vispatching help

[foq]

I'm trying to vispatch my maps but I have a problem... some of them are already vispatched and some arent. I really don't want to go through 197 maps and look for water and see if they are vispatched or now. Is there any quick way I can tell if a map is vispatched or not, or is there a way I can tell the vispatching program not to un-vis already vispatched maps?

[=peg=]

if you just want transparent water for on-line play, then just use r_novis 1 and r_wateralpha 0.3 (or whatever value below 1) and forget about vispatching..

this wont let you see players beyond the surface of the water tho.. unless the map was originally vis'ed for transparent water or the map is vis-patched as such server-side..
(even if you vis-patch your map client-side this wont make you see players beyond the water surface unless its also done server-side)

[foq]

I knew someone would say that...

I want my maps to actually be vispatched so that when I run my own server I can see things under water (and other players can too)

[=peg=]

then why didn't you mention that in your first post?

i suppose LH's hmap2 tool might come in handy combined with some .bat script or other from of automating the task (shell scripts / python)

read the readme

you'd still need to find out yourself what maps to re-vis tho (and make a backup of those first)

[foq]

That's the problem, I don't want to go through 200 maps and see what's vispatched and what's not

It really sucks cause map downloads come mixed. For example the rquake maps. Some of them are vispatched and some arent. Some of them don't even have water (but the do have teleporters, which that is technically water too). I'd go nuts looking through 200 maps like that. What I need is a way to tell the vispatch utility (this one, id1vis at quake terminus) not to un-vispatch maps that have already been vispatched, or a quick way to tell if a map has been vispatched or not without actually playing it and looking for water, or another program that has some sort of option to skip over already vispatched maps.

Aha, I found some other program. Apparantly it does vispatching the "long way" and doesn't un-vis maps.

[=peg=]

ah so you are talking single player maps... well.. since the vis info on most large single player maps is actually very valuable (in terms of fps) i would not even bother screwing around with it.. bare in mind that it could literally take days to compute the vis data on some maps.. so you really don't want to re-vis em all.. and just ripping out the vis data all together will basically have the same effect as r_novis 1 so... yeah.. why bother really :/

[foq]

Ah so what if it takes days.

I found a program called rvis and it was looking like it was going to work nice for me until I tried using it on one map (ant.bsp one of the rquake maps) and it said it encountered a problem and needs to close.

What I want to do is run all my maps through the vispatching process so I can have proper transparent water. The only thing r_novis 1 does is give me clear looking water instead of solid looking murky water. I can't see objects that are actually in the water, like a health box or another player. Simply running r_novis 1 is not good enough because if there was a health box under the surface of the water I wouldn't be able to see it even though the water is clear because the map isn't vispatched.

Basically what I need is a program that:
1. Works =/ (rvis was such a letdown when it "encountered a program and needs to close")
2. Won't un-vis already vis'd maps

irregardless of the amount of time it takes. Besides, what you are talking about with maps taking days to compute is really old info that came from the days of the origional Pentium. I'm sure my setup will do it much faster. Not that I expect it to be instant, but faster than days for one map

[foq]

Ok.. so I used BSP2PRT.exe from the rvis package to get the .prt file needed to perform the vispatch. I then used Bengt Jardrup's vis.exe. It pooped out a .vis file. What do I do with that?

Apart from that the map I vispatched (which I looked at in-game to make sure it wasn't vispatched yet) came out vispatched after this (long) process. Still dunno what the .vis file is for

[=peg=]

irregardless of the amount of time it takes. Besides, what you are talking about with maps taking days to compute is really old info that came from the days of the origional Pentium. I'm sure my setup will do it much faster. Not that I expect it to be instant, but faster than days for one map

actually.. i was talking about todays pc's.. trust me.. vis can take ages on some maps.. and i do mean ages..

you won't need the .vis
that is, running vis.exe on a .bsp (with the .prt available along side it) will store the vis data inside the .bsp once its done..

well if you really want to go though all that trouble then all i can say is.. good luck sir!

mind you.. i feel that if the map-author wanted you to be able to see beyond the surface of the water he/she would have vised it for transparent water in the first place.. but.. ah well thats just my take on the matter..

