Does Service Pack 2 slow you down?
Microsoft released Service Pack 2 and millions installed it. Did boat anchors come with the enhancements? We threw 108 benchmarks at Service Pack 1 and Service Pack 2 in an attempt to declare a winner.
Does the old saying go “newer, better, faster” or “you can only rework the plumbing so many times”? Service Pack 2 brought enhancements to Windows XP including a very handy pop-up stopper and the much needed firewall security. A new Service Pack does improve upon some features of an operating system besides plugging up the holes but do the improvements come as extra weight or is some code jettisoned in favor of newer, leaner code?
A comparison between the Service Pack 1 clean install and Service Pack 2 clean install shows 4 extra processes and 42 extra megabytes added to SP2 on the test system.
Service Pack 1
Service Pack 2
Benchmarks
The test systems
- AMD Athlon 64 3800+ Processor (32-bit mode)
- MSI K8T Neo2 MS-6702E ver 1 ATX mainboard
- ATI 9800 PRO 256 MB Video Card Catalyst 4.9 drivers (Default settings)
- 2 x 512 MB Corsair PC3200LL TwinX DDR RAM in DIMM 1 and 3
- LG 8x DVD+/-RW.
- 120 GB Seagate SATA Hard Drive
- Samsung 950p 19″ Monitors
- USB Keyboard and MX700 Optical Mouse
- Retail HSF
- AMK PC67 PC case (window, fans, cables, loom)
- Enermax 465 Watt FC PSU
- Windows XP Professional Service Pack SP1 – Listed in graphs as AMD 3800+ (SP1)
- Windows XP Professional Service Pack SP1&2 – Listed in graphs as AMD 3800+
Programs used
Sisoft Sandra 2004- FutureMark
3DMark 03 - FutureMark 3DMark 05
- Quake
III Arena - GL
Excess - SpecviewPerf 8.1
- Serious Sam SE
- Splinter Cell (Chinese Embassy timedemo)
- Unreal Tournament 2003 flyby benchmarks
- Unreal Tournament 2004 UMARK ver. 1.2 (16 bots, high-quality, br-colossus)
- Aquamark3
- X2
Demo - Wolfenstein Enemy Territory (Railgun demo)
- Adobe After Effects 6.0
- N-Bench 3
- PC Mark 04
- DOOM III timedemo
Catalyst settings: let application decide for Anti-Aliasing/Anisotropic filtering, quality texture preference, quality midimap detail level, no smartshader or truform. AGP aperture
was set to 256 MB. Windows visual effects
was set to ADJUST FOR BEST PERFORMANCE. Pagefile set to 1024 MB min/max. System Restore set to off.
Individual performance will vary with any particular or specific timings or
tweaks enabled by you. All programs were benchmarked with initial monitor
settings at 1024×768@75Hz. Your own mileage may very.
3DMark 2003
3D Mark 2003 was originally designed to measure performance specifically in
shader-heavy titles.
3DMark 2005
3D Mark 2005 takes advantage of more DirectX 9 functionality that many new game titles are using.
Aquamark3
Aquamark3 is a benchmark from Massive Development. For the most part
it is a DirectX 8.1 benchmark though it is run with DirectX 90c installed. Four
measurement sets were used. The first has high and low detail with Anti Aliasing
and Anisotropic filtering turned off. The second has high and low detail with
Anti-Aliasing (6x) and Anisotropic filtering (16x) set at max.
GL Excess
GL Excess is an OPENGL benchmark that is optimized for DX8.1.
Quake III high quality
Quake III still continues to hang around. This benchmark is one that
most can’t just let go of and it retains grandfather rights in the community.
Many of today’s games are based upon the Quake engine. The accuracy of this benchmark is becoming questionable with frame rates consistently around or over 300.
Serious Sam
Serious Sam uses OPENGL.
UT2003 Flyby
UT2004 Benchmark
Wolfenstein Enemy Territory: Railgun timedemo
Wolfenstein Enemy Territory uses an improved version of the heavily
modified Quake III engine from Return to Castle Wolfenstein. The Railgun time
demo results were recorded.
X2 Rolling Demo
X2 – The Threat is a teaser with a benchmark option for Egosoft’s
upcoming release. It does not use pixel shaders.
SplinterCell (Chinese Embassy Timedemo)
Splinter Cell is a DirectX 8.1 based game that is involves very GPU-intensive texture rendering.
Sisoft Sandra 2004
NBench-3
Specviewperf 8.0
SpecviewPerf measures the rendering performance of systems
running under OPENGL incorporating eight standard SPECopc applications viewsets:
- 3dsmax-03 , based on SPECapc for 3ds max 3.1 configured with the Open GL driver; includes three models containing an average of 1.5 million vertices each, and tests performance of scenes with different levels of lighting.
- catia-01, based on Dassault’s CATIA, with models containing up to two million vertices.
- ensight-01 , based on CEI’s EnSight engineering and scientific visualization application, covers both display-list and immediate-mode workloads.
