i7 at 4Ghz on eVGA MB = Stutters ?

[Bob Scott 7,715]

Driver 182.50 is the FSX golden driver right now for most system using the GTX 275, 280 and 285 video card products

This is an important point right now...the last few major nVidia driver releases are simply awful for FSX...major stuttering without remedy. I haven't seen any mention of what version is in use...I am using 182.52 right now on my 285GTX. Most of the video cards on the shelves have a driver version included that's a disaster in FSX. Get the 182.50 drivers from www.guru3d.com or the nVidia site.0Artur0, with your CPU at 3.2 GHz, you should run 1600 MHz RAM (160 BCLK with a 10x memory multiplier), CAS 6 is best, CAS 7 is acceptable, CAS 8 and up will slow you down.RegardsBob ScottColonel, USAF (ret)ATP IMEL Gulfstream II-III-IV-VColorado Springs, CO

[Tomaz Drnovsek 440]

Thank you for the info about RAM.My driver version is 190.38 64bit Windows 7.. so you say I should uninstall those and install older version (182.50 for Vista)?Edit:OK... I opened the case and checked the RAM... it says: MUSHKIN 3x2GB HP3-12800 1.65V 8-9-8-20... is this the poor RAM you were talking about? I don't know why Easytune utility said it's Samsung something something (screenshot a few posts above)? I'm confused. I have time till tuesday to return RAM or exchange it. Please help!

[Guest Nick_N]

Thank you for the info about RAM.My driver version is 190.38 64bit Windows 7.. so you say I should uninstall those and install older version (182.50 for Vista)?Edit:OK... I opened the case and checked the RAM... it says: MUSHKIN 3x2GB HP3-12800 1.65V 8-9-8-20... is this the poor RAM you were talking about? I don't know why Easytune utility said it's Samsung something something (screenshot a few posts above)? I'm confused. I have time till tuesday to return RAM or exchange it. Please help!

its the memory SPEED and the memory TIMING combination thats killing the performanceYou are running 1066 @ 8-9-8The absolute highest timing I can see 1066 running with results is 5-5-5 which is a heck of a lot faster than 8-9-8Even so.. at CAS 5-5-5 that 1066 memory speed is still not allowing everything i7 can deliver to the application, therefore as Bob pointed out, if you want the RIGHT memory speed and timing for your system which will in fact deliver the highest perf to the application, then you need DDR3 1600 CAS6 memory and you need to set it up in the BIOS so it runs the correct timinghttp://www.newegg.com/Product/Product.aspx...N82E16820226050which is 6-7-6-18 timingNow.. thats just part 1. Part 2 is getting a video card that will run with that i7 and memory correctlyAnd yes, 182.50 is THE driver to use for the GTX series cards.. most find 182.50 the golden FSX driver.. how it will work in W7 I dont knowW7 is not what I consider a OS to move to for FSX performance.. not only is it still in its testing phase drivers are just now being developed for it. Regardless I have gotten at least 40 people OFF Windows7 and back to either Vista64 or XP x64 for FSX in the last 6 months due to perf issues. As far as I am concerned, until W7 is out and has been allowed the proper time for driver and other development I dont see ANY reason to use it, at all. The forums are full of hyped horse manure about W7. The OS will not in any way, shape or form run FSX faster or smoother than XP x65 or Vista64if someone says its does.. their old OS/FSX install was not right, period

[Guest Nick_N]

I know people do not understand memory speed and timingThey see the difference between the timing values 5-5-5 and 8-9-8 and assume.. hey, its only a difference of 3, so it cant be that bad or they read some story at a hardware site about how memory speed/timing does not matter, which is total hogwash!Well, in order to help people understand it without going into engineering, here is an analogy that puts this issue in perspective for non-tech individualsThe speed limit on a highway is 65MPH To FSX:Running 1066 @ 8-9-8 is like driving 30MPH on that highwayRunning 1066 @ 5-5-5 is like driving 50MPH on that highwayRunning 1333 @ 5-5-5 is like driving 65MPH on that highwayRunning 1600 @ 6-7-6 is like driving 75MPH on that highwayRunning 1700-1740 @ 6-6-6 like driving 85MPH on that highwayRunning 2000 @ 7-7-7 is the SAME as running 1700 @ 6-6-6 so if high speed DDR3 2000 memory is rated at 8-8-8 its SLOWER than DDR3 1600 @ 6-7-6!The higher the memory SPEED and the lower the memory TIMING the better the perf result as long as you understand the SPEED/TIMING rulesYour targets for best performance per speed are:DDR2 800 3-3-3DDR2/3 1066 4-4-4DDR3 1333 5-5-5 DDR3 1600 6-6-6 DDR3 2000 7-7-7

