HyperThread - FSX - i7 -and YOU

[Alvega 2,119]

So if it was your intent to come into this thread to insult someone with a twisted, out of context statement and by suggesting something provides benefit across the board without proper testing and research to confirm... you have indeed accomplished those goals!

You say I insulted YOU? Well, since are so sensible i'm not even going to argue with you.I know you are the only one who can do proper testings and research. :( Cheers.

[Guest Nick_N]

You say I insulted YOU? Well, since are so sensible i'm not even going to argue with you.I know you are the only one who can do proper testings and research. :( Cheers.

That makes even more sense.. ! I guess asking the forum and posting this around the net for feedback http://www.simforums.com/forums/forum_post...D=167132#167132 in order for users to look at this on their own systems equates to me being the only one who can evaluate its use on their systemSo far you have misquoted me out of context and insinuated quite a bit more in nothing but pure speculationApparently you have a problem with something. I don’t know, or CARE, what it is however I would appreciate it if you would leave your very well established personal issues out of this thread.Thank You

[Guest Alphahawk3]

F18 here pops mach1+ on the deck, no stutters and no terrain blurs.. love it especially canyon runsI will say this about the multithread and flight loading.. it makes sense based on what was posted about SP1 work=================================During loading, we run the DEM loader on threads. You'll see good balanced usage across all cores, as well as about 1/3 faster load times on average.=================================OK so that does not really mean chit to the game in progress although with Adams changes to how the 'world' and terrain with the thread work they added are used may, which is one of the primary reasons I want to check this out with as many i7 users as possible. It may load the flight like a rocket and that actually makes sense but I don't care about flight load.. I want to know about in-game changesToo many freaking MYTHS get posted about tweaks and the only way to know if JOB SCHEDULER really makes a difference is the RIGHT mindset in what is REALLY happening and of course reports from other so I can compare those reports to what I see here when I end up spending quality time testing all this.The forums are full of MIRACLE tweaks which are BS and I do not like posting something works unless I know they really have teeth

OK...spent most of all day playing...testing this out. I have an i7 920.....Thermaltake CL-P0508 110mm CPU Cooler.........6 gig DDR3 1600 Mem...two 500gig HDD with one dedicated to FSX.....9600GT video card.....Asus P6T Deluxe M/B. The 920 is clocked right now to 3.6GHz...with a very basic use of the M/B settings.....the BCLK is at 180 and the DRAM freq is at 720MHz.....NB freq 2880MHz tripple channel....Win XP 32 bit. My system is setup according to Nicks setup and use guide for FSX exactly except for the defragger....I use Ultimate Defrag and imho I have the defrag setup correctly. Before I started this I had a sound problem that caused stutters at certain times. I can say after todays findings that is now gone and the only stutter I get now or the past several hours is on the rare touchdown. My FSX settings are pretty much maxed out except for autogen=normal for this test.....both traffic sliders at 54%...water at Low 2. Weather was set at Fair weather. My test locations were Seattle....GEX....FEX.....FSGenesis....Alice Springs...with FTX. When turning on HT and setting Affinity Mask at 255 all cores...threads..were in use with thread 0 always being maxed out with the load spread evenly out over the remaining threads with CPU usage showing on average around 80% with a low of 24% and high of course 100%....which is where the stutters were at the worst....as one guy said when i7 stutters it stutters hard. The same happened when plugging in the different masks. The next set of numbers would drop off thread 0 and thread one would be maxed out and the rest spread out over the remaining 6 threads. It was the same result as I went down the threads until I got down to using 3 threads and it was just not usable to me. As a matter of fact with HT on and the mask set at 255 I created lots of hard stutters at 3000 ASL just cruising along with no ATC...just flying. All of my tests with HT caused me stutters....bad ones. So to try a more level playing field I took the clocking out and set all settings in BIOS to default and ran everything again. Same results....stutters....not all the time clocked or not clocked but you did not have to wait long for them to come along. The FPS did not change in any of these test except when the stutters started and then they were all over the place. I stayed locked at 30. So I clocked again and tried one more time...same results. Then I turned off HT and set up the mask for 4 cores....15 being the number I use for all four. Without the Affinity setting the 920 would use all 4 cores with core 0 being usually maxed but the remaining 3 would be giving low usage most of the time. There were times that I would hit 100% but rare and usually 24% to 46% but no stutters....still locked at 30 FPS. Then when I used the Affinity mask it was a whole new ball game. My total CPU usage very rarely dipped below 50% and most of the time staying between 65% to 80%. It would hit 100% at times but not much...but here is the thing...smooth as silk. Raised FPS to 60 and locked. Making turns at Alice Springs and not even a micro sttuter..same at Seattle. Now my FPS did not stay at 60 but varied between 45 to 60 but as one knows that bottom number could be a lot lower but it happens so quick one cannot see it...but the key for me is performance. My flying is much better with 4 cores and using the Affinity Mask in the fsx.cfg file than without. Up until yesterday or maybe the day before I had not even thought about Affinity Mask...FSX is supposed to use all 4 cores after SP1...and one can see it does.....but on my PC with Affinity Mask in the cfg file it is showing me with hard numbers that my CPU is being used more than when there is no Affinity Mask in cfg. My load times are faster using the Affinity Mask setting with HT on or off. I am aware of all the myths and have seen a ton of videos on the 920 being clocked on air to 4GHz...but I am never going to try that. I read and saw a lot of so called gurus all over the NET saying how a 920 could save you a 1000 bucks just by clocking to 4GHz. I never bought that even before I bought the chip. I have spent about 12 hours today doing this and these are the facts....I have Nicks guide printed out and went back over my system with it to make sure it was all still set up the same as it was before I started this...and it was. On my PC HT is a no go as it causes stutters. But the Affinity Mask setting has merit for the 4 cores. Now maybe down the road someone can explain why the Affinity setting in task manager is not the same as when you put it into the cfg file. As I said earlier my clock is mostly with everything set at auto...so am not sure where that plays into this but when the CPU was at stock HT still caused the stutters for me. Anyway I need to go back and thank who ever put us onto this because it has helped me. Regards Nick and thanks for all your input that helps me enjoy my hobby.

