I’m quite new to RawTherapee as well, but I come from years of forgetting a camera can even make jpegs.
Well, my little bit of advice, coming from what I could unexpectedly but gladly rediscover with the last camera I recently got (X100V): it’s OK to be OK with the OOC JPEG
The raw development should be, I think, mostly to get out of the camera something that the camera development is not giving you already, but that you alone can get.
The above magnified view of your camera JPEG (with your brightening) on the left, and RawTherapee with the bundled profile “Standard Film Curve” for your camera on the right (with no other adjustments at all), shows how the camera JPEG has already been surpassed in terms of detail.
I applied the Samyang lens profile plus vertical perspective correction. This will definitely introduce some softening off-centre, but for most other lenses this will not happen.
One last thought from me, then I’ll let someone else have a turn.
Trying to replicate camera JPEG results is the whole wrong way to go about raw file handling. This will be more obvious when the JPEG is well off from what you want. If you think a particular JPEG is highly satisfying, the job is basically done, yes? So surpassing it to any great degree is futile.
But in many other cases camera JPEG files fall short to various degrees, or are just a starting point for where you want to ‘go’ with making the picture, so raw will give a vastly better result in those cases.
The short answer is either 120 cd/m2 or 160 cd/m2, I don’t recall which. It is a recommended level for my graphics working environment, which I follow the best advice and keep relatively dim. I set it with a monitor calibration device.
The longer answer is that chosen monitor brightness depends not only on personal preference, but also on the general room brightness, any direct light shining on your monitor, and the brightness of the viewing frame (for example my view of this page on pixls has a bright background, too bright for ideal photo viewing). An introductory article: What is the ideal brightness of a photographer’s monitor? | BenQ Asia Pacific
Also, your brightened JPEG has significantly blown highlights in a large area of sky (see black area in image below). That is a bad sign.
It’s not my jpeg! I haven’t shared any files in this discussion. I’m just commenting. As I say, I do find your image a bit underexposed. My monitor is set to 160 cd/m2.
Sorry, I thought you were the OP. My comments stand, though, as general advice. The OP might be a relative newcomer to the topic of brightness, and has brightened his photo quite a lot. Mine might be a bit dim depending on taste and usage (such as viewing on a white-page website), but his brightening is clipping the image.
Thanks again - I have been super busy at work and have not had time to digest all comments. Pretty sure my jpeg was not that bright - but will go through, check all comments and learn.
Would like to learn more about brightnness - my brightened jpeg seems right to me. I bought a factory callibrated monitor to try and get this right - I would be interested in what makes it too bright (I take the point that I have burned out some detail). I have found some of the edits a bit dark for my taste - perhaps my monitor is not high enough quality? ( Dell UltraSharp Monitor: U2518D)
I could not find an appropiate Samyang profile - where did you get this one? The trees in the centre are very high fidelity. I also thought a bit underexposed for my screen.
I have found this super-insightful. Sorry it has taken me so long to re-engage. I have learned:
Camera jpegs are OK, not all pics have to come from raw to be good. (Heaven help the partners of people who only shoot raw I say!)
People come up with quite different optimal raw images - so there is a range of tastes. Mine therefore are OK
RAW editing seems to be quite complex and there is quite a learning curve - and you need to love it to want to impove a 95% correct image - I can see that it would come into its own rescuing a problem image.
I still want to spend time jpeg-ifyng some of these raw image edits to see if I can get a better jpeg from them. I honestly thought that I must have stuffed up my lens settings on the day (can happen on the manual Samyang - f22 is truly junk - and you only find out a week later at home !). I thought this sequence of images (there are many more) was unprintable on canvas - but I no longer think that.
I gave a few thoughts on brightness in the middle of this post, link.
I don’t want to give the impression that there is an absolute right and wrong about brightness levels. It is obviously and ultimately a question of personal preference and interpretation. Although, having said that, if large areas of your processed file are clipping in one of the colour channels, that is a bit of a warning sign.
One can have a calibrated monitor from the shop, but if one is using it in a brightly lit room, one will tend to brighten the photos quite a bit. Conversely, if the same monitor with the same calibration is used in a pitch black room, one will be turning down the brightness quite a lot to make it “look right”. The same goes for the on-screen wallpaper or background that might be visible when editing photos: if it is very bright or very dark, it will affect what do you think looks right. However, that is why there are guidelines about the room illumination levels to use while editing photographs, and also for the grey levels on the screen backgrounds.
It is there in RawTherapee for me, under Lens Profiles/Manual/Camera=E-M5/Samyang/7.5mm ƒ/3.5 UMC Fish-eye MFT. I can’t recall if it is there by default, or if I found it somewhere when I started using that lens about 6 years ago.
I’m not using software that has sophisticated color manipulation, and that’s what this image seems to need. With global tools, if you deal with the yellowish green foliage, it’s at the expense of the sky…
I see this a bit in my Nikon-captured greenery; the matrix camera profiles yield more yellowish greens than the recent LUT profiles I’ve been able to make from camera-measured spectral response data. I really think this has to do with characterizing more precisely each channel, so that the high red band doesn’t mix too much with the mid-band green…
.
