I have a 128 MB ATI 9200,
Windows XP,
80 GB Hard Drive,
512 MB PC2700 RAM,
And a ASUS P4PE motherboard that is expandable to 2 GB RAM

Would it be worth the money to upgrade that to 2 GB ram, or how much more should I put in it ???

If it is, how much would it increase my system performane?

Thanks alot

1024mb is more than enough the higher the PCXXXX the better it is if your motherboard can support DDR400 then go for the PC3200 ram if not then 2700

XP utilizes RAM better than the W Dos programs. 98SE did not handle RAM above 512 very well.  XP does better. According to the experts I've read.  I did read of a RAM test in Gr. where XP was the OS and RAM increase 64 to 128 resulted in a very large performance increase. RAM from 128 to 256 was a marginal performance increase.  THG had a recent article on RAM, check it out, that asserted that XP (NT system) would perform much better if it had more than 512 MB RAM.  Asserted but did not test to be true.  If more RAM gave significantly better performance would objective tests not be shown on the major manufacturer's web sites?  Any tests shown? No?  Opinion:  512 is fine for FS9.  Maybe CFS3 would benefit from more RAM.  Why?  Because the HD light comes on infrequently in FS9 and more frequently in CFS3.  CFS3 is very rich in scenery and AI objects which require more RAM or virtual memory.

Looks like your system is plenty good enough.  Your card is not.
See Toms Hardware Guide for several articles on 9200 tests.
Summary from March 2003 test of 9200, 9600, 9800:
"Don't hold your breath for any surprises where the Radeon 9200 is concerned, however. This chip offers nothing new over its predecessor, aside from an AGP8x interface."
If you go for more RAM, post your impressions.

Define a "slide show" pls.  Do you mean stutter?
My app FR is set to 12.  No stutter.
Substantial testing shows no difference 12, 20, 30 or max of 40 my system.  At least none that I can see.
I believe the fixation on FR stems from a "fact" that most users have as a deeply held belief.  If the FR is set to a lower value, the user "knows" that the rendering is not as good as it would be at say 50 FR.  Not that anyone can see the difference; it is a DHB.
I read a recent test where 2004 was included.  The test system was the latest P4, 3+, hyperthread, etc with 5900 and 9800 cards.  Just a "typical" simmer set up.  The app was maxed in settings and the variables were card anitialiasing and anisotropic settings.  In this super duper system FR's were driven to a (barely) acceptable 30 FR for the ATI card and FX was the "winner".
I find comedy in the most unexpected places.

There isn't a system out there that will give FS9 60FPS at high app and card filter settings. None.  Kowledge that the standard is 28 for TV  likely influences the conclusion that 28 or more is a requirement in FS9.  Most systems will not give FR of 28 or more with high app and filter settings.  Won't happen.  FS9 is designed for compromise hence the FR slider.  High FR or 3D filter settings but not both.  MS:"Limiting FR frees up computer resources" for Bi/Tri, antialiasing and anisotropic filtering.  Filtering which is designed to enhance realism but carries a high card resource consumption penalty.  The 9200 card is likely the weakest point in the system.  And the obsession with FR will not improve the result.

I can't believe Bear said that!

A comparison! I think not!

How the heck can you call two systems comparitive when one of them is running a GF2 video card and the other a RADEON 9700.

The reason the system running FS2002 got "hammered" is because the poor old GF2 card just couldn't handle antialiasing the clouds.

Stop it Nickle, your confusing people.

O h .......... and where did I say anyone got 60fps in FS2004?  

---------------------------------------------------------------------

This is a FS2002 pic I just shot. The scenery is modified with terrain mesh, my custom airstrip complex and custom ground textures. The glider is a fairly complex model, a nice addon.

The clouds are custom thermal scenery as well as stock FS2002 clouds.

The resolution in the sim is 1024 x 768 x 32, all graphics sliders maxed. My video card options are fully set to Quality except my antialiasing is set to 2x and the anisotropic filtering is OFF.

The Frame rates on my sims are set to unlimited.

The Video card is a lowly GF4 TI4200. The computer runs an XP2600+ cpu on an nForce2 chipset with 1024mb PC2700 DDR in dual channel mode.

The CPU, FSB and RAM all run at a standard 333mhz.

Flying here is very smooth with around 30 FPS in the nicely detailed virtual cockpit. As you can see, the external view shows over 40 FPS, as does the 2D cockpit view.

I have a balanced "budget" system that is purpose built and maintained for use in flight sims, and I'm very happy with it. I just hope FS2004 isn't going to slow me down too much when I get it.

FPS does matter.  8)

The pic shown by Orenda is really nice.  Looks like the scenery was substantially enhanced.  There is a cost to the enhancement in the form of more polygons and texture processing required by the MB/chipset and CPU.  And the graphics card must render more detail.  Doesn't come for free.
I tested 2002/2004 Along the Rockies to check the FR difference.  With card settings FS and FA of 4X and Bilinear; comparable app settings, unlimited FR, 2002 was around 50 and 2004 around 30.
12 FR was not good for senic stuff.  20 was Ok for me both apps.  Depends on rate of change of scenery and altitude.  At 30 bank 20 FR was Ok.  Higher bank angles generate higher scenery changes and higher FR requirement.  So if the purpose of the flight is ever tighter circles then higher FR is better until the inevitable. Most of my flights are instrument and scenery is less important.
FR does matter; it's that it isn't the only setting that matters and 2004 is designed to operate smoothly at the 20 FR default.