Valve HL2DM Mapping Competition Submission

Hey all, just wondering if anyone has posted off their mapping
competition maps yet. I'll be posting mine tommorow, so hopefully it
will get there in about a week without any problems or delays.

I was also wondering if anyone has a clue about what files need to be
included with the BSP, as Valve have given little to no information on
submission requirements. I was going to follow standard map release
protocols (text file on map/author, etc), but don't have a clue as to
what files need to be or can be included since its a new engine.

It will be interesting to see if Valve request the VMF files for the
winning entries, so they can change the maps if they want. Since
they'll legally own the winning entries I'd imagine they'd get rid of
any references in the map to the author. I've placed a onetime text
message in the map that is triggered at one of the RPG sites, stating
the map name, author and my email address. Has anyone else done similar?

Unfortunately I'm still a ways away from finishing my map, and I doubt I'll finish it in time. I still have to make the 3d skybox, place physics objects, and do weapon and playerstart placement, plus whatever optimization that will inevitably be required.

I was planning on sending only the .bsp and a .txt file myself.

I wouldn't place that message in your map pimping yourself if I were you, very unprofessional- da Vinci didn't scrawl his signature up Mona Lisa's nostril :razz: But I would agree with Yak that all you need are those two items- I presume the txt file should be kept to a bare minimum.

Deadline's the 15th, isn't it? That gives another couple of days before it needs to be posted, really...

Do you actually POST it somewhere? I was under the impression that you had to SEND it to them Via snailmail.

Yes, you have to post them to Valve HQ. I wouldn't have said that I'll need to post mine in a few days if I can upload it somewhere within a few hours of the deadline :razz:

I wanted to join the contest, but really... I think I'm just stuck with a designers block during the past few weeks...

Just every idea I came up with either didn't work right or I lacked the
right textures or other content. I couldn't get myself to seriously
start on a map.

Engine coding, however, is just very interesting at the moment... and time-consuming. :smile:

According to the post office website, airmail should get a small
package (eg. a CD) to the US in 3-5 days from the UK, so I'm crossing
my fingers that if I don't have my entry done and sent by friday, it
might still get there if I send it on monday. I still have a crap load
of work to do on mine, things are not looking good :sad:

I have a crap load to do on mine too! It's bulls**t that they gave us like 1.5 months or less to make a GOOD map... Anyway, I have changed my project from DM_Coast to DM_Frankenstein. I started the NEW map from scratch 2 days ago I am about 33% done with it. I am REALLY on a role, I have the opposite of mapper's block!!! It is modelled after Frankenstein's castle mixed with Castle Dracula from "Van Helsing" It has the lab with lots of complex electrical zaps and lighning that strikes randomly outside and lights up all the windows. It has the tesla coils and all that other stuff. I will not post it up on the maps forum until it's at least 80% done. I have LOTS of entities to work out with my traveling env_beams...

I have a crap load to do on mine too! It's bulls**t that they
gave us like 1.5 months or less to make a GOOD map...

How dare they set such an unreasonable time limit on the competition
they're forcing you to enter and not giving any prizes for!

lol

Architecture courses' most common quote in use here me thinks...

"If only we had one more week..."

Hehe, yeah...

Though I think deadlines can force me to finish something, wich rarely
happens with my maps, so 1,5 months isn't really my problem.

Oh well, good luck to those that do enter! :smile:

That reminds me, I need to print up the submission sheet and burn my
map to the disc. I'm going to include my VMF file. Maybe
they'll fix the stupid building003l cubemap error that keeps popping up
(even though I ran buildcubemaps). Also, just incase some
smart-ass gets the bright idea of submitting my map for me.

I might even spend a couple of extra bucks to make it 2 or 3 day delivery, just to ensure that it arrives on time.

The 1.5 month deadline is pretty rough. Thankfully, I have no job
and no life. Also, my map is small. That helped.

Well the in game message is in there now, but of course I don't know if it will work, as I've had some trouble with it. It's pretty standard in a lot of custom maps Lep, I would have put an unobtrusive egg in there somewhere, but the fact there are no letter decals and you are not allowed to include custom textures leaves me no choice.

Anyway, I'd imagine they want the vmf files of the winning entries, to change something if they need too, especially considering they'll own it. They can always just take it out.

