Frames

[Buzz313th 0]

endless talk ! my opinion is (and thats actually what i did): I got myself an i5 2500k, got ryan advices, read a couple of topics (around 100) about fsx, ovc'ing , Vcore, cpu temp etc...I MADE UP MY MIND and set up my values in the bios: My computer is a winner. AND THX RYAN!!! AS LONG AS YOU ARE WISE, you'll always do well. If my cpu burn, i'll get another one thats it.

LOL, thanks for the compliment...You will probably do just fine... But remember that CPU's just don't all of a sudden fail.. They slowly get there. And your CPU will never run better than when you first get it.JB Edited January 10, 2012 by Buzz313th

[Turbine777 13]

The below is something I wrote a while back that I copy and paste quite often in PM's. In this case though I did tailor it to go with this discussion best I could. I also added a couple things that I think are very important for younger folks that are trying to give good advice but may lack the ability to take their eyes off someones specs.One of the hardest things someone can attempt to do is answer a question like this. Many are correct in what they're saying but lack untold answers. In other words, if I had built you and I identical systems and left you or anyone alone with it for a small period of time the two identical systems could easily have a 90% difference in fps, easily.You don't list any other addons running or sliders you have set. Most that get a new comp automatically assume their new 2012 computer will handle maxxed sliders, they are wrong. Sure at some dormant airport somewhere and even some paywares I can get FSX to tell me I have 100+ fps (lies), but then again, that airport or surrounding area doesn't show me the effects of most of my sliders either. If you did decide to max your sliders, take Ryan's advice on the ones that hit hard.Other addons also have their breaking points such as UT2 set to 50mi and then having the "AI use Jetways" option enabled. Couple those things with a maxxed scenery complexity on two identical systems where one isn't sharing identical addons and you have the hardware having 0 importance to the question of fps.Lastly, and most importantly, I am amazed at how many simmersgamers have so little knowledge on basics of computing. Not a bad thing at all, but a very important difference that also makes hardware specs mean nothing. Sure they have meaning but people never assume that person understands system resources. The average simmer that I talk to that seeks my advice are normally retired gentlemen that want the best rig they can have for FSX. These people, and many people that sim in general don't know or think that certain things partain to them and their comp, but to give an example of the most comonly overlooked things people do in getting falsely blinded by meaningless specs...Twice now in the last month or so, I've helped a couple people resolve their FSX mysteries on my forum and one on the phone. All were falsely diagnosed by myself giving the same basic advice that you read above. It's hard to see certain things in specs and I'm not even talking FSX settings. People look to specs 1st, then addons, then FSX tweaks leaving out the more basic and most important things. I had two of these people show me screenshots of certain things. Both times I was picking up my jaw off the desk since they both had progams running next to their system clock practicallly past the middle point of their desktop. Both of these people also had virus scammers running in the background set to their default resource hungry settings. One of these people simmed around the time I do and happened to have defrag running a scheduled defrag during that time... Another had Vista with that useless sidebar running and craptastic programs like "Weather Bug"! He also had windows update set to automatic which if you sim and you have that set that way, smack yourself in the face for me, please. :lol:Not saying all of this partains to yourself, but people should never go by hardware, ever! It's the last thing that makes or breaks simming most of the time (well unless the hardware is really terrible). The assumption someone with great hardware knows these things as common sense is where we all fail to better answer these types of questions. 99% of the time around here I see people mention hardware and others just saying a comparison based on their own specs. Chances are these people too could have windows update running, weather bug, or a million other things they just don't understand.Not to preach, but I do have 13 computer certifications and was 1/4 through my MCSE years ago, so on the arrogant side I can say confidently, that I DO, know what I'm talking about.Oh one more thing... Another issue I saw here in my own family not to long ago. My family all have computers built and tweaked by me. When I visit on the rare occasion I always spend my time recleaning crap from their bottom bar. They all insist they follow my advice that I reiterate every visit but my last visit added another important one to the list.When you run windows update, or if it runs in the background. It's process (trustedInstaller.exe) can bring a good system to its knees. It can and will also corrupt ones cache if running FS or anything more than a few years old. This is one major reason why it should never be running without you telling it to run..Net also is a big one. If you ever install a .Net update, it's process is going to be hard at work your very next reboot and may run for 15+ minutes. It should be left alone and left to do its thing.So insure I got my main point across, before people seek help for something that may seem odd or running slow, they should first do the following per FSX session:

