Hi

 

My computer is now about 4 years old and I desperately need to upgrade it, unfortunately money is very tight at the moment and it will have to wait.

 

I currently have a Intel Extreme QX9650 CPU running at 3.0 GHz with 4 GB Ram, P5N-D mobo, Win 7 x64. When flying the PMDG 737NGX in bad weather (AS2012) and near a large 3rd party airfields like Heathrow or Gatwick (UK2000) and all sliders approx half to two thirds. I struggle to get 12 - 14 FPS and not very smooth. I have tried various tweaks from this forum and NickN, but feel my problem is the CPU speed and age (not a shiny new i7).

 

Can anyone tell me if overclocking my QX9650 from 3.0 to 3.6 GHz would give me any appreciable benefit to FSX. I have read various posts on the internet about people getting 4.0 GHz from this CPU and more, but I don't want to fry anything so thought 3.6 would help. As I have never overclocked before are there any safe..ish settings I could try. I've looked on the internet but there seems to be so many different opinions on what to tweak and how much.

 

I would be far more happier trusting the FSX community with any advice than some of the extreme answers I've seen elsewhere.

 

Glen

That CPU was in my main FSX machine for years.

 

Out of the box I set it to 1.3875 volts and 4GHz, worked great never had an issue.

i read 1.3875v was the Intel recommended limit so that is what I used.

That was a Intel Extreme cpu made to overclock.

 

I used a 3rd party air cooler but as I recall it came with a big fan as well.

 

FSX loves CPU speed...

 

That machine is now a family pc so I run it stock to save power.

I'll see if I can find my OC settings.

 

Edit found them

Cpu ratio 10.0

FSB 400

PCIE 100

Cpu voltage 1.3875

 

the cpu ratio x FSB (front side bus) that is key with the product being the cpu speed.

Both can be adjusted on that CPU, I would set the cpu voltage at 1.3875 or less and try 10x330 , 10x360, ...

 

 

 

Great thanks

 

Is that all I need to change, the overclockers forums seem to make it sound a lot more complicated.

 

Glen

Good luck, I edited my original post with more info.

 

 

Remember that you'll overclock your RAM as well when/if you raise the FSB. As you have an unlocked CPU it should be dead easy to overclock. By just raising the CPU ratio and leaving the FSB untouched you won't change the RAM speed for a much easier overclock.

By just raising the CPU ratio and leaving the FSB untouched you won't change the RAM speed for a much easier overclock

 

What about the PCIE and CPU voltage ?

PCIe should stay the same as long as you just use the CPU ratio.

To make things simple, every individual CPU chip has its own limitations in what frequency it can reach for a given voltage.

It will also have a limit to what frequency it can reach no matter how much voltage it gets.

Our cooling will also limit how much voltage we can feed the CPU without overheating it.

As always, increasing voltage is always a risk we take, but as long as we have good cooling and don't go crazy with the voltage its really low.

You can probably get a slightly higher clockspeed without adding any voltage at all. But with only minor voltage increases you can get even higher. It's when the voltage increases aren't yielding much speed you need to decide if the increased risk is worth it.

Hi Glen,

 

I'm still running a Q9650 (my original QX9650 died and my belief is that overvolting it killed it.) Intel gives 'absolute maximum [voltage] ratings in their data sheet, and that is 1.45v however I don't understand exactly what they are pointing at--not sure if they are talking about a voltage setting in the BIOS or VID or what have you. From the QX9650 data sheet, here is their warning regarding voltages:

 

 

 

At conditions outside functional operation condition limits, but within absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. If a device is returned to conditions within functional operation limits after having been subjected to conditions outside these limits, but within the absolute maximum and minimum ratings, the device may be functional, but with its lifetime degraded depending on exposure to conditions exceeding the functional operation condition limits.

 

At conditions exceeding absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. Moreover, if a device is subjected to these conditions for any length of time then, when returned to conditions within the functional operating condition limits, it will either not function, or its reliability will be severely degraded.

 

In any case, I ran my QX9650 at around 1.4v much of the time and although temps were excellent (<62C at peak load) the thing died after a few years of that. Rather than do a whole upgrade, I simply bought a Q9650 and run it at 3.8Ghz at 1.35v. My memory runs fast w/ low latency, and so I still enjoy quite good performance, though I must be mindful not to fly PMDG stuff in dense terminals. But I do use a host of high end scenery from ORBX & FSDT, fly lots of 3rd party birds and have no need to upgrade. I'm holding out for Haswell, and there is rumor that it may be designed to cope w/ much higher voltages and hence perhaps will be a great over clocker, plus it's the tock and so involves generally more significant architecture changes and hopefully better per clock perf.

 

By all means--overclock that QX9650! Just be careful on the voltage & temps issue. I maintain low latency in the memory subsystem contributes to smoother performance.

 

Noel

I recently overclocked my Core 2 Duo for the same reasons, and I can tell you that it is pretty simple to raise your speed.

 

Simply increase the FSB or multiplier to increase the GHz, and then run stress tests to make sure it's stable. If it's not, raise the Vcore for the CPU. You may also raise the voltage for the Northbridge (named "MCH" in most BIOSs).

 

Make sure the PCIE is 100 MHz.

 

However, make sure you keep the temps in order. If you're running hot, but you're stable, then you have the opportunity to decrease the voltages until it becomes unstable and then you raise it again.

 

It is simply balancing the voltages with the speed.