I have a i7 920 running stock (for the moment) and tested a bit with the AffinityMask=254 setting. Loading took LONGER, fps was exactly the same, smoothness (hard to test) was also about the same... I tested with a complete reboot of the computer in between tests. Maybe you have to OC to get any results...? (Driver etc. is setup according to NickN's posts.)

I see absolutely no benefit at all on my system. In fact I see great degradation in the visual quality of FSX. Leaving my FSX configuration as is and disabling cores has one ugly result the BLURRIES which was what I expected. Overall framerates were still about the same however visual quality, namely ground textures, go right down the drain. I experience the same differences I had in FSX between running FSX on a dual core and running FSX on a quad core.

I have a i7 920 running stock (for the moment) and tested a bit with the AffinityMask=254 setting. Loading took LONGER, fps was exactly the same, smoothness (hard to test) was also about the same... I tested with a complete reboot of the computer in between tests. Maybe you have to OC to get any results...? (Driver etc. is setup according to NickN's posts.)

Yes.. I have also seen several reports of load times LONGER and not shorter as well.. I also saw that here on one of my i7 test rigsI have threads running around the net for i7 and AffinityMask testingone is here at AVSIM http://forums1.avsim.net/index.php?showtopic=241915So far no one in the threads I have running has reported HT ENABLED does anything for i7 with any AffinityMask setting except stutter... however, I am seeing a pattern for i7 with HT DISABLED in the BIOS and AffinityMask=15 in the FSX.cfg file. Try that and see if the result is different from the edit not being in the config at all and HT OFF in the BIOS

I see absolutely no benefit at all on my system. In fact I see great degradation in the visual quality of FSX. Leaving my FSX configuration as is and disabling cores has one ugly result the BLURRIES which was what I expected. Overall framerates were still about the same however visual quality, namely ground textures, go right down the drain. I experience the same differences I had in FSX between running FSX on a dual core and running FSX on a quad core.

Yes sirThat is what I get as well with SP2 using the affinitymask entry on just about every system I have tuned which is why I originally left it out of the tuning list

Don't mean to interrupt, but the easiest way to find what cores will be used (if you can read binary code) using the affinity mask is using the calculator.1. Open the Windows Default Calculator (Start > All Programs > Accessories > Calculator)2. Go to View > Scientific3. Enter in a regular decimal number e.g. 1, 2, 3, 4, 5 ..... 12, 13, 144. Just below the result box look for the line the Group Box that has Hex, Dec, Oct, Bin5. Click on the Bin (Binary) radio button and the number will be changed to binary code. This way you can see for yourself what cores will be used, and you do not have to search the forums for this thread if it gets old.I tried to figure out what the number would be for the last 3 cores to be used would be and tried to enter in 0111 in the calculator while in binary mode but the calc wont accept 0s. But anyway I get 111 and the affinity mask would be 7. Because 1110 is 14 and 111 is 7 (7 is obviously different than 14), give AffinityMask=7 a try and see what results you get. :(...And I strongly agree with Nick N.Although I have an overclocked AMD Phenom 9550. I know that the Intel Hyper-Threading will probably cause more problems than benefits. Hyper Threading just takes 1 Core with 1 Thread (A thread is like a processor) and splits them into 2 virtual threads. Intel designed this to help with multi-tasking, but common folks. On a 4 Core processor why would you need hyper-threading, its already enough processing power and more cores is just overkill and can cut your performance in half, because multicore processing is no easy task (If you program you know this, I'm only 15 though so I'm still learning how to program). FSX was most likely designed for quad-core processors at the most, so eight threads will ruin efficiency, and if I am making a hypothesis here, but if it was designed for 4 threads with two threads doing something (calculating physics?) and the other two threads processing scenery, that would mean (this is a postulate) 4 threads are processing something (physics processing?) and 4 threads are processing scenery. Lemme tell you guys that believe in this whole Hyper Threading fiasco something. With 8 threads, it will increase the load on processing, because with the programming, 4 cores will be processing the same data, then the program has to check results and make sure that there are no faults with the result, increasing load (and in this case load is NOT a good thing).BTW if you look at Alvega screenshots on page 1 and notice the frame rate in the first screenshot with 4 threads processing, the frame rate is 28.1 FPS. With 8 threads processing, the FPS is 25.0. So maybe my hypothesis is correct, and Nick is most likely right.(I know the screenshot is hard to read and the first screenshot looks like a frame rate of 20.1 which is lower than the 2nd but it is 28.1. Look at the 0 in 25.0 in the second screenshot and compare it with the 8 in the first screenshot, and you can tell the number in the first screenshot is an 8 because of the distortion at the top and no distortion at the bottom of the number. Just wanted to state this because someone will probably come in and say its 20.1. :()

http://forums1.avsim.net/index.php?s=&...t&p=1533999

What I also found amazing is those running WINDOW MODE making the same statement about performance after disabling Vsync... when Vsync DOES NOT work in Windows Mode, at all!The psychological result of seeing the frame counter move over the refresh rate allowed all sorts of perfomance 'sightings' to occur all over the net

Windowed mode > not all normal full scene Aniso/AA filtering is applied in windowed mode vs. 3D which also isn't apples to apples. Also to add: running FSX on Systems that run Vista bloat or other systems that are cluttered with several Services running is misleading when disabling cores and can become just a wild goose chase.Best to learn how to set up your OS to run FSX with min servises and other tweaks as need(like 3gig switch for 32bit OS etc.) and then do the Tweaking... its a bit more accurate and will help gety rid of bogus observations and such (not meant to offend original poster).Just thought for food.Paul

I tried to figure out what the number would be for the last 3 cores to be used would be and tried to enter in 0111 in the calculator while in binary mode but the calc wont accept 0s. But anyway I get 111 and the affinity mask would be 7. Because 1110 is 14 and 111 is 7 (7 is obviously different than 14), give AffinityMask=7 a try and see what results you get. :(

The cores are numbered from right to left, not from left to right.That is why the "last three cores" is 1110, not 0111 :(

Can someone explain to me why people are setting their affinitymask to 8 cores on their i7 while the i7 only has 4 cores and 8 threads?

Can someone explain to me why people are setting their affinitymask to 8 cores on their i7 while the i7 only has 4 cores and 8 threads?

Those doing it, are doing it in vain. FSX does NOT use HT. People mostly either use HT-off and 14/15 or HT-on and 84/85.

Those doing it, are doing it in vain. FSX does NOT use HT. People mostly either use HT-off and 14/15 or HT-on and 84/85.

So I should use 84/85 on my i7 720QM?edit: i just tried 255 and in FSX (while moving around) all 8 threads get used

Any particular reason a year+ old thread got dredged up when there's already one (Word Not Allowed's) discussing the current state of this tweak?

Any particular reason a year+ old thread got dredged up when there's already one (Word Not Allowed's) discussing the current state of this tweak?

the FFTF=0 one?