Hello all With the new PMDG 737 about to be released I think I may need to upgrade my old core2duo PC to a nice new shiny i7-2600k. What I will be replacing is the MB with most likely an Asus P867-Pro. The CPU with an i7-2600k and a cooler to match. I should be able to overclock with these MB and CPU? I've never really tried OC before so it is mostly a mystery to me at the moment. The GPU with an NVidia 570 and the memory. The problem is, I'm not really sure what sort of memory I need. All those clock speeds and numbers just confuse the hell out of me so if anyone with a bit of knowledge could recommend some sticks that would be a great help. I figure I will be needing 8GB of memory. The case (an Antec 900) and PSU will remain and I already have a new 1TB HD to put in. The OS will be W7 64bit. Thanks in advance for any advice.

Hey Ant-I cant help with you with all the RAM and stuff, but if you want to save some $ go for the i5 2500k, unless you REALLY!!! want hyperthreading (which FSX doesnt even use). Best of luck with the system though.

Any mid range dual channel 1600Mhz CAS 8 or 7 kits will do just fine. Preferably 1.5V but regular 1.65V will work okay. One big debate these days is rather or not 1.65V DRAM is healthy for SB processors. For me, yes.

http://forum.avsim.net/topic/340377-choosing-the-best-ddr3-for-sandy-bridge/

In all honesty, it makes very little difference. Dazz did some FSX specific testing a few months ago that showed maybe 2 to 3 percent gain from fairly low-end to high-end memory. With memory you want more MHz and lower timings. The first number in the sequence, #-#-#-##, is the most important and is often referred to as CAS latency or CL. Anyway, two really common sets of memory are: 2x4GB 1600 (7-8-7-24), often referred to as 1600 CL7or2x4GB 2133 (9-11-10-28), often referred to as 2133 CL9 Performance-wise, I would recommend 2133 CL9. Price-wise, the extra money may be better spent towards something else such as cooling. So it's up to you. Regarding 2x4GB vs 4x2GB... 2x4GB = two 4GB modules, so only two sticks of RAM. 4x2GB = four 2GB modules, or four sticks of RAM. Four sticks is generally going to require more voltage than two sticks and it may even be unable to run at rated specs unless you buy it specifically as a set of four sticks. Just stick with 2x4GB and you will avoid a lot of potential hassle. Mushkin Ridgeback 2133 (9-11-10-28)Mushkin Ridgeline 2133 (9-11-10-28)GSkill Ripjaws X 2133 (9-11-9-28) P.S. I also agree with going the 2500k route instead of the 2600k

Thanks guys, that's some handy information there. Special thanks to Corey for his explanation. i will look into the various prices of the sticks Corey mentioned. On the matter of the 2500k versus 2600k I had read some posts about how the 2600k is probably overkill for FSX but I guess I'm trying to future proof myself a bit here (as much as you can future proof yourself with PCs). If MS Flight gets released sometime next year then this machine will be the one to run MS Flight and with any luck MS Flight may have been built with hyperthreading in mind.

Thanks guys, that's some handy information there. Special thanks to Corey for his explanation. i will look into the various prices of the sticks Corey mentioned. On the matter of the 2500k versus 2600k I had read some posts about how the 2600k is probably overkill for FSX but I guess I'm trying to future proof myself a bit here (as much as you can future proof yourself with PCs). If MS Flight gets released sometime next year then this machine will be the one to run MS Flight and with any luck MS Flight may have been built with hyperthreading in mind.

With any luck Flight will be GPU bound and it won't really matter too much if you have a 2600K or a Q6600 Anyhow, you're going to love your new rig Anthony.

If you can afford it the I'd recommend you definitely buy the 2600k. I've tested both the i7 and i5 directly in the same system and found the i7 to provide the better overall experience. FPS difference is negligible, but I found the i7 provided a slightly smoother flight in dense areas of scenery, probably down to the extra cache.

If you can afford it the I'd recommend you definitely buy the 2600k. I've tested both the i7 and i5 directly in the same system and found the i7 to provide the better overall experience. FPS difference is negligible, but I found the i7 provided a slightly smoother flight in dense areas of scenery, probably down to the extra cache.

