Apparently in Eevee it was decided that the minimum value prevails while in Cycles it is the maximum value.
Nothing of the sort was “decided”. EEVEE calls the GLSL clamp, and that’s it. This is not an EEVEE vs. Cycles issue. This is an EEVEE cross-platform consistency issue which, unfortunately, comes from outside of Blender codebase altogether.
If you want to cite intended behavior please at least first make an effort to actually verify what that intended behavior is.
You’re missing the point. I’m not complaining about a bug report that was closed. I don’t care if it’s closed or stays closed or whatever else. After all, it’s not even a Blender bug.
What I do care is that reports are actually looked at and analyzed, and not dismissed based on conjecture, especially when it’s incorrect. That’s just unprofessional, (un-professionaly-courteous, if you will) and reflects poorly on development.
dismissed based on conjecture, especially when it’s incorrect.
It’s unfair for you to suggest that. Germano knows what he’s doing. He makes mistakes sometimes, but he isn’t flippant about it. There are better ways to resolve this – send a DM on blender.chat. To me, it seems more unprofessional to go to a public forum for something like this, without at least making an attempt to clarify things with the triager. They do make mistakes, but they also take time to look at and analyze the problems.
Besides, look at the title of this thread: “Guys, please don’t invent reasons out of thin air”. I don’t think that reads like a request.
You’re helping by submiting reports, and even protesting a closed bug, but please do so on the proper channels. This isn’t really appropriate for devtalk. And similar threads have been opened in the past – Some of the bug triage staff are very inexperienced Browse through it, It’s worth the read.
The tracker is public. There was no point in taking this private now - it’s all already out there, for all to see. It’s not a personal “between you and me” issue, it’s a request to all the devs and everyone who’s involved in bug reports. You’re continuing to miss the point. There are indeed better ways to “resolve” this - by not creating this situation in the first place, which is exactly what I’m asking, and which is why I’m asking it here. And it is not much to ask, really.
“Please don’t do X” - is not a request? Well, I don’t know what is then.
This is absolutely a personal issue. You’ve publicly accused Germano of being unprofessional and dismissing your report without investigating it. It makes no difference that you’re extending the unfair characterization to the rest of the traige staff. You shouldn’t say something like that without backing it up.
I wouldn’t be so harsh with you in my response, except that I’ve seen this thread pop up again and again with small variations. Please read the other thread, you’ll see what I mean. I would take this a lot more seriously if it were someone else’s report you’re citing. But this problem never seems to occur with anyone else’s report!
Professionalism isn’t personal, I’m sorry you didn’t know that. Without backing what up? I’ve already explained everything: the root of the problem is in the GLSL clamp function (which apparently produces undefined results while being defined a certain way). Read: you can get the same file do different outputs with the same inputs on different platforms, or even on same platform with different driver versions. And all of that can be “correct”. Which is what the report is about, which is why it named the way it’s named. Which is what is stated in the report. Which is what Clément immediately alluded to as well. 'Tis not remotely about anything that Germano tried to conjecture in his response. Like, at all. There was no “decision” about picking different values for Eevee and Cycles. I don’t see how else must I explain this.
Worst thing is - he didn’t even need to write that. Close for now, CC, done. That is exactly my point.
…I’m sorry, you want me to point out similar behavior in someone else’s task as well? D9180, that’s all I’m going to say.
Also, now you’re putting words in my mouth. I’ve explicitly stated that the status of the report is not the issue here, pertinent information is. I don’t think you should continue in this fashion.