DM-Dawn[pics][beta][download]

[Harsh Criticism]

Name: DM-Dawn
Version: Beta
Compatibility:V2.1 + Titan Expansion
Description: Dawn strikes over an abandoned city block where power was recently restored, where careful timing brings powerful rewards.
Concept: I wanted a visually pleasing rooftop level with an organic look, and I think it came out fairly close to how I envisioned it.
Comments: My first UT3 map!
Credits: Thanks Epic for the assets! Design/Layout is all me. Props to Odedge for his feedback.
Homepage: None yet.
Download: Filefront: http://www.filefront.com/15733941/DM-Dawn.zip/


[screenshot]http://i226.photobucket.com/albums/dd213/oddman08/Dawn.jpg[/screenshot]

[screenshot]http://i226.photobucket.com/albums/dd213/oddman08/Dawn3.jpg[/screenshot]

[screenshot]http://i226.photobucket.com/albums/dd213/oddman08/mymap19.jpg[/screenshot]

What I would like to get out of this beta test is advice on weapon/pickup locations as well as bot pathing. There are still several things with
the map that I am polishing up, such as the invisible walls as well as some visuals and collision snags as well. I think the way I did BSP is frowned
upon, so I may go back and replace a lot of it. Admittedly, there was not a lot of planning that went into this map, and is more of a visual level
than a balanced one I believe, and that is something that I am taking into account in the future. So to reiterate, I would love advice on pickups
to help make my map a more fun/balanced experiece, and I am working on polishing up everything else while this beta is in effect!

Since this is my first map, throw as much constructive criticism at me as possible. What I did right, what I did wrong, as well as advice for future
level design. I should note that I have a beast PC and I have no performance problems, so I would like some feedback there as well.


Also, adding the screenshot's and using packages are still a little foreign to me, so I apologize if it doesn't work for anyone and would appreciate help there as well. It works on my PC though! If there is anything wrong with the format of this post, I apologize as it is my first.

Map is currently being optimized and revisions are being made.
Changes made since Beta:
Drastically reduced the number of BSP from 1103 to 346.
Reduced number of meshes that cast shadows where these shadows could not be seen.
Reduced the amount of visible terrain that was being rendered that the player could not see significantly.
Applied Unlit material to BSP surfaces that could not be seen by the player.
Decreased the speed of the train, as well as added a little bit of extra time between runs.
Fixed spawn points so that the face in the correct direction.
Added the correct type of Biorifle ammo.

Still to come:
Retouch of visuals that don't make sense or are in the wrong spot.
Adding a slight amount of visuals to the outside of the map to make it more realistic
Experimenting with heightfog and post processing volumes to give the map more depth.

Note: File updated!

Be sure to check out my other map, CTF-Archive (linked below).

[Freaky1]

Looks cool,dl'ing now.Not good a picking out all the little stuff but will let you know how well or not it runs on a server.

[Odedge]

Begin rambling...

1. The map has a nice look to it over all. Although practice make better, some times the theme of your map is a bit confusing. Some times it feels modern, other times you get that strong asian feel to it, while other times you get a bit of a futuristic vibe to it.

2. As far as pathing, I didn't notice really any big issues. But after looking at the map in the editor, keep in mind this... Bots will only go where you allow them to go. So when you view your paths in the editor, ask yourself if the bots have enough choices in a given area. While they may be able to get from point a to point b following a straight line you gave them, would the game play be better if you added a few extra path nodes to give the bots more choices.

3. For player starts, make sure they are pointed in a direction that makes sense, hopefully pointing towards a weapon base or something similar.

4. That big lift uses a mesh that just doesn't look right. In addition it doesn't make any sounds when it moves up and down.

5. In one area, when you are on top of a roof, you can look out in the city and see a road just disappear into the ground. While the back drop areas of a map can be tricky to setup, they need to at least look like they are "real" to the environment where the player is at.

