[ltp] Re: What to take care of when buying a Lenovo? revisited
Micha Feigin
linux-thinkpad@linux-thinkpad.org
Fri, 15 Apr 2011 13:31:34 +0300
On 14/04/11 18:59, Stefan Monnier wrote:
>> Three years have passed since you said this, does your advice still
>> stand?
>
> I can't speak for him, but with regards to politics, the situation
> hasn't changed w.r.t nVidia which is still completely uncooperative.
> ATI/AMD on the other hand is now very cooperative for its GPUs (not sure
> how it compares to Intel nowadays, but judging from the quality of the
> drivers it seems to be in the same ballpark).
>
> I don't know if the situation in terms of power usage has changed much.
> Based on technical specs of recent AMD and NVidia chipsets for
> nettop-level systems, it seems like the difference between Intel and
> AMD/NVidia has been reduced, but practice tends to be very different
> from theory (especially when you add a Free Software requirement into
> the mix), so don't take my word for it.
>
>> Has the situation improved with nVidia drivers?
>
> There are now 2D Free Software drivers for most NVidia chipsets and they
> work OK. But this is *despite* NVidia rather than thanks to them, so
> it's now on par with AMD/Intel's support and progresses slower.
>
>> If one were to buy an T420s today and had to choose between IntelŽ HD
>> 3000 and NVIDIAŽ QuadroŽ NVS4200M, which one is the one to get? Is
>> there a chance nVidia drivers will improve to a point in the next
>> three years that it will be worthwhile to have an nVidia optimus card
>> running under Linux? And lastly, is there any harm in getting the
>> laptop with the nVidia chip and disabling it for the time being if the
>> price of the two T420s is nearly the same?
>
> I'd recommend you vote with your wallet and stay away from NVidia.
>
Although my experience is that the NVIDIA closed source drivers are much better
than the ATI ones, so it's also a question of whether you insist on open source
drivers, which are 2D only for NVIDIA, or are willing to work with the
proprietary ones.
>
> Stefan
>