Summary of findings:
- Minor lift in Northeast Corridor performance in the tested Early Clocker scenario following the update, though the long pause around the 13-14 minute mark was still there in all its glory.
- Major reductions in sustained slowdowns on Horseshoe Curve in the tested scenario after the update - sections that would bring things to a crawl for a period of several seconds to a few minutes pre-update were no longer performance-killers. Momentary pauses throughout the scenario were still evident but appear reduced in number.
- Very minor increase in the Oxford-Paddington scenario's performance post-update. This scenario is notable for producing higher FPS than the other two scenarios used in the tests, yet exhibiting more momentary (approx. half-second) pausing, and the update did little to change this.
- The Multi-Core setting had little effect on frame rate in any of the tests, either before or after the Feb 9 update.
The most interesting revelation IMHO is the contrast between the major changes in the performance of the Horseshoe Curve scenario and the much more minor changes in the NEC and Oxford-Paddington results. It hints at a sort of route/scenario/feature-specificity to the update's performance optimisations, and has me wishing I'd cloned the drive so I could load the older version to do a broader sweep of routes/scenarios.
Hope to finally retest with an SSD soon, pending approval from the Department of Finance.
Northeast Corridor Early Clocker tests
Test #09: pre-update, multi-core disabled
Test #11: pre-update, multi-core enabled
Test #16: post-update, multi-core disabled
Test #18: post-update, multi-core enabled
Horseshoe Curve The Champion tests
Test #08: pre-update, multi-core disabled
Test #10: pre-update, multi-core enabled
Test #15: post-update, multi-core disabled
Test #17: post-update, multi-core enabled
Oxford-Paddington Rush Hour tests
Test #12: pre-update, multi-core disabled
Test #13: post-update, multi-core disabled
Test #14: post-update, multi-core enabled
Test configuration
SYSTEM CONFIGURATION
Hardware: Intel DH67BL, Core i5-2500K, 8 GB (2 x 4 GB) Kingston DDR3-1333, MSI HD 6850 Cyclone PE OC, WD Green 2TB
Software: Windows 7 64-bit, latest DH67BL motherboard BIOS & drivers, Catalyst 12.1, FRAPS 3.4.7 (frametimes enabled)
Catalyst: default 3D Application Settings, manual fan control enabled and set to 100%
BIOS: CPU and inlet fan speeds fixed to 100%
disabled Microsoft Security Essentials Real-Time Protection
disabled startup of unncessary progams
switched off/unplugged unnecessary peripherals: network cable, printer, USB sticks
TS2012 CONFIGURATION
Launch window: TSX enabled, basic settings radio button set to Low then Custom to enable 8x AF only, 1920x1200 full screen, 1x1 monitor, no command line or audio options checked unless noted otherwise
In-game options: graphical detail set to High, Depth of Field unchecked, Adaptive Bloom unchecked
