Overclock.net › Forums › Benchmarks › Benchmarking Software and Discussion › 3DMark 11 Physics (CPU) score declined by 82% in 18 months with identical hardware and speeds
New Posts  All Forums:Forum Nav:

3DMark 11 Physics (CPU) score declined by 82% in 18 months with identical hardware and speeds

post #1 of 29
Thread Starter 
I ran multiple 3DMark 11 tests in September 2012, just days after I reformatted my hard drive. Then, today, I ran 3DMark 11 again. My graphics scores were about the same, even though the clock speeds have changed a bit between the two dates, but my physics score (which tests my CPU) is 82% lower than it was back in September 2012, even though all of the speeds and everything are exactly the same.

September 2012 3DMark results
March 2014 3DMark results

As you can see, all the speeds (except GPU) are identical, the hardware is all identical, etc. I ran multiple tests in September 2012 and the results were all very close. I also ran multiple tests today, and the results were also all very close (extremely low.) The tests were all performed with no background processes. I restarted my computer before all tests, as well.

These are the only differences I can think of between the two tests:

1) The September 2012 test was done on 3DMark 11 version 1.0.3; the current test was done on whatever the latest version is, I just downloaded it.
2) The September 2012 test was done on the non-Steam 3DMark 11 application; the current test was done on the Steam version (but it seems to start up 3DMark 11 as its own process anyhow.)
3) The September 2012 test was done within a few days of reformatting my hard drive and reinstalling Windows. I have not reformatted since.
4) Graphics card driver is different.

As for #1, the patch notes for all subsequent versions after September 2012 say they do not affect benchmark results.
As for #3, I also did a 3DMark 11 test before reformatting in September 2012 and before overclocking my CPU, and my physics score was still above 6000: pre-reformat stock CPU 3DMark 11 results here.

How is it possible that stock CPU speeds, before a reformat, can give me over 500% higher physics score on 3DMark 11 than my current score? What's going on?

Thanks.

(I'm not sure if the links I provided will work; if they don't, I can screenshot it.)
post #2 of 29

The only thing that I can think of is CPU degradation.

It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
post #3 of 29
Thread Starter 
Is there any other sort of benchmark or something I can test to determine whether this is unique to 3DMark 11?

I mean, I play (and have played) games all the time and I have not noticed any increase in performance problems over the last 18 months. Would a 3DMark 11 Physics score this low certainly be causing me a lot of performance problems in modern games, if indeed it is indicative of my CPU's true capabilities?
post #4 of 29
I doubt it's hardware related, my bet is on driver related problems.
One thing you can try is to downgrade video driver AND physX driver to a lower version, like v314.
The physx driver has had some updates lately which may have something to do with things, even if not running the physx on the GPU.


Edit: followup question: you sure you weren't running the physx on the GPU before ? Nvidia can do physx on the GPU itself, maybe you were running them on the GPU before, and now are running them on the CPU instead. Check the settings in Nvidia control panel and/or the test settings.
Edited by hexaq - 3/17/14 at 2:33am
post #5 of 29
Quote:
Originally Posted by Aelius View Post

Is there any other sort of benchmark or something I can test to determine whether this is unique to 3DMark 11?

I mean, I play (and have played) games all the time and I have not noticed any increase in performance problems over the last 18 months. Would a 3DMark 11 Physics score this low certainly be causing me a lot of performance problems in modern games, if indeed it is indicative of my CPU's true capabilities?

 

Not unless you changed the NVIDIA Control Panel to use your CPU as the PhysX processor. 3DMark is only doing the PhysX on your CPU as a test. It's being forced.

It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
post #6 of 29
Thread Starter 
I tried the physics test again three times, each with a different PhysX setting in my Nvidia control panel. One for CPU-only, one for GPU-only, and one for auto-select, and the physics score was basically the same for all three.
post #7 of 29
Running 32bit or 64bit.
More phyics score 1/4's when I use 64bit 3dmark 11.
post #8 of 29
Quote:
Originally Posted by Aelius View Post

I tried the physics test again three times, each with a different PhysX setting in my Nvidia control panel. One for CPU-only, one for GPU-only, and one for auto-select, and the physics score was basically the same for all three.

 

Yeah, that's because it's a CPU test. It has nothing to do with the setting in your NVIDIA Control Panel. Like I said, that's for gaming. You know, normal things - not benchmarking.

It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
post #9 of 29
Thread Starter 
I'm still confused as to whether or not my CPU is dying or if this is just some quirk my 3DMark 11 results but my CPU is actually functioning normally. Is there any other way I can test to see if my CPU is actually performing 5x slower than it did 18 months ago?
post #10 of 29
Quote:
Originally Posted by Aelius View Post

I'm still confused as to whether or not my CPU is dying or if this is just some quirk my 3DMark 11 results but my CPU is actually functioning normally. Is there any other way I can test to see if my CPU is actually performing 5x slower than it did 18 months ago?

 

What I'm trying to say is not that the CPU is slower, but perhaps it's just not as stable as it once was, thereby screwing up the performance so much that you get bad benchmark results. What voltage have you been using or are you using for this overclock?

It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
It's a computer!
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5GHz (1.368-1.384V fixed voltage) ASUS P8P67 EVO B3 (UEFI ver. 1850) GTX 780 ASUS DirectCU II (1228 / 6300, 1.180V) G.SKILL Ripjaws X 8GB (2 x 4GB) 1866MHz, CL9 
Hard DriveHard DriveOptical DriveOptical Drive
250 GB Samsung 840 EVO (OS) 3 TB Toshiba P300 (storage) Samsung SH-S243N 24x DVD Burner Samsung SH-S203N 20X DVD Burner 
CoolingOSMonitorKeyboard
Thermaltake Frio Win 7 Home Premium x64 SP1 Retail AOC G2460PG (24" 1920 x 1080 144Hz G-SYNC) Filco Majestouch 104-key Cherry MX Blues w/NKRO 
PowerCaseMouseMouse Pad
Corsair HX650 (Bronze, ordered on 12-12-2009) CM 690 Intellimouse Optical (1.1A) 1000Hz polling rate Basic, but premium round 
AudioAudio
X-Fi Titanium HD Klipsch ProMedia 2.1 (with 16 AWG Monster Cable... 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Overclock.net › Forums › Benchmarks › Benchmarking Software and Discussion › 3DMark 11 Physics (CPU) score declined by 82% in 18 months with identical hardware and speeds