Author Topic: IndieGoGo campaign ended and future  (Read 152 times)

Offline Rotacak on: May 26, 2018, 02:54:25 PM

  • Administrator
  • Hero
  • *
  • Posts: 693
  • Likes: 14
  • Gender: Male
    • View Profile
    • Awards
Hello,

first I want to thank to all backers. There was only few backers but with high pledges. I never expected that anyone will choose these high pledges. That was big surprise for me. You did really something for this game. Also most of you will get VIP status forever, which is not implemented yet in game but it will be impossible to get it anytime later.

Campaign duration was 2 months. During that time we fixed lot of bugs and improved some things. But most important bug still remains - server crash bug inside Unreal Engine. There is no point to play this game when server crashing so often. This is not common deathmatch game where "nothing" happens, so every crash will demotivate players. So instead of previous plans (create admin system, add support for custom maps, ambush mod...) was developing mainly targetted to fixing crashes. But that one crash cannot be fixed. So I decided to upgrade Murnatan to most recent version Unreal Engine and try it if it will still crash.
Sadly - I cannot test it for that crash. Game never crashed when you played it alone. And crashed totally randomly with more players. Sometime after few minutes of match and sometime after hour. So only solution how to test it is release new version in public and test it there.

Current state:
Currently I trying to update game from Unreal Engine 4.15 to 4.18. Most recent version is 4.19 but there is totally reworked skeletal mesh support and that is currently impossible to change. Game is already upgraded sucessfully to 4.18 and in dev mode works "ok" (few things doing problems but it is playable). But packaged shipping version for distribution is of course fucked up and every new packaging will take 6 hours (or two/more days if it is first packaging).

Future:
Now I trying to build some working version with UE 4.18. Then I will release it.

Next, if it will still crash with same bug then I will try to update it to 4.19. But if that skeleton component will affect some core mechanics then I will have to stay in 4.18. And don't know how to solve that bug. Maybe code some bots to simulate players and try to finally replicate crashing on dev machine, then fix that bug or rather change that what causing it. Or completely redone whole game in recent UE version :o

Next, I would like to implement (not in this order):
- dumb bots (basic walking and attacking)
- admin system (log in, vip, bans)
- support for custom maps
- ambush mod

But this implementation will take time. I need also start my life. Make some money, buy something for self, maybe find woman, make children, divorce etc. Don't know yet.

But good thing is that core of this game is good and this development is not dead. Only decreased mainly to one developer (me) and slowed down.

It is sad because if there could be 6 months of developing more then everything could be different. And with about another 1-2 years could be released final version. I tried to get money from investors, from game companies, from Unreal grant, from Europe donations, from various sources but nobody care. And I really do not want to spending time with talking, writing, planning all these bullshits for investors and for nothing. Oh man, how much time I already spent like that.

Well, Murnatan looks like a whole life project :)

Offline Loki #1 on: May 26, 2018, 04:16:34 PM

  • Hero
  • *****
  • Posts: 705
  • Likes: 3
  • Gender: Male
  • Awards
    2
    4
    • View Profile
    • Awards
Thank you for your and your colleagues' work. I wish you luck in hunting that reason for a server crash. I've played Murnatan again today, there was just one player playing but I jumped in nonetheless to test the aliens. They're much better now (though not perfect) so it's more playable for aliens.

Your goals are rational and important, but one thing that really bugs me is that the wallwalking is laggy. Is that fixable with some simpler fixes? It's smooth most of the times but sometimes there's that moment when I get choppy with wallwalking. Also, wallwalking seems to be faster than normal walking, unsure if that's intended.

It's sad that it all turned out the way it did and that it couldn't have been finished, as optimization and other, more extensive gameplay tweaks would make the game a perfect Q3>UE4 'port'

Thanks a lot for the game, I wish you luck in life and in developing this game further!

Offline Rotacak #2 on: May 26, 2018, 05:43:51 PM

  • Administrator
  • Hero
  • *
  • Posts: 693
  • Likes: 14
  • Gender: Male
    • View Profile
    • Awards
Loki: Wallwalking is buggy in Tension because that map is constructed only by brushes, not objects. And that causing some problems. But it is not perfect also in other maps. It is complicated thing, not easilly fixable. But one day it will be solved too. And it should not be faster that walking. I will test it.

