Although the effect is always far more pronounced late at night when there's no light around except for that emitted by the screen I occasionally become disturbed at the processing I applied to pictures in the past, particularly often those visible through the Year Ago thumbnail. They can't have looked that bad at the time or I hope I wouldn't have uploaded them. I'll have to check them again in different lighting conditions, though. I did start noticeably buggering about a bit more with the post-processing last August, seemingly spending far too much time darkening shadowed areas in the name of increasing the contrast a bit more between the light and dark bits. I started calibrating my screen every couple of weeks at the start of the year when I started being occasionally more bothered about white balancing but only use my eyes rather than one of those automatic thingbies which dangles over the top of the screen so it's probably not massively accurate and only works under the lighting conditions under which it was performed. Anyway, what started me noticing it (apart from clicking through a few Years Ago to see if anything referencable is approaching) was the fact that after looking for nice leaves on the way home (after accompanying my co-trainees back to the office in the taxi rather than walking today) and finding this one on Meadow Lane but then not processing it for a few hours after watching Let the Right One In again (with a break in the middle to watch popular television wizard Derren Brown's latest gullible-harvest) was that anything I try to do to it makes it look shiter rather than emphasized. MIght have another go in the morning when the ambient light is different but to avoid making it look the victim of HDR processing I'll probably not save the results.
Last day of training went ahead as planned today, though further evidence of the training-provision company's woes were evident in the absence of most of the flipchart marker pens and an even greater lack of other people anywhere in the building. Today was mostly catch-up, swiftly covering topics not so far covered and sorting out the problem with there being no branches nor sales in yet sales targets for Denmark in the sample data being used. It was getting a little bit irritating towards the end when training-blokey started mentioning the existence of and half-typing some parameters into some useful-smelling features only to stop half-way through without advising us of the vital next step. Luckily he'd handed out the course feedback forms by this point. As usual the usefulness will depend on the time available to continue to poke about with the relevant software over the next few months, but at least this built slightly upon some previous training eighteen months ago which I've only just got round to being able to make some use of but can still mostly remember.