I recently got a new Lenovo Thinpad T61:
Processor: Intel(R) Core(TM)2 Duo CPU T8300 @ 2.40GHz
Ram: 2 GB (not sure what the latency is).
HDD: 160 GB 5200 rpm
It has the 965 chipset (with the IGP X3100 graphics card - supposed a lot better than the GMA 950, but that's not saying much).
Since the X3100 has absolutely no benchmarks published, I've finally done a couple myself.
First of all: glxgears (which everyone says is not a benchmark)
$ glxgears
5471 frames in 5.0 seconds = 1094.008 FPS
5618 frames in 5.0 seconds = 1123.475 FPS
5606 frames in 5.0 seconds = 1121.063 FPS
It sucks (my Pentium-m 1.6 Mhz laptop with a GeForce FX Go5200 gets ~2100 fps).
That's probably why Eric Anholt (http://www.anholt.net/) wouldn't post any info on performance, it's not so good. Also, as everyone keeps telling me glxgears is not a benchmark, fine...
So here's some UT 2004 deathmatch - DM-Rankin, at 800x600@32 (not that it matters - lower resolutions and color depth did almost nothing (although fullscreen does help - a little - the last row is fullscreen):
5.563165 / 12.212557 / 45.396217 fps -- Score = 12.217196
5.556774 / 12.213729 / 46.879482 fps -- Score = 12.218822
5.914872 / 13.538128 / 115.044312 fps -- Score = 13.543859
The score is the avg. frame rate per second - more or less (they seem to do some minor adjustments). It still sucks, although it is playable (with less bots).
//benchmark script (from here):
/opt/ut2004-demo/Benchmark $ cat benchmark.sh
#!/bin/sh
ut2004-demo dm-rankin?spectatoronly=1?numbots=12?quickstart=1?attractcam=1 -benchmark -seconds=77 -ini=default.ini -exec=/opt/ut2004-demo/Benchmark/Stuff/botmatchexec.txt
//end script.
//Here is the botmatchexec.txt:
showhud
ship
//end snip
If you lower the number of bots, it's a bit better, I think...
Bashmark makes me think that the issue isn't a CPU limitation, but then again I'm not entirely sure what bashmark is testing:
//start bashmark snips
$ bashmark
#######################################################
: T E S T : :S C O R E : : R A T I O:
:-----------------------------------------------------:
:Cpu, Integer : : 4746: : +342%:
:Cpu, Floating point : : 3954: : +421%:
: : : : : :
:Memory r/w (cached) : : 7418: : +517%:
:Memory de-/alloc : : 1607: : +146%:
: : : : : :
:Multithreading : : 1903: : -24%:
#######################################################
: S Y S T E M I N F O :
-------------------------------------------------------
2x Intel(R) Core(TM)2 Duo T8300 @ 800.000MHz, L2 3072KB
Linux 2.6.24-tuxonice-r4
GCC 4.2.3
92KB binary size
#######################################################
: R E F E R E N C E S Y S T E M I N F O :
-------------------------------------------------------
Reference system was Geno's pc with:
Athlon XP 1800+ 1575.631MHz, 256KB
Linux 2.6.11-ck1
GCC 3.4.3-20050110 (compiled with standard cflags)
glibc 2.3.4 (with nptl)
128KB binary size
Scores gathered on March, 30th. 2005 with bashmark 0.6
//end bashmark snip
Hard drive reads (I really don't care about write speeds unless I'm setting up RAID):
//start hdparm snip
# hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads: 3084 MB in 2.00 seconds = 1543.44 MB/sec
Timing buffered disk reads: 130 MB in 3.04 seconds = 42.73 MB/sec
//end hdparm snip
My 3 year old laptop (with 7200 rpm's as opposed to 5200 rpm's like my current laptop) gets about 8 MB/sec less.
Power usage is pretty good (via powertop):
16.3 W (3.4 hours) at idle (albeit with firefox and my wireless sending kernel interrupts like mad). The screen doesn't dim at all when on battery power - so it should actually be a bit better than this. My old Pentium-M laptop gets about 21 Watts at idle (which is pretty good, imho).
My OLPC get's less than six, and my server (with similar architecture) gets about 8 Watts at idle - and only a couple watts more at 100% CPU usage. Whatever you have to say about the OLPC, hardware-wise it's phenomenal in what it does at that level of power usage.
Long story short, this thing is weak on the graphics department, but otherwise it seems pretty quick. It also works quite well in Linux
The gaming stuff is most important since I'm getting back into that sort of programming...
more later...