• Before running FSX, no matter what comp you have and to just seek the best performance you should reboot. Run nothing when it comes up not even FSX.
• Open task manager and take a quick 2 second look at your processes. If your cpu usage isn't 95% or better consistantly for "System Idle Process", start googling your processes and shut those dumb free addon things down running in your task bar (No, they do not need to be near your system clock to work when you want them).
• Once sure you're stable and windows is really at idle, then load your flight.
• Other than the above, they should also have their FSX.cfg tweaked like the OP did with Boyjotes tweaker, Great site and utility he gives there.
• Never overclock until you're absolutely certain your system is already running your programs (like FSX) smoothly. So many people that overclock have no idea what they're doing or how to be sure of this.
• Defragging should also be very high on your list (just not while you're simming...) :(

I follow my own advice above all the time and never have a hard time pinpointing where the issue is by doing so. Sure a good understanding of FSX settings, and tweaks does help, but understanding windows is more important in most cases. I've never reinstalled FSX to fix an issue. I find the issue and I fix the issue 99% of the time (the other 1% I leave the issue until it falls into the other 99%).Last thing to add to the OP. In your original post you said the airport you were at but not where you were. Meaning at the jetway or at the runway. That could be a 10fps difference easily to others not knowing that or wrongly assuming you fly as they might directly from the active.Hope this helps.

[rocketfs 2]

It's my understanding, that both High frequency and temperatures will degrade the silicon in a chip. As this happens, the chip will start producing eroniuos errors and start correcting itself. It will use clock cycles to correct and take those clock cycles away from FSX. This was my first sign as I started seeing my lower frame rate limit drop out significantly over a few months.

That a CPU can degrade by overclocking is clear.But: " It will use clock cycles to correct and take those clock cycles away from FSX"I don't buy that sir. No way you would see any of that.Bert Van Bulck

[Buzz313th 0]

But: " It will use clock cycles to correct and take those clock cycles away from FSX"I don't buy that sir. No way you would see any of that.

I didn't make it up, I read it. It didn't specifically say "It would take clock cycles away from FSX". One of the articles, maybe not one that I posted, explained that as the cpu degrades it will produce more eronious errors. These errors are corrected by the OS/CPU by using system resources. If the errors can't be corrected, then system instability occurs. It also mentioned that the higher cpu load, frequency, heat and voltage is, will determine the frequency and quantity of errors.Look it up yourself JB Edited January 10, 2012 by Buzz313th

[FFT170 0]

I've tried just about every tweak out there to improve frame rates in FSX. The one that did it for me was disabling Hyper-threading on my I7 920. Since then I never get frames below 40 what ever situation, in cruise I am getting 150-300 with either the 737 NGX or MD-11. The only slider I have to check is the Water effects that I keep at 1x since it does not provide anything worthed to me in my simulation. I have 6 gb ram GTX560 card P58 MB.Good luckLuc

Edited January 10, 2012 by FFT170

[dazz 69]

I didn't make it up, I read it. It didn't specifically say "It would take clock cycles away from FSX". One of the articles, maybe not one that I posted, explained that as the cpu degrades it will produce more eronious errors. These errors are corrected by the OS/CPU by using system resources. If the errors can't be corrected, then system instability occurs. It also mentioned that the higher cpu load, frequency, heat and voltage is, will determine the frequency and quantity of errors.Look it up yourself JB

What kind of overclocking experience do you have Jack? if done properly, risk is minimal.Keep your Vcore in spec and your temps in check and you can have a massive overclock last you for years.Intel and AMD "overclock" themselves. They release for example an I7 920 @ 2.66GHz and overtime they release a 3.2GHz I7 960, or compare an 1055T @ 2.8GHz vs 3.3GHz 1100TSame chips, different clocks. Should I7 960 or 1100T users be concerned about degradation? no, not at all Edited January 11, 2012 by dazz

