Jump to content


Benchmark STEP - Windows 10

  • Please log in to reply
2 replies to this topic

#1 miatamaniac



  • Members
  • 2 posts

Posted 05 August 2015 - 01:49 AM

CPU: Intel i5 4690K

VGA: EVGA GTX 980 SC 4GB x2 in SLI (Overclocked to +125 Mhz Core, +500 Mhz Memory using MSI Afterburner)

PSU: EVGA 1200w Supernova

Mem: 16GB G.skill DDR3 2133

SSD: 240GB Intel 730 Series x2 in RAID 0 (dedicated game SSD)


Individual STEP Core data per step not obtained for GPU 2 (PhysX).


Nvidia Inspector set to framerate limit of 75 FPS to avoid game breaking graphics issues (initial Core tests peaked at 144 fps Vsync).


No noticeable (~4 FPS) drop in FPS until STEP Core nearly complete (step 2.M).  VRAM steadily rises to about 2600MB by step 2.H and maintains this value approximately until ENB is loaded.  Initial 10% jump in GPU Load at step 2.D then load maintains steady percentage at ~ 44 - 46.5%.  GPU Temp starts at ~ 42° C and rises to ~ 47° C when STEP Core is completed.  Temperature drops slightly to ~ 46° C when Extended is completed and rises back up to ~ 50° C with finalized mod list.


I'm including a spreadsheet of all benchmark data obtained and my current mod list and load order.


Hope this helps.


Edit: Updated with overclock values.

Attached Files

Edited by miatamaniac, 06 August 2015 - 01:37 AM.

  • 0

#2 miatamaniac



  • Members
  • 2 posts

Posted 06 August 2015 - 01:49 AM



Updated mod list to remove conflicting mods between STEP and installed RealVision supporting mods.  Also, reorganized mod list to better follow STEP structure.


Mod list is now as follows:


STEP Core + STEP Extended with the following mods removed:

  • Relighting Skyrim
  • Improved Vanilla Mountains
  • More Realistic Hair
  • Superior Lore-Friendly Hair

The following RealVision recommended mods have been removed:

  • Cinematic Fire Effects 2 HD
  • Vivid Landscapes - Dungeons and Ruins and supporting patches
  • Immersive Roads
  • Fences of Skyrim

The following new mods have been added:

  • Hearthfires Display Case Fix
  • Sexy Vanilla Armor (Standalone)

New mod list and load order attached.

Attached Files

  • 1

#3 echo451



  • Members
  • PipPip
  • 101 posts

Posted 05 September 2015 - 02:08 AM

Benchmark was taken using the opening carriage ride.

Measurement was from the start until the first autosave. The first few 0 fps were removed.

Three samples were taken. 

No Mods, Core and Core + Extended mod sets. Nothing else was included.


Samples were taken using the skyrim performance monitor. This required the SKSE scripts be enabled so the FPS measurements could be logged. So No Mod is a very small lie.


SPM also includes a nifty on screen summary to amuse you while you eat dinner and watch a repeating scene over and over and over.



Quick summary 




Not sure the above will show in a post. Can you show a picture in a post or only a link? i'm guess not given to what that could lead to. 


~ a 20 fps hit between no mods and core or core extended.

The most interesting number was the MODE which is a measure of the most frequent number in a series.

No mods has 60 fps, core and core+extended have 37.


Other numbers like GPU VRAM and MEMORY were very similar with the curious fact that core had lower average VRAM usage. Not a big difference ~ 200 MB. Summary is in the attachment. 


For me anything less than 40 FPS is annoying. I can walk thru it but running is painful to watch.  

Evaluation of image quality and gameplay are still a mystery.


Also DynDOLOD is in an unknown state. Not sure of the impact this will have good or bad.


There seems to be no difference between core and core+extended.


System specs are as per my signature.


Attached is an excel spreadsheet with the raw numbers, a plot and the stats.

Attached Files

  • 0

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users