UT99.org | Unreal Tournament Community

CloudZone destroys everything out of pawns, including decals, etc. While these actors quickly goes in bDeleteMe, accessing them in a moment will deliver a crash... especially OFF-Line... As explained already CloudZone is not for CTF mapping as AlternatePath is not for other games - and here is nothing like that even if that Actor is aiming CTFGame. In a recent AMC map when I did some fixes I had to quit using said CloudZone because it was causing only troubles and I replaced it with a normal zone but having setup for damaging - in a CloudZone there is nothing to do in boolean variables regarding to damage - code is executed by zone without anything else in account.

FraGnBraG wrote: ↑Tue Jan 07, 2020 5:21 pmI recall cloudzones were not recommended for other reasons I think they would prevent spectators or something like that.

Yes, even spectators will be destroyed!

Beautiful map. I am simply jealous of your mapping skills. I wish I could make big maps like this!

Thank you all, for the compliments and insights.

As for the CloudZone, I went with it on purpose, since I really wanted to "destroy" anything that entered the zone, and not exactly "damage" to death, namely any projectiles that entered it so they would look like they went away to oblivion and never hit anything, which is the main purpose of a CloudZone in the first place.

Of course, I am aware that it will destroy pretty much anything that enters it, that's why I extended the CloudZone into a MyCloudZone and MyLevel'd it with custom code to ignore my own actors for instance (otherwise the sun and other effects wouldn't work, since they were always getting destroyed).

I did forget about the flag dropping there though, or rather I incorrectly assumed that the home would just spawn a flag actor and that, if that was destroyed, it would spawn another one.
It's been many years since I looked into something like that, and I forgot that the flag home and the flag itself are the same.
This is something that will be fixed in the next release (this was an alpha after all).

Que wrote: ↑Tue Jan 07, 2020 4:10 am The problem here is that UCC freezes and even after leaving it running "Compressing" for hours on end it does not seem to compress these files.

I just tried it out myself, and I am also having issues compressing it.
It passes stage 0 (in about 3-4 seconds), but then it gets "stuck".
Although it's doing things (using a core thread fully, and using 110MB of RAM), but "ucc compress" is likely to be so inefficient that it probably struggles to compress a large map like this.

So I would advise to not bother too much about it, or perhaps to wait for what maybe Higor has to say, since he's the one who looked into that sort of thing the most iirc.
Furthermore, if the compression ends up taking hours or a day to complete, I wonder how much time it will take to decompress once downloaded...

Also thought and wandered about the decompression of the map.. I wonder if this problem happens with CTF-FacePalm..

Barbie wrote: ↑Tue Jan 07, 2020 9:34 amA typical problem of a CloudZone: it destroys everything that enters it

I can attest to that...

...CloudZone killed me too, and I was in spectator mode!

Thank's alot.

Maybe changing this map into a DOM version would be interesting.


(on DOM there is no flag drop)

Hey Ferali, have you fixed the death clouds of greedy flag keeping yet?

Finally got around to playing this thing. The outdoor looks stunning. Nicely done. The interiors look nice, too, for the most part. Don't really like the yellow pipes beneath the floors but that's just preference.
I've had no real issues playing the full version, got around 45fps average with noticeable dips in the outdoor area, especially when quick turning. Haven't checked for polycounts but I think the slowdown is caused by emitters and effects rather than pure polycounts.
Gameplaywise, 16 players crams the map full. It's still playable, but feels more on the campy side. Lower bridge is hitscan heaven, although dropping down from above is a good way to get some line of sight cover if things heat up above. Haven't had any issues finding a gun, and bots gave me a challenge (haven't played for months, though, so take that for what you will ). I did have issues with some of the lights, though. UT is notorious for flares to shine through character models, and all the flare effects in base seriously messed with my aim, or even my ability to note where the enemy is. It looks nice, but for me it's also a gameplay issue.

Regarding zones - didn't Lava or Slimezones also reset the flag? I any case, I have been using a custom zone made by Wolf called NuclearReactor (he did write it for DM-FoT-Styx). I think he used the CloudZone as a base, but I've used it in Selket and never had anyone complaining about it.

Yep, it's up there! ^w^

Thank you very much for your review Swanky!

CloudZones destroy everything, but I used my own in the map (a direct extension of CloudZone), so I just need to tweak it a little bit to not destroy flags and some other stuff.

EvilGrins wrote: ↑Fri Feb 21, 2020 3:40 am Hey Ferali, have you fixed the death clouds of greedy flag keeping yet?

No.
I haven't been active at all in UnrealEd lately, and currently I am using this alpha as a test example for something else, and I would like to only release a beta after that.

According to STATS GLOBAL there exists 3 FPS eaters:
1. Actors.
2. Occluding AKA static light.
3. And render (which usually include occluding as I understand).
So for better FPS need reduce actors count and use less static lights.

Main part is actors. So better focus on this if you want improve FPS of course.

Map looks incredible. I'll be trying it out this weekend.
What FNB said about painzone is what I remember too.

ss of one of my ctf maps that returns flag.

Thank you.

I just want to remind everyone however that this map is meant mostly as a benchmark at this point in order to solve bugs and improve the engine performance in future patches.
So it's not really "playable", mostly due to the poor performance, and also since some of the visuals get in the way of proper gameplay.

I do intend to eventually release a beta with some performance improvements and the cloud zone fix, but only after 469 is publicly released, that's why I never actually ended up doing it just yet.

This map is also a benchmark for myself, in order for me to see where the real limitations in map making actually are, in case I end up doing a more serious map for actual gameplay (I have a particular map in mind, not sure if I am actually going to develop it though, especially since it only makes sense with the development of a specific new mod as well).

That is to say, that I am perfectly capable of creating a map with a similar level of visuals, and make it actually playable lag-free, but it would just take a lot more time as I would be required to create a more serious and optimized package for all the effects (which I already thought about), and to create some textures/meshes to replace some of the minor BSP details (to also optimize the BSP).

So I do really appreciate all the advice on performance improvement and fixes, but this is a map which has been built like this on purpose, to push the engine itself further so that both me and the UT patch devs have a good baseline to compare with and improve upon.