6. Be careful about the details. You have a fence that curves into a building, but it runs into a window. You have plants sticking through a small fence. If a player can't see this stuff, don't worry about it. But the more this happens, the more the player is reminded that they are playing a map, rather than fighting in this world.

7. I would choose a different sky dome or material for it. When you look straight up, you can clearly see the stretching of the texture to the top center point of the sky dome.

8. The map needs ambient actors as it's way too quiet (minus the weapon fire). Just keep in mind that ambient sounds should tend to be a bit more subtle than "in your ear".

9. Make sure that it's reasonably possible for some one to grab the shield belt on the train tracks and make it to the other side. If it's something that has to be timed perfectly, that might be a bit too much risk vs the reward. Also, the train should make sounds as it passes by.

On a related note, it looks like the train tracks are just floating in the sky which isn't very believable for the environment the player is fighting in.

10. Some subtle fog (using a height fog) may help give the surrounding areas (the city and mountains) a bit more depth.

11. Out of curiosity, why do you have a directional light toggleable in the map?

12. FYI, you need to use the Biorifle_Content for the ammo, don't ask me why they have 2 of them when only one works.

13. As for game play, I am more of the "traditional DM game play" player in which I like maps (for the most part) to have that "arena" layout, even if it's not obvious to the player. The more "realistic" it is, the less fun it tends to be.

Big and flat areas tend to not be any fun as they don't give the player much choice in how they can fight. In addition, too long of a hallway or walkway can provide the same limitation.

Also, scaling is a big part of the equation too. That lift for example, is big enough to place 15 people. While a map can be big, the scaling shouldn't necessarily be big too. "Mini-me" maps are the one exception as that's the whole point. But where you fight in those maps doesn't always have to be big and open and flat.

There are a few areas in your map I like as far as game play, but the rest isn't that exciting. Don't take it it too personally as game play tends to be a very subjective topic.

14. The one area this map needs work on is the optimization part. Given it's fairly wide open nature (you can see big areas of the map at a time), optimizing it is very important. The area by the cars (looking across the map) really hurts your frame rate.

I would read a few tutorials on the subject (you can check out my list of UT 3 tutorials in my signature). But to get you started....

a. If a player can't see it, it shouldn't be there. This is primarily for the terrain in your map. There are many, many parts of it that the player will never, ever see and there is a good chance it's being rendered. You should "erase" these sections with the terrain visibility tool.

Also, you only need to place stuff into the map that the player can see. Essentially, your map is like a movie set and should be built in a similar fashion.

b. If a player can't see it, make it easy to render. This applies for materials and lighting/light maps. On BSP walls that the player will never see, it doesn't need a "fancy" material. Even the default material is too fancy for these areas. In the EngineMaterial package (which always loads in the editor), there is a "BlackUnlitMaterial" that should be used for these BSP surfaces.

In addition, these surfaces should also be set not to accept light and the light map resolution should be at the highest number it can (just to be on the safe side). This way, when you building lighting, the light/shadow info will not be calculated for these surfaces.

This will help on build times, file size, and performance. If you understand how the engine occludes other actors, you can even apply the "RemoveSurfaceMaterial" in the same group for any BSP face that won't really occlude anything.

Another thing is to set any static mesh that will cast a shadow that the player can't see, to not cast it. This will have similar benefits as setting a BSP surface to not accept lightmaps.

c. Looking at your map (with only BSP visible) and in wire frame, I noticed that there are a lot of little BSP squares, why did you do this? BSP takes more time to calculate than static meshes, so having a bunch of tiny brushes instead of a medium size one is not beneficial.

From what I understand, the engine can draw only part of BSP, so having "bigger" pieces (with out going too crazy) is probably a better way to go.

I am probably repeating myself, but it's important to understand how the engine "sees" your map and compare it to how you "see" the map as a player. The two different ways of seeing the map should be as close as they can. So remember than your map is like a movie set. The parts the player can see, should be nice and pretty. The parts they can't see, should be basic and efficient to render..

