One of the un-documented adjustments in the standard change log is the adoption of a new installer. We feel the 1.3.1302 installer is MUCH improved over the 1.1.1282 and 1.1.1286 versions. Rather than the previous installer, which was all console-based and offered very few options, the new installer offers an entirely revamped GUI similar to that found in the Linux UT2k3 and UT2k4 demo. The setup options include the install path, link path, PunkBuster client/server files, and startup menu entries (KDE/GNOME/CDE). Overall, this simple yet intuitive interface is improved tremendously when compared against its earlier counterpart.
A plague with previous versions of Doom 3 has been the inability to properly listen to in-game audio. However, with the 1.3 release, these unfortunate problems have been fixed. In the system menu, new audio options have been added in the 1.3 release. The options now available are for selecting the type of speakers (surround and stereo), enabling/disabling EAX 4.0 HD, and selecting the sound backend (BEST, OSS, and ALSA). Keep in mind, EAX 4.0 HD can only be enabled if your PC sound card and drivers are compatible with these specifications. Overall, these options are very welcome and should prove to be a valuable asset if you previously experienced audio conflicts when playing Doom 3 under Linux.
Below is our testbed configuration we used for the Doom 3 performance testing.
|Processor:||Intel Pentium 4 530 (3.0GHz)|
|Motherboard:||DFI LANPARTY UT 915P-T12|
|Memory:||2 x 512MB Corsair XMS PC4400|
|Graphics Card:||Gigabyte 6600GT (GV-NX66T128)|
|Hard Drives:||Hitachi 80GB SATA|
|Optical Drives:||Lite-On DVD-ROM|
|Cooling:||3 x 120mm case fans|
|Power Supply:||Thermaltake Purepower 460W|
|GCC (GNU Compiler):||3.4.3|
|Graphics Driver:||NVIDIA 1.0-7174|
The benchmarks used in this review were all using the timedemo1, which ships with Doom 3. Benchmarks were performed by entering timedemo demo1 into the console. As with all pure graphics testing here at Phoronix, we ran the demo first as a trial run and to allow the benchmark to cache into memory, followed by a second and final run where the average frame-rate was recorded. For comparison purposes, we also used Doom 3 1.1.1282 and 1.1.1286 to compare the results against those attained by the 1.3.1302 release.