[tfm 169]

Hi TimI posted a reply here on what we 'know' for sure and why the affinitymask hook can help with procs prior to i7http://forums1.avsim.net/index.php?s=&...t&p=1532737I am waiting to hear back from Phil so when I do look at this I have it right in my mind what I am seeing on the screen and how what is presented on the screen is being translated by the application through the hardware.EDIT: and the key word used by Intel which may be allowing this to work is the new design they refer to as "uncore"It was the tech docs in the description of the processors in that they are treated as somewhat separate in use from the single 'core' for which they are contained, and, the self-contained cache for each which sparked my interest in this when applied to 'multicore' type threading.. in that what you said is very possible based on the i7 design change.

Thanks for looking into this properly, Nick. I look forward the results.Tim

[Guest Nick_N]

OK...spent most of all day playing...testing this out. I have an i7 920.....Thermaltake CL-P0508 110mm CPU Cooler.........6 gig DDR3 1600 Mem...two 500gig HDD with one dedicated to FSX.....9600GT video card.....Asus P6T Deluxe M/B. The 920 is clocked right now to 3.6GHz...with a very basic use of the M/B settings.....the BCLK is at 180 and the DRAM freq is at 720MHz.....NB freq 2880MHz tripple channel....Win XP 32 bit. My system is setup according to Nicks setup and use guide for FSX exactly except for the defragger....I use Ultimate Defrag and imho I have the defrag setup correctly. Before I started this I had a sound problem that caused stutters at certain times. I can say after todays findings that is now gone and the only stutter I get now or the past several hours is on the rare touchdown. My FSX settings are pretty much maxed out except for autogen=normal for this test.....both traffic sliders at 54%...water at Low 2. Weather was set at Fair weather. My test locations were Seattle....GEX....FEX.....FSGenesis....Alice Springs...with FTX. When turning on HT and setting Affinity Mask at 255 all cores...threads..were in use with thread 0 always being maxed out with the load spread evenly out over the remaining threads with CPU usage showing on average around 80% with a low of 24% and high of course 100%....which is where the stutters were at the worst....as one guy said when i7 stutters it stutters hard. The same happened when plugging in the different masks. The next set of numbers would drop off thread 0 and thread one would be maxed out and the rest spread out over the remaining 6 threads. It was the same result as I went down the threads until I got down to using 3 threads and it was just not usable to me. As a matter of fact with HT on and the mask set at 255 I created lots of hard stutters at 3000 ASL just cruising along with no ATC...just flying. All of my tests with HT caused me stutters....bad ones. So to try a more level playing field I took the clocking out and set all settings in BIOS to default and ran everything again. Same results....stutters....not all the time clocked or not clocked but you did not have to wait long for them to come along. The FPS did not change in any of these test except when the stutters started and then they were all over the place. I stayed locked at 30. So I clocked again and tried one more time...same results. Then I turned off HT and set up the mask for 4 cores....15 being the number I use for all four. Without the Affinity setting the 920 would use all 4 cores with core 0 being usually maxed but the remaining 3 would be giving low usage most of the time. There were times that I would hit 100% but rare and usually 24% to 46% but no stutters....still locked at 30 FPS. Then when I used the Affinity mask it was a whole new ball game. My total CPU usage very rarely dipped below 50% and most of the time staying between 65% to 80%. It would hit 100% at times but not much...but here is the thing...smooth as silk. Raised FPS to 60 and locked. Making turns at Alice Springs and not even a micro sttuter..same at Seattle. Now my FPS did not stay at 60 but varied between 45 to 60 but as one knows that bottom number could be a lot lower but it happens so quick one cannot