Phenom X6 1090T FSX performance

[dazz 69]

I managed to get my CPU to a stable 4.126GHz and FSX is running really smooth. Temps are in the high 30's under load, 29C iddle. Room temp is 75F.The reason I was not seeing a difference upwards of 3.8Ghz was that I forgot to overclock the CPU/NB link. It went from 1800Mhz to 2685Mhz for a 3.5 fps gain.I'll keep experimenting since the temps are really under control. I'll keep you guys posted.

Ramiro, what are you using to monitor CPU temps? those temps can't be right, or is it watercooled?

[Ramirin 0]

Ramiro, what are you using to monitor CPU temps? those temps can't be right, or is it watercooled?

I'm using AMD Overdrive for temps monitoring (Not sure if it is good or not).And yes, I am using a Corsair H50 watercooling system on an Antec 900 case.

[dazz 69]

I'm using AMD Overdrive for temps monitoring (Not sure if it is good or not).And yes, I am using a Corsair H50 watercooling system on an Antec 900 case.

AMD overdrive is not providing actual core temps. Phenoms don't have sensors in the cores so you need HWMonitor for example to monitor CPU temp. Be careful cause those "core temps" are usually much lower than the actual CPU temp.What's your motherboard? depending on the model the sensor you're looking for will be labeled "CPU" or "TMPIN0" or "TMPIN1" in HWmonitor

[Ramirin 0]

My mobo is an Asus Crosshair IV Formula. I will try HWMonitor as soon as I have a chance and report my findings. I suspected those temps were too good to be true.

[Ramirin 0]

AMD overdrive is not providing actual core temps. Phenoms don't have sensors in the cores so you need HWMonitor for example to monitor CPU temp. Be careful cause those "core temps" are usually much lower than the actual CPU temp.What's your motherboard? depending on the model the sensor you're looking for will be labeled "CPU" or "TMPIN0" or "TMPIN1" in HWmonitor

HW Monitor reports all 6 cores. Idle temp is 29C and maximum load temp reported is 46C running FFIV benchmark.Should I look somewhere else?

[dazz 69]

HW Monitor reports all 6 cores. Idle temp is 29C and maximum load temp reported is 46C running FFIV benchmark.Should I look somewhere else?

And in Prime95? monitor your CPU temp while you run Prime95 and if it goes above 62ºC stop it

[Ramirin 0]

Maimum temp while running Prime95 was 59C. Does this mean that I already hit the wall because of temps?How representative of real life usage is Prime95? Because most games and apps, including FSX don't push the CPU past 50C on full load.

[dazz 69]

you're ok IMO.What full load Vcore?NB freq?Vcpu/nb?

[fghdgdfdfgfgf 111]

Maimum temp while running Prime95 was 59C. Does this mean that I already hit the wall because of temps?How representative of real life usage is Prime95? Because most games and apps, including FSX don't push the CPU past 50C on full load.

Prime95 running a 2 core operation has roughly the same current load as an FSX load (peak) with lot's of eye candy @ 1680x1050.

[fghdgdfdfgfgf 111]

Running AMD 970 BE OC to 4.2 ghz. 1.47VCC 49c solid. No stutters FSX all addon aircraft and scenery. Runs great. Also have a Intel I7 2600K OC at 4.2 ghz. Runs at 58c. I have seen no performance diff between the AMD 970 and Intel Sandy Bridge. Save your self a couple 100.00$ and buy some nice addons or upgraded hardware with the difference! Benchmarks mean very little as you will never see the difference in FSX.. Could be patient and wait a mo. or so for the new line of AMD CPU's named " Bulldoze", sound meaner than a river crossingl:). Promise to outperform intel SB significantly and I'm sure at far less cost!

Your english is good- but, lets see some benchmarks. :Big Grin:

[SpiritFlyer 987]

I have seen no performance diff between the AMD 970 and Intel Sandy Bridge. Save your self a couple 100.00$ and buy some nice addons or upgraded hardware with the difference! Benchmarks mean very little as you will never see the difference in FSX

Hi Torkermax,Sure, it is great that you love your AMD machine, and it does sound like a good one. Why not run a real, objective comparative analysis to accurately compare it to others by running the Mark 11 Benchmark exactly as per test instructions? Stating: "I have seen no performance diff between the AMD 970 and Intel Sandy Bridge" seems rather a tall statement, especially when accompanied by a most convenient: "Benchmarks mean very little as you will never see the difference in FSX." It is a lot better to show and tell, than tell alone. I hope you participate in the interest of the growing factual FSX performance knowledge base the test provides. If so, here is the link you need:MARK 11 FSX BENCHMARKKind regards,

[dazz 69]

I'd just like to warn those waiting for Bulldozer not to get AM3+ boards and wait for the second generation socket if you want to go with AMD no matter what. The first bechmarks leaked show an 8 core bulldozer performing like a 6 core 980X, meaning that they'll be good in multithreaded stuff, but the IPC hasn't improved much, so for FSX it looks like a terrible option considering it will need upgrading mobo and CPU.

[TechguyMaxC 516]

I'd just like to warn those waiting for Bulldozer not to get AM3+ boards and wait for the second generation socket if you want to go with AMD no matter what. The first bechmarks leaked show an 8 core bulldozer performing like a 6 core 980X, meaning that they'll be good in multithreaded stuff, but the IPC hasn't improved much, so for FSX it looks like a terrible option considering it will need upgrading mobo and CPU.

To which Bulldozer benchmarks are you referring? I have yet to see anything other than benchmarks proven to be fake so I'm curious.

[dazz 69]

To which Bulldozer benchmarks are you referring? I have yet to see anything other than benchmarks proven to be fake so I'm curious.

http://www.hwbot.org/newsflash/1317_salt_obr_and_bulldozer

[TechguyMaxC 516]

http://www.hwbot.org...r_and_bulldozer

Prime and Cinebench?