That’s a possibility, indeed, and not a bad thought. The splash would still seem unresponsive during that wait, but it would be better than nothing, I do think.
Well, I have some options to think about, at least (and I’m still open to more, if any have them)–thank you all! 
In fact, there are such problems in other games. The problem is that the 3D camera’s culling objects does not work during loading. And it is the first friezes - this is something other than the work of culling objects that suddenly appeared in front of the camera.
I don’t know if it is possible to recalculate visible objects in Panda on demand.
Also, do not forget about optimization tools, lod, occluder culling and portal culling.
That’s a good point: one of the sources of this initial lag is, I suspect, the processing of those first few frames, with just about every texture and model new to the graphics memory and cache.
I don’t have LOD active (well, I sort of have it implemented, but not significantly used in the current levels–not “proper” LOD, at least), but I very much have portal culling and, in the case of the third level, grid-based culling. No occluder culling, in all fairness–in my experiments, it didn’t help, as I recall. Still, I think that I have support for it if called for.
A significant update to the Redux Demo has been posted!
The link is the same as before, but for the sake of convenience let me re-post it:
This update includes a number of changes, some of the most-salient of which are shown in this short video:
Too long, didn’t watch:
I hope that you enjoy this new version, and let me know what you think! 
Found a bug in the latest release. When playing in fullscreen in 1920x1080 resoultion (my display native resolution), all objects start to flicker. Here’s the video:
Yup, that’s one that I’ve had reported in the past–it should hopefully be fixed in the next version!
Thank you very much for the report!
A new demo has just been released: the “Depictions” Demo!
The first post has been updated, but for the sake of convenience, here’s the link to the demo, and the new trailer:
This version, version 0.9.0, includes a variety of changes:
Hello, @Thaumaturge. I am happy to see that you’ve added new content, since I am a fan.
I am especially excited about these new small locations.
Thank you very much, on all counts–that’s heartening to read! I hope that you enjoy the new demo, should you give it a try.
The library puzzle is too MASSIVE… I understand what keywords I need to search, but there’s a lot of bookshelves.
How far have you gotten, if I may ask? Have you reached the point at which the alphabetical ranges of the shelves are shown? (i.e. the point at which a given shelf might be labelled “So-and-so to That-other-guy”.)
[edit] I just noticed that I hadn’t changed the thread-title; I’ve updated it now!
I mean the second floor. The first one is easy. The second part is really huge to complete it in any reasonable amount of time. I haven’t found a single book in this part yet. Though the first clue seems pretty obvious - search for “magic workers”, “magic”, “labourers”, etc. Am I correct? Sorry, the forum engine has no ability to set spoiler text, as far as I know.
Aaaah, I see–I should have guessed that it was the second floor that was the problem. Sorry about that! ^^;
Let’s see… It looks like on this forum we can sort-of make spoiler-tags via the “Hide Details” option–click on the cog on the right, then select “Hide Details”.
You’re slightly off in your objective here: these are histories, and you’re looking for a point in time–the point at which the people of Ketel became refugees.
Thus you’re looking for any documents that reference Ketel, specifically.
Still it is a huge amount of shelves, I think it’s overkill to have 3 time periods in one shelve. But, of course, this is your choice, if you want to make your game really hardcore.
Also, notice the frame drop - it happens after a quick load. But I must admit, it almost always gets back to 60, except when you quck load from the battle.
If you die in the the battle, and then quick load to the time before battle, it causes a memory leak, and after N reloads game stops working at all.
It is a lot of shelves–but it’s possible to significantly reduce the number that you have to visit.
As to having quite so many, I mainly want to disincentivise players from brute-forcing their way through it, just starting at one end and searching every single shelf. The hope is that this will prompt players to think about whether there isn’t a less-tiresome way of finding the relevant location.
That said, please forgive me if I’m being or seeming at all dismissive–I’m not rejecting the possibility that there’s a problem with this puzzle!
It’s very reachable, indeed.
Remember: traversal is a big part of this game, and so it’s quite liberal in what you can climb or stand on!
Hmm… A temporary frame-drop after a quick-load isn’t too surprising–after all, the game has just loaded in a bunch of stuff. Still, it may be worth looking into!
Thank you for reporting it. 
Ooh, that’s bad! Thank you for letting me know–I’ve made a note to look into this! ^^;
But since the first clue is just “it happened earlier”, and books on the first level don’t tell any dates, it can be only brute-forced, isn’t this correct?
Not at all.
In short, there’s a method that can be used to drastically reduce the number of shelves that you’ll likely visit before discovering the target-point. In a sense, part of the puzzle is hitting on this method.
They are unrealistic–but that’s intentional: the game is intended to be somewhat stylised!
(Unless there’s a bug there that I’m not seeing?)
I just posted a minor bug-fix-and-polish update. Specifically, this new version–version 0.9.2–includes the following changes:
The link remains as before, but here it is again for the sake of convenience:
