Highlight Tone Priority Question mk3

MaDProFF wrote in post #4011766
PacAce can you please explain why, or where you have seen any info that makes you think this.

I checked 2 images taken in raw, and opened them in DPP and CS3, (ACR), one image with HTP on and one off in camera, and don't see anything that tells me anything different.
One thing for sure though, DPP reads and interrupts the raw data different from CS3, DPP seems to compress the range of the raw data in the histogram, from the shadow to the highlights, and it does it to the raw data from an image taken with and with out HTP, where as CS3 histogram seems to have a wider dynamic range, with HTP on and off, though with it on it seems to cap the highlights more.
Need other opinions on this, and I only checked a couple of images, I would be better off I think finding images where Highlights were more blown

What I posted here are based on what I've read in the 1DmkIII white paper and observation from making test shots with HTP enable and disabled and processing them in DPP and ACR, as well as simulating HTP (with HTP disabled) and processing the image to get results identical to those shot with HTP turned on.

could it be that DPP, just reads the raw data different form ACR, and CS3?

MaDProFF wrote in post #4011904
could it be that DPP, just reads the raw data different form ACR, and CS3?

Yes, and that's exactly my point. DPP knows how to process the raw images shot with HTP enabled and ACR does not. But the data they both see are the same. Just "interpretted" differently. If you look at an HTP image in ACR, it'll look like the image has been over exposed with the highlights blown out. With DPP, the exposure will look normal and the highlights won't be blown. And if you run on a Mac, OS X will show the same image as being underexposed.

I took some shots tonight and looked at the images in DPP and Lightroom. As Leo said, the images don't look very different in LR if you don't do anything to them. However, if you slide the recovery bar across, the HTP image can retain all the highlights with virtually no impact on the rest of the picture, unlike the non-HTP image for which most of the highlights cannot be recovered.

I think using recovery must do a similar job to the automatic htp adjustment in DPP. DPP shifts the histogram to the right for HTP images. The recovery slide in LR seems to have a similar, though probably not identical, effect.

I'm happy now and will continue to leave HTP on the "my settings" tab in the camera

(Note: this was just a quick test with a few shots and not a rigourous scientific experiment!! All other factors were equal in the camera and set to zero in DPP and LR, to help isolate the impact of HTP on the shots; except for white balance, which was set to "as shot".)

EDIT: On looking at the pictures closely, interestingly there is more detail in the highlights and better gradations using LR than using DPP, having fiddled with the settings to maximise the detail in both (ie LR with the recovery slider all the way to the right, and DPP with brightness set all the way to the left.)

Did you take JPGS or RAW?

Personally I am going to be rather upset if the statement is true for RAW images, as for the life of me I cannot believe Canon would spend a lot of time, money, and effort, into making the ability of HTP on the MK3, 40D and prob every other Camera that they introduce now, where it is a software adjustment only applied after you move the image onto a computer and only if you use DPP, especially as most people never use DPP, and use Adobe, or other editing packages.

For me CS3 shows a much better image in a RAW converted image than DPP does, have you ever opened a raw image in DPP, and used the convert to ADOBE in a tiff format, the tiff image in CS3 is appalling.

MaDProFF wrote in post #4017203
Did you take JPGS or RAW?

Personally I am going to be rather upset if the statement is true for RAW images, as for the life of me I cannot believe Canon would spend a lot of time, money, and effort, into making the ability of HTP on the MK3, 40D and prob every other Camera that they introduce now, where it is a software adjustment only applied after you move the image onto a computer and only if you use DPP, especially as most people never use DPP, and use Adobe, or other editing packages.

For me CS3 shows a much better image in a RAW converted image than DPP does, have you ever opened a raw image in DPP, and used the convert to ADOBE in a tiff format, the tiff image in CS3 is appalling.

Hi MaDProFF, if you are asking the question of me, then I took RAW. Perhaps I didn't explain very well. HTP works very well with Lightroom (and ACR) IMO.

My results show that HTP in fact works better with processing in lightroom than it does if processed using DPP. But you need to move the recovery slider somewhat to the right to see the full effect. This is no big deal as it doesn't affect anything much other than the highlights themselves.

That's just my observation of my own pics anyway. I probably won't do much more on this as I'm not a "tester".

I agree with you about the ACR conversion being better than having to go through DPP. What my investigation showed that not only is it not necessary to go through DPP to get the benefit of HTP, but that it's not desirable to do so. The better quality image when you use HTP in the camera, is got from processing the HTP image using LR (or ACR) directly and adjusting the recovery slider.

As I said in my previous post, if you don't use HTP then you can adjust the recovery slider all you want but you won't recover all the highlights. If you DO use HTP, then it DOES show the detail in the highlights.

Hope that helps.

Thanks, I also have Emailed Canon direct to actually get a proper Workflow of how HTP is processed through the taking of a photo with HTP set to on, though they make take weeks to answer

Since HTP increases noise in the shadow, so it is obvious that the HTP is just a software game and there is no evolution on sensor DR tech. so pratically, my 30d has this HTP

farrukh wrote in post #4017334
Since HTP increases noise in the shadow, so it is obvious that the HTP is just a software game and there is no evolution on sensor DR tech. so pratically, my 30d has this HTP

Practically, pinhole camera has 1/8000s shutter speed.
You just have to manually open and close the hole fast enough.

Although not technically correct (40D and 30D have different sensors), you are right: if you have enough time and knowledge, you can fiddle 30D files to get HTP effect.

But just the same as with high shutter speed on pinhole camera, I somehow prefer when it’s done automatically.

farrukh wrote in post #4017334
Since HTP increases noise in the shadow, so it is obvious that the HTP is just a software game and there is no evolution on sensor DR tech. so pratically, my 30d has this HTP

From my examination of the effect on the histograms, HTP is not something that can be emulated by other means. It looks to me that it mainly affects the highlights rather than the shadows and therefore is quite different to merely lowering exposure or manipulating the curves. Without HTP, more of the highlights are blown and cannot be recovered by any means. I have not noticed any marked effect on the shadows.

Just what I see - YMMV.

Hermeto wrote in post #4017398
I somehow prefer when it’s done automatically.

Off course its good to have luxury of automatic. I'd mostly prefer it manually though.