xtom

Members
  • Content count

    345
  • Joined

  • Last visited

  • Days Won

    2

xtom last won the day on January 1 2016

xtom had the most liked content!

Community Reputation

117 Excellent

About xtom

  • Rank
    Advanced Member
  • Birthday 08/10/1977

Profile Information

  • Gender
    Male
  • Location
    Ireland
  • Interests
    Gaming, Game Dev, Design

Recent Profile Visitors

3,100 profile views
  1. I think 4096 is plenty big imo. You can always load different areas/maps to expand bigger and keep dev easier than trying to custom code something to handle a massive fps terrain + all it's content. Recently played Vampire the masquerade which loads areas in city, yes it's an old example but the gameplay is no lesser because of the map loading except maybe the loading time. I think 4096 is much bigger than what that has game too. But if you want to compare to latest AAA openworlds you would need a comparable dev team aswell just to come close. I guess if you are making all procedural game it's a different story but in general 4096 is plenty big for an indie game imo.
  2. With lots of mechanics might need to introduce them gradually over multiple levels. First level(s) should be easy, not too hard on the difficulty curve/frustration. Maybe checkpoints. I like the way that video has red barriers on some of the ends making the level not too difficult. What about a putty ball that splits in half if a saw hits it, or powerups that change from soft to hard ball to beat certain enemies or puzzles. So much you could do.
  3. Maybe a turnstile/roundabout where you have to time your exit carefully. And some rubber walls that bounce you more. Didn't realize so many things could be added to a marble game.
  4. Sounds good, maybe have a jump/bounce to hop over gaps and obstacles. And teleports to other parts of level.
  5. That's cool what modelling program do you use?
  6. I would prefer a collapse all option as it would make for less scrolling and easier to browse the scene tree in general. A right click move to filter option within the scene tree itself would be more intuitive and less disjointed to having it as a properties selection but I don't know what is possible and what isn't.
  7. I use the filters, they are a very useful addition but still even when there are lots of objects under custom filters the suggestions mentioned here would also be very useful to have. Anything that speeds up map making/organising is good.
  8. Thanks, next update will be out soon! The houses and bungalows are already converted from csg to mdl in my dev version of the game. I did my best to remove hidden faces in blender. The most complex is the house model with interior walls and floors which turned out 907polys. The non-entry buildings are about 350poly. So in the game a house is now one mesh/entity + any windows/doors/furniture and the benefit I hope is being able to repeat more times with lesser impact on performance, plus they can be hidden out of view beyond the fog using the view range setting. If it was just one or two of the houses then I wouldn't bother converting from csg but for 10,20 it cuts down on a lot of separate csg pieces on the map.
  9. Whoops this is my fault, feel free to close. Name on my player prefab was wrong.
  10. I think latest beta files may have broken trigger scripts from working.
  11. Yeah it's a pain moving things up into different filters when there's a lot of stuff. Even a collapse all would be a nice addition but all the other suggestions would be very nice too. A big +1 for scene tree improvements.
  12. One More Day - Performance Updates So it’s been a while since my last blog update on One More Day but recently I had a look to see if I could do anything with improving the performance of the game. As it turns out there were still some optimisations to be made that have helped boost performance. V0.1.6 on the Game Launcher Here are the main performance related changes I’ve made which were recently published to the Game Launcher: I changed some of the furniture models to lower poly versions. Some of these models were about 3000polys and are now replaced with lower poly versions with no real visual difference in-game. So about 10 or so models that were repeated more than once around the level were shaved in poly detail by about a third. I halved the draw distance on camera range and increased the fog a bit to compensate. I believe this helped more for when you are some distance outside the town but less so when you are in the town. I removed some rogue geometry in the map (wasn’t much), and also a few objects that weren’t really needed. I went back over every object to ensure it’s viewing range was as minimal as I could afford without creating too much pop-up. This is something I have always done but still sometimes you end up missing a few things. I spaced out some detailed areas on the map (ie. the power station got moved out further) to try to reduce the number of objects being shown at any one time. I made sure physics shapes set in model viewer were as simple as possible and removed collision from some small objects. I removed some workshop objects that had very big textures. All these changes stacked on to the latest version of Leadwerks which also includes performance improvements have given a noticeable performance improvement to the game. One thing I had to add in (sparingly) was some grass using the vegetation system as I felt it added nice detail to the level and was worth a small performance hit. That’s what makes optimising the game hard, trying to balance between good visuals, content and good performance. Unfortunately sometimes sacrifices have to be made. But even with the grass added the new version is running better than before. I would like to add volumetric spot lights for light coming in the windows but for now I resist that temptation as they would add up to a significant performance hit. Maybe I will add them as a high graphics option in the future. Other Improvements As well as performance improvements the game download size is now smaller thanks to the newly added ogg support and also removal of some unused files that were being published. One thing I noticed when publishing a game in Leadwerks is that if you have files with the same filename in two different folders both files will be packaged even if only one of them is used. So there were some unneeded texture files etc. being packaged from unused workshop content. Going forward I will have to keep an eye out for duplicate filenames. More performance in the works The following changes are in my current dev build which hasn’t been released yet but are showing even more performance increases: Converted the CSG houses and bungalows to models. Simplifying and reducing object/entity count. This should allow me to add more houses to the map with less of a performance impact. Further reduced the camera range and increased the fog to compensate. This allows me to reduce the view range on many bigger outdoor objects and buildings from Max to Far whilst minimizing popup and also reduce the vegetation view range. This will result in a much more foggy look to the game (see screenshot) which will create a slightly different atmosphere than previous versions and with less viewing distance it might make approaching zombies a bit more dangerous. I identified SSAO was making quite a performance impact so I have disabled this for now and might re-introduce it as a high graphics option when I get around to creating an options menu. For outdoor map SSAO is probably not as important or worth the hit but it is nice to have when inside the buildings. After all these performance improvements and v1.7 is released I am optimistic that most people should get at least a 30fps game experience with OMD. Fingers crossed!
  13. To get the most from social media you need a constant active and personal presence on them. More so than auto feeds or auto posts. Very time consuming which is why bigger companies have community managers or a social media person. I still think they are worthwhile though even if updates/interactions are low. I like/follow all the games and game devs I'm interested in and it's handy to get their latest updates pushed to me if I don't frequent their website/forum. I would say at the least they are good for free advertising of new releases, special offers and dev posts/blogs. But yeah it can be a time sink with small returns.
  14. Really like the chm format/layout. Very fast to browse and find what you are looking for.
  15. Yeah, good entries. I've tried a few already and downloading some more to try. Lot's of gameplay variation which is cool.