...rambling has ended.

[Harsh Criticism]

Odedge,

Awesome feedback, was exactly something that I am looking for as a rather new designer. Some little things I had already knew about when posting this map (lack of ambient sounds for example), but gameplay and optimization are two things that I really appreciate knowledge (or opinions) on.

Well, it's my first map, and that looks to be an acceptable list of things to take care of!

I sorta knew my BSP (and I even mentioned this up above) wasn't the best way to go about it. This came about as a general lack of knowledge about optimization, and due to the way I blocked in the level it came out that way. I realized that about half way through making the level that I would most likely have to go back and replace a lot of it.

I would say most of the "errors" found in this map are a result of a lack of my poor planning. Also, this map was still a learning experience (I learned and still am learning a lot), and in the future I plan on avoiding a lot of the things I did.

But again, this is prime feedback and I will not overlook it. I do plan on revising this level too. If you have spare time, take a quick glance at my second level CTF-Archive (linked below), and let me know what you think of that in terms of progression. It's a small map, and it only took me about 25 hours or so over 9 days to do.

Thanks a lot!

[Odedge]

I sorta knew my BSP (and I even mentioned this up above) wasn't the best way to go about it. This came about as a general lack of knowledge about optimization, and due to the way I blocked in the level it came out that way. I realized that about half way through making the level that I would most likely have to go back and replace a lot of it.

It should be rather easy to replace and should help out with the optimization aspect as that's a lot of BSP to render at one time.

I would say most of the "errors" found in this map are a result of a lack of my poor planning. Also, this map was still a learning experience (I learned and still am learning a lot), and in the future I plan on avoiding a lot of the things I did.

I hear you. When I made my first "real" Unreal based map, I ended up doing the BSP around 3 times for various reasons. It was a rather large map, so it did take some time.

If you have spare time, take a quick glance at my second level CTF-Archive (linked below), and let me know what you think of that in terms of progression.

How "done" is it. I am more of a "map is basically done" beta tester. I don't like presuming certain things will be done (like this map still needed ambient sounds). I want to basically see the "finished product" before I test it so all things about the map are fair game.

Thanks a lot!

You're welcome.

[Harsh Criticism]

In my opinion, it is 90+ percent complete. More than this map, anyway.

[1XTreme]

Just a fast word of encouragement; map looks promising! (has a nicely flowing look judging from the screens)

[Odedge]

In my opinion, it is 90+ percent complete. More than this map, anyway.

Took a quick run through it, when you're closer to 99% let me know and I will give you my thoughts.

[Harsh Criticism]

In the process of optimizing my map. I have reduced the number of BSP from 1132 (ouch I know) down to 502 so far and there is a noticeable difference in even how the editor performs. Hopefully this is not imagined... Much more to go, however.

[Odedge]

In the process of optimizing my map. I have reduced the number of BSP from 1132 (ouch I know) down to 502 so far and there is a noticeable difference in even how the editor performs. Hopefully this is not imagined... Much more to go, however.

Sounds like progress. I am just guessing, but you probably use less, but it's hard to tell with out looking at the map in the editor. Keep it up!

[SecondSoul]

Look's really unique and well done!
PS3 cook when final or in future?

[Harsh Criticism]

I haven't looked into how to cook for PS3 at all yet, but I will when things are final and I am happy about it, yes.

[Odedge]

I haven't looked into how to cook for PS3 at all yet, but I will when things are final and I am happy about it, yes.

Unless you have "special" stuff added to your map, cooking it for the PS3 is fairly easy. Check out this thread when you are ready.

[rathalos150]

Very good looking industrial themed map. Will d/l after work.

[Harsh Criticism]

Changes that have been made to the map have not yet been uploaded! Will hopefully be done by Sunday before Modern Warfare 2 comes out and I get distracted.