I am reading David Ziser's excellent book on photograpghy where he recommends Highlight Tone Priority be enabed even for RAW shooting. I did some research and got a headache reading all the posts about Highlight Tone Priority in RAW and found a lot of conflicting info (including that it changed ISO and/or exposure in camera) so I decided to actually experiment (novel approach ) I don't want to start an endless thread on this subject yet again but I thought I would share what I actually experienced. I shot a high contrast scene with my 5D2--our outdoor shower in bright sun and shadow with a almost white stone bench in full sun. With and w/o HTP enabled the camera (on P) exposed at 1/320 at f11 shooting RAW. Looking at the LCD screen, there was a dramatic difference with the HTP enabled shot showing no blinkies on the bench while the non HTP shot was shown as completely blown out. I then downloaded the images to ACR which showed the exact same exposure settings for both--same ISO, same shutter speed, same f stop The ACR settings for exposure, brightness, blacks, recovery, etc were also exactly the same. The images also processed exactly the same. My conclusion is that HTP does absolutely nothing to a RAW image BUT, since the LCD shows a jpg image, an effect does appear on the LCD screen. Therein lies the dillemma. The LCD screen with HTP turned off will show blown out highlights which are recoverable in RAW and therefore not really blown out. Turning on HTP therefore arguably makes the LCD jpg image a better indicator of the RAW exposure. I therefore have decided to leave HTP on even though it has no effect on the RAW file so the LCD does not induce me to under expose more than necessary if the blinkies start showing up. Just wondering if anyone else is taking this approach? (I would not be surprised if the Canon RAW processer reads whether or not HTP is enabled and processes accordingly but that was not a concern for me)
UPDATE: I did one more experiment and clicked on Auto in ACR. Interestingly, ACR set the Recovery slider higher on the HTP enabled image so it appears ACR may read the file for HTP.
Personally I don't see the point of it when shooting RAW, I know what kind of recovery I can get in RAW files from both of my cameras, and LCD screens are so rubbish in sunlight anyway that relying on what I see on the screen is a big mistake anyway, if I went by what I saw on the LCD I'd vastly underexpose because EVERYTHING looks blown on a normal exposure when the sun shines on the screen anyway.
What I'm basically saying is..........histogram FTW.
I understand your point but the histogram is not easy to read regarding blown out details so I do rely on blinkies. However I know I can recover a lot so I do not adjust the exposure unless there are signifigant blinkies Uing HTP decreases the insignifigant blinkies and makes the lcd less misleading.
It can be hard, but if you know what to look for, you can get a pretty good idea of what's going on, you're very very lucky because I've done you a very professional diagram!
In the top histogram, see how the stack on the right-hand edge is near the bottom edge, and heading downwards, in JPEG you'd get a few bits of blown whites, but in RAW you're likely to be able to recover pretty much everything.
Whereas in the bottom example, the highlights have a lot of value right up to the very end of the graph, there's likely to be a lot of blown data in there that can't be recovered by RAW.
If this sounds patronising I apologise, but it's essentially how I read my histogram and I find it very effective, even my 6 year old 1Ds has a good 2 stops of highlight recovery, you have to get it pretty far out/shoot into the sun to lose image-destroying amounts of highlight.
You need to use DPP to use HLTR fully. DPP will open the image with camera parameters as defaults. Aftermarket RAW converters like LR, Aperture, etc., can't read Canon's software correctly. Personally I prefer the tools in Aperture and can get better control of contrast with a few quick tweaks. However, it may be worthwhile if you shoot JPEGs or mainly use DPP.
My understanding is that HTP works by applying the analog gain of one stop below rated ISO level, then applying a tone curve to compensate. This results in an image with a longer tonal scale in the highlights (because they have been "underexposed"), at the cost of somwwhat higher shadow noise (because they have been "boosted").
For me, on a pracctical level, whether or not it "affects the RAW bits" is immaterial. My experience is that I cannot recover highlight detail in a RAW image to nearly the same degree without significantly underexposing the image. Since I don't want to be bothered with these exposure games, I tend to enable HTP in high contrast lighting situations (even low light). The resulting photos do seem to have a less digital, more film-like look.
As far as I can tell, ACR/LR supports HTP as well as DPP these days; I haven't seen any surprising results from having HTP enabled when using either processor.
I use ACR and do hot have any HTP supported cameras, so if I wanted HTP, I would just underexpose the image and modify the tone curve manually.
It is not rocket science, and it is something I do to a minor degree on most of my images (I like a tone curve with a roll off at the top, just looks right to me).
I think Gary is mistaken. HTP is just a processing tag. You can accomplish the same thing w/o underexposing by doing it manually.. Regarding the graph, nice job but the chin in the midtones on graph 2?
I always, always shoot RAW.
If HTP adjusts the JPEG and automatically recovers highlights, then it is a bad thing for me. This is why.......
I prefer to nail the exposue in camera. I check the image review to monitor exposure periodically when shooting. If HTP is adjusting the jpeg used for review, then it is giving me false feedback regarding my exposure. The closer the exposure when shooting gives me the maximum to work with in post processing, should I need it.
craigy wrote:
I always, always shoot RAW.
If HTP adjusts the JPEG and automatically recovers highlights, then it is a bad thing for me. This is why.......
I prefer to nail the exposue in camera. I check the image review to monitor exposure periodically when shooting. If HTP is adjusting the jpeg used for review, then it is giving me false feedback regarding my exposure. The closer the exposure when shooting gives me the maximum to work with in post processing, should I need it.
Craig.
