Not sure if there is any way to improve this, but I am using trilinear filtering in game and 16xAF in the video driver. Water reflection just looks so marginal compared to games like Far Cry & Crysis. Any way to improve the shimmering in dense forest areas in the NW and also with water reflection? Pretty marginal for this day and age. I don't have this issue in FS9.NoelQX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed

Maybe MipBias=4 setting in fsx.cfg?

Also make sure anisotropic optimization is off. Note only turn off the optimization option, not the filtering itself!!

Thank you Tom and Geofa. Are you talking about in the nV driver regarding "anisotropic optimization?" Also, I have SP1 only installed. Do you recommend SP2 for XP users?QX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed

I run Sp2/AcellerationAs for shimmering and the best FSX visual resultsInstall the latest beta of Nhancer and set it up as I laid out here=============================================Install and run the latest version of Nhancer. It does NOT need to boot with Windows or have a tray icon since the service runs in the background and will properly influence the settings. When it is first run select NO to have it run with Windows or have it place an icon on the tray. With any driver above 163.xx Nhancer can no longer delete the profiles as it did in the past, instead the way you OVERRIDE the FSX and FS9 profiles is do the following: NOTE: you can see IMAGES of the following settings at the BOTTOM of this thread: http://www.simforums.com/forums/forum_post...?TID=25186&PN=1NOTE: With the Nhancer latest BETA version (which I also use now).. PROFILE DUPLICATE is NO LONGER REQUIRED http://forum.nhancer.com/showthread.php?t=269NOTE: WITH THE NEW BETA, FOR VISTA, DURING NHANCER INSTALL ALLOW THE SERVICE TO INSTALL WHICH IS NEEDED TO OVERRIDE UAC. WITH XP YOU CAN DISABLE THAT DURING NHANCER INSTALLa. Scroll down the profile list until you find the Microsoft Flight Simulator 2004 (or) Microsoft Flight Simulator X profile. Click on it once to highlight it. Once it is highlighted, click the DUPLICATE button. You will get a message, select YES to complete the duplication. What that will do is DEACTIVATE the Nvidia profile and ACTIVATE your new duplicate. b. With the new profile in the list, make sure there is a bullet in PROFILE on the right. Go down the list on the right and select the following: AA: For FSX select SUPERSAMPLING and 8xS (use 8xS Steve)NOTE: Slower cards and systems in FSX, set to MUTISAMPLING and 4X or SUPERSAMPLING and 4xS NOTE: FOR THE BETA VERSION THIS HAS CHANGED... SET AA TO COMBINED 8xSUnder the next box, check GAMMA CORRECTION DO NOT ENABLE THE SECOND SUPERSAMPLE FEATURE IN THIS AREA UNCHECK "ENHANCED AA" if it is checked by defaultAF: CHECKED Bullet in: 16X VERTICAL SYNC: CHECKED, Bullet in: FORCE ONOPTIMIZATIONS Tab: Texture Filtering CHECKED Bullet in - High Quality Trilinear Opt CHECKED, Bullet in - Off Aniso. sample opt. CHECKED, Bullet in - Off Negative LOD BIAS, CHECKED, Bullet in - Clamp Check the next three items but do not change their values COMPATABILITY TAB: Place a CHECK in all header values but only change the following if needed: Force Mip-Maps = NONE Conformant Texture Clamp = ON Hi Dynamic Range HDR (Enabled)All others should be at their default values. Under the CPU Multicore Support, REGARDLESS OF THE TYPE OF CPU YOU USE: FOR WINDOWSXP SET CPU TO SINGLE FOR WINDOWS VISTA SET CPU TO MULTICORE Repeat the above for any other game profile as you may need it. c. In the main Nhancer Window, set the bullet to GLOBAL Set the SAME settings on GLOBAL as you did above. Close Nhancer and reboot. ============================