[Tomaz Drnovsek 440]

Thank you for your help!One more thing...Also, with the latest nvidia drivers, get rid of nhancer and all the all tweaks...Is this true?

[Guest Nick_N]

Thank you for your help!One more thing...Also, with the latest nvidia drivers, get rid of nhancer and all the all tweaks...Is this true?

Not for me it isnt... still use Nhancer (2.5.1) and driver 182.50 for i7/GTX 285 and FSXI would use the same setup I posted in the tuning thread Nhancer is about visual quality.. Some people cant see the forest from the trees Some can: http://forums1.avsim.net/index.php?showtopic=257161

[Jim Padgett 11]

I started reading this thread with doom and gloom and then remembered my experience with memory.........It was long ago and I can't quote actual numbers... but I remember it this way..I upgraded my memory from 90ns memory to something like 9ns memory and could not see any real difference in performance... I'm sure benchmark scores would showa big difference but in the day to day use of my PC I saw no real difference. Now moving from 5400 rpm hard drive to a 7200... was a big obvious difference.I related the memory speed to something like increasing you MPG on your car... If I upped MPG from 10 to 15 MPG get a 50% increase, but If I go up it 10 mpg from 30 mpg the same increaseis now only 25% ... I feel memory timings and such are much the same...If your already going so fast it is hard to comprehend the actual speed then the poor old human will never really notice it.I have the slow as molasses in January memory CS 9 stuff.... but this review doesn't show any real difference in actual performance.. http://www.overclockersclub.com/reviews/cr...i_channel/5.htmNow does this mean you won't see a difference in FSX? I don't know, but my gut feeling is that it is not the bottleneck it is supposed to be.

[Tomaz Drnovsek 440]

These tests show little or no difference in performance. I don't know how this applies to FSX.

[Jim Padgett 11]

I don't know... and it just seems to be that shooting for $$$$ ram is not the best bang for the buck unless you are shooting at performance numbers..Lets face it... the number 4, 5, or 6 olympic competitor may not have won, but they are still world class competitors in their own right.I just can't see where the timings involved in any memory that is running in the Ghz range can really matter that much for a simulation running the frames per second FSX is.So even if you had FSX maxxed out and running at 60 frames per second.. your worried about memory timings in the Giga Hertz ranges having a real affect?I'm not disputing those who know more than I do, but I just don't see it making THAT much of a difference...I would love to see some numbers comparing FSX with different memory timings and speeds....

[Tomaz Drnovsek 440]

I would love to see some numbers comparing FSX with different memory timings and speeds....

Me too!

[tfm 169]

Me too!

Did you see this? Note that there is an update within the thread at post #5, showing a further result with even faster RAM settings:http://forums1.avsim.net/index.php?showtopic=255816Since then, I have also done a test with the i7 975 at stock settings and with RAM at 1333MHz: the average FPS was only about 26fps, whereas the lowest fps with an overclock to 4.4GHz was well into the 30s. Personally - and I think those numbers will help to illustrate why - I do not subscribe to the theory that overclocking delivers only marginal benefits. Compared with the performance of a 4 or 5 year old PC, the default settings of an i7 975 are certainly impressive, assuming the CPU is combined with a decent GPU. At stock speeds you should be fine, if you just want to use the default hang glider or the Cessna and if you just have the default scenery. But if you want to use the "heavy" add-ons with high settings in demanding situations (ie, with complex third party scenery and weather) then you stand to gain considerably from overclocking - provided, of course, that all the components are "in balance" so that nothing is unduly bottlenecking anything else. The extra few frames per second when on short finals can make the difference between a tolerably smooth and immersive experience and an intolerably stuttery and jarring one.Tim