CRUNCH TIME!!!!!!!!!!!!!!!!! AHH WHERE DID ALL THE TIME GO I STILL NEED
TO ADD DEATAL, hey just a question, how big of a map is too big?

CRUNCH TIME!!!!!!!!!!!!!!!!! AHH WHERE DID ALL THE TIME GO I STILL NEED
TO ADD DEATAL, hey just a question, how big of a map is too big?

If 4 people are playing and you have trouble finding one another: I would suggest scaling the scope of the map down.

<DIV class=quote>
<DIV class=quotetitle>? quoting Crono</DIV>
<DIV class=quotetext>

If 4 people are playing and you have trouble finding one another: I would suggest scaling the scope of the map down.
</DIV></DIV>

huh?
a map the size of boot_camp should be avoided, but I see no reason why people can't make a map with suggested size of 12-20 players, especially given how many servers run with that playerload.

My map has turned out pretty big, and can comfortably contain about 8-14 players, with room for more. It'd also be a pretty fun one on one or 4 player sniper arena. To give you an idea of size its been compiling for 12 hours in VIS and its only up to 2 notches past 8, still a while to go, so looks like I'll have to post it tommorow.

Who was the bright spark that thought posting the entries in would be a good idea? Idiots.

Well I think it's a good way to limit the amount of entries they
recieve to the "serious" ones. If entrants just e-mailed a link and
such, they would probably be flooded up the brim and take months to
reach any sort of judgement. I guess it can be thought of as a way to
keep out the "riff-raff", so to speak.

I figured it was something along those lines. The problem with that though is that it gives US citizens an advantage of a few extra days. It would only take 3 days tops to arrive if posted in the US, but posted here in Australia it'll take 4-6 working days to arrive. Seems a bit unfair, I would have loved the extra day or two.

12 hours vis and still at 2 bars past 8? You'll be there for about a week! You do realize that vis gets exponentially slower as it goes along... A map I was just compiling took the following times:
50% - 5 mins
80% - 15 mins
90% - 1 hour
90..% 2 hours
<sleep overnight, wake up the next morning>
90...% - 10 hours
100% - 14 hours

What the hell kind of maps are you people making? The snarkpitcrew map only took 10 hours to compile. My current HL2 map only takes 12 minutes to vis and it has one big open area that lets you see almost 70% of the map at once.

Seriously, dm_casual took 6 minutes to compile. RAD ran for about
4 minutes of the 6. VIS was a breeze, especially compared to the
maps I did for HL1, dyshysteria took an hour and delirium3 took 25
minutes.

12 hours and you're not even done? Maybe it's time to rethink the strategy.

<DIV class=quote>
<DIV class=quotetitle>? quoting Yak_Fighter</DIV>
<DIV class=quotetext>What the hell kind of maps are you people making? The snarkpitcrew map only took 10 hours to compile. My current HL2 map only takes 12 minutes to vis and it has one big open area that lets you see almost 70% of the map at once. </DIV></DIV>

pitcrew took just a shade over 2 hours on my comp. perhaps its one of those "AMD vs Intel" things again :sad:

when i made false_accretion, it took me about 10 hours to compile on my K62 400, it took Doomdog less than 1 on his PII 450.. same map and only 50 mhz difference in speeds on chipsets.
shrugs

but Orph, you didn't try compiling after I tried to double the size of the map :wink:

<DIV class=quote>
<DIV class=quotetitle>? quoting Yak_Fighter</DIV>
<DIV class=quotetext>but Orph, you didn't try compiling after I tried to double the size of the map :wink: </DIV></DIV>

well, it would help if you added that small omission before. :wink:
doubled huh, so now its the size of Texas instead of Vermont?

Echo takes 2 minutes to compile fully...but then its still nowhere near finished :sad:

I first started compiling my map on my second PC (server if you will), and in the final stages it started taking up to 12 hours getting up to %90.
I thought that was rubbish and decided to try it on my main PC, it took under 3 hours for a full compile.
Now for the system specs:

Server: AMD 2600+, 1GIG DDR, VIA KT400A

Mine: INTEL P4 3GHz, 1GIG DDR, Intel 875P

My P4 put the AthlonXP to shame!