>I run Sp2/Acelleration>>>As for shimmering and the best FSX visual results>>Install the latest beta of Nhancer and set it up as I laid out>here>>=============================================>>Install and run the latest version of Nhancer. It does NOT>need to boot with Windows or have a tray icon since the>service runs in the background and will properly influence the>settings. When it is first run select NO to have it run with>Windows or have it place an icon on the tray. >>>>With any driver above 163.xx Nhancer can no longer delete the>profiles as it did in the past, instead the way you OVERRIDE>the FSX and FS9 profiles is do the following: >>NOTE: you can see IMAGES of the following settings at the>BOTTOM of this thread:>http://www.simforums.com/forums/forum_post...?TID=25186&PN=1>>NOTE: With the Nhancer latest BETA version (which I also use>now).. PROFILE DUPLICATE is NO LONGER REQUIRED>http://forum.nhancer.com/showthread.php?t=269>>NOTE: WITH THE NEW BETA, FOR VISTA, DURING NHANCER INSTALL>ALLOW THE SERVICE TO INSTALL WHICH IS NEEDED TO OVERRIDE UAC.>WITH XP YOU CAN DISABLE THAT DURING NHANCER INSTALL>>>a. >Scroll down the profile list until you find the Microsoft>Flight Simulator 2004 (or) Microsoft Flight Simulator X>profile. Click on it once to highlight it. Once it is>highlighted, click the DUPLICATE button. You will get a>message, select YES to complete the duplication. What that>will do is DEACTIVATE the Nvidia profile and ACTIVATE your new>duplicate. >>b. >With the new profile in the list, make sure there is a bullet>in PROFILE on the right. Go down the list on the right and>select the following: >>AA: >For FSX select SUPERSAMPLING and 8xS (use 8xS Steve)>>NOTE: Slower cards and systems in FSX, set to MUTISAMPLING and>4X or SUPERSAMPLING and 4xS >>NOTE: FOR THE BETA VERSION THIS HAS CHANGED... SET AA TO>COMBINED 8xS>>Under the next box, check GAMMA CORRECTION >>DO NOT ENABLE THE SECOND SUPERSAMPLE FEATURE IN THIS AREA >>UNCHECK "ENHANCED AA" if it is checked by default>>>AF: CHECKED >>Bullet in: 16X >>VERTICAL SYNC: CHECKED, Bullet in: FORCE ON>>>OPTIMIZATIONS Tab: >>Texture Filtering CHECKED Bullet in - High Quality >Trilinear Opt CHECKED, Bullet in - Off >Aniso. sample opt. CHECKED, Bullet in - Off >Negative LOD BIAS, CHECKED, Bullet in - Clamp >>Check the next three items but do not change their values >>COMPATABILITY TAB: >>Place a CHECK in all header values but only change the>following if needed: >>Force Mip-Maps = NONE >Conformant Texture Clamp = ON >Hi Dynamic Range HDR (Enabled)>>All others should be at their default values. >>Under the CPU Multicore Support, REGARDLESS OF THE TYPE OF CPU>YOU USE: >>FOR WINDOWSXP SET CPU TO SINGLE >FOR WINDOWS VISTA SET CPU TO MULTICORE >>Repeat the above for any other game profile as you may need>it. >>>>c. >In the main Nhancer Window, set the bullet to GLOBAL >>Set the SAME settings on GLOBAL as you did above. >>Close Nhancer and reboot. Thank you Nick, I had already set this up this way ages ago when I was communicating regularly with our friend Sam. The only diff thing is the SET CPU issue. Ok, now why is this recommended?Also, I contacted ASUS to see if there is any issue with using the second matched pair of DIMM slots (doesn't show this as acceptable in the manual), which WILL work with the OCZ Flex 2 DDR2-1150. The guy said should be no problem. I took SLOT A1 and A2 out with my current memory and it is indeed fine. I think this is worth the try, plus it can be RMA'd if there is no appreciable difference, so I'm looking forward to trying this when it arrives in a couple of days.Now here we go: my Mushkin DDR2-800 at 4-4-4-10 does 59.3ns/8350mb/s; the OCZ DDR2-8000 at 5-5-5-12 did 54.3ns/9300mb/sec. Any predictions re this new DDR2-9200? Maybe around 50ns or so? It is rated 5-5-5. I still don't understand the real deal on CALWI, but methinks 50ns and 10000mb/sec must be close, no?QX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed

I am still trying to figure out what you are using to base your assessments that one is/is not better than the other.If you are not using an established baseline benchmark test method, simply flying over things and saying "I don

>I am still trying to figure out what you are using to base>your assessments that one is/is not better than the other.>>If you are not using an established baseline benchmark test>method, simply flying over things and saying "I don

