PAIN's new Map01 has way over 5000 sidedefs after only two rooms built.

BTW: I or my maps somehow always get mentioned by someone whenever there's a thread on this subject, hehe

Share this post


Link to post

Somebody asked me what TeamTNT is up to? Okay, at the present time we are in the testing phase of "Daedalus: Alien Defense". The maps are basically completed and right now we are tweaking it. There are some scripting issues and architectural things that we need to correct before release. We are trying to squash all the bugs through a extensive testing run. We don't want to release a buggy product to all of you out there. A while back we released a very beta of Hub 1 of the project just so you could see what we were up to, we know it was very buggy but the community was asking what was up and we gave them a taste of it. Here are some quick notes...

1. It will use ZDoom
2. It is designed for single play
3. All new textures
4. Hub format...ala Hexen. You'll be able to go back and forth
between levels within the hub.
5. This is a continuation of "Icarus"
6. Scripted events/e.g. Switches on one level can affect another
level.

For further information check out our website: teamtnt.com

Cadman - Member TeanTNT

Share this post


Link to post

Personally, i think people are going for overkill on map sizes.

Architecture can be made varied and interesting at the same time without a few hundred linedefs per room.

God help us when the Doom III mappers take up arms.

Share this post


Link to post
Spike said:

Personally, i think people are going for overkill on map sizes.


I agree. I tend to find really big maps (ones that take over an hour to finish) a bit laborious after a while, especially if they're not that good. Not to mention the long loading times for savegames....

I try to make maps that take about 30-35 minutes or less to finish. (except 'special levels' like the one I made which takes 40 minutes)

I also agree about detail, you don't need 10,000 sectors in order to have nice looking architecture in maps. In fact a lot of the time, only a few sectors (10-20) is sufficient for each piece, depending on the application. (eg. industrial/tech maps generally require less sectors to make good architecture, whereas rocky/hell/gothic maps require more)

If the map is good however, I like it regardless of size/detail.

Share this post


Link to post

The map i've just created ( reznor.demon.co.uk/olympus.zip , jdoom 1.13+ only) has 338 sectors, 2588 linedefs and 2245 vertices.

This is pretty much the average size for my maps, tho it doesn't usually rise above 3-4000 vertices, etc.

Share this post


Link to post
Shaviro said:

Frack (C) specs:

Frack total specs so far:

Sidedefs: 39604
Linedefs: 21422
Vertexes: 17520
Sectors: 4931

:)


whoa, that's amazing

Share this post


Link to post
Shaviro said:

Frack (C) specs:

Sidedefs: 23121
Linedefs: 12517
Vertexes: 10000
Sectors: 3128

Frack (B) specs:

Sidedefs: 16483
Linedefs: 8905
Vertexes: 7520
Sectors: 1803

Frack total specs so far:

Sidedefs: 39604
Linedefs: 21422
Vertexes: 17520
Sectors: 4931

:)


woah buddy, what source port are you using for Frack? many of them (and of course the vanilla .exe as well) assume signed shorts, not unsigned shorts, for sidedef/linedef indexes.

Share this post


Link to post

Cadman- how long do you think the testing phase for Daedalus? I hope it doesn't require a high-end PC, because if so, my P2 333mhz is screwed.

Share this post


Link to post

Meh. Map07 of Nimrod is over 2 meg in size. and have over 23k sides.. although Both Phobia and the discontinued Phobia Exodus are exceeign that, actually I had to cut Phobia Exodus in 2 before I tired of doom editing totally.

Share this post


Link to post

Testing is going good...but we still have numerous issues to address. No, you won't need a super-duper machine to run it. We want to make sure that just about anybody can play it, that's part of the testing. Gotta check skill levels, scripting, texture alignments and the balance of weapons, ammo, health and artifacts. We want to make sure you guys get a good playing experience and not some buggy megawad that crashes everytime you load it.


Cadman - Member TeamTNT

Share this post


Link to post
Cadman said:

Gotta check skill levels, scripting, texture alignments and the balance of weapons, ammo, health and artifacts.

I don't ordinarily do this, but in this case I'm willing to make an exception and volunteer my services :)

Share this post


Link to post
esayeek said:

woah buddy, what source port are you using for Frack? many of them (and of course the vanilla .exe as well) assume signed shorts, not unsigned shorts, for sidedef/linedef indexes.


I use ZDOOM.

Share this post


Link to post
esayeek said:

woah buddy, what source port are you using for Frack? many of them (and of course the vanilla .exe as well) assume signed shorts, not unsigned shorts, for sidedef/linedef indexes.

*All* source ports assume signed shorts, so there's no way a single level can have any more than 32768 segs/sidedefs/linedefs...

I really wish that port authors add support for using the unsigned range. 65K would be an infinite level better than the current bottleneck.

Share this post


Link to post
Fredrik said:

*All* source ports assume signed shorts, so there's no way a single level can have any more than 32768 segs/sidedefs/linedefs...