[Buzz313th 0]

What kind of overclocking experience do you have Jack? if done properly, risk is minimal.Keep your Vcore in spec and your temps in check and you can have a massive overclock last you for years.Intel and AMD "overclock" themselves. They release for example an I7 920 @ 2.66GHz and overtime they release a 3.2GHz I7 960, or compare an 1055T @ 2.8GHz vs 3.3GHz 1100T

I've manually overclocked my last 3 systems. An E8400, q9650 and this one the i5. All on different Motherboards. I am not a Benchmarker and only OC for FSX. With that being said, I guess I am very new to OCing.Outa curiosity, did you read any of the articles from the link's I posted, or search for "Silicon Degredation cpu"?IMHO, you can credit or discredit my statements. I am only repeating what I have read from what seems to be very credable sources.After thinking about it, I'm not saying anything that people haven't already heard, except for the idea that CPU's don't hold 100% integrity throughout their life until one day when they finally croak. My statements are simply reiterating from another source, that CPU's wear out and when they do they are slowly dropping in overall performance and the harder you push them, the quicker they will wear out.Here lemme grab the link incase you didn't see it.http://www.anandtech.com/show/2468/6And I quote from this linkAnd highlight the error parthttp://www.tomshardware.com/forum/300350-31-overclocking-life-expectancyBest answerI'll just post my standard wall of text for these questions.Why does OCing dmg a CPU?This is a bit hard to understand if you don't know some basics about electron orbital theory and quantum tunneling?I'll assume you know nothing so I'll try to keep this as simple as I can. You'll have to take my word on a few facts though.Normally, electrons stay around their atom's and don't go wandering off. So in a CPU, they'll stay in one transistor and not move to others. However, if you've learnt about quantum mechanics, you'll know it's actually possible for electrons to escape from energy wells, even infinitely deep ones, it's just very uncommon. In a process known as quantum tunneling, electrons can pass through solid matter and be ejected out the other side.Now, a transistor in a CPU is made from alternating + and - doped and undoped silicon. Once in a while, an electron will escape and bury a couple atoms into an adjourning transistor, and if this happens enough times, eventually all the way through to the adjourning transistor before coming back to it's orbit.Keep doing this and eventually an electron doesn't come back, but stays attached to an atom in the adjourning undoped section of silicon. Over time (usually years), this tunneling causes a hole to be formed between two adjourning transistors and allows free electron flow.This bypasses the "gates" between the transistors and as a result, the computer will misread this resulting in an error.This process is called silicon degradation and eventually results in a complete CPU failure.Now, as to where overclocking comes in.If you know about electron orbital theory, the more energy an electron has, the more likely it is to leave it's orbit and tunnel. IE if your CPU is running hot, or has a considerably higher voltage going through it, electrons tunnel in much higher numbers. As a result, the more you OC, the faster you make those tunnel which cause silicon degradation.In addition, if you increase the voltage enough, you can actually physically destroy the silicon lattice of the gates within aprocessor. Don't make me explain this cuz I can't without lots of math.Now, on to OC and HeatIn a CPU boosting F, has a very minor, almost insignificant heat increase.It's v increase that dramatically increases heat.I'll just quote myself againPower Dissipation = PD in WattVoltage = VoltFreq = HzC= Capacitance in FaradsTotal PD in Watt = C x F x V^2As C doesn't change (ok it technically does, but for the sake of keeping the math simply we can assume it doesn't)If you actually plug in numbers and graph the function, the heat increase due to a freq increase is minute compared to the heat increase from a v increase, as one increases exponentially, the other linearly.Indeed, the more you increase the V, the less the F part of the equation is relevant to the total temp.Looking at real world data, look at the power usage increase in Tom's i5 efficiency article.http://www.tomshardware.com/review [...] 500-7.htmlEach bump was a constant 10mhz clock speed increase, but due to the exponential nature of the voltage increase contribution to PD, the graph is not linear, and power usage does not increase until you start seeing large v increases.Power usage directly translates into heat.As for actual temps, it's more complicated than purely based on power dissipationCpu temperature = (Total PD in Watt) x (HSF's Thermal Resistance inC/W) + (Ambient Temp in Celcius)For comparison purposes the resistance and ambient can be considered constant (technically not true once again, as resistance changes slightly with temp, and ambient increases with more heat output).In your specific case, the answer is not so much the 200mhz F increase, but how much v increase you'll need to attain it. If there is no v increase, life of the CPU will be minimally impacted.There is no easy way to tell how each chip is affected as due to imperfection in the manufacture process, the degradation rate vs v or f graph would be unique to each chip.JB Edited January 11, 2012 by Buzz313th

