Picture style...why is it recommended to set sharpness all the way down?

Hello, I have found a few sources (sorry, I dont have them listed) where it is suggested to modify the picture style by turning Sharpness all the way down, contrast almost all the way down and saturation down 1 or 2 notches. Then to handle all the settings during post. I came away from these sources thinking, if I recall correctly, that changing sharpness in post was much better than letting the camera do it. Same for the others.

So, I have been doing that...I also ONLY use the latest DPP for the moment...need to invest in LR or PS soon and well this is working fine for me.

However, I am getting a little discouraged with this approach, especially on the sharpness for the following reasons.
1) When I am shooting, when I check an image (chimp), and look for sharpness, the image is not going to be as sharp as it could be due to my picture style modification.
2) There are brief periods of time when I want to use Wifi and send a few images to my phone to share with others, these too are not that sharp since the JPGs are not as sharp as they could be due to my picture style modification.

So, is this just a compromise I have to live with? Or is there something I can do for best of both worlds?

The main reason for doing this is to attempt to move the displayed histogram closer to the RAW data saved in the CR2 file. Unfortunately Canon doesn't provide the option to display a histogram based on the RAW data, which many more advanced photographers would like to see. The histogram shown on the camera is based on the fully processed JPEG based on the in camera settings, even if you only record in RAW. This difference is probably not much if you are using DPP, since from what I have seen it is still not very good at dealing with highlights.

Many other RAW converters seem to have algorithms that will work with the images highlight details in a much better way. So pixels that in the Canon conversion that have clipped will prove to actually have detail in them, when processed in a different program. If you add using ETTR, Expose To The Right, to attempt to maximise the Dynamic Range of the sensor, something that again DPP is not the best at doing, this is why you might need to see a close approximation to the RAW data. Of course it does have the disadvantage that you can no longer check critical focus using the camera's screen, or use the out of camera JPEG directly. Sometimes I will go the full way and also use what they call UniWhiBal. This WB setting sets the WB up so that all of the colour multipliers for the four sensor filter streams, Red, Green, Green, Blue, are set to 1. The big disadvantage of this is that because in camera there are actually as many green pixels, as there are red and blue combined the image ends up looking green in the preview/JPEG. You can set the Uni WhiBal by simply making a very over exposed image, so that all the pixels are saturated. If you then ignore the warnings and use this for a custom WB you will get Uni WhiBal.

I use Lr, and since I started with a Canon DSLR that predated DPP, I never used the previous Canon RAW option. I now run with Faithful Picture Style, with the other controls at -4, 0, 0, -4. When I import them to Lr without adding any sharpening my Lr previews are now sharper and more saturated than the camera preview. Usually this is the other way around, and is often commented/queried about by new Lr users, as Lr initially uses the camera preview until it can build it's own. As it builds it's previews you see them change over.

If you are not shooting ETTR, or otherwise really worried about pushing the limits of the sensors highlights I would not worry about this. You will probably lose close to a stop of highlight detail if you look at the RAW data, and then process in a converter that is good at making use of it. At the moment that doesn't seem to be a problem for you, and you are using DPP anyway, so go with in camera processing that will get you a close to useable JPEG so that you can convert in camera and use straight away via WiFi. Remember that when the camera writes the CR2 file it is still writing all of the unprocessed RAW sensor data to the file for you to be able to process later, regardless of what the preview image looks like. For example if you shoot using the monochrome picture style, then open the image in any converter but DPP you will see a colour image. It is only because DPP is able to use the same algorithms that the cameras use that you see the camera's processing as the default conversion in DPP. You can still change all of that later, including the picture style.

So at the moment don't worry or overthink. Use your normal settings, that will produce a nice out of camera JPEG, and use those. As long as you are saving CR2 files you have all the sensor data, and can still use that in other converters later, you just might not have quite so much highlight detail still waiting to be utilised. At least you shouldn't have to worry about overexposure this way, the overexposure highlight alert blinkies will show up early not late for you.

Alan

kezug wrote in post #18437358
Hello, I have found a few sources (sorry, I dont have them listed) where it is suggested to modify the picture style by turning Sharpness all the way down, contrast almost all the way down and saturation down 1 or 2 notches. Then to handle all the settings during post.

Short answer - stop doing it.

I may be wrong but it seems from your post that you read about a technique and then adopted it without understanding its purpose. You have found that it doesn't work well with your workflow so stop using it.

The above settings are used by people who are using a technique call ETTR (Exposing to the right). If you are not doing it then there is no point in using the above settings.

Not if you need a JPG to upload immediately.
Also, depending on what format you shoot....
1. If you shoot JPG then the resulting files are less suitable for adjustment in post so you want to get them right in camera - the above technique is not suitable in this situation.
2. If you shoot RAW or RAW+JPG the in camera settings have no effect on the RAW data and will (as you found) render the JPGs useless for your intended purpose.

In conclusion there is no good reason for you to be using the technique, so stop.

When I shoot JPG, I process my photos using the Picture Style. SOOC is what I want.

When shooting in low, or crappy light, I shoot raw with the Picture Style set no Neutral and have all sliders at 0. I believe/hope that this gives me the most accurate histogram. I tend to push the histogram to the right with +EC. That gives me the most data to work with in post. Canon builds about 1 stop of headroom into raw data/metering/however you look at it.