see it...but the key for me is performance. My flying is much better with 4 cores and using the Affinity Mask in the fsx.cfg file than without. Up until yesterday or maybe the day before I had not even thought about Affinity Mask...FSX is supposed to use all 4 cores after SP1...and one can see it does.....but on my PC with Affinity Mask in the cfg file it is showing me with hard numbers that my CPU is being used more than when there is no Affinity Mask in cfg. My load times are faster using the Affinity Mask setting with HT on or off. I am aware of all the myths and have seen a ton of videos on the 920 being clocked on air to 4GHz...but I am never going to try that. I read and saw a lot of so called gurus all over the NET saying how a 920 could save you a 1000 bucks just by clocking to 4GHz. I never bought that even before I bought the chip. I have spent about 12 hours today doing this and these are the facts....I have Nicks guide printed out and went back over my system with it to make sure it was all still set up the same as it was before I started this...and it was. On my PC HT is a no go as it causes stutters. But the Affinity Mask setting has merit for the 4 cores. Now maybe down the road someone can explain why the Affinity setting in task manager is not the same as when you put it into the cfg file. As I said earlier my clock is mostly with everything set at auto...so am not sure where that plays into this but when the CPU was at stock HT still caused the stutters for me. Anyway I need to go back and thank who ever put us onto this because it has helped me. Regards Nick and thanks for all your input that helps me enjoy my hobby.

So to sum up.. HT was a no-go but adding the affinity mask to the config with HT disabled, helped if I read this correctlySo I assume right now your AffinityMask setting is = 15 for 4 cores?As for the affinity mask edit.. I somewhat explained what is happening and why here http://forums1.avsim.net/index.php?s=&...t&p=1532737The edit sets the hook for FSX to allow it to select the cores for thread spawn calls. By default Sp2 should do this automatically on all recognized physical cores. Why you would see a difference with all 4 enabled in the FSX.cfg is just one of those mysteries at this point where this is a setting that does not net a linear result in use and can not be classified a good or bad edit to use. Its a try and see setting.So far in the spots I have planted the request for information around this, the same report keeps coming back,.. HT Enabled produces faster load times with absolutely no in-game perf change. The smooth flight issue is also not consistent as most are reporting neutral or worse conditions ensue with HT enabled no matter what the mask setting and essentailly i7 is responding the same as C2/Q with affinity mask and no HT... hit and missWe shall see if that remains the norm as more and more chime in on the testTo be quite honest.. every system I have personally visited (sat in front of) running SP2/Acceleration where stutters were a problem and the person who owned the system had solved most of them through the affinity mask edit, I was able to remove that edit and tune the stutters out of the system using sliders and settings. I have yet to come across a system running SP2 that needed that edit however I can see where the edit is a quicker way for people who may not have a lot of experience around MSFS tuning can use it to basically accomplish the same goal and there is nothing wrong with that.So once I get all the information/reports supplied and go through this I will add the affinity mask use to the tuning guide I posted for all procs.Thanks for looking at this.. every report helps! :(

[Guest pete1_leeds]

excellent post, thanks for the hard-work and time put into sharing the info.im going to give this a try in a few days, just waiting on a new GPU before i start tweaking things with my new I7.Will of course post my findings Regards

[Guest Nick_N]

excellent post, thanks for the hard-work and time put into sharing the info.im going to give this a try in a few days, just waiting on a new GPU before i start tweaking things with my new I7.Will of course post my findings Regards

