Crumbling Hospital WIP
Hey all,
I'm currently working on a hospital environment to slot into my dissertation on horror in games, and the level I'm making is along the lines of the classic Silent Hill "otherworldly decaying hospital" style. Basically everything needs to look heavily weathered, and I was hoping to get some feedback about whether or not I'm going in the right direction with my textures. They don't currently tile, but here are the walls that I've been working on:
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
Any and all criticism is appreciated!
I'm currently working on a hospital environment to slot into my dissertation on horror in games, and the level I'm making is along the lines of the classic Silent Hill "otherworldly decaying hospital" style. Basically everything needs to look heavily weathered, and I was hoping to get some feedback about whether or not I'm going in the right direction with my textures. They don't currently tile, but here are the walls that I've been working on:
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
By saarlemsanders at 2012-02-08
Any and all criticism is appreciated!
Replies
Also, either flatten the blue and white areas to the same plane, or break them up more. Right now it sort of looks like the blue is slightly extruded, but not enough to actually affect the silhouette, so just flatten it down.
To answer SgtNasty, I'll be making the textures tileable in the near future - at the moment these are just the basic wall textures without tiling. Also the blue is slightly extruded - to me it felt too flat when the detail was added with a normal map, but I'll be sure to flatten it down again and put it up on here for comments.
Anyway thanks again guys, I'll post new images when I've made the changes, and I should be adding some new floor textures and objects on here soon.
They're looking great! But yeah definitely learn vertex painting, it's a pretty standard practice these days.
The colour of the exposed wall seems to elicit a different response from everyone I ask - who'd have thought that a blank wall could be controversial? Anyway I'll try and even up the colours of the exposed areas, especially in the bit where the crack spans both halves of the wall.
You could use vertex painting, but just having a decal sheet with these decals is an option as well. Vertex painting is going to be a little more expensive performance wise, and your going to be a little less flexible with how it looks. Your going to be restricted to using tiling textures for vertex painting, and you can use a mask to kinda get the peeled paint look on the transitions, but it might not be as varied as you can get with decals.
As far as what you have goes, the decals on the top half seem to break away to a slightly different colored material underneath, you might want to have that look the same.
It might take a bit of time but it will definitely be worth it in the end!
Vertex painting is technically the same on all engines, but the way of getting there depends on the engine, but just googling "Vertex painting UDK" should get you a grasp of the concept.
Here you go:
http://eat3d.com/free/vertex_painting
Short video packed full of everything you need to know.
Thanks for the useful crit - much appreciated! I've been bogged down in the distinctly less interesting side of this project (the dissertation-writing part) for the past few days, but I should be posting new images by Saturday or Sunday.
@gsokol - I've been working on getting the textures to tile, cos I'm going to be making a few more wall variations and I need to basically texture a series of corridors with these maps. I've been working on sorting out the differing colours as well. I should have images up by the end of the weekend, along with a new texture or two.
@KarlWrang - Thanks for the advice! I can pull all of the paint crack alphas straight from my PSDs, because the majority of the cracks are done using alphas anyway.
@WarrenMarshall - Thanks for the link!
Hopefully I'll be able to show some more image updates soon - juggling four modules means that I can't work on this as much as I'd like to.
I'm a bit late posting this due to my other work, but here's an update of my latest textures:
This should give a proper impression of how the textures will look in my level. I've modified the bits of exposed wall so that the top matches the bottom, but the textures don't tile yet - I was preoccupied with making as much headway as possible on a variety of textures.
Let me know what you think of the new textures, as well as any other issues that you can see with the modified walls. I'll post new stuff as and when it's ready. As usual, any and all crit is welcome!
@Habboi My full dissertation title is currently "How Do Game Artists Create Effective Environments for Horror Games?" Sound is actually an area that's currently lacking in my write-up, but it's obviously really important to the level, so any help would be greatly appreciated. I'd also be really curious to see how you tested your theory, so it'd be great to look at your findings if possible!
By saarlemsanders at 2012-02-22
By saarlemsanders at 2012-02-22
I've made the changes switz suggested on the first floor tile - if people think that looks more suitable then I'll roll it out onto the other tiles as well. I've also spent some time on the ceiling, so any comments about that would be helpful.
Basically I was looking to see how sound effects are used in horror games and how each sound effect type affects the player. It's a strong subject as we all know sound in horror games probably contributes over half of the scare.
So I studied the well known horror games and broke each sound effect into a category and named them so that I could create a playable level that demonstrated each of these sound categories in different scenarios.
I got about 20 people to play my level on a laptop and I had three tests, a video camera recording their faces throughout the whole thing, an optional heart rate monitor (a sports one) which worked quite well but only 15% used it as it had to be on their chests which made some of them uncomfortable and the third test was a questionnaire at the end asking them stuff like which part scared them the most and so on.
I then compared all three tests and made a HUGE excel spreadsheet which helped me see results. To my amazement there was a pattern and it made writing up my results section real easy.
Obviously I can never share the video files and I don't want to share my dissertation in case a google search brings people towards it for stealing but I can give you my excel sheet to show how much work went into it.
Also I'll share the questionnaire and a text analysis I did. There was this website that made any emotional words BIG for analysing.
http://upload.habboi.co.uk/habboiwork.rar
Let me know when you downloaded it, I wish to remove it.
I looked through all the information and it's interesting that, based on what's written, you used quite a few scare effects that are similar to what I was planning - it's good to know that they worked on your test subjects. With any luck I'll be linking my test on here when it's ready.
