Voici un copier-coller tiré de Simtroplis, c'est intéressant... même si c'est en anglais... :
Maxis has posted three threads about performance optimisations for patch 2 which are copied below:

Hello SimCity fans,
With all the talk of performance improvements and the upcoming SimCity 4 update I wanted to share with you some of our findings. Keep in mind as you read this information that your experiences may vary. Different cities and different machine configurations will affect the results but generally speaking the performance of SimCity 4 has been improved. Read further to understand more about these improvements.

Performance Improvements in SimCity 4 Update 2

The main focus for SimCity 4 Update #2 was to significantly improve game play performance, thus addressing the biggest outstanding issue with the retail game. Listed below are performance improvement test results as well as a description outlining the changes that were made and the tangible results that a player would perceive when playing the game.

Test Results

In order to understand the underlying mechanics of the results below, here are the parameters and definitions of the categories:

* Two cities were used for testing. The first city was a large, well developed city with a 500k population and stage 8 buildings. The second city was also a large city with a population of 400k with horrible traffic congestion.
* Simulation is defined as the time it takes for one sim year to pass.
* Responsiveness is the average performance of zoom, rotate, and movement by clicking on the mini-map. This is from the time the action was invoked to the time the game was able to scroll again.
* The graphic settings for each machine was left as originally set by installation. That means for the minimum spec machine, all options were turned very low; the high spec machines had all the options set to highest quality. If compared side-by-side, there would be a difference in graphics quality between the min and the high spec machines; however this is part of the optimizations we made in order to achieve better responsiveness from the game.

Min Spec Machine
500 MHz P3, 128M RAM, W98

Retail
Simulation= 38 Sec, Responsiveness= 4 Sec, Framerate= less than1fps

Update 2
Simulation= 32 Sec (16% gain) Responsiveness= .8 Sec (80% increase) Framerate= 9 fps (Up to 900% increase)

Low Spec Machine
1 GHz P3, 256M RAM, XP Pro

Retail
Simulation= 73 Sec, Responsiveness= 3.5 Sec, Framerate= less than 1fps

Update 2
Simulation= 21 Sec (Up to 350% Increase) Responsiveness= 2 Sec (100 - 200% increase) Framerate= 7 fps (Up to 1000% increase, avg of 500% Increase)

Low Spec Machine
833MHz P3, 512M RAM, Win2k

Retail
Simulation= 71 Sec, Responsiveness= 3.6 Sec, Framerate= less than 1fps

Update 2
Simulation= 22 Sec (Up to 350% Increase) Responsiveness= 1.4 Sec (100 - 200% increase) Framerate= 10 fps (Up to 1000% increase, avg of 500% Increase)

High Spec Machine
3.06GHz P4, 1 Gig Ram, XP Pro

Retail
Simulation= 24.5 Sec,Responsiveness= .8 Sec, Framerate= 2 fps

Update 2
Simulation= 25.5 Sec (0 - 30%) Responsiveness= .8 Sec (0 - 50%) Framerate= 15 fps (Up to 1500% Increase, avg of 250% Increase)

High Spec Machine
2.6GHz P4, 512M RAM, XP Pro

Retail
Simulation=29 Sec, Responsiveness= 1.9 Sec, Framerate= less than 1 fps

Update 2
Simulation= 22.3 Sec (0 - 30%) Responsiveness= .9 Sec (0 - 50%) Framerate= 15 fps (Up to 1500% Increase, avg of 250% Increase)

continued with next post..

Au lieu d'un copier coller tu aurais mieux fais de faire un lien comme ca on aurait pu traduire avec google

Désolé, j'y penserai la prochaine fois...