I've finished my map and I sent it 2 ago from now, I'll update my map and post the best screenshots tomorrow (going to bed).

Architecture courses' most common quote in use here me thinks...

"If only we had one more week..."

One more week? Architecture!? What happened to the "3 more years even though I won't be a doctor OR have a masters" :razz:

Lazy building designing youth of today, eh?

Edit: I doubt i'll be finished on time, far from it. I'm aiming for a
polished first release rather than rushing it to meet the deadline.

I haven't read the contest rules, mostly because as a non HL2 owner, I really have no vested interest, but is the deadline a "receive" deadline or a "postmark" deadline? The latter seems much more fair to me.

In the rules it says that all entries must be recieved by 11:59 PM on January 15th.

Makes me wonder if someone's going to rush in there 30 seconds before midnight to hand in their entry.

In that case, if I had a better connection, I'd offer to do some mailers next week for the international brethren concerned about arrival time. Unfortunately, the time saved by having me mail it would currently be lost by the time spent having me download it. I'd expect there's others around here that could help a brother out, should it come to it.

Dude, our avatars are dancing in synchronization. That's pretty awesome.

And my Athlon XP apparently beats out many P4s. It's all in your
settings. You have to use the tools to their abilities. The power is
already under the hood.

Especially with the spec's you people are posting there must be
something horribly wrong with your mapping. It just doesn't make sense
that it'd take that long. I've made maps that max out on 4-6 of the
limits and they still take a maximum of an hour on my old 1.33GHz
T-bird. And my 2500+ Barton compiles any source map I've thrown at it
in under 2 minutes so far. (The comparable HL1 maps take about 30 mins,
so it cut time in half roughly.)

I reckon people aren't using detail brushes and hint brushes to full
effect - if you make the world as simple as possible for VIS to
calculate by ensuring leaves are decently sized and well laid out then
its ultra quick. Echo might be be very big but VIS compiles it in 2
seconds. I took a look at another map made by a relative newcomer to
mapping, and managed to cut compile times down vastly by spending a few
minutes optimising it. You can help cut down RAD times by applying
nodraw to any faces that will never be seen but would still require
lighting calculations being made for them (ie. faces inside the hull,
but never in view of the player).

Its most likely I will need a few days yet on Echo, so if anybody in
the US is feeling generous I might appreciate a more direct send on.

The reason my map is taking so long to compile is because:

1. The whole map is essentially in one giant outdoor area, and therefore can see most of the map at once.

2. Most of the surrounding area is constructed using displacement surfaces.

3. Despite optimizing the map in every way possible, it is still essentially inside a giant sky box (of course, not a kill box).

I had this same problem and similar compile times with my snarkpit mapping competition entry Paralaxion, as the whole map was contained in a large open box and you see the entire level at once. I should have some screens up soon of the basically finished product. Obviously I would have liked at least another month to completely perfect the map, but its finished enough to play well with but a few minor glitches, and hopefully no big ones.

I know a lot of people are preferring to skip the competition and produce something really polished, but I think the whole point of the rediculous deadline so soon after the full SDK release is to see if people are capable of producing high quality work in a limited space of time, with a loosely assigned theme. If this is the case there might, might, be the possibility of job offers to those who can perform such a feat, and also have a previous portfolio of high quality work to back it up. Thats my hope and dream anyway. I'm probably dreaming, but there is a possibility I might be right.

And anyway, if your map doesn't win you can finish polishing it later.

Indeed, detail brushes are amazing things. I used them for trims,
entire buildings, walkways, stairs. The whole gas station in my
map was all detail brushes. Also, the whole broken building,
which is actually about 7 or 8 different brushes, is all func_detail.

I first started compiling my map on my second PC (server if you
will), and in the final stages it started taking up to 12 hours getting
up to %90.
I thought that was rubbish and decided to try it on my main PC, it took under 3 hours for a full compile.
Now for the system specs:

Server: AMD 2600+, 1GIG DDR, VIA KT400A

Mine: INTEL P4 3GHz, 1GIG DDR, Intel 875P

My P4 put the AthlonXP to shame!