[dazz 69]

I've manually overclocked my last 3 systems. An E8400, q9650 and this one the i5. All on different Motherboards. I am not a Benchmarker and only OC for FSX. With that being said, I guess I am very new to OCing.Outa curiosity, did you read any of the articles from the link's I posted, or search for "Silicon Degredation cpu"?IMHO, you can credit or discredit my statements. I am only repeating what I have read from what seems to be very credable sources.After thinking about it, I'm not saying anything that people haven't already heard, except for the idea that CPU's don't hold 100% integrity throughout their life until one day when they finally croak. My statements are simply reiterating from another source, that CPU's wear out and when they do they are slowly dropping in overall performance and the harder you push them, the quicker they will wear out.Here lemme grab the link incase you didn't see it.http://www.anandtech.com/show/2468/6JB

I know any amount of overclock will decrease the chip's lifetime. What I'm saying is that you can overclock staying within spec and degradation will be negligible.Too many people just use the auto OC features in their boards, or Windows based apps, but if you take the time to do things right, there's absolutely nothing to worry about

[ZachLW 37]

This is true, Jack. But degradation, unless running extreme voltages, isn't noticeable for a long time. My old E8400 (still running at my original OC of 4.1GHz at 1.35v) is running strong for a friend in his FS9/BF3 machine. It's going on 5 years old now and runs FS9 like a champ! Keep in mind the base clock of the E8400 is 3.0GHz.Degradation is going to happen at stock clocks too, btw. It's the nature of the technology. Edited January 11, 2012 by ZachLW

[Buzz313th 0]

This is true, Jack. But degradation, unless running extreme voltages, isn't noticeable for a long time. My old E8400 (still running at my original OC of 4.1GHz at 1.35v) is running strong for a friend in his FS9/BF3 machine. It's going on 5 years old now and runs FS9 like a champ! Keep in mind the base clock of the E8400 is 3.0GHz.Degradation is going to happen at stock clocks too, btw. It's the nature of the technology.

Ok fair enough then... Lemme paint a picture and you tell me if you think this was Silicon degredation...My old Q9650, running at 3.6 for 2 years ran great Vcore at 1.20 and core temps never exceeded 55c. Then I started noticing that I wasn't getting the same performance as before on FSX. Addons the same as I always tried to do tests with apples and apples instead of apples and oranges. I picked up a new water cooler and better case (Cooler) and overclockable memory (was running 1to1 on the FSB to memory before). Took the OC to 4.05 with Vcore @ 1.28 and max core temps never exceeding 58 on prime 95 and FSX 52c. Wow, the comp ran better than before. Got the NGX, was completely happy. After two months, started seeing performance come down. Upper frames were great, but lower FPS was dropping significantly as well was average and the experience started getting choppy. Started getting an FSX crash here and there, maybe a Blue screen once in a while. Reinstalled windows, reformatted, changed gpu's, memory, dowclocked, upclocked upped voltage, memtest, the whole gamut. Started doing some research. Found those articles, decided to change out the CPU and see if it was infact wearing out. Borrowed a Q9650 from a freinds video editing machine that was about as old as my computer, but has been running at 3.0 for about 2 years. Turned on FSX and walla... Same performance and better than I had years ago. Didn't change a damn thing except swapped out to another processor same exact model.I forgot to mention that when my cpu was at 4.05, I had to up vcore a couple hundreths of a volt every couple weeks to keep it stable. as vdroop was getting a bigger margin.Decided to buy my i5 settup and take the Q9650 and make a FSX MP server.What do you think the cause of the lower performance and increasing rate of instbility was?JB Edited January 11, 2012 by Buzz313th

