747-400X Wing flex

[Guest pellelil]

Something like this is hard to test as you might see highter FPS in some instances and lower in others. The 30% drop you have noticed is that "sitting still looking at a parked plane" or is it in flight?. Sitting still I see one core at 100% and the orther doing (almost) nothing. Whereas if I fly at x4 the last cores get their load as well (so in this case I guess additional cores is a plus).I've got a QE6700 (Quad Core 2.66 GHz), which I run on VISTA 64 bit with 4GB, and 2 monitors (1600x1200 and 1200x1024). I only use the 1600x1200 monitor for FSX (windowed) but use the other for Navigraph, FSCommander and other stuff.Sitting still at EKCH parked at the gate I see the 30 FPS that I have locked FSX to. But flying into major US airports I do see a drop below 20 - but not as much as the plane gets unflyable (while waiting for the PMDG 744 for FSX I am currently "setteling" for the Level-D 767).Since I only fly big jets I have turned down the autogen, but in stead raised the scenery detail (to have more buildings at the airports). I'm not using "Light Blume" (or what it's called) since it has a huge impact on my FSP, and running under VISTA 64 bit Ultimate I found that I could increase my framerate from 18-20 to 33-35 just by disableing Aero and the other screen-candy in VISTA).[link:fs2crew.com]http://fs2crew.com/betateam.jpgPelle F. S. Liljendal

[Guest darrenthe747400]

I feel a bit silly now because I read Phil Taylor's instructions wrong. I was supposed to put affinitymask=3 but oddly enough as I mentioned before I saw a drop in FPS when I did that.Pelle, to answer your question about my FPS on the ground or in flight I always run the exact same flight to test my FPS. I start on RWY 01L @ KSFO at 12:00 noon in the payware 767. Then depart straight out over the SFO bay then over Oakland then to Sacramento. I feel this is a great benchmark test becasue there is a lot of autogen, a lot of water, a lot of hills etc. I feel the true FPS test is over dense cities, not over the great plains. Right now my FSX runs pretty darn well. I am getting 20ish FPS over dense scenery and I don't have any blurries anymore after I tweaked my fsx.cfg file. I must say though I am a bit disappointed to hear that you are not getting much better FPS than me considering you have one of the best processors in existance. My rig is somewhat similar to yours in that I use Vista Ultimate 64bit, 4GB Crucial Ballistix RAM, and I am using a GeForce 8800GTS with 640 MB video RAM. What kind of graphiocs card are you using? I am now thinking I will keep my current rig for 1 more year rather than buy a new processor at the end of this month and wait for the next generation of CPU's and video cards to come out. I don't want to spend $2000 on new computer parts for a minimal FPS gain. I want to build a computer that will run FSX in ALL of its glory.Darren

[phoenixphire 50]

Now that is some good news. It just keeps getting better and better. Can't wait.

[Guest pellelil]

I've got a 8800 GTX and I'm running at 1600x1200 (which resolution are you running at?). Perhaps I haven't tweaked my FSX.CFG as hard as you have done. As I recall the only thing I've changed is the AffinityMask !?Have you tried disabling Aero? and if so did you see any improvements?[link:fs2crew.com]http://fs2crew.com/betateam.jpgPelle F. S. Liljendal

[Sartanius 399]

Hi guys... I've been reading this thread with great interest. Note in my signature that I recently sprung for a beast in the pursuit of and preparation for decent FSX performance with the upcoming Queen.In reading Phil Taylor's blog he states (and I loosely quote) that the affinity mask cfg statement is only needed if the processor is mainstream and can't be recognized by their code. Your results interest me in that I'd certainly have expected your Intel cores to have been "on the list" since they're mainstream, therefore not needing the "affinity" statement. I'll add the statement to my cfg tonight and see if there's anything else to squeeze out of my quad core in the event it isn't on the list. As it stands now with the Level-D 767 I'm averaging low 20s when ready for TO at KSEA 34R with autogen disabled, airlines and cars at 10% and scenery settings full right. At altitude over deserts I'm seeing avearge frames in the upper 30s to low 40s.Also, when FSX starts on my rig Aero is automatically reduced to "basic" which I just assumed as "off". I'll fully disable it tonight as well and see what gains there might be.DougDell XPS 720H2C (3.73GHz) | 2GB DDR2 @ 1066 | 300GB RAID0, 10k RPM (C: ) | 500GB RAID1 (D: ) | 2 x 768MB nVidia 8800GTX (FW 158.45b) | SB X-Fi Sound | MS Force Feedback 2 | Vista Ultimate | DirectX 10.0

[Guest pellelil]

FSX do detect the cores and will run across them all automatically (at least this is my experience), however if you want something different then you'd have to manually alter your FSX.CFG file. I've setup mine to only utilize 3 cores ("leaving 1 for the rest"), and Daren is setting his FSX to only use 1 of the cores (dual core) since this on his system gives better performance than running it on 2 cores - my guess is that it's becuase the CPU can better utilize the Cache and/or perhaps because you won't have multiple threads blocking each other (more cores/more threads has its advantace, but also some disadvantages - so its a matter of finding the correct ballance).[link:fs2crew.com]http://fs2crew.com/betateam.jpgPelle F. S. Liljendal

[Guest MauiHawk]

>Darren,>>aren't you supposed to set AffinityMask=3 if you have two>cores? >Please correct me if I am wrong since I have never tried this>tweak, but I searched Phil Taylor's blog for it and found>this:>>http://blogs.msdn.com/ptaylor/archive/2007...aks-in-sp1.aspxThat's certainly my understanding. I think all 2 does is tell it to use the second core rather than the first.Darren, open task manager when running with that settings, then after flying for a bit, go back and check the CPU usage graphs. Are both up toward 100%?