Do I need a bigger spec computer

It could. I spoke with some skilled C++ programmers early this year about that. Basically, it would require every skilled C++ modder to work together to make it happen. It would be a momentous task, to say the least...

I'd donate a couple hundred $ to that cause
 
Nice to see it is possible, even though unlikely then.
It's funny, slowness has always been the bane of civ games, right back to the original I used to play on the Amiga, that was just as slow! :D
The greatness of the mod makes up for it's slowness. Playing vanilla civ4 feels so plain now.
 
Guess my mediocre pc should crack you up, currently reaching turn times of 10 minutes and up!
Running dual 3ghz, geforce 8600 gt magic 512mb, 3.2gb ddr2ram 667mhz windows xp on the GEM Map on turn 551 with 36 know civs :P

Adjusted storm stuff to disable 'recurring'..
Want to remove it now. :P
 
Take it up with EmperorFool, Jdog5000, and Kael... They are probably the most skilled here...
It is not like Civ is an abandoned game. Unlike Simcity there will a Civ 5 some time.

@SaintRaw you should play with less civs. I only play a max of 14 and it runs a lot faster.
 
It is not like Civ is an abandoned game. Unlike Simcity there will a Civ 5 some time.

Speak for yourself. I expect a new Sim City or RollarCoaster Tycoon before I expect Civ5.

Firaxis hasn't even started development. They're working on some console game. :(

It will be at a minimum 3 years, but more likely 5+. We better make the most of the game engine.
 
Speak for yourself. I expect a new Sim City or RollarCoaster Tycoon before I expect Civ5.

Firaxis hasn't even started development. They're working on some console game. :(

It will be at a minimum 3 years, but more likely 5+. We better make the most of the game engine.

by that time we might be at octo cores being standard :eek: so getting it multi core compatible would be a godsend to modders and players alike
 
A couple of observations:
  • I use windows 7 x64 with 4gb RAM. After upgrading to Win 7 I have yet to see a MAF. I also play with static leaderheads, nobuildinggraphics, single-unit graphics, low res textures -- basically all the tweaks I've found to improve speed. But upgrading to windows 7 x64 with 4gb made a HUGE difference.
  • I also OC'd my rig (dell xps 630 allows for some OC) from 2.4 to 3 (quad-core Q6600). This helped a lot. I paid around $700 on dell outlet for what I thought was a short-term fix (since my machine died) into what's turned into a very nice computer.
  • I was playing it in window mode, but I noticed that Windows 7 kept turning down the priority of Civ when it was minimized (e.g., I was surfing the net), and turns were taking significantly longer than when running full-screen. So I would venture a guess that windowed vs. full-screen impacts more than just things like Alt-tab and mouse-scrolling. I also confirmed that when I started seeing it change the priority on me!
  • I downloaded a process manager and set "rules" for processes, including setting the affinity to cores 2-4 (not 1), and to run at high priority (realtime caused a few issues). I used to use a command line for this in XP, but after watching my CPU monitor and a few other things, I'm pretty sure those command lines to startup Civ aren't very effective under win7.
  • I tend to play 180x120 maps with around 20 civs. I've disabled storms (OS79 fix) and don't play with barbarian civs (seems to slow it down). Playing with more civs == longer turn lengths

My recommendations:
  • Create a sticky with all the "performance" advice dealing with hardware, OS, settings -- and ROM settings (storms slow things down, increasing # of civs increases turn length, etc.).
  • A major focus on ROM 2.9 should be on efficiency: maybe getting those C++ experts Afforess mentioned (EmperorFool, etc...) to help us make ROM more efficient, helping to identify redundant code, maybe even moving some of the features out of XML and into SDK or python if that'll help performance.
 
Our only hope is if there are plans to make Call To Power 3 ... I still prefer Call To Power over vanilla civ 4.
 
Speak for yourself. I expect a new Sim City or RollarCoaster Tycoon before I expect Civ5.