HTP is a 1 stop push, so If you want the camera to underexpose by 1 stop, it is great
HTP prevents me from setting ISO 100, and I shoot raw, so HTP is disabled.
I wish the histogram reflected the actual raw file, rather than the embedded preview jpeg. I can see how HTP together with the "blinkies" would be handy for quickly nailing the exposure to the right, but I don't want to give up full ISO range.
I believe highlight tone priority is more than just a processing switch.
I just tried it with a 1D4. Took pictures of the same scene with bright sunlight reflecting off a white fence and white house. With and without highlight tone priority. RAW.
Loaded them into ACR 5.6. I see a difference in the histogram. The regular one's histogram is shifted more to the right. There is more "headroom" for highlights that can be pulled back via the exposure or recovery sliders on the one with highlight tony priority enabled. It's very subtle, but if you're trying to capture details on a highlight in high contrast scenes, I see a difference for RAW files processed in Adobe's Camera Raw.
I don't miss not having 100 speed. I rarely shoot 100 unless for some reason I want a slower shutter speed. In my tests, prints shot at 100 are not distinguishable from prints shot at 200. So I go with 200 as my default.
HTP is nothing more than just a processing flag and a file that has been underexposed by a stop.
ACR will read the flag and under the hood boost the exposure by a stop.
The reason you see a difference in histograms is because with HTP on, its ACR's exposure boosting algorithm at work and with HTP off its the natural response of the sensor (or whatever Canon does in the ADC).
Try this experiment. Take one image with HTP on, load it in ACR. Take another image, this time with HTP off, underpose by a stop. Open this one in ACR and set the exposure to +1. The histograms of the two files should be identical. It is on every camera with HTP I've tried so far (except for the 1D4).
Oh and this was confirmed by a post from Thomas Knoll (in the Adobe user-to-user forums) back during the 1D3 days when he first added support for the camera in ACR.
According to Mr. Chuck Westfall, HTP affects RAW data:
Highlight Tone Priority (HTP) is available with all current EOS models excluding the Rebel XS/1000D. HTP has no effect on the actual dynamic range of the image sensor. It's just an alternative method of image processing that preserves more highlight detail than Canon's standard processing, without significantly altering midtones or shadows. The effect of HTP is enhanced by Canon's 14-bit A/D converter, which provides finer tonal gradations than the previous 12-bit system. HTP is a Custom Function with a simple on/off setting, and the available range of ISO speed settings is slightly limited when it is on. Take a look at the following web page for some sample images that show the capabilities of Highlight Tone Priority: http://www.imaging-resource.com/PROD...RE_HTPMODE.HTM ALO does not affect RAW image data. It is at its best for in-camera JPEGs shot in extremely contrasty lighting conditions. Examples would include backlit portraits and urban landscapes on sunny days, where the tops of buildings are brightly illuminated by the sun but subject matter at street level is in heavy shadow. HTP affects RAW data as well as in-camera JPEGs. It is very useful in high-key shooting conditions such as wedding photography and certain kinds of sunsets. With the exception of the EOS 7D, ALO can be combined with HTP with cameras that have both features. ...Show more →
thedigitalbean wrote:
The reason you see a difference in histograms is because with HTP on, its ACR's exposure boosting algorithm at work and with HTP off its the natural response of the sensor (or whatever Canon does in the ADC).
So you are saying that HTP depends on the raw processor and has nothing to do with in-camera processing other than the camera underexposing the image by one stop?
I don't see anything necessarily at odds between what thedigitalbean is saying and what Chuck Westfall stated.
Raw data is being "affected" when HTP is turned on since the image is being underexposed by one stop and this underexposure is baked into the Raw file. There is also a flag, which alerts the Raw development software of the underexposure but the actual Raw data was affected.
ALO, on the other hand, is purely a processing trick (like a levels/contrast adjustment) applied on top of the Raw data, either in-camera during the conversion to Jpeg, or in Canon's DPP software during post-processing. Even if ALO has been selected in-camera, you can turn ALO off in DPP when processing the Raw file. The ALO "flag" comes in to DPP as an initial setting - not a permanent one.
eric_m wrote:
So you are saying that HTP depends on the raw processor and has nothing to do with in-camera processing other than the camera underexposing the image by one stop?
) I don't want to start an endless thread on this subject yet again but I thought I would share what I actually experienced. I shot a high contrast scene with my 5D2--our outdoor shower in bright sun and shadow with a almost white stone bench in full sun. With and w/o HTP enabled the camera (on P) exposed at 1/320 at f11 shooting RAW. Looking at the LCD screen, there was a dramatic difference with the HTP enabled shot showing no blinkies on the bench while the non HTP shot was shown as completely blown out. I then downloaded the images to ACR which showed the exact same exposure settings for both--same ISO, same shutter speed, same f stop The ACR settings for exposure, brightness, blacks, recovery, etc were also exactly the same. The images also processed exactly the same. My conclusion is that HTP does absolutely nothing to a RAW image BUT, since the LCD shows a jpg image, an effect does appear on the LCD screen. Therein lies the dillemma. The LCD screen with HTP turned off will show blown out highlights which are recoverable in RAW and therefore not really blown out. Turning on HTP therefore arguably makes the LCD jpg image a better indicator of the RAW exposure. I therefore have decided to leave HTP on even though it has no effect on the RAW file so the LCD does not induce me to under expose more than necessary if the blinkies start showing up. Just wondering if anyone else is taking this approach? (I would not be surprised if the Canon RAW processer reads whether or not HTP is enabled and processes accordingly but that was not a concern for me)
? 