In all of the material I have read about setting the picture style I mentioned above, I dont recall any mention of ETTL. It was all for better control in post to get the colors corrected and the sharpess to be done better, rather than in camera.

I shoot RAW+JPEG, but I always process the raws to how I like it and then export to JPEG for final use. I rarely, if ever use the JPEG SOOC.

So, for example, if I set Sharpess to 3 in my picture style, wont that sharpness of 3 show up as the setting in my my RAW and JPEG?
If so,
Is a sharpness of 3 done in camera, going to be the same sharpness of 3 I would set in DPP (assuming my picture style had sharpness of 0 in camera)?
Will my RAW be any different of sharpess setting of 0 or 3 in camera vs 0 or 3 in post using DPP?

kezug wrote in post #18438187
In all of the material I have read about setting the picture style I mentioned above, I dont recall any mention of ETTL. It was all for better control in post to get the colors corrected and the sharpess to be done better, rather than in camera.

I shoot RAW+JPEG, but I always process the raws to how I like it and then export to JPEG for final use. I rarely, if ever use the JPEG SOOC.

So, for example, if I set Sharpess to 3 in my picture style, wont that sharpness of 3 show up as the setting in my my RAW and JPEG?
If so,
Is a sharpness of 3 done in camera, going to be the same sharpness of 3 I would set in DPP (assuming my picture style had sharpness of 0 in camera)?
Will my RAW be any different of sharpess setting of 0 or 3 in camera vs 0 or 3 in post using DPP?

ETTL, the way I use it, gives me blacker blacks, more detail in the shadows, and less noise in the final product. It is not related in any way to Picture Style. Using N(0,0,0,0) gives me a further to the left, and lower histogram than I'd get with say S(4,2,2,1,1), or whatever. THAT allows me to push ETTL without clipping important highlights.

Whatever sharpness you set in your Picture Style will be read by DPP and applied to the raw file and the jpg file. The sharpness setting does NOT affect the raw FILE, only DPP's representation of that file.

DPP offers different settings than Picture Styles. As such, I don't see it as reasonable to expect the same output from the camera and from DPP.

Raw+JPG makes no sense to me if you are using DPP. Why settle for ONE specific PS setting, when any combination you can imagine is available through DPP?

I shoot both and use DPP, this way I can use a majority of the jpg files as is, but can go to the raw when I need corrections or better quality results. Faster this way...

Completely valid. I shoot raw when I know I'm going to need raw data, and JPG when I'm certain (?) I won't. I have been caught with my pants down more than once, though, leaving me to work a less than stellar JPG.

kezug wrote in post #18438187
In all of the material I have read about setting the picture style I mentioned above, I dont recall any mention of ETTL. It was all for better control in post to get the colors corrected and the sharpess to be done better, rather than in camera.

Better control how?

Most RAW processors ignore in camera settings but as you are using DPP the in camera settings will have an effect on your RAW files. This is because DPP (being Canon) uses the same processing engine as the camera does. That means it will read the in-camera JPG settings and apply them to the RAW.

But all that means is that when you first open the RAW file it will be soft/lacking colour which doesn't seem to be what you want. So how exactly is it helping you?

You guys keep saying ETTL which is an automated flash exposure control.

Do you mean to be saying ETTR? Expose to the right.

Some folks do ETTL, expose to the left, it is a thing... Canon shooters have a hard time of it given the substandard DR in the shadows on older bodies like the 6D, 5D3 and before, 6d2, 7d etc. Now that you have reduced noise in the shadows, you can expose to the left a bit and pull up highlights. I would rather maximize how much of the scene fits into the available DR, then only do surgery at the 2 edges where needed.

BigAl007 wrote in post #18437554
The main reason for doing this is to attempt to move the displayed histogram closer to the RAW data saved in the CR2 file.

I do this for this reason but leave the sharpening turned up to evaluate it in the preview image. I understand why contrast and saturation affect the LCD histogram, but does sharpness really affect it? If so, why?

CyberDyneSystems wrote in post #18438385
You guys keep saying ETTL which is an automated flash exposure control.



Do you mean to be saying ETTR? Expose to the right.

Nice catch, Jake. I did mean ETTR. Note to self, no more POTN before my morning coffee, or lunchtime beer.

If you are shooting RAW, even using DPP, the in camera settings are completely irrelevant to processing the RAW file later. With RAW ALL of the processing settings, all the way back to the WB can be changed in post, with no detrimental effect whatsoever. Admittedly unlike other RAW converters, DPP is fully aware of the in camera processing and uses those as the default conversion. Still even in DPP it is very easy to reset any setting to anything you like, without causing any intrinsic loss of quality in the image. Non Canon converters like Lr for example can usually only pick up the camera's WB settings, and even though you may use the same colour temp and tint values, the results may vary a bit from processor to processor, since they all use different algorithms to do the conversion.

If you need SOOC JPEG files then you cannot use any of the techniques that would make using the low settings necessary, so as already mentioned stop using them. Set the camera up so that you can have a useable image out of the camera. As long as you are saving a CR2 file for every image you still have the ability to start the processing again from scratch, and could even save the low contrast settings as a DPP recipe and apply them to the image before you start processing them, if you really want to. Not that I would bother.

Alan

If you want to really level out your histogram, just shoot the lens cap and set your white balance from that. Your shots will be green but you can correct everything later during raw edit.