I really wish that port authors add support for using the unsigned range. 65K would be an infinite level better than the current bottleneck.


I agree! Go bother randy =)

Share this post


Link to post
Shaviro said:

I agree! Go bother randy =)

AND you have to change all the utilities - Editors, nodebuilders, that kind of stuff. It's NOT gonna happen soon:))

Share this post


Link to post

Actually Quasar said someting once about fixing that in eternity.

Share this post


Link to post
deep said:

AND you have to change all the utilities - Editors, nodebuilders, that kind of stuff. It's NOT gonna happen soon:))

I don't see why there would be any problems on the editor side... seems like simple hacks to me.

Share this post


Link to post

Cadman - Member TeamTNT [/B]

Uh, do you know about the plutonia II project?

Share this post


Link to post

Speaking of Plutonia II, does anyone know what the current state of it is? (as Metabolist's site never gets updated)

Share this post


Link to post
Ultimate DooMer said:

Speaking of Plutonia II, does anyone know what the current state of it is? (as Metabolist's site never gets updated)

It's close to getting finished and should be released soon.

Share this post


Link to post
RailGunner said:

Uh, do you know about the plutonia II project?

I don't think Team TNT made Plutonia. Rather two people who worked on some Team TNT stuff named Dario and Milo Casali.

Share this post


Link to post
Fredrik said:

It's close to getting finished and should be released soon.


YAY. A lot of megawads have been released lately.

Share this post


Link to post
Fredrik said:

I don't see why there would be any problems on the editor side... seems like simple hacks to me.

LOL - I forgot - if you don't see any problem then there isn't one. It's just a simple hack. Hehe.

As a code issue: what is the value of -1 now? If you made it unsigned, what happens to code that checks for signed numbers?

If you can code C, just do some simple stuff if gt/lt code with signed numbers (compare to -1) and see what sort of messages the compiler throws out when the code changes to unsigned. You can probably do this in any language (although the others may not complain). The code just won't work the same.

I'm not saying it can't be done. I could probably convert fairly fast, but it's not trivial work. Remember - ALL utilities that use the basic level values as signed (like an index value - which they do) or error condition (which they do) need to be changed.

Share this post


Link to post

I think you may be right on this one, even though to me it seems like all you'd have to do is write a function that converts the numbers. At least for the level editor. On the nodes builder sides it seems a whole deal trickier.

-1 is being used to determine that a linedef doesn't have a second side's sidedef. You could keep that and have -32768 to -2 for extra sides :P

Share this post


Link to post
Fredrik said:

I think you may be right on this one, even though to me it seems like all you'd have to do is write a function that converts the numbers. At least for the level editor.

Only makes for more work. You have to find ALL uses of the numbers and make it a function. Ugh.

That solution runs slower and ends up being MORE work than just fixing the issue in the first place. Here's how I would do it.

I make it unsigned. The compiler will automatically bitch at most of the places where it needs to be changed. Then I manually change some code that I know won't complain, yet won't work. I always let the compiler do all the work it can:) C/C++ is WONDERFUL that way. So much better than anything else I've used.

On the nodes builder sides it seems a whole deal trickier.

Much simpler and not tricky at all. Not much code. Probably take me an hour.

-1 is being used to determine that a linedef doesn't have a second side's sidedef. You could keep that and have -32768 to -2 for extra sides :P

Not exactly. If you made a sample program you'd see that your idea would just make it worse.

It's not hard to do, just tedious finding all the locations. So you make it unsigned, check for 0xFFFF (not -1), change a few routines that don't like negative numbers (like 0x9934) and all better.

All I'm was pointing out is the fact that the change is much more than a quick hack. Good design leads to good code. Hacks just cause trouble:)

Share this post


Link to post

All I'm was pointing out is the fact that the change is much more than a quick hack. Good design leads to good code. Hacks just cause trouble:)

Though with DOOM, you'd have to hack in this case...

Share this post


Link to post
Fredrik said:

Though with DOOM, you'd have to hack in this case...

You mean DOOM port? Stock DOOM would never work.

DOOM port is not any harder or easier than the editor. Mostly it's familiarity with the code though and how it operates.

My original statement still holds - quick hacks are usually ok for teeny tiny changes - not for something as pervasive as modying a global concept.

Share this post


Link to post

the problem tho is that there are about umptillion doom ports and utilities all expecting signed shorts.

this is one of the many things i think randy heit should have addressed when he decided to adopt a different map format for zdoom, but he didnt.

Share this post


Link to post

TeamTNT, the team did "Evilution" for Final Doom. Dario & Milo did "Plutonia" but were members of TeamTNT at the time. Originally the maps were going to be released free onto the net but Romero got wind of it and wanted to look at the maps and he liked what he saw. We had to modify the maps to meet ID standards and the people involved in the project got a little money for it. Nothing we could buy a Ferrari with though! Heh!


Cadman - Member TeamTNT

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now