Ahhh, but can any of ya'll hand-fly an ILS to a squeaky landing in the PMDG 744 in a 20 kt cross-wind - WithOut looking out the window? Talk about Ram Timing? I'd be thankful to time-it so I could simply Ram the runway ever now and again. With the smoothness of that Core2 at 3.6, I can feel the airplane's weight, the float of another notch of flaps, the drag of the gear Doors (even before the gear gets out). I had a Grand Canyon run going the other day on the 42" and a pal behind me playing copilot. Cranking down the Colorado, right down on the river. I heard a thud behind me. My copilot had fallen out of her chair. I was busy. I provided an encouraging word. What I see, what I feel in real gameplay is where the rubber meets the road (hopefully not at ram-speed!), least for me. That IS what matters. It's the entire point of all of this. A novice would have to spend $1K for a QX to do that otherwise, but a $200 Q6600 - with the knowledge we've acquired - lets Everyone enjoy this sim. This hobby has sooo many interesting facets. Anyone remember where we started? The technical LOD we've progressed to in this Forum is simply amazing. It has brought this game to life. Thanks to all.

Excellent prose as always.Signed, The NoviceQX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed

The sticks you ordered have a spec of 1150MHz DDR2 @ 5-5-5-18 which is the SAME as their REAPER edtion product and I suspect without the water cooling on the FLEX, they wont get you much betterSo if they dont tighten up, here is your CALWIhttp://forums.avsim.net/user_files/188025.jpgand your going to come up with the same results... the myth resultsIf those sticks wont do DDR2 1200 @ CAS5 or DDR2 1066 @ CAS 4, your not going anywhere with results, just like with the rest of the sticks you have tested and without 1T even if you hit CALWI you will still be seeing limited results unless that bandwidth will jack up past 1200I dont care what Everest shows for latency The CALWI is the most important, followed by tRD and then 1T. You could have a Everest latency of 52ns with CALWI, tRD and T1 in line and it will completely outperform 52ns with everything out of whack and FSB jacked results. The bandwidth (speed)after you have hit those jewel numbers is the perk and multiplies the result so as I have said in another thread, you are prettly much throwing money at nothing with DDR2. Right now the memory companies are putting their money on DDR3 so the magic 'bins' of chips are just not there. They may start introducing those again this summer with the DDR3 carrots but right now, the choices are really limited.I have stated several times higher FSB is not needed IF you hit the numbers... the reason you see better with higher FSB is because you have not even come close to CALWI yet and FSB is what you are using to make up for it. I dont how else to say it Noel. Truth is DDR2 wont out perform correctly set up DDR3 and if you really want to see a difference I would not waste my time with that product any more and move into DDR3, especially since they are cracking DDR3 1600 @ CAS6 now. But remember.. anything you buy now in the way of motherboard will need to be replaced if you intend to run Nellie when she hits the streets.I cant make any crystal ball estimates about what memory may or may not do.. that all depends on the sticks, your motherboard chipset and the BIOS of the board. There is no such thing as estimating past engineering specs without defined data as to proven results... just like there is no data to review about what FSX skipped behind the scenes to display the result on system A compared to system B.

>I dont how else to say it Noel. Truth is DDR2 wont out perform>correctly set up DDR3 and if you really want to see a>difference I would not waste my time with that product any>more and move into DDR3, especially since they are cracking>DDR3 1600 @ CAS6 now. But remember.. anything you buy now in>the way of motherboard will need to be replaced if you intend>to run Nellie when she hits the streets.I heard you loud and clear when you said DDR2 won't outperform DDR3 set up correctly. My question has always been, is there affordable DDR2 that would be worth the expense and trouble. So, the comparison is between my mushkin, and something else with reasonable cost, not your system Nick. As I say, I already bought your arguments. However, you have hinted that perhaps something less than DDR3 might help. You made some recommendations. That is the impression I got from your narratives. Never doubted "my" DDR2 couldn't compete with your DDR3.The 9250 arrived, and with plenty of experimentation with diff STRAP settings, various combinations of DRAM freq and FSB, attempts at tightened timings, etc. The best configuration is:CPU CPU Clock Motherboard Chipset Memory CL-RCD-RP-RAS LatencyCore 2 Extreme QX9650 3933 MHz [ TRIAL VERSION ] X38 Dual DDR2-1093 5-5-5-15 CR2 53.7 ns53.7ns in Everest. I haven't tried this yet, but my guess is that this will not make an appreciable difference in my experience of FSX. I don't even know if it is stable with this setting. This memory is not particularly high performance. I will end up sending this back, but it was fun playing with it nonetheless. I am intrigued by how I end up with an optimal CPU frequency, that is very much less than my CPU's top end. I think you made mention of this. I could not come up with any configuration that would allow latency of less than 55 with a CPU speed over 4.1 GHz.I am actually very impressed with the mushkin I got. Not half bad considering it can handle all 8GB in stable condition. Latency is 59.4 as optimized as it will permit. No, won't be buying Nellie. I just bought this and it works well, even with the mushkin memory. No need to upgrade so soon. Again, when you compare AMD's onboard memory controller and Intel's solution, it is interesting to see how other caches or pipelines end up having Intel outperform the entire system in the end, at least that's how i interpreted an article on this.

