You can see my current video card in my sig, all monetary concerns aside what benefits would I see from upgrading to a 640mb 8800gts? Mainly I would like to be able to use lightbloom and max water textures, and see less blurries in the distance, everything else is maxed out and I am getting good fps. I also have to lower my aa setting to use 1680x1050 resolution, I would like to get better fps with aa on using this resolution.

You would enjoy that card then....it should allow increased AA and AF at higher resolutions than your current 7950...That said, I do not believe it will help much with light bloom and increased water effects. Although people say bloom and water effects are gpu oriented, they still require TWO passes from the CPU to run them....Don't get me wrong, I think the 8800 would help a little with the bloom and water, but not nearly as much as getting a faster CPU...And I'd definitely go for the 640mb model.One thing though, I was recently reading at Toms Hardware and all the VGA charts showed that the 320 vs the 640mb version didn't have a lot of difference in FPS/etc in various games. I don't know if you want to spend the extra 100 bucks - like I said, I would, but I'm starting to rethink my thinking (LOL wordiness) based on the VGA chartsGood luck!

http://www23.tomshardware.com/graphics_200...2=707&chart=275Here's the chartsThe extra memory is really only crucial if you're running high res or multiple monitorsJim

If people would only do some research on THG, people would really steer away from their so called "trustworthy" reviews. :-roll

What, you are saying Tom's isn't well-respected????also, if you recommend another site or link why not post it?????

Regarding the card I just plopped into my rig....I also read Tom's article on the ATIX1950 Pro...and from what I am experiencing first hand in performance PROMISED by his review of the card...TOM'S HARDWARE was totally on the money.My two cents...and a few found FPS thrown in...lol.Mitch

Ryan,THG in the past has done some things that make it unable to have give them a credible name nowadays, with not going into details. just avoid it is my best advice.If you need credible reviews there are plenty of others where www.anandtech.com is one of them.But again to each its own, If you trust them thats fine... From me its just a view from someone who actually works in the IT business.Im not telling anyone to what to do, just to give some helpful insider tips.

Some of you might be interested in the article from the front page of hardocp which compares two XFX 8600-series cards (a factory-overclocked GT and a factory-overclocked GTS) against a Radeon HD 2600 XT. All are 256MB cards. The GeForce cards have GDDR3 memory, while the Radeon uses GDDR4. Hardocp initially tested in FSX at 1280x1024. None of the cards could reach "playable" frame rates with AA turned on. With AA turned off, the overclocked GT achieved an average fps of 37.8 (with 24 fps as the minimum, and 62 as the maximum). The GTS achieved an average of 39.1 fps with 20 and 68 for the minimum and maximum, respectively. The HD2600XT was unable to achieve "playable" frame rates at 1280x1024, so the tests were done at 1024x768 with AA disabled. At this lower resolution, the card achieved an average fps of 35.6, with 20 and 63 for the minimum and maximum. The tests made it very clear that the HD2600XT is far from an ideal choice for FSX users. The reviewers were able to further overclock the GT card to 690MHz core and 1720MHz memory (versus the "factory overclock" of 620/1600 at which the above tests were done). Unfortunately, they didn't report FSX results at the highest overclock. Too bad, because the effect on fps would have been interesting to know. It's also a shame they didn't do any comparisons with cards that have 512MB of video RAM. Incidentally, the CPU in these tests was an E6600 running at stock 2.4GHz. Unlike the so-called "reviewers" at flightsim-dot-com, the hardocp folks were knowledgeable enough to use FSX with the SP1 patch applied. It seems you can get quite good results with a 256MB video card costing a little over $100, if you're willing to live without AA.Here is a direct link to the article in case anyone wants to take a look-see: http://enthusiast.hardocp.com/article.html...W50aHVzaWFzdA==It's nice to see a "hardcore gamer" site like hardocp using FSX in its comparison tests, and apparently doing a good job of it. Personally, I would trust hardocp and anandtech far more than I would ever trust tomshardware.

>>>I would trust hardocp and anandtech far more than I would ever trust tomshardware. Wise words :) thanks for adding your post.