I've
finished my map and I sent it 2 ago from now, I'll update my map and
post the best screenshots tomorrow (going to bed).

cause u got a weak mobo for the AMD amd owns intel

and also intels are for servers not amd

<DIV class=quote>
<DIV class=quotetitle>? quoting ReNo</DIV>
<DIV class=quotetext>Its most likely I will need a few days yet on Echo, so if anybody in the US is feeling generous I might appreciate a more direct send on.
</DIV></DIV>
You do realise that getting someone in the US means they will have to forge your signature on the print-out form you're meant to attach (which Valve will be able to tell by the postmarks etc.?)

Argh, I forgot about that :S I'll just have to hope the post office
website estimate of "3-5 days" is kept to and sending it on monday
morning gets it there on time. Might go for that "airsure" thing that
makes sure it receives priority handling at both ends too.

I first started compiling my map on my second PC (server if you
will), and in the final stages it started taking up to 12 hours getting
up to %90.
I thought that was rubbish and decided to try it on my main PC, it took under 3 hours for a full compile.
Now for the system specs:

Server: AMD 2600+, 1GIG DDR, VIA KT400A

Mine: INTEL P4 3GHz, 1GIG DDR, Intel 875P

My P4 put the AthlonXP to shame!

I've
finished my map and I sent it 2 ago from now, I'll update my map and
post the best screenshots tomorrow (going to bed).

cause u got a weak mobo for the AMD amd owns intel

and also intels are for servers not amd

Would you mind if I punched you in the face ... hard?

Yeah sure give it ur best shot.

Yeah sure give it ur best shot.

I'll meet you on 10th and Yamhill tomorrow.

<DIV class=quote>
<DIV class=quotetitle>? quoting Crono</DIV>
<DIV class=quotetext>

<DIV class=quote>
<DIV class=quotetitle>? quote:</DIV>
<DIV class=quotetext>Yeah sure give it ur best shot.
</DIV></DIV>

I'll meet you on 10th and Yamhill tomorrow.
</DIV></DIV>
hold it.. it takes me 2.5 days to get there and i have got to see this.. do not start without me.

<DIV class=quote>
<DIV class=quotetitle>? quoting Mouse</DIV>
<DIV class=quotetext>Indeed, detail brushes are amazing things. I used them for trims, entire buildings, walkways, stairs. The whole gas station in my map was all detail brushes. Also, the whole broken building, which is actually about 7 or 8 different brushes, is all func_detail.
</DIV></DIV>
Detail brushes blow ass. All they do is reduce compile time, which means jack to me at this point. While making my map I determined that they still split faces of other brushes if you line their edges up with the edges of world brushes. I have all these 'detail' trims and rubble that still split up all the faces of the walls and ground around them, so they're pretty worthless in my book.

Detail brushes cause geometry splitting? Damn, that's disturbing.

"SuperCobra" said:
cause u got a weak mobo for the AMD amd owns intel
and also intels are for servers not amd

I take it you have an AMD? How can AMD own intel if intel are better for servers?
I would use my 2600 for a my main pc and my P4 for a server if the 2600 wasnt so slow at synthetic processes, Intel and AMD a good for different things. :smile:
My 2600 was originaly my main pc which I bought coz it was CHEAP, and then I won my P4 system at a lan.
I can throw whatever I like at my P4 and it never let's me down, ever.

<DIV class=quote>
<DIV class=quotetitle>? quoting Tracer Bullet</DIV>
<DIV class=quotetext>Detail brushes cause geometry splitting? Damn, that's disturbing. </DIV></DIV>
Tell me about it. I wasted two days worth of development figuring that out, and it basically f**ked my map in the ass.

Can't you just use the "entity report" to select all of your func_detail ents, and quickly change them all into func_wall(s)?

I did change most of them into func_brushes (HL2 equivalent of func_walls), but that is problematic. Func_details cast shadows, func_brushes don't, so I either have to accept face-splitting or no shadows for my chopped up ceilings. Func_details don't have any entity information whereas func_brushes have a bunch of entity info that I'm not using, which I think causes unnecessary FPS drops. I determined that because of the switch to func_brush I managed to lose about 10 FPS in some areas. That was partially balanced out by the fewer face splits, but damnit, I can't optimize the stupid thing when all I'm doing is increasing FPS by 10 then reducing it by 10 with something else.