Would like to know if there are any video cards that would run this game on max settings. Please feel free to add a setup is necessary.  
 
Note: I've read the reviews on this game while performing on today's top notch gaming systems but none proved to pass the test at running flight simX on max settings.
 
Feedbacks appreciated!

Well, you don't a big card, a 7600 could do the job very well, but I've heard that the ATI cards can do a better job than the NVIDIA. I personally use an Nvidia 7300LE with several add-ons and I get around 25fps.
 
Regards
 
Roberto

http://www.amazon.com/Microsoft-Fl [...] B000GCJ6MK
www.gamespot.com/pc/sim/microsoftf [...] eview.html
 
Those are two of several reviews that tested this game on High end cards.  
 
Should i invest in a SLi system with two 8800GTS? But keep in mind that tomshardware's VGA chart already did benchmarks and SLi looks pretty bad if i'm going to run on Max settings. This game MUST be played on highest possible settings.
 
Currently, this is what the game is running on
 
EVGA Geforce 7300GT 512MB DDR2
P4 @ 2.8GHz
768MB DDR Mem
 
My PCI-x slot is only 4x based.

Here's a benchmark that shows two 8800 GTX cards working together to achieve a lousy 22.6 fps at 1920x1200. (The same setup gets 54 fps in Oblivion, at the same resolution).
 
http://www23.tomshardware.com/grap [...] &chart=353
 
A 7600GT can get 20 fps at 1280x1024:
 
http://www23.tomshardware.com/grap [...] &chart=291
 
It depends on your resolution. This game is badly written and you won't get it to reach many fps with any hardware. Sorry... Maybe wait until the 9800 GTX is out (around Christmas, apparently) and then see.  
 
8800 GTS SLI is a bad idea. It's smarter to get a 8800 GTX and a 680i-based mobo, and later add another 8800 GTX if it's really necessary. As shown by the benchmark even that doesn't do much for Flight Simulator.

Actually FSX is more about a fast CPU. Get a fast c2d or c2q.

Q6600, after July 22. If that's still not enough overclock it. Good luck  

I think it's more of a GPU problem rather than a CPU cuz tomshardware system specs are one of the highest for testing those cards.
 
 
Thanks for the replies!

You need a better CPU more then anything. This game does as it was already mentioned rely more on the CPU then the VGA card.
James

 
stupid game is single-threaded only, maybe with the patch it will support more cores, it's a great game but the optimization is pure crap, they should learn more from cry-tek or Valve.

A GF7600 won't run the game anywhere near max, and neither will a single GF8800 or HD2900, the game is a beast.
 
It's stil primarily CPU bound, but change the settings enough and you kill performance for the added visuals.

Personally I'd side with an HD2900 setup based on the dev's comments and the nature of the visuals in the game, but I think if you get either a GTS or an XT you'll get yourself on your way to deciding whether adding a second card for more power is worth it to you or not.
 
Really once you reach the upper level of visuals you will then start to be more CPU bound, but you need to get beyond that initial power level to reach a good baseline.
 
Expec the DX10 patch to change performance characteristics again when some of the effort can be offloaded off the CPU.

 
No it's not. SP1 already added multi-core/thread support.

 
I have to disagree here.  Comparing cards running fsx is really a matter of splitting hairs but HardOCP did  a thorough review and did better with both the gts and the gtx over the xt.  They were able to run higher quality settings with both Nvidia cards.  
http://enthusiast.hardocp.com/article.html?art=MTM0MSwxMSwsaGVudGh1c2lhc3Q=
 
FSX is one title where I really notice higher AA.  The lines of an aircraft against a blue sky can't be looking all jagged.  
The other problem with the 2900 in those tests is the minimum fps it hits.  With fsx you are already brushing up against unacceptable frame rates so minimums are more important than averages.
 
Of course this could all change when the dx10 patch is released but until then none of us know.

 
I have to disagree here.  Comparing cards running fsx is really a matter of splitting hairs but HardOCP did  a thorough review and did better with both the gts and the gtx over the xt.  They were able to run higher quality settings with both Nvidia cards.  
http://enthusiast.hardocp.com/article.html?art=MTM0MSwxMSwsaGVudGh1c2lhc3Q=
 
I understand what you're saying, but I'm not talking about right now exclusively, I'm talking about DX10. And right now all three cards are equal except for the intial weak AA for the Ati cards (do you still think the GTS has as much of an AA lead as it did when that article was written? do you think nV has improved their geometry and vertex performance since then?).
 