I should have also added that I would not, under any circumstances, trust a "hardware" review by flightsim-dot-com. These were the guys who kept pushing Pentium 4 -based machines, when every other INDEPENDENT review showed significant performance advantages (at the time) by the Athlon64. The so-called "reviewer" at flightsim, "Capt. Mason," actually claimed that Athlon64 performance was irrelevant because there was no 64-bit operating system that could run it! (Note: flightsim had an exclusive marketing deal with Dell at the time, and those were they days when Dell refused to offer any AMD chips. Conflict of interest, much?) At the same time, the capable folks over at SimHQ were running rigorous comparison tests which showed HUGE performance gains in FS for the Athlon64, using 32-bit WinXP. As much as I dislike tomshardware, I dislike flightsim even more. They deliberately misled their readers and proved to me that they're not trustworthy, at least not when it comes to hardware reviews.

Words of wisdom! :-) Sooo true... It's OK to look at THG and other 'trustworthy' sources, IMHO, just don't make purchasing decisions (solely) based on their infos. Besides, THG critics need to read their 'bias' too, otherwise it would be an opinion based on nothing. Two THG goof-ups for example: They suppressed information that GF8800s are 'only' DX10.0. If their intention was to inform, why would they withhold this info? Or, 2-3 weeks before Intel's new C2Ds appeared - THG obviously isn't aware of Intel's roadmap ;-) - they ran an article about an insane P4 overclock. Those who knew the roadmap must have been kind of surprised to see this 'article' or 'review'. @ Mike, with the resolution you're running, you should notice a nice little difference between your current GPU and the one in spe. Just keep in mind it isn't extremely future proof. Also, that it snaps another 100-odd MB from your system RAM. So, if, the best experience will likely result from adding 1-2GB RAM at the same time. AIM to you earlier, I think 2GB is history for an enjoyable & hi-end FS experience. 2GB was good for FS9. ;-) I'll attach a shot of Vista32 with 3GB, look how much free physical memory is left. Cheers and kind regards Jaap

Thanks jaap. I did go ahead and buy 2 more gb of ram earlier today, so maybe that will help with blurries a little bit too? My main reasons for thinking of getting a new gpu would be to lessen the blurries and increase the visual quality without changing fsx settings much because autogen and pretty much everything else is maxed out and I am getting good fps. The only things turned off are lightbloom and water is only at low 2x, thats really what kills fps for me. My fps are not really a concern because my cpu seems to be handling it well, but I would just like to increase the overall visual quality. Also if got a new gpu now I would be able to get a fair amount still for my 7950 on ebay, but later down the road people arent gonna want a new dx9 gpu. But then again when fsx gets dx10 compatiblity it will be dx10.1 right?

>But then again when fsx gets dx10 compatiblity it will be dx10.1 right?Not necessarily, and I higly doubt it because it would require massive recoding. DX10.1 is an incremental update and to this point Microsoft has been working on a DX10 update and contrary to the scare purported via the internet, DX10 cards will still work with DX10.1 just not some of its new functions.

hahah I remember that...I think I also posted about that in the forums and so did others....I believe it was locked hahahI remember thinking , "wow the why don't they ever advertise AMD..." and at the time the AMD chips were serving the Intel chips on a silver platter lol....Yeah, what I should add is what others kinda said, don't buy based on one review...shop around/research...I, personally, haven't had any problem with THG.... but then again I'm not a uber l33t guru either :)

Indeed, AFAIK, an upcoming title (HalfLife2?) will use 10.1 for a couple of face mimics... I doubt those without the moving faces will miss much. Maybe I should have added that those measurements show how much FSX consumes in dense areas like NY, SF, LA, etc. At medium to high settings. Today's my screenshot day :-) so I'll attach a Taskmanager screenie* from XP64 with a quad. If you look at the previous screenshot with a twin, you'll see that peaks are shorter on CPUs 2-4 than a single secondary core on a twin. As I would dare to judge, this seems to pay back in terms of smoothness. Performance in terms of fps OTOH, will still mainly come from the fastest core. @ Mike, some blurries seem to be a FSX-feature. Type3 is apparently by design. ;-) Adjusting the zoom to .8 or .7 cures this type of blurries for me. It's a pity there aren't 1.5GB modules! Most FSX users would be fine with 3GB. Kind regards Jaap * shot was taken after 2 minutes into FS, therefore some RAM readings are still fairly low (system cache for example).