[wanabflyer 53]

After thinking about it, I'm not saying anything that people haven't already heard, except for the idea that CPU's don't hold 100% integrity throughout their life until one day when they finally croak. My statements are simply reiterating from another source, that CPU's wear out and when they do they are slowly dropping in overall performance and the harder you push them, the quicker they will wear out.JB

When it comes down to it though, at the rate guys upgrade their rigs these days, whether it be for FSX or any other game, before your CPU has kicked the bucket you're already using another one. Probably along with a nice new graphics card, motherboard, etc, etc. Edited January 11, 2012 by wanabflyer

[ZachLW 37]

I highly, highly doubt 1.20 or 1.28 volts damaged your CPU. It could have been dust accumulation on the cooler or in the case. Ambient temperature changes due to the season change (5GHz is now feasible again for me now that it's the cooler months). Typical junk with an aged OS install. Etc, etc.I'm not saying it's impossible. Just simply highly unlikely. Those are low, low voltages for such an overclock on that chip, btw (which could be another answer to the question you posed). The Q9650 typically took in the 1.35v range to overclock to 4GHz+. Edited January 11, 2012 by ZachLW

[Buzz313th 0]

I highly, highly doubt 1.20 or 1.28 volts damaged your CPU. It could have been dust accumulation on the cooler or in the case. Ambient temperature changes due to the season change (5GHz is now feasible again for me now that it's the cooler months). Typical junk with an aged OS install.I'm not saying it's impossible. Just simply highly unlikely. Those are low, low voltages for such an overclock on that chip, btw (which could be another answer to the question you posed). The Q9650 typically took in the 1.35v range to overclock to 4GHz+.

But thats what those articles are talking about. It's not only voltage that will cause it.. It's freq and heat as well. Have a look at that article and then also search for "quantum tunneling".I wish it was dust and seasonal change.. I checked that stuff before replacing the gpu and reinstalling the OS. Trust me, the last thing I wanted to do was build another computer. I just couldn't understand why the computer would slowly lose performance with everything being the same. I read all the OC forums where if you ask about CPU's wearing out ppl will tell you your crazy... They would say that it's a solid state device and they dont wear out, that it was impossible. That sooner otr later it will fail all together, but it would never wear out.I believed that until I read the articles, still a bit skeptical, that is until I did the cpu swap.JB

When it comes down to it though, at the rate guys upgrade their rigs these days, whether it be for FSX or any other game, before your CPU has kicked the bucket you're already using another one. Probably along with a nice new graphics card, motherboard, etc, etc.

Very true.. But for me, my system upgrade cycle is every 3-4 years. So whether it's important to others is really not my concern, or I won't lose sleep over it.. But for the small few who don't want to replace the rig every year or so, have a look at those articles and make your own decision.Last thing I wanted to do is force an opinion down anyones throat, especially an opinion thats not even mine. I'm just trying to pass on info that I felt was credable enough to pass on to others.JB

[Buzz313th 0]

I'm not saying it's impossible. Just simply highly unlikely. Those are low, low voltages for such an overclock on that chip, btw (which could be another answer to the question you posed). The Q9650 typically took in the 1.35v range to overclock to 4GHz+.

I asume your not using Load Line Calibration for your overclock, or maybe you are. My load voltages were the ones I mentioned and I used Load Line Calibration in the bios to be able to maintain an even Vcore, instead of what most people recomend, which is turning it off. I read the articles as well on the old dual core chipsets that LLC was bad as people thought the VCORE would spike as it came down to idle from load, but I did more research and found that it was very well implemented on my old motherboard and it was safer than not using LLC as without it, you would have to bump your idle vcore way up to accomidate the vdroop to get your target vcore at load.Lets just consider my old Q9650 to be a bad chip. That would probably explain most of it. :( JB Edited January 11, 2012 by Buzz313th

[ZachLW 37]

LLC is a must for higher overclocks, and the norm for SB chips.Either way, it wouldn't explain why you had such low voltages. You said yourself you had BSODs with those voltages.