Same clockspeed on both chips? Do you empirical data which demonstrates your claim? Because the extra cache hasn't been demonstrated to make a meaningful difference in anyone else's testing in the FSXMark11 thread, nor in any other consumer application tested in a review yet seen on the web...

I'm not making any "claim", rather an observation, so just take that for what it is. As I said, there's was no real framerate difference, but the simulator did operate in a smoother fashion in high load areas. I'm not sure how you propose I quantify or measure this? The FSXMark11 thread is totally meaningless for comparison for two reasons. Firstly because it's run on a default sim, meaning that the hardware isn't pushed to it's limits at all. It's only when reaching the limits that any extra performance can become apparent. Secondly because the comparisons drawn are taken from varying systems, which puts far too many variables into the mix to make a direct comparison of only the two processors. My comparison was made on the same system, on the same installation of FSX. Can't get more direct than that. Just out of interest, do you understand how a processor works? If so then you should realise that any extra cache will most certainly be utilised in a program with a poorly optimised memory subsystem like FSX. Like I said though, the difference is marginal but certainly noticeable in certain situations. Nothing that I can make a nice chart with for everyone to look at, rather you'd have to go out and buy both processors to see the differences for yourselves.

I'm not making any "claim", rather an observation, so just take that for what it is. As I said, there's was no real framerate difference, but the simulator did operate in a smoother fashion in high load areas. I'm not sure how you propose I quantify or measure this? The FSXMark11 thread is totally meaningless for comparison for two reasons. Firstly because it's run on a default sim, meaning that the hardware isn't pushed to it's limits at all. It's only when reaching the limits that any extra performance can become apparent. Secondly because the comparisons drawn are taken from varying systems, which puts far too many variables into the mix to make a direct comparison of only the two processors. My comparison was made on the same system, on the same installation of FSX. Can't get more direct than that.

Right, so it's a feeling, not a fact.

Just out of interest, do you understand how a processor works? If so then you should realise that any extra cache will most certainly be utilised in a program with a poorly optimised memory subsystem like FSX.

I sure do. And the engineers at Intel and AMD do far better than you or I, and there are many publicly available documents from these fine folks which state that increases in cache size beyond a certain level offer little to no performance gains across the vast majority of applications.

Like I said though, the difference is marginal but certainly noticeable in certain situations. Nothing that I can make a nice chart with for everyone to look at, rather you'd have to go out and buy both processors to see the differences for yourselves.

I have no problem with your preference in processors, I just can't agree with your assertion that larger cache size has any factor on FS performance with current processors, all else being equal.

That's fine you're entitled to your opinion, albeit a very uninformed one given you haven't made the comparison yourself. I'd love to have this conversation again should you ever get your hands on both chips.

That's fine you're entitled to your opinion, albeit a very uninformed one given you haven't made the comparison yourself. I'd love to have this conversation again should you ever get your hands on both chips.

You haven't offered anything in the way of evidence to support your *opinion*, and everything that is available contradicts you both in theory and in practice. You don't have any ground to stand upon.

Listen, I have no interest in proving anything to you. I made a comment based on my opinion of something, and it was you who decided to take it upon yourself to tear it apart and analyse it, demanding proof. Nobody needs to provide evidence of an opinion by definition, and anyway for the second time, how is it that you would like me to quantify a smoother running simulator?! Again, take the comment for what it is and nothing more, and come back to me when you at least have first hand experience of these processors. As far as I know I'm the only person here that has done this direct comparison, therefore it's breathtakingly arrogant to be so certain in your self appointed judgement. Either way, I couldn't care less, my opinion is still there for anyone who chooses to listen or not.

Listen, I have no interest in proving anything to you. I made a comment based on my opinion of something, and it was you who decided to take it upon yourself to tear it apart and analyse it, demanding proof. Again, take the comment for what it is and nothing more, and come back to me when you at least have first hand experience of these processors instead of acting like a massive know it all. As far as I know I'm the only person here that has done this direct comparison, therefore it's breathtakingly arrogant to by so certain in your judgement. Either way, I couldn't care less, my opinion is still there for anyone who chooses to listen or not.

I wouldn't have cared in the first place if this were simply an opinion thread, but you are offering purchase advice so it is entirely worthwhile for the original poster to have a counter-opinion.