I thought it could be fun to see if the one of the most debated topics in here could be proved useful or not.
I had to reinstall my machine last week, and then I've tried to use the Blackviper Service tweak as described here (main page is under construction).
I created 2 hardware profiles so I can boot up with default settings and another labeled gaming where almost all services have been disabled.
The machine in use:
CPU: AMD 3200+ Barton
RAM: 1024MB Kingston PC3200
GFX: ATI 9500 PRO, 128 MB RAM, AGP
MB: ABiT NF7-S
HD: Hitachi SATA, 250 GB, 8MB, 7200 RPM
Running Windows XP SP2 with ESET NOD32 AntiVirus
This picture shows the upstart usage of ressources with default profile:
This picture shows the upstart usage of ressources with gaming profile:
Gaming has the following services started and nothing else:
COM+ Event System
DCOM Server Process Launcher
Event Log
Network Connections
NOD32 Kernel Service
Plug and Play
Remote Procedure Call(RPC)
Security Accounts Manager
Security Center
Shell Hardware Detection
System Event Notification
Windows Audio
Windows Firewall
Windows Management
Windows User Mode Driver Framework
Workstation
The difference in memory used is app. 22 MB.
Benchmarking:
Both benchmarks where run like this: 1 one run to get machine "warm", then 2 runs where I noted down the scores, and took the averages of the 2 last runs.
Synthetic benching with CPUBench2003 beta2 version 1.5:
Realworld benching with CS: Source built in Benchmark tool.
Default: 85,06 FPS
Gaming: 84,23 FPS
Conclusion:
Gaining 22 MB isn't much when you have lot's of memory, but on machines with less memory (ex. 128 MB) Windows XP might be runnable.
As for the performance associated with this tweak. It's almost non existing much to my surprise this tweak only pays off a little in synthetic benchmarking. In a real world application tweaking like this actually gave worse performance.
The security is also a concern for some people who use this tweak, and there's no denying that it's impossible to hack a service that isn't running, but relying on this for security is a bad idea, also because it seriously affects the usability of the machine.
Instead use a firewall and antivirus.
Anyways I'm looking forward to hear from all you people who have an opinion about this.
-E
EDIT** : I might have been unclear on this, but the main goal was to see if the extreme/gaming tweak had any realworld use/gain for the avarage user.
I had to reinstall my machine last week, and then I've tried to use the Blackviper Service tweak as described here (main page is under construction).
I created 2 hardware profiles so I can boot up with default settings and another labeled gaming where almost all services have been disabled.
The machine in use:
CPU: AMD 3200+ Barton
RAM: 1024MB Kingston PC3200
GFX: ATI 9500 PRO, 128 MB RAM, AGP
MB: ABiT NF7-S
HD: Hitachi SATA, 250 GB, 8MB, 7200 RPM
Running Windows XP SP2 with ESET NOD32 AntiVirus
This picture shows the upstart usage of ressources with default profile:
This picture shows the upstart usage of ressources with gaming profile:
Gaming has the following services started and nothing else:
COM+ Event System
DCOM Server Process Launcher
Event Log
Network Connections
NOD32 Kernel Service
Plug and Play
Remote Procedure Call(RPC)
Security Accounts Manager
Security Center
Shell Hardware Detection
System Event Notification
Windows Audio
Windows Firewall
Windows Management
Windows User Mode Driver Framework
Workstation
The difference in memory used is app. 22 MB.
Benchmarking:
Both benchmarks where run like this: 1 one run to get machine "warm", then 2 runs where I noted down the scores, and took the averages of the 2 last runs.
Synthetic benching with CPUBench2003 beta2 version 1.5:
Realworld benching with CS: Source built in Benchmark tool.
Default: 85,06 FPS
Gaming: 84,23 FPS
Conclusion:
Gaining 22 MB isn't much when you have lot's of memory, but on machines with less memory (ex. 128 MB) Windows XP might be runnable.
As for the performance associated with this tweak. It's almost non existing much to my surprise this tweak only pays off a little in synthetic benchmarking. In a real world application tweaking like this actually gave worse performance.
The security is also a concern for some people who use this tweak, and there's no denying that it's impossible to hack a service that isn't running, but relying on this for security is a bad idea, also because it seriously affects the usability of the machine.
Instead use a firewall and antivirus.
Anyways I'm looking forward to hear from all you people who have an opinion about this.
-E
EDIT** : I might have been unclear on this, but the main goal was to see if the extreme/gaming tweak had any realworld use/gain for the avarage user.