Like I said, I'd bet on the HD2900 based on the dev's comments, but any of the 3 will do fine for now.
 
BTW, something I never figured out, but maybe you could enlighten me, because I often find myself lost for words when it comes to [H]'s recent reviews posts.  
Just how did increasing resolution from 1600x1200 to 1920x1200 with the same settings on the HD2900 (and the other two) INCREASE it's performance? Just something I found interesting, and I'd love to know the answer to  to understand how lower setting and performance is 'max playable' by their benchmarking standard? Is it that they don't have a 1920x1440 or above monitor to check above 16x10 and see what's what?  
 
And if in their Apples to Apples 8XQ was actually the same performance as the 8XCSAA, then why not enable it instead of claiming it? Especially after opening with this statement in the AA portion of the review;
"Looking very closely at 8X AA it appears that ATI’s 8X MSAA is superior in image quality to NVIDIA’s 8X CSAA mode. But when you look down and compare NVIDIA’s 8xQ mode (which is true 8X MSAA) the image quality is comparable. "
 
Is it just me or are we no longer talking about apples to apples anymore?
 

 
I agree, minimum is important, but not as important in FSX as it is in a first person shooter. The difference of a few frames at the low end in what otherwise looks like identical hystograms doesn't convince me that the difference is that great. As for AA I wouldn't trust [H]'s comments on what they could and couldn't do, considering what I mention above.  
 

 
That's true but based on the type of DX10 exploits that FSX will focus on (vertex and geometry) and the areas that the HD2900 is strongest at (complex vertex and geometry) and what that will help offload (the CPU), I'm favouring the HD2900 in this one. I could be wrong, but like I said both the GTS and XT are good, I just prefer the HD2900, and I think if you think about the future of FSX, you might understand why I do even if you don't agree, regardless of what [H] wrote in their review of the launch contender.

 
No it's not. SP1 already added multi-core/thread support.
 
yay great didn't know that, i always loved the flight simulator games now with the usage of the both cores maybe i will see a better framerate.

make it simple, make it 8800 GTS 640MB!

For FSX I'd take an HD2900XT over a GF8800GTS-640 if I was planning on keeping the card into the DX10 patch lifespan.

This is just another case of of M$ picking the wrong dog to run with.  They have back themselves in a corner just as the did with HD DVD.  Despite Sony's involvement Blu-Ray will crush HD's inferior technology in the coming years.  FSX is the same way rather coding the most efficient way they favored ATI method which hasn't lived up to expectations.  I personally don't care if it's in the DX10 spec or not good coding is all about efficiency which something M$ is incapable of.

 
What are you talking about? It hasnothing to do with favouring ATi or nV in coding, it has to do with playing to the strengths of what your applications needs to do. Flight Sims have huge vertex/geometry loads, and that's just it. Whether it were ATi, intel, nV, S3 or SIS who performed better at that, that would be the horse to go with for this one game because that's what this game will require most. It will also be somewhat texture heavy depending on situation so there may be lots of offset benefits, but the area which will benefit the CPU load from DX10 will be bother geometry/vertex and efficient materials handling (which will reduce texture loads). So really once the game arrives we'll know the differences, but unless M$ actually did the opposite of what you say and coded for nV's hardware then the benefits should favour the HD2900, although this might be one of the very few titles that that would be the case.
 

 
I wouldn't disgree there based on their initial issues with the CPU side of their house, but it's inefficient for both companies, so the architectural benefits will have to work out which can overcome the bottlenecks the best.

I've seen the newer reviews and agree that ATI has made improvements but there is still weakness with AA.  Here is Tweaktowns summary of the 7.6 drivers with regards to AA.

http://www.tweaktown.com/articles/1123/7/page_7_benchmarks_high_quality_aa_and_af/index.html
These guys have been very kind to ATI in their reviews.  Heck they even put the xt up against the Ultra and concluded it was a better buy; no f'n kidding.  
 

 
Will in the case of the gtx they did reduce the settings but overall I can only guess it is because the game is so CPU limited.  If you look right here at Tom's VGA charts you see