Thanks Nick for the technical insights. Learned alot, but still not enough in some areas. Here is a very direct request: can you please define CALWI? Not what the acronym represents, but in simple terms what relevance does CALWI have?As an intereting side note: I decided to replace my dimms with all four of my mushkin PC-6400. I left the stock timings at 5-5-5-18. I then discovered I could run a FSB of at least 440MHz with my NB volts up from 1.25 to 1.33. After playing with a few other settings, I discovered I can get almost identical EVEREST scores with this memory. In fact, it runs at a faster DRAM freq than the PC-9200 would, or at 1103 using a 266 strap. Is there a generally safe guideline for northbridge voltage? Right now I only have a 5% increase which can't matter. I have excellent cooling over the entire guts as described. I'm guessing Everest scores have some value, though you seem to be saying Everest benchmark has little to do with CALWI, the mysterious and yet to be defined other benchmark you seem to be focusing on.Pretty weird tho that my mushkin really is doing as well with Everest and Sandra. In fact, it may prove better. The mushkin rep thought I had mis-binned modules, ie better than rated.Here is my everest result with newly configged mushkin 8GB:CPU CPU Clock Motherboard Chipset Memory CL-RCD-RP-RAS LatencyCore 2 Extreme QX9650 4180 MHz [ TRIAL VERSION ] X38 Dual DDR2-1100 5-5-5-18 CR2 54.5 nsI'll add a few FSB and see where it hits a wall.QX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed

I was/am not making "you" to "me" comparisons. If I wanted to play that game I would not use the P35 system I used in the test and instead post results off the board and memory I am using for my regular tower which is not on the market as of yet and runs DDR3 2000 @ CAS7, 1T and tRD 6. To do so out of shear bragging rights would serve no purpose and would result in breaking NDA agreements. This is not and never was any kind of competition so let

>I was/am not making "you" to "me" comparisons. Nor am I Nick. You may have wonderful technical expertise, but I have never been focused on any competition Nick. Please, it's easy to follow technical information I am seeking . . . Which reminds me, again I don't get an intelligent answer with a very simple question posed: can you please define CALWI? Not what the acronym represents, but in simple terms what relevance does CALWI have? >Optimal FSB for correctly set up DDR2 is between 400-430 and>DDR3 450-470 and that will change soon. What do you know about this: is it risky, in terms of damaging my modules, to leave DRAM voltage on AUTO? I think it must be because my lowly DDR2-800 boots and runs stable with these settings:FSB 440/STRAP 266/DRAM FREQ 1103/stock timings @ 5-5-5-12/4.18Ghz or so I forget.If I set DIMM volts at 2.1 let's say, it will not boot. If I leave it to auto with this insane FREQ of 1103, it will boot and is indeed stable! I'm guessing the explanation is the BIOS using AUTO sends more voltage to the DIMMS, true? Sounds like it could fry my modules. It turns out Everest scores, 3DMark06 scores, all are the highest with this configuration, but I don't want to destroy these DIMMS as I think they are quite good despite their low end. So, again, does leaving DIMM v on AUTO imply this kind of risk? As it turns out, the modules ARE NOT hot at this setting, because of my very effective cooling solution as posted.Thanks . . .QX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed