Posts by spence

Updated the map on the workshop today with some small gameplay fixes (easier for players to navigate inside the airplane if a hostage gets dropped, fixed a lot of nav mesh issues, etc). Also bumped up the brightness of the environment/ambient and basement lights even more since some players thought it was still a wee bit dark.

Quote from mmortall

Very great and nice work. Now many time does it takes?

You're asking how much time it took to create? About 4-5 months in the spare time I had from studies. I was pushing myself to become proficient with some new tools (dDo) and learn some new things like animating props so there was some extra time spent on these aspects whereas the layout, prototyping and design phases took significantly less time

Quote from Beck

• Are you planning any advanced routes through your level?
  
  • For example, you can dodge off of a wall to get to a higher location.
  • Some sloped walls can be "surfed" when dodged into, any plans for these? They could be quite fun when moving down in the level.
  • There's gameplay videos of the above two points on youTube, I had a quick search but couldn't find them. Might be worth having a little dig for them.

These are some good points for people who aren't experts with UT (I didn't know about the sloped walls thing for instance). Thanks!

I too will be starting a level once I sync up the UT Game...later this week when new PC parts arrive I think (trying to move around the viewports at ~10fps is a little rough)

I've only been doing game development professionally for a couple of years, and only with independent companies. They've certainly been interesting and fulfilling experiences which have reinforced my desire to continue building worlds and environments for games. I decided to go back to university after losing my job last year, though, and I don't regret it. Programming is something I've never had much experience with (I'm one of those designer-types), so I chose to undertake a computer science degree. I don't think I would have been unable to learn the same programming concepts on my own, but the way the teaching is structured makes it easier for me to have a better understanding of these concepts more quickly, and having a schedule set by someone else helps to keep me focused (so I don't start shifting my goals around because there was a delay in completing one, for instance).

I don't know if I would have had the same level of comprehension if I was trying to teach myself programming concepts while also working on design or art for a game. Since I already have some professional experience with the latter, I'm working on building up these skills in my spare time away from studies, and when it's all done I'm hoping I'll have a better understanding of art and design, plus some new knowledge that can be applied to making games. I don't think I would have been able to learn as much if I was employed full-time, at least not in the same time frame, so this way of thinking always keeps me motivated through the times when I wish I could be working again with a really solid team on something exciting

I do think there is immense value to what you accomplish outside of work (be it through a hobby, education, etc) even if it's not readily apparent. Sometimes this can be more valuable than the work itself (hmm, maybe the employers won't like this one, but I think in a lot of cases it's true). Before I started working on games professionally I was studying and training to become a commercial pilot. The discipline of planning and executing a successful flight has it's own relevancy to games development (like doing things correctly the first time, even if they are time-consuming), even though the specific procedures and skills are of little use unless I'm working on flight simulators or something.

Hey, that's another reason I like making games - when you make a mistake, there is significantly less risk of physical injury or death

Ah right, I didn't know that things like coordinates count towards this limit but I suppose it makes sense because it is just a map of sorts in the end. Honestly I think that trickery like this will always be inherent to some degree (hell, you could probably pack two normal maps into one RGBA and derive their blue channels in the shader to free up some samples - there is a DeriveNormalZ node in UDK and UE4 for these kind of instances or you can do it with some math too) but I agree that it is a kind of shitty solution. I think you are right about why they use meshes for a lot of terrain stuff because in the public demos I've seen and picked at for UDK (the castle level for instance) the landscape implementation has always been super basic terrain layers with the more complicated bits on static meshes/multiple materials, but I guess the real proof will come from analysing the new landscape demo.

I was curious why there is a hard limit and dug around a little and it seems like this is a limit not set by UE but by shader model 3.0...

Is that assuming that you are going to have separate maps for everything (ie each layer has a separate diffuse, normal, spec/gloss, etc) since that would be 3-4 texture samples per layer?

You can do things to reduce samples like packing b+w masks into alphas or single channels of a single texture then using only that channel as the mask (in combination with component mask nodes). If you are using a lot of grayscale/b+w maps/masks you can cut your samples down easily this way. You can derive your spec masks from a base texture or one channel of a normal map and then use math nodes to adjust/tweak the map at will, saving you extra samples. I don't know about referencing other material solutions as a parameter though. Someone with more experience with UE might know. If I remember there is a node where you can paste in raw HLSL shader code generated from the material editor so maybe this would be useful.

The nodegraphs get quite messy I find in my (limited) experience but you can use comment boxes/nodes to help with this. When you move the comments around the nodes move as well.

Can't wait to check this out when I get my house/internet connection sorted out. Looks great from the screenshots!

Thanks for sharing, it's always good to have some useful Youtube tutorials referenced because a lot of them tend to be overly lengthy/unhelpful and it sucks wading through loads of them trying to find the good ones some of the basics like material instancing are probably equally useful for people picking up UDK as well.

Nice, this is looking great so far!

Those are some really nice prizes for the winner!

As someone who worked on this project as a designer through the early stages, I have one piece of advice to anyone entering: go wild with the theme of your entry!

As a small team we were relatively limited in the range of environments that could be created, but as a lone designer (or a small team focusing on one map), you have no such restrictions aside from the time limit!

Love the site marking at A, that's brilliant!

Haha, you're totally right! I can't unsee it either!

I doubt I'll change it though, to be honest. I had a lot of issues with the CSM shadows causing really awful artifacts depending on angle, especially on the plane's fuselage, so I did loads of tests tweaking light_environment angles by 1-2 degrees each compile until the artifacts (mostly) disappeared. I don't know if I want to go down that road again

Although looking at it in-game it's starting to annoy me now, maybe I will sort it out for the next version

Many thanks, this is fantastic feedback!

I pushed an update to workshop this afternoon to resolve some of the clipping issues that text_fish mentioned (the scaffolding, around the plane doors, and some of the open doorways - some of the passages are still a bit narrow but at least noone should get stuck now) and also boosted the brightness in the terminal area and the basement/tunnels which should make it easier to see players in these areas.

You're absolutely spot-on about some of these visibility optimization issues. This was a real struggle because of the wide open nature of the ramp/tarmac areas, but your review has given me some ideas of how to improve this. The key has been to balance these optimizations with vvis compile time - right now it's around a comfortable seven minutes (even with all the sloppy leaves inside of the plane) whereas some optimizations I tried before release caused the time to jump to ~45 minutes which was not only unacceptable, but caused little reduction in leaf rendering or any notable framerate gain. It's been difficult to assess performance on my own (aside from basically testing what is being rendered and what isn't) since I've been building and testing on a desktop replacement that's nearly five years old, so I'm really thankful for your input here

I wanted to go (and also to gamescom) but even the student pass is a bit pricey for my likes. Maybe by next year I'll have saved up enough to afford it

Cheers!

Please do share any thoughts you have regarding gameplay; it wasn't designed for competitive play (and really was just designed as a map to build art skills around a quick layout) but I'm of course open to changes that will make it more accessible and fun to players.

Thanks! I'm going to lighten some areas up in the next build I think, particularly the basement. I'll sort out the police tape as well..I had to replace the ropes with displacements at the last minute because the rope lighting kept going almost entirely dark on a full final compile (and only when I built the entire map...)

Are there any areas in particular that were frustrating to maneuver through? Maybe some of the single doorframes need to be double doors to accommodate easier movement of players?

I'm sure I missed clipping something but I tried to make sure I got all the doorframes and other fiddly bits that were annoying to move around. The front doors of the plane were the big ones I noticed to fix for the next build and I'll go through and clip all the scaffolding I missed but if there was anywhere else that stood out to you and you can post a screenshot or setpos coordinates (or have a description of where), I would be extremely thankful!

Just released the airport map I've been working on for the past few months on CSGO Workshop:

Workshop Link

If you check it out and have any feedback about how it looks or plays or find exploits or anything weird, I'd love to know about it, so I can improve the map and release some future updates! There is still some polish to do but I wanted to release to the public before worrying too much about really fine tweaks since they don't impact on gameplay so much.

Quote from Evert

Haha cool stuff

Where can I find info about this masking? I guess somewhere on the wiki, but I couldn't find it.

There is a little information here: https://developer.valvesoftware.com/wiki/$blendtintbybasealpha

I think the best practical example however is the L4D2 vehicle sheet. The tint has a multiplicative effect so the diffuse has to be appropriate for this or else things start to look weird.

Quote from cashed

Haven't tried the tint yet, but I don't know if source lets you with a diffuse. Right now I have 2 models with skingroups on them.

You can indeed, static props have a Color RGB field You can also mask where the tint is applied.

Great idea for outside of the play area.

fuck yeah, lego

Yeah I have to agree with this, it's engaging when both teams are working together and playing the objectives but it gets frustrating if most of the team has their heads elsewhere. I've had some fun moments hanging out on the roofs doing my own thing sniping and ziplining around, but it seems like this isn't contributing to the general effort as much as having my feet on the ground with everyone else, because it's too easy for someone to gtfo with the package on a bike or something and therefore the moments in which I have the most fun are contributing to the frustration of other players or to our team's loss

The map seems really small for a Battlefield game (there is reasoning behind this, sure, but still) and I'm not entirely sold on the gunplay either. I skipped out on BF4 so I can't comment there but in BF3 I felt the combat was a little more thoughtful and at least made me consider the decisions I made as I moved through the environments. Here I feel comfortable just sprinting and jumping all over the place and being careless with my ammo. It usually works out in my favour, but it doesn't feel as satisfying.