for any sample of 5DM3 banding with DPP. The banding is certainly less for the DPP conversions vs ACR but either way the 5DM3 banding that is there with either raw processor cleans up well with Dfine2 so the point is kinda moot.

My point is not really to criticize dpp. there was a test by someone that showed indeed a bit less banding with dpp, but a lot less detail, and also, by default, it does not push the darkest shadows. I dont actually have a problem with my 5d3, but again, I rarely had it with 5d2.I find if I push the shadows enough, usually the details fall apart, with or without banding in the shadows. my main use will be for landscape, and I exposure bracket, and honestly find I can push my exposure easily by about 1.5 stops, with addtional stop or so of pushing shadows.

LR4 academic right now was 64 bucks from newegg (yesterday). very very much worth the price. I will not be upgrading cs5, as lr4 now takes care of 99% of my processing. but its awesome that canon includes a good program with even their entry level cameras.
--
http://razzi.me/kevindar/photos

Out of curiosity, have you tried converting that file with Aperture or ACR?
I have neither with me at the moment so I can not try it myself.

Thanks.
--

ron purdy dot com

"there was a test by someone that showed indeed a bit less banding with dpp, but a lot less detail, "

I am guessing that person had too much NR or or some such. I find DPP to have as much detail as the best of them, and more detail than most (when the correct settings are used.)

Thanks.
--

ron purdy dot com

I have no idea about DPP but Nik Define2.0 works quite nicely on this one and that integrates right into LR4 for those who are interested. I haven't seen anything from the 5DIII that Define2.0 won't clean up. I didn't even tweak it, just ran the defaults.
--
Flickr: http://www.flickr.com/photos/26158506@N07/

processed with DPP. looks OK to me.

If you want to push the shadows transfer the file to photoshop and use highlights/shadows .You get no banding that way.Period.

Forget lightroom if you want to push shadows.

Use lightroom for everything else!!!!!!

involving sunlight in order to reproduce banding the easiest if merely reproducing banding is your primary goal. Sunrise or sunset are really good times for this. Most landscape shooters who have been doing this a while spend a lot of money on GNDs and manage to "muddle through" the huge dynamic range of such conditions settling for a little clipping in the highlights (e.g. the center of the sun). Others use HDR techniques but I think such blending degrades resolution a bit. Still others think they can use "software filters" as their only filter and are probably the ones who end up with banding complaints.

Having said this, occasionally I am caught shooting in these conditions without my accessories and capture a decent composition marred by some amount of banding from PPing. It would be nice if the shadows were as recoverable as my D3x images but this is a very occasional happenstance and nowhere near enough to dump my Canon gear for a straight Nikon setup.

On those occasions when I have ended up with banding on an image I just had to keep, switching to DPP didn't help.

ron purdy said:

I am trying to reproduce the banding which I am seeing on the web.

Click to expand...

--

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.