 |
|
notyoueither
|
 |
of naught
Aug 2001 time: 22:18
|
|
No (sorry Ming) what is needed is a P4 2.2G with 1 Gig of RAM.
I'm sorry, but Firaxis has stated that they included the mongo maps to satisfy the demand for them, EVEN THOUGH they would perform like pigs on many of what would have been good computers a year ago.
So they built at least one aspect of the game with the future in mind. What are they going to do? Reduce the maximum map size? Dumb down the path finding?
I have a Duron900 (effectively), 256M DDR, Matrox G550. Large maps work well for me. I tried Huge maps twice. They worked well too, only I don't want to go through that much territory to conquer the world.
Last edited by notyoueither on 01-04-2002 at 13:55
|
|
|  |
 |
|
watorrey
|
 |
Rochester, NY, USA
Oct 2001 time: 00:18
|
|
With all the complaints about the speed issue, i haven't tried a huge map yet. Though i will soon just for the experience.
PII 266 @ 300 (512 L1 cache), 224 RAM, 66MHz bus, Diamond Viper II w/ 32 meg, SB Live sound, Win98. I run seti@home when not playing and work units average about 18 hrs for me. (seti is a good way to find out how well your motherboard is tuned)
I always play large maps, usually with 7 AIs. The main waiting period for me is when i can actually see the other civs units moving. More civs = more units moving = more pathfinding.
Since i have such a low end sytem, i assume that what i see on a large map is similar to what many see on a huge map with a better sytem. Could i be right?
|
|
|  |
 |
|
Arrian
|
 |
Kneel before Grog!
Jul 2001 time: 00:18
|
|
I do wish the game didn't slow down so badly (for me) on huge maps. Large maps are generally ok, a little slow at the end. Marla's map got insane. Utterly, completely insane. 45 minute turns in the Industrial Age. The map's nice, but not that nice.
400mhz
64 MB RAM
So yeah, my computer isn't nearly top-o-the line, and the game works well on Normal Maps (aside from the fact that I did have to shut off the music, which stuttered and skipped and slowed down the game). Eventually I will cave in and get a better system and things will probably be fine. By then, they'll be up to Pentium VI or whatever... with God knows how much speed & memory.
-Arrian
p.s. I play with "friendly moves" and "animate auto moves" off. This occasionally leads to real surprise at an AI attack, but it does speed up the game.
Last edited by Arrian on 02-04-2002 at 00:10
|
|
|  |
 |
|
DrFell
|
|
quote: Originally posted by notyoueither
So they built at least one aspect of the game with the future in mind. What are they going to do? Reduce the maximum map size? Dumb down the path finding?
|
Well, you could code the pathfinding and other regularly used time-consuming AI routines in assembly instead of in a high level language. Although you wouldn't notice the difference if the routine was run once, when it's run hundreds of times the slightly shorter time it would take to run could speed up things considerably.
|
|
|  |
 |
|
dubwai
|
|
There are a lot of reasons that Civ II runs slowly on large maps. The foremost being that after the civs have expanded a bit there is a massive amount of information needed to process the game. Some optimization can be always be done, but there is a limit for the minimum amount of memory needed to store that information. No amount of whining is going to change that. The main consideration when trying to solve this is your RAM. When the system runs out of memory it will start pagin out a lot of memory to the hd. This means that it will have to read and write to the hd constantly while processing. Using a NT or even worse DOS based OS you are talking about extremely stupid paging algorithms, so the game may have to page memory in and out several times in what would normally be a very quick process. So if you have any more room for RAM expansion do it, otherwise it might be a good time to get a new machine. They've never been cheaper.
|
|
|  |
 |
|
dubwai
|
|
I meant Civ III not Civ II of course.
|
|
|  |
 |
|
Ethelred
|
|
Mike Breikutz has said the main thing slowing things down on larger maps is the pathfinding algorithim. Its not a bug from what he said. The time the algorithim takes exands exponentialy with map size. A lot of algorithims expand exponentialy as the number of choices expand.
Also I doubt the number of Civs matter as much as the number of units. Certainly if the pathfinder is the main time user anyway. On smaller maps the number of civs might be the determining factor as less time is spent on units and time spent on dealing with other civs become more significant.
|
|
|  |
 |
|
Ethelred
|
|
quote: Originally posted by steelehc
was 12 minutes, 19 seconds.
With an entire 256x256 map filled with cities. I don't understand what people are b1tching about here.
Specs:
Athlon 1400
1024MB PC-2100DDR-RAM
2x Maxtor 80GB 7200RPM drives in a RAID 0 array
GeForce 2 Ultra 32MB DDR
Steele |
I sorta think the time you had there is exactly what people are *****ing about.
Of course with you takin up 2/3 of the map the AI isn't spending a lot of time doing things.
Showoff.
|
|
|  |
 |
|
Sir Ralph
|
 |
Long live teh Schwampel!
Dec 2001 time: 06:18
|
|
I think the # of cities matters pretty much. Check the time the computer needs to recompute things when you found/capture/raze a city at a huge map. The more cities are at the map, the longer takes it. Not only linear, but may be even exponential.
My old specs: PII/350, 128 MB SDRAM, RIVA TNT 16 MB, SB AWE32. Standard maps were ok, large ran quite slow, huge and oversized (Marla's) were absolutely unplayable. My only try at Marla's ended in the early industrial age with 40 minutes AI moves. It's history, as this machine is now a DSL router .
My new specs: Athlon XP 1800+, 512 MB DDRAM, GeForce2 64 MB, CMI8738 onboard. Sufficient for me as I'm not much into ego shooters. First test, in the mid-medieval age with 12 (of total 16) civs on my continent, AI moves take about a minute, even though I REX'd and ICS'd like a madman. I think after finishing the first "huge" game I'll give Marla's another try.
|
|
|  |
 |
|
Ethelred
|
|
quote: People are complaining about times of 35-45+ minutes. I have never seen anything approaching that. |
You have four times the processor power and 16 times the ram of the person that had 45 minute turns. Somehow I don't think they had have something ELSE bizzare going on to get such long times.
|
|
|  |
 |
|
Ethelred
|
|
Sorry I haven't a time machine to check on this. Where did you purchase yours? I might be interested in buying one myself. It would come in really handy if the energy cost isn't too high.
|
|
|  |
 |
|
falsfire3401
|
|
Winterpeg, Canada
Mar 2002 time: 23:18
|
|
The only time I've had *any* issues with turn times is on Marla's gargantuan earth map.
I've played games on random-generated huge maps, 16 civs, that lasted right up and into the 2000's in modern times, and the turn times never exceeded about 30 seconds or so. And I don't have *that* powerful of a machine:
Athlon 1.4GHz
512MB regular SDRAM
2x40GB 7.2k drives in RAID-0 <--Civ3 on here
1x20GB 5.4k drive (emergency backup & storage in case RAID array ever fails)
VIA KT133A chipset m/b
Windows 2000 SP2
Tons o' tasks running in the background (just after opening windows BEFORE running any programs I'm already using 175MB of physical RAM)
And on Marla's map, the turn times, while annoying, were still bearable, at around 45 seconds/turn in the year 700AD (late part of middle ages era)
|
|
|  |
 |
|  |
All times are GMT. The time now is 05:18. Apolyton Time is 00:18. |
top of page
|
| archivepost |
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|