Back in 2012 in Unreal Engine 3 there was spider physics and that was perfect. But nothing like that is in UE4 :( And my question "can you release sources of spider physics?" was ignored.

Offline Machine Medic #3 on: May 26, 2018, 08:18:58 PM

  • Hero
  • *****
  • Posts: 643
  • Likes: 12
  • Gender: Male
  • Been there. Done that.
    • View Profile
    • Awards
The only other suggestion I can add at this point is to look at possibly going open source and/or accepting volunteer help on an NDA/closed source basis.  There are players in the GrangerHub community who seem to be capable coders...

...even if they don't exactly possess a unified sense of direction.  Or foresight in general.  Most of their 'solutions' are in need of a problem.

Code: [Select]
"Hey, why not give all structures 3x health and delete the repair feature?  Humans camp too much!"
"Sure, that sounds like a great idea!  While we're at it, let's give the humans portal guns and grenade launchers... because the aliens camp too much?"
"Brilliant, but grenade launcher?  We're going to have to give the tyrant 1200hp then...  Let's do it!"
"Hey, want to give all humans ckits as sidearms and let the aliens devolve into grangers, so that when either team's spawns are dead they can just switch to ckit and rebuild somewhere else?"
"Yeah, that'll make forward bases easier, woo!  But won't that mean the game will never end?"
"Yeah, but we can just make it so build points aren't recovered when structures die."

¯\_(ツ)_/¯
« Last Edit: May 26, 2018, 09:10:42 PM by Machine Medic »
2x 980ti @1.3GHz, 5820k @ 4.4GHz, 32GB DDR4 @ 2.4GHz, SSD 950 512GB & 850 1024GB.

=\Co-Leader of Contra Clan/=

Offline Rotacak #4 on: May 26, 2018, 10:34:27 PM

  • Administrator
  • Hero
  • *
  • Posts: 693
  • Likes: 14
  • Gender: Male
    • View Profile
    • Awards
Machine Medic: That could be also maybe somewhat possible but I doubt that anyone will want to dig it into code and whole Unreal Engine system. I think that in Q3 can anyone do many things but when you do not know Unreal Engine, then you will be totally lost, even if you know c++ well. And also synchronizing all source changes would be impossible.

Offline ViruS #5 on: July 03, 2018, 05:49:16 PM

  • Member
  • ***
  • Posts: 232
  • Likes: 1
    • View Profile
    • ZdrytchX's reference website, the tremulous-related link heaven.
    • Awards
The raw surface of the gamelogic should be sufficient to help smash out basic physics and related bugs without a need for leaking potential security and the alike. Synchronising source changes isn't needed with the availability of patch submissions however it does mean that the user's end must always be updated and patches needed to be tested for potential clashes if the user's end isn't updated or if an update rolled between the user finishing their work and a new update that is related to the section of the gamelogic (e.g. an update may rework the routing order of the marauder zap, user fixes the way the marauder zap hit detection system)

Speaking of which, we need at least a publically announced regular dev game going to help boost the population as it is.

Regarding bot AI, I can provide some advanced concepts (that no other game from what I know has yet to combine) to the aiming system so they won't be so aimbotty but it will require the use of the latency compensation system similar to quake's unlagged or a similar memory system to feed off. I have not seen how the UE3/UE4 system works so it may or may not be easily applicable.
Quote from: lamefun
Therefore, Tremulous 2 might have the future (or might not, since it isn't developed by a corporations, depends on what Global Government will decide), but Unvarnished doesn't.
Quote from: Khaoz
Sucks to be those Unvarnished guys.

Offline Rotacak #6 on: July 04, 2018, 01:50:55 PM

  • Administrator
  • Hero
  • *
  • Posts: 693
  • Likes: 14
  • Gender: Male
    • View Profile
    • Awards
About synchronizing code: it is not so easy like in Q3 and such engines. In UE is souce code connected with assets more closely. For example when you want to change default HP of player, then you have to do it in UE editor, not in C++ code. I think that for working simultaneously on more versions is possible, but merging it together is almost impossible.

About AI: UE have own AI system and should be quite easy for default game. But I think, no, I know that for this custom game it will be much more work/problems.