[Mindf!3ldzX]

2002 HOPw News - Clan HOPw

Behold the power of QuakeOne.com community member Peg! No wait,that link came from google,not anyone in this post as of yet?

BEHOLD, THE POWER OF GOOGLE. NEWBS!

How long did this take?
In real time, all the VISsing took place in approximately 3 days.
However, we had several computers working in parallel, two of
which were P2-300s (oc'd 266s), so if the equivalent work was done on a
single P-166, it would equal at least a week, probably two. We used
rvis (well, at least *I* did), and that cut the compilation time nearly
in half. For reference CTF3.BSP, from the CTF verison 3 pack took
approximately 4 1/2 hours to compile. That's *with* RVis.

I was rude to peg so I edited the post. but seriously, come on Peg.... that guy said 3 days @ several PC's,two of which was P2-266 oc'd to 300. I've got like,50 or more of those 266mhz's cpu's in just one of my core's ( in terms of processing power),not to mention those computers was probably running a leet 256mb ram (leet for the time,eh?hah).

[ORL]

Well Mindz, Vis time varies on the map layout. In the case of the original ID1 maps, they are simple, not very detailed with rooms closed off from view from one another. Vis likes simple details, thats why it didn't take long.

On the other hand, Very large maps with many small and large details with rooms that can see one another, like Masque of the Red Death, Marcher Fortress and Castle of the Dark ages take much much longer to vis. In fact, Castle of the Dark ages took 1218 hours to full vis, the record so far.

If your still skeptical, I can send you a small map of mine and you can do a vis on it. Although I can't guarantee how long it will take, but I'm pretty sure it will probably be longer than a week.

[Mindf!3ldzX]

your missing my point, the work has already been done.

[foq]

1218 hours.... thats like 50 days. What were thay using, a super nintendo? 50 days to vis one map... heh. With a modern Core 2 duo I doubt it would have takes 50 hours

Just in one full day I've vissed almost all the rquake maps. All that's left are the Dredd maps and Rocket Arena maps (Which shouldn't take very long)

Hey ORL, if you want you can send that map to me and I'll see how long it takes to vis on my setup. I've got a Sempron LE-1250 at 3GHz and 4GB RAM running at DDR1000 (latency 5-5-5-15)

[=peg=]

not that i feel the need to proof my point to the ignorant but.. ah well here is just another example of a vis process taking ages, well 17 days actually but still..

anyways.. i was just trying to prevent some frustration on foq's part, but if it all worked out then good job!

[foq]

Nowhere do I see the system specs of the rig he used to vispatch that map. For all I know he's running an origional Pentium like the example Mindz posted.

Unfortunately, nothing is perfact and the vispatching tools are no exception. Some maps, when used with bsp2prt, report that something was "clipped away". I wish I knew what that meant and what it did. Additionally, when I run "intro.bsp" which is the rquake start map where you shoot the icon of the map and the box lifts up over the portal, bsp2prt.exe says it encountered a problem and needs to close. That and yes doing the fullvis takes a long ass time to do. It's been running for a full day if not more and I still don't have all the rquake maps done (although 48 are completed already). If I can do 48 maps in a day I just don't see how one map would take 17 days unless it was running on exceptionally slower hardware than mine (and mine's nothing special).

Again, I am merely sporting an overclocked AMD socket AM2 single core @3GHz. If I had a dual core this would be done roughly twice as fast since I would have 2 processors to vispatch with... quad core would be roughly 4 times as fast, provided the are the same processor type and clockspeed.

This process doesn't bother me nearly as much as it would going through all those maps and seeing which are vispatched and which aren't.

After the maps are fully vispatched I can upload them if you guys want...