Alright, something's fishy here. When Apple announced that the
new MacBook Pro has two NVIDIA GeForce chips -- the 9400M and the 9600M GT -- the focus was on what that means for battery life. Absent any mention of Hybrid SLI, we assumed that was all, but PC Mag has posted some eyebrow-raising benchmarks comparing the new MacBook Pro to HP's Pavilion HDX16t, which also features a 9600M GT. While the MacBook Pro test model fell behind the Pavilion in most benchmarks due to its slower processor, its
Crysis framerate beat that of the Pavilion by 24.1 frames per second -- 41.9 over 17.3. That doesn't make a lot of sense, unless you look at benchmarks of a desktop with NVIDIA's similar GeForce 9300 chipset and a GeForce 8500 GT -- turns out
Crysis runs 12.63 frames per second faster (29.19 over 16.56) in Hybrid SLI than it does on the 8500 GT alone. Is the MacBook Pro running in SLI mode when
set for performance? We don't have confirmation of that, but we'll put it to the test in our
forthcoming review -- until then, feel free to grab a grain of salt while freaking out anyway.
Update: Sorry, folks -- NVIDIA's just
posted a support doc that says the MBP doesn't support Hybrid SLI in either OS X or Windows -- and when running Windows, it's locked into using the 9600M GT. We're not sure where that
Crysis boost is coming from -- GDDR3 vs GDDR2, perhaps -- but we'll dig deeper in our review. Stay tuned.
Read - PC Mag (MacBook Pro benchmarks)
Read - Hot Hardware (NVIDIA GeForce 9300 desktop motherboards benchmarks)
posted October 15th 2008 3:43PM
Aren't Nvidia windows drivers (Bootcamp) giving more informations about this ?
Does anyone know what settings they ran the game at? Depending on the settings that better ram in the macbook card could make a huge difference.
Did anyone else realize the data in this chart doesn't make sense at times. Like the last column. It says that the "lowest score is best", but a middle time is bolded as the "winner" even tho it is 10 seconds slower than the fastest time. There are several places in the chart where this happens.
What gives? When they can't even find the lowest number, I get a bit nervous about the overall quality of the tests.
What I don't understand is that if you look at those results on PCWorld's site. Other than the gaming graphics, the MBP is slower than all the other computers listed even WITH the same processor (ie Dell Studio 15). And the Dell has esata and HDMI (not that I'm a fan of Dell in general). How can Apple justify the price of almost $1000 difference for the same basic specs (again - aside from video)??? I was anxiously awaiting the hardware news from Apple but this big price drop they mentioned never happened, the hardware doesn't seem justifiably more expense (it's a brick of aluminum - so what - it's a laptop).
http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php?p_faqid=2243
This support doc states that currently it is not possible two have the two gpu's work at the very same time in the new Apple notebooks.
But, they also say: "Apple's hybrid graphics technology is supported under the MacOS X operating system version 10.5.6 and higher only." Am I reading this wrong or are they saying that when 10.5.6 is released it will support the method of two gpu's working at same time?
Maybe the Benchmark testing dude is dyslexic like my mate and saw 41Fps instead of 4fps
Hi. Snow Leopard will support it, I'm sure. Windows 7 will be launched as well and will support SLI for different cards a few months later. Apple Pro notebooks sales will go off the roof (especially when it launches the 17" which probably has two next generation GPUs and a quad core).