Has anyone tried changing the bulb?
Issue with extreme shadow flickering and shadow contrast on planet surfaces post patch 6, is it just me?
- Thread starter Rheeney
- Start date
Apologies for not getting back to you with some flickering shadow videos (I blame Psykit and her latest race for keeping me off world this week).
I did just want to highlight some great examples of flickering (and strangely behaving) surface shadows that just happen to feature in today's video from the @Buur Pit tho.
I think these are particularly good illustrations since the Buurs will not have actively sought out bad examples, quite the opposite I would imagine, they'll want their videos to look as slick as possible and I imagine must really hate this stuff and have presumably been unable to avoid it.
Last edited:
Yeah some of it is just unavoidable still sadly. Thanks for posting @Alec Turner
sallymorganmoore
Senior Community Manager : Elite Dangerous
Thanks both
Zac's been doing a grand job of continuing to look into lighting since the initial post to bolster efforts so again, all this reporting in is amazing. Thanks again!Made a video about planetary shadows in 6.01:
Source: https://youtu.be/aulwaR765C8
And here's one about shadows in hangars flickering:
Source: https://youtu.be/juJns6k3q3A
And here's one about shadows in hangars flickering:
Weird, I've never(?, certainly rarely, if ever) seen such.
Maybe play odyssey more, because this whole thread + this issue on the issue tracker has plenty of evidence https://issues.frontierstore.net/issue-detail/41717
I play Odyssey several hours almost every day. However, I play it on Linux via steam thus Proton and Vulkan (on a 1080).
I DO see minor wobble on shadows, but that's just lowish resolution shadow maps. Those videos are something else completely.
Also, I use High in the "civilized" (hah!) areas and Ultra in the black.
No, I will not rest assured. I have seen plenty of elongated shadows, but they did not behave that way as far as I can remember. However, over the last two weeks, I spent a lot of time mining tritium for my carrier (when not visiting the sites along Hyford's Trail (as I call it)).
Do note that I have not denied that others do see these problems (that would be silly considering the videos). Rather, I was expressing my surprise because I'm pretty sure I would have noticed (and remembered) them if they were happening on my system.
Last night, after seeing those videos, I thought I saw some weird shadow behavior in a station (similar to that video), but as I watched carefully, it turned out to be just the fog from the ship lift (I was in a FS ship, prior to disembarking). I have been on the lookout for weird shadow behavior since seeing those videos.
Last night, after seeing those videos, I thought I saw some weird shadow behavior in a station (similar to that video), but as I watched carefully, it turned out to be just the fog from the ship lift (I was in a FS ship, prior to disembarking). I have been on the lookout for weird shadow behavior since seeing those videos.
I wrote "rest assured" (and state it again with full confidence) because it's not something that happens from time to time or in particular edge-case circumstances, but wherever the ground has to project shadows. It's a fundamental issue with how those shadows are rendered relative to the ground generation tech regardless of detail settings, so as I said it's not a matter of opinions or anecdotal experience. If you never (or very rarely) saw this, it's either not happening on Linux, or you are not remembering it. I'm not attacking you mind, so sorry if I come a bit harsh on the subject, but on the contrary I'd find really interesting if you could in fact confirm that the issue is not present, or in a much more limited way, in the game running on Linux.
It is not regardless of detail settings because part of detail settings is shadow map resolution. I do see jagged shadows sometimes (like just now with the toast rack at Crippen Port (sun at a "good" angle)), and jagged shadows will result from the shadow map having too low a resolution with poor interpolation, but I have not seen anything like those dancing shadows shown in the videos.
But this matters only in that it is somehow a "doesn't work for me" vs "works for me" thing, and there can be many reasons for it, including bad config files, use of uninitialized memory, corruption, ... And I know all too well about "works for me" being a dev myself and having to fix far too many bugs that worked for me but not the people using my code.
In addition, the jaggies on low-res shadow maps will craw as the view shifts, but I would not expect the results shown in that planetary approach unless the shadow map was of ridiculously low resolution.
I was really pointing out how this is one of those few cases where it's not a "works for me" vs "doesn't work for me", it doesn't work, end of. That's why there's plenty of proof and the issue went at voting stage. And it's not an issue with shadow maps resolution, the clipping and flickering have nothing to do with that. The jagged lines are because of clipping, not resolution. Low res shadow maps will always have blurred edges, regardless of how blocky they are due to resolution, that's indeed what you were seeing at that toast rack, the issue reported in this thread is totally unrelated to that and only has to do with how the proc-gen ground projects shadows on itself.
Well, obviously you won't believe me (don't blame you really, as I can't provide proof), but I do not see those dancing shadows in the report. Thus my surprise at seeing them in the video. That (many) others do see them is even more obvious.
I really don't get why you're so upset that somebody who has done a lot of playing hasn't encountered them, it's not like it diminishes the importance of fixing the issue, whether it's bad shadow maps or bad geometry (which very much can result from detail settings (but with the settings having buggy effects)).
I really don't get why you're so upset that somebody who has done a lot of playing hasn't encountered them, it's not like it diminishes the importance of fixing the issue, whether it's bad shadow maps or bad geometry (which very much can result from detail settings (but with the settings having buggy effects)).
But I'm not upset. And it's not a matter of "believing" someone or not. It's actually three replies I'm evaluating the possibility it might not happen in the Linux version and why, all the while pointing out it definitely happens on pc outside of any anecdotal experience.
(Also why it's so hard to take any message exchange in this god forgotten place as anything less than a last stand at noon, I'll never understand the mental stance
)
(Also why it's so hard to take any message exchange in this god forgotten place as anything less than a last stand at noon, I'll never understand the mental stance
)Well, that's just it, that's what I want to know, too: why do I NOT see it (assuming I actually don't, but I'm pretty sure I would have noticed, as I DO notice the other lighting bugs all too well). I even did a planetary approach to check. While I didn't see the problem in question, I certainly did see a shift in shadow resolution as I got close to the surface (a few t0s of km, think I hadn't entered glide yet). but it was rock-solid before and after the shift. Of course, the terrain wasn't all that similar to what's in that video, so it doesn't really mean much.But I'm not upset. And it's not a matter of "believing" someone or not. It's actually three replies I'm evaluating the possibility it might not happen in the Linux version and why, all the while pointing out it definitely happens on pc outside of any anecdotal experience.
(Also why it's so hard to take any message exchange in this god forgotten place as anything less than a last stand at noon, I'll never understand the mental stance
As for the last stand at noon... I guess there's been so much of it going around I get gun-shy
Anyway, as I mentioned earlier, I am now keeping an eye out for these things, as I can't rule out being too busy paying attention to other things (like blue "maybe here" maps (and I just realized I've done the vast majority of my planetary approaches in analysis mode... to mapped planets, so maybe the issue was just hidden)).