Firaxis hasn't even started development. They're working on some console game. :(

It will be at a minimum 3 years, but more likely 5+. We better make the most of the game engine.
and they're working on Civilization Network for Facebook... wasn't it going to beta next year... So no Civ 5 for couple years I guess.
 
I hope next civ uses hexagonal tiles instead of squares, terrains look more realistic, cities would have more tiles to utilize and more strategic possibilities would be possible. :)
 
I hope next civ uses hexagonal tiles instead of squares, terrains look more realistic, cities would have more tiles to utilize and more strategic possibilities would be possible. :)

And they give modders all the same tools as we have now. I don't want to go back to CivIII where hardly anything was moddable.
 
And they give modders all the same tools as we have now. I don't want to go back to CivIII where hardly anything was moddable.

That's why most Civ freaks were modding CTP 2 which was as simple as modding a acrobat text page! Ages Of Man is still a must have mod for it. I found Civ 3 just awful and gave my copy away shotly after purchase. Diplomacy was more adaptive but espionage was just horrible! :) But you are correct, developer tools that work with the engine make modding easier, less tedious and help avoid many hard to catch bugs which many people miss.
 
, developer tools that work with the engine make modding easier, .

Even easier:

Give us the game engine! Come on Firaxis, it's not like we can run with it and steal it, it's over 5 years old...
 
Yep. UT3 engine was published this week! :eek: So why not publish Civ4? ;)

Edit : Ups. I just see. It's not the engine, but a Develpoment kit. :(
 
Java isn't any faster than C++.

What Firaxis needs to do is write it with multi-core support. That would make it immensely better.
I second, third and fourth this.

Nevertheless, under the assumption that a Civ5 will have multi-core support (anything else would be just a slap into the face), do you think it would be optimized for a certain number of cores?

Not being a programmer I don't have the slightest idea if you could make it scalable (? - I mean that it coud identify the number of cores present in a given system and then adjust itself to that number).
 
I second, third and fourth this.

Nevertheless, under the assumption that a Civ5 will have multi-core support (anything else would be just a slap into the face), do you think it would be optimized for a certain number of cores?

Not being a programmer I don't have the slightest idea if you could make it scalable (? - I mean that it coud identify the number of cores present in a given system and then adjust itself to that number).

I did a bit of reading on multithreading (That's making code use multiple cores), and you need to specify the number of cores, ATM. Multicore support in modern games like Crysis still only use 2 cores, even if you have more.
 
I did a bit of reading on multithreading (That's making code use multiple cores), and you need to specify the number of cores, ATM. Multicore support in modern games like Crysis still only use 2 cores, even if you have more.
I think python also supports multithreading (not sure how good support) and I've been thinking that maybe some python guru could convert Civ 4 user interface to use threading which would make interface faster?!?
 
I second, third and fourth this.

Nevertheless, under the assumption that a Civ5 will have multi-core support (anything else would be just a slap into the face), do you think it would be optimized for a certain number of cores?

Not being a programmer I don't have the slightest idea if you could make it scalable (? - I mean that it could identify the number of cores present in a given system and then adjust itself to that number).

It'll probably made for 4 or 8 because it is going to be a while before it comes out and by then octo may be standard

I think python also supports multithreading (not sure how good support) and I've been thinking that maybe some python guru could convert Civ 4 user interface to use threading which would make interface faster?!?
even if only python did multithreading it could major speed up the game if you could multithread callbacks
 
I think python also supports multithreading (not sure how good support) and I've been thinking that maybe some python guru could convert Civ 4 user interface to use threading which would make interface faster?!?

That would be nice, the civilopedia is pretty slow...

The only problem that I see is that for both C++ and Python Multithreading, you need to add new API for the multithreading functions...

That would not be fun.

It would be nice if we could convince Jdog or EmperorFool to start a project though...
 
Back
Top Bottom