[Guest Nick_N]

Did you see this? Note that there is an update within the thread at post #5, showing a further result with even faster RAM settings:http://forums1.avsim.net/index.php?showtopic=255816Since then, I have also done a test with the i7 975 at stock settings and with RAM at 1333MHz: the average FPS was only about 26fps, whereas the lowest fps with an overclock to 4.4GHz was well into the 30s. Personally - and I think those numbers will help to illustrate why - I do not subscribe to the theory that overclocking delivers only marginal benefits. Compared with the performance of a 4 or 5 year old PC, the default settings of an i7 975 are certainly impressive, assuming the CPU is combined with a decent GPU. At stock speeds you should be fine, if you just want to use the default hang glider or the Cessna and if you just have the default scenery. But if you want to use the "heavy" add-ons with high settings in demanding situations (ie, with complex third party scenery and weather) then you stand to gain considerably from overclocking - provided, of course, that all the components are "in balance" so that nothing is unduly bottlenecking anything else. The extra few frames per second when on short finals can make the difference between a tolerably smooth and immersive experience and an intolerably stuttery and jarring one.Tim

People like Bob Scott and myself who have been around the electronics game since Apollo understand why there is a difference and why a title like FSX FEEDS on that differenceOthers who claim they are technically trained (and they may be to some extent) do not understand the underlying science and engineering behind memory speed/timing and its relation to the entire system or the application in play. Hardware sites live for controversy.. its their bloodline for internet traffic and I have found more often than not if they are not feeding a line of controversial BS they have simply made a mistake in assuming their test bed was correctly configured and the software used is telling them everything. That assumption trickles down to the reader as well. Some people can see the forest from the trees, you happen to be one of them Tim :(

[tfm 169]

...Some people can see the forest from the trees ...

On the subject of arboriculture, as far as I'm concerned, there's a solitary FSX tree in the thick of a forest of other applications (usually modern DX9 or DX10 games) by which review sites judge the usefulness of memory speeds. I'm only interested in FSX; and for FSX, my own experience as well as the experiments I posted in the other thread have satisfied me that the RAM does make a difference on my kit. Obviously (as I said in the other post) whether it's a worthwhile difference or not is a matter of personal opinion.To take one example: I got a copy of the HAWX game free with a video card recently and ran the built-in benchmark on it using different RAM settings. The results - which were pretty counter-intuitive in some respects - did not demonstrate any clear relationship with RAM speed. But that only tells me that HAWX, no doubt like many other modern games, is less sensitive to RAM speed than FSX. So what?I wish more people would actually run some tests with FSX on their own PCs and share the results, so that people interested in getting an extra boost from their kit can get some realistic idea of what's in store for them depending on what particular component(s) they change. Obviously, not all tests will be equally useful, because differences in method will affect the weight that different tests should receive. But we can all judge that for ourselves and improve the process with a little friendly feedback: that's what the forums are good for. It's no good leaving it to the review sites: they're not interested in FSX and besides I agree with you that it is difficult to be confident about the neutrality of much of what's written on them.BTW I hope your summer's going well: I was on your side of the pond last week and had a great time in Virginia and Washington DC - too hot for the locals apparently but my sun-deprived family lapped it up.Tim

[AlphaInfinity 13]

Has anybody asked this guy for his FSX settings? I don't care how fast the machine is, with incorrect scenery, cloud draw distance, FSX.cfg settings, it will make ANY hardware out there stutter right now. Don't forget those AI settings! NickN has a great thread out there for optimizing your settings. You may want to take a hard look at that thread(if you havn't already) and make sure you play with those sliders in FSX, make the suggested changes to the FSX config and even give Nhancer a try. And these guys are absolutely correct about the newer Nvidia drivers. I had to revert back to the 182 drivers after giving the 190x drivers a try. They truly do reduce performance in FSX. I did that and now I have pretty smooth flight(except around large airports).I am not saying that hardware isn't the issue. I am saying that you may want to try changing the gasoline and increasing the tire pressure before you open the hood to do an engine rebuild.