Thanks Pete! The more info collected from different users.. the better :(

[David Roch 1,980]

Well I was a bit skeptical about the Affinity Mask entry + HT.But I wanted to know!!I made the same flight, after rebooting the machine every time in order to be in the same conditions.My conclusions on my machine (it can change on another PC!)1) HT doesn't help at all (with or without AMask). 2) Affinity mask (n= 15) does reduce stutters, that's obvious on my machine. :(

[Guest Nick_N]

Well I was a bit skeptical about the Affinity Mask entry + HT.But I wanted to know!!I made the same flight, after rebooting the machine every time in order to be in the same conditions.My conclusions on my machine (it can change on another PC!)1) HT doesn't help at all (with or without AMask). 2) Affinity mask (n= 15) does reduce stutters, that's obvious on my machine. :(

Thank David!Starting to see a pattern appear.. thats the 6th report I have seen now about AffinityMask=15 and i7 and stutters being tamedI did not include the AffinityMask=N without HT in the original request so the fact that it has come up on its own places good merit to the result.I am going to feed this to a few other forums and see if it holds upI have a sneaky suspicion that i7 HT does nothing in FSX but the virtual cores are seen by the software due to the Uncore design.. and it may be causing thread collisions in which limiting FSX to only work with the 4 physical cores in the proc is correcting.When I sit down with this I will force stutters on my system and see if AM-15 makes a difference too.

[jackcnd 0]

2) Affinity mask (n= 15) does reduce stutters, that's obvious on my machine.

It reduces stutters with HT on , off, or it does not matter?

[Guest Nick_N]

It reduces stutters with HT on , off, or it does not matter?

From his post I must assume HT off because his first line was: 1) HT doesn't help at all (with or without AMask)at least that is what I got from his post and that configuration does tie in with other reports

[jackcnd 0]

Here's a saved flight that produces stutters with the i7, despite changing settings.http://members.shaw.ca/HD_IMAGES/files/rjttstutter.zipIt uses just a C172 flying out from RJTT , an area where very few have extras installed.The flight was saved mid flight, auto pilot is on.Stuttering for me starts right off the bat.After a couple minutes when the stuttering stops,try turning auto pilot off , and make a sharp turn,which produces stuttering again during the turn.Love to hear if any others are able to run it smoothly and suggested settings...If that area is too busy, maybe others can post a suggested stutter testso we all can be on the same page when we report on this issue.

[Guest Alphahawk3]

Here's a saved flight that produces stutters with the i7, despite changing settings.http://members.shaw.ca/HD_IMAGES/files/rjttstutter.zipIt uses just a C172 flying out from RJTT , an area where very few have extras installed.The flight was saved mid flight, auto pilot is on.Stuttering for me starts right off the bat.After a couple minutes when the stuttering stops,try turning auto pilot off , and make a sharp turn,which produces stuttering again during the turn.Love to hear if any others are able to run it smoothly and suggested settings...If that area is too busy, maybe others can post a suggested stutter testso we all can be on the same page when we report on this issue.

Jack, I did your test. Lots of stutters....less as you get higher. Keep in mind my settings are maxed out. That is one ugly congested place. Yes lots of stutters in a tight turn....but also less as I gained altitude. I don't fly there nor will I now after seeing it. I use GEX....FEX....FTX...and those are some good scenery's. Seattle is a busy place but no stutters. Before I tried the Affinity settings I would have some stutters in a hard turn where I normally fly....now none. Just for sake of testing I remarked out my Affinity settings and tried your flight again. It was not flyable for me....herky jerky. Put the Affinity settings back in and micro stutters...annoying...but flyable.....but I got to tell you while the airport looks nice the scenery is horrible there. So to me it is a question of does it remove or reduce the stutters....period. It does not matter where you fly. I just proved...to me anyway...that without the Affinity setting I could not fly that flight....with it the flight is doable but I would not. I did not see much reduction in FPS there....just stutters. Hope this made some sense. Regards

[David Roch 1,980]

It reduces stutters with HT on , off, or it does not matter?

HT off

[jackcnd 0]

Alphahawk3 - thanks for the feedback, so HT off and affinitymask=15 works best for you. What area is a good test to show that for you?David Roch - What area are you using to test?Nick - can you comment on the RJTT flight I posted, does it stutter on your setup?Overall while the i7 clearly offers higher fps for me, when it stutters its more noticable, would like to solve that somehow...