- light-07 , based on traces of Discreet’s Lightscape radiosity application.
- maya-01 , based on traces of Alias’ Maya 5.
- proe-03 , based on SPECapc for Pro/ENGINEER 2001, measures two models in three modes — shaded, wireframe and hidden-line removal (HLR).
- sw-01 , based traces of the Solidworks 2004 application from Dassault Systemes.
- ugs-04 , based on SPECapc for Unigraphics V17, tests performance based on an engine model containing 4.1 million vertices.
The following two tests are targeted mainly towards CPU performance and will
show if any “flaws” are in board design affecting the ability of the
CPU to crunch through the data. While in render mode the two test programs virtually
bypass ram and GPU.
Adobe After Effects 6.0
Adobe After Effects is a tool to produce motion
graphics and visual effects for film, video, multimedia and the web. It is primarily
a 2D application using imported graphics or digital footage or self generated
effects. A test project was created combining many video footage
files, resizing and rasterizing effects, text animations and multiple layer
effects. This “average” combination was felt to best demonstrate advantages
and/or disadvantages that a real world user may experience rather than isolating
and benchmarking a particular effect.
There is no official benchmark for After Effects
but render tasks can be timed to show specific results. Rendering, or the task of building
and compiling frames, is primarily CPU intensive and After Effects generally bypasses
the video card to rely solely upon the processor for speed. The time taken
to a render 900 frame Quicktime 720×486 movie file (animation codec, millions of colors, best quality) shows how fast the processor is working on the
given task.
Which wins?
Service Pack 1 tallied 63 “wins” out of a total of 108 scores. Service Pack 2 took only 42 possibilities with the rest nullified by matching scores. It can be said that, on average, Windows XP Professional with only Service Pack 1 installed is faster 2/3 of the time. The table below is the sum of all resolutions for each category.
All resolution scores combined
|
SP1 score
|
SP2 score
|
SP 1 Percentage faster than SP2
|
|
3D Mark 2003
|
31684
|
30888
|
2.6%
|
|
3D Mark 2005
|
12689
|
12399
|
2.3%
|
|
Aquamark
|
AA off, Aniso Off Details vlow |
546
|
542
|
0.7%
|
AA off, Aniso Off Details vhigh |
217
|
239
|
-9.3%
|
|
AA 6x, Aniso 16x Details vlow |
400
|
393
|
1.7%
|
|
AA 6x, Aniso 16x Details vhigh |
173
|
205
|
-15.5%
|
|
Doom III
|
130
|
130
|
0.0%
|
|
GL Excess
|
66274
|
66082
|
0.3%
|
|
N-Bench 3 (1024×768 and 1600×1200 scores combined) |
Integer 1 (Landscape) |
6148
|
6094
|
0.9%
|
Integer 2 (Mandelbrot) |
5796
|
5796
|
0.0%
|
|
Float 1 (Flock) |
4950
|
5045
|
-1.9%
|
|
Float 2 (Rigid Body) |
4977
|
4919
|
1.2%
|
|
CPU Overall |
5467
|
5463
|
0.1%
|
|
Ninja 1 |
8122
|
8069
|
0.7%
|
|
Ninja 2 |
6914
|
6873
|
0.6%
|
|
Grand Touring |
8469
|
8423
|
0.5%
|
|
Star Fighter |
7258
|
7281
|
-0.3%
|
|
3D Overall |
7690
|
7661
|
0.4%
|
|
Overall Score |
6578
|
6562
|
0.2%
|
|
PCMark 2004
|
4781
|
4761
|
0.4%
|
|
Quake III high quality v.11
|
1640
|
1592
|
3.0%
|
|
Serious Sam
|
586
|
592
|
-1.0%
|
|
Splinter Cell Chinese Embassy
|
449
|
445
|
0.8%
|
|
UT2004 – Umark utility – 16 bots BR-Colossus – High Image Quality
|
585
|
577
|
1.4%
|
|
UT2003 – flyby
|
845
|
868
|
-2.7%
|
|
Wolfenstein Enemy Territory (railgun demo)
|
434
|
451
|
-3.8%
|
|
X2 Rolling demo
|
106
|
107
|
-0.9%
|
|
Sisoft Sandra CPU Arithmetic
|
18662
|
18746
|
-0.4%
|
|
Sisoft Sandra CPU Multi Media Benchmark
|
41543
|
41519
|
0.1%
|
|
Sisoft Sandra Memory Bandwidth
|
11740
|
11704
|
0.3%
|
|
SpecviewPerf 8.1
|
98.7
|
98.1
|
0.6%
|
|
After Effects 6.0
|
2.11
|
2.07
|
-1.9%
|
The test PC equipped with Windows XP Professional Service Pack 1 was an average of 0.5% faster than the same hardware with Service Pack 2 installed. The percentage difference between faster and slower is insignificantly small. Less than 2 or 3 percent in some benchmarks could be dismissed by most but when gamers are looking to squeeze every extra frame out of their machines…it may count.
Digital food for thought.