Frame Per Second

[G0-Gerbil]

Use the hint brushes to break down the map into tunnels. It is the obvious tool to make a map run better.

No they aren’t. I see hint brushes touted so often as some miracle cure for framerates, and yet very few people I’m aware of know how they really work or use them properly (despite the tutorials out there, some of which I might add also get it wrong).

If you can’t make your map run properly without hint brushes and you don’t know how hints work, then I’d suggest re-designing your map.

In other words, unless you really know what you are doing with hint brushes, don’t map with the assumption you’ll get a magical frame-rate increase from them at the end of it all - you have to create your map with hint brushes in mind to get the benefits of them.

[SCDS_reyalP]

                                  Total:$626US (All prices from Pricewatch)

Wow, ONLY $626 ? To play poorly designed custom map on a free game, OMG!!! sign me up!!! Yes, I realize the AMD stuff is a couple hundred cheaper, but that is still a lot of money. Guess my $300-$500 wasn’t so far off.

Never mind that the lowest prices from pricewatch are often from dodgy vendors (i.e. $20 dolllar ‘shipping’ for a stick of ram, poor service, impossible to RMA etc… then there are the ones who are flat out fraudulent…)

If you think hint brushes are magic FPS dust you can add to your map, think again. The only places they will give you big improvements is in a few pathalogical cases.

[pack]

NOP has a good point there. It’s only a $150 for a 2.5 gig Pentium (I don’t deal in Athlons). People really should start upgrading their systems. I think a good indicator for map requirements would be to look at hardware requirements for recent shooters like UT2003, Tron, Max Payne 2…so you know what today’s standard is for low end. But seriously though, just because it’s the same engine that was used 4 years ago doesn’t mean it should work on the same hardware. If your computer means anything to you, you will go and upgrade it. Otherwise I think you should give up on shooters and get going on those wargames whose only graphics is one big map with a bunch of hexes and some numbers where your units are supposed to be.

"people should invest money so that they can play my map "

[G0-Gerbil]

Poor blushing_bride’s thread - totally hijacked

For speed testing, there’s no real way you can know what will work on other specs without getting someone else to test it.
Using the FPS from the official maps is a rough guide (and I use it as such myself), but it’s not the be-all and end-all of game speed (which can either be CPU-limited, rendering-limited, or sometimes MB->card).

Just try to keep your map as fast as possible, at least until you find people with a range of specs to test it out - then simply get their feedback and see where you need to make improvements / can add more detail.
Then get out another demo and so on and so forth.

BTW I don’t really mean a public beta - these should be a closed alpha really - if you can arrange a private session on a server even better

[LodeRunner]

I don’t know when I’ll get it back online but I have an old Celeron 333 with a first generation TNT card and 320 MB RAM. That ought to be a good low end test system. I’ll tag you if/when I get it back up and running (I fear I have a bad Mobo and vid card though…and I’m broke, so don’t expect it really soon…). I think my server (see below) might be a good test bed too. I just need to get ET on it now…

My mapping comp:
2x Athlon MP 2000+
1 GB DDR2100
Tyan Tiger MPX mobo
1x 80GB ATA100
2x 40GB ATA100
GeForce4 Ti4200 128MB/DDR
Windows XP Pro

My server:
Pentium III Coppermine 800
512 MB PC 100
1x 20GB ATA66
1x 13GB ATA66
Radeon 7000/VE 32 MB/DDR
Windows 2000 Server

[FireFly]

The map i’m currently working on is a huge outdoor map. Because the art of using Hint-brushes remains a mystery to me (after reading all available tuts on the net) I had to ‘chop’ my map into 4 smaller parts, having them connected by tunnels.

Although r_speeds now rarely exceed 20K i’m fully aware that this doesn’t mean my map will run smooth on other people’s pc.

The only real way to find this out is to release your map as a Beta and get some feedback.

Also try to imagine what will happen to your map (or better what will happen with the fps /r_speeds of your map) while there are 32 people playing it, airstriking the whole damn place…

For this reason I also try to predict where all the fighting in my map will occur and where the choke points are. Those places will have to do without any ‘eye candy’ , complex structures or fps-eating shaders.

[Computertech]

This is a good tut for explaining the hint brushes:
http://countermap.counter-strike.net/Tutorials/tutorial.php?id=2
And a very simple one here:
http://www.nibsworld.com/rtcw/tutorial_detail_and_hint_brushes_part2.shtml
Same principle used in SD maps, dividing up the level, according how big and how many things are being drawn at one time. I think i’ll go by what SD does on their maps then what ppl pretend to know.

[SCDS_reyalP]

This is a good tut for explaining the hint brushes:
http://countermap.counter-strike.net/Tutorials/tutorial.php?id=2

You should be aware that BSP and VIS in CS do not work exactly the same as it does in q3 engine games. That tutorial has some good illustrations, but the details are not exactly the same.

[Computertech]

It is still based on the Quake engine, including this game. To date, id’s game engines have powered blockbuster titles from other producers, including Jedi Knight II: Jedi Outcast, Soldier of Fortune, Half-Life, Half-Life: Counter-Strike, Medal of Honor: Allied Assault, Star Trek: Elite Force, Anachronox, and more.

[G0-Gerbil]

Cheers for the ID promotional material

The engine that powered half-life all that time ago is quite probably almost entirely different code to the Q3-powered engines.

The fact you were happy posting that counterstrike tutorial suggests to me you are one of these people I mentioned - who think they know how hints work but don’t, and worse, mislead other people.

BTW - the counterstrike thing did have one bit of useful info for us mappers:

Don’t let r_speeds go over 600

I say it again - hinting is not a miracle waiting to happen.