What you did sounds a lot like the audio version of my dissertation, so knowing how you tested it really helps - testing has been a concern for a while, cos it's hard to tell when people are scared by particular things. I was similarly considering the questionnaire and pulse rate method, but I didn't think of adding the camera, which should help a lot.
Thanks for all the help - if I have the appropriate Harvard citation then I can reference your dissertation in the write-up of my product, cos I need to write about the sounds I'm using. I'll post updates as and when I get them so that you can get an idea of what I find too.
Criticism is welcome.
Anyway I can't really crit it much until I see it in its proper environment with the final lighting but the texture is nice and detailed. Although now that I think about it, there's damage on the wood just above the metal bit at the bottom but the metal is unharmed?
Other than that keep it up.
I've tried to give it appropriate lighting for my scene (a faint yellow glow from a dim fluorescent light fitting) but obviously there'll be some differences once it's in the engine.
I'm going to add scuffs to the metal when I get a chance so it hopefully won't seem as separate from the damage above it.
Thanks for the crit, I'll post more stuff when I have it.
The "Can you see me" with the arrow is a neat idea, but I think it looks a little...flat. it doesn't look like someone painted that with their fingers, or hair or whatever, it sort of looks like the paint can tool in paint.
Also, i'm not sure if you fixed it, but the wall sections don't mesh well together in the screenshot you took, i'm not sure if that's on purpose or not but I just noticed it.
Your texture work is really great! I would be interested in seeing how you painted the chips in the paint on your door. could you explain that? I'd be very interested
Addressing the issues, I agree about the writing. I may try writing on an appropriate material, scanning it in and layering it onto the door that way, so that it looks more like 'real' writing.
The walls aren't tileable yet. Based on my schedule I decided to try and lay down a base of assets for level-building before going back and refining them. Once I've made the light fittings I'll have the bones of the level, then I'll make the walls tile.
For the paint chips I made my blue paint layer and my door base layer, then I applied crack decals to the opacity in the paint layer. I got my original details from the Dirty Texel brush set, then mixed and matched bits of the different paint decals to make what I needed.
The video quality is considerably less HD than it is on my PC - I'll see what I can do about uploading a better video when I get my Camtasia and compression software working again (damn those expired trials).
I got it done for my deadline, which was my final uni deadline, and after all that I decided to be away from the hospital scene for a while. Anyway I wanted to post it here for anyone who was interested, and I also wanted to thank all of you who helped and gave me crit during the process. I got a First in the end, so clearly your advice was a big help! Now, I'll be moving onto other 3D projects. Expect my Tyranid Fleshborer WIP to appear very soon.
Once again, thanks a lot for the help!
I worry that this environment may not have enough visual interest to classify as "horror". There are things like the door that sunk into the wall but I feel that's more of a tech artist than an environment artist (I could be disputed on that), which isn't the subject of the dissertation. On that note, I just watched a two minute video of a player walking through a hallway... and not much else. That's worrysome.
Personally (and this is just personal) I'm not a big fan of blood text. I feel it's a cliche cop-out that honestly just doesn't make sense. I knew some guys who pulled an elaborate prank on some girls at school by creating what looked like a devil worshiping ceremony on the ground in blood. The pattern they created varied in line weight (like a quill constantly running out of ink) and there was a pool of blood next to the pattern from where the pig was laid down and the left over blood pooled out. The key to the girls' fright was that the blood was still wet and smoke was coming out of the candles when they arrived, as though they had just interrupted something and the crazies were probably still around.
Changing topic, I worked on an environment not too long ago that included a fly through. My professor said, "Why would you end your video on one of the least interesting parts of your environment? End it in a way that the last thing someone sees is a strong image." So we changed the entire direction of the fly through. I worry that this video suffers from the same problem.
As I mentioned before, that was 2 minutes of narrow hallway walking. Frankly, that's too much. At the end I expected a scare; I thought the reason the walking was so long was because by the end I had let me guard down. But there was no "cheap scare" with something popping out and a loud scream (also cliche).
That said, there is some strong work in here. I hope you don't let my negativity get you down; your wall textures are very strong. All your corners looked like they made of had some warping? Regardless, you've demonstrated you clearly have the texturing skill so I'd suggest putting that to use with more simple models. Maybe the player has to walk through a room and ends up with more hallway on the other side? Just something to interrupt the monotony.
Really what I wanted to do was just...make an environment. So some form of detailed, self-contained scene. The problem is that my grade was heavily dependent on testing 'scaryness', which required me to develop a whole level, complete with sounds and scripted events. That's partly why it diverged so much from the 'environment' tag into more of a general (diluted) experience. Incidentally, the reason it was two minutes long is because it needed to be long enough to test peaks and troughs effectively, which did come across successfully in about 50% of the subjects (who I measured with a heart rate monitor).
Is your story about the blood prank true? If so, I commend the extreme commitment to scaring those girls. That prank really goes above and beyond. But yeh, again, blood text wasn't really what I wanted, but more just something that needed to be there because I didn't have time for the complex things I wanted.
In regards to the fly-through, the walking through hallways came across very differently to some of the test subjects; the game-savvy ones weren't scared, but the horror-illiterate ones found it unnerving to play through.
Anyway I don't intend to come back to this for a long time; I decided to treat it as a learning experience to improve my texturing and UDK skills, rather than something to go in my portfolio. That said, I will be making a horror environment for my portfolio once I complete my current plans. In the meantime I do appreciate the comments, and if you have any useful tips for my current project then I'd like to hear them. If you're interested, this is the link: http://www.polycount.com/forum/showthread.php?p=1680115#post1680115