<div dir="ltr"><div>I don't think you have to worry about pointer size creating any sort of material difference. It is there, but the impact is very low. On the plus side, 64-bit code generation tends to be significantly more cpu-efficient with the 8 extra registers.<br><br></div>Personally I prefer 4GB of ram for a small laptop, but the more important feature is the M.2 SSD. To be useful as a laptop (and not a chromebook) that 16G SSD has to go. I threw a 128G (or larger) SSD into both of my Acers. And since swap will be on the SSD even having only 2G of ram isn't going to cause any real problems. The storage is by far the most important piece.<br><div><br>-Matt<br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Nov 28, 2014 at 3:54 AM, Carsten Mattner <span dir="ltr"><<a href="mailto:carstenmattner@gmail.com" target="_blank">carstenmattner@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I'm sorry for confusing people. There's an inherent overhead of the larger<br>
address space due to bigger pointers and some software will have<br>
a larger footprint due to that. Therefore depending on what you run<br>
you may have less free memory and the 2GB might be not as much<br>
as it was in 32bit mode. Same amount available but more used for<br>
book keeping if you want to explain it that way.<br>
<br>
On Thu, Nov 27, 2014 at 6:30 PM, Matthew Dillon <<a href="mailto:dillon@backplane.com">dillon@backplane.com</a>> wrote:<br>
> I'm not sure I understand what you are saying. The Acer chromebooks are<br>
> 64-bit haswell cpus. Why would there be less memory available in 64-bit<br>
> mode?<br>
><br>
> -Matt<br>
><br>
> On Thu, Nov 27, 2014 at 3:42 AM, Carsten Mattner <<a href="mailto:carstenmattner@gmail.com">carstenmattner@gmail.com</a>><br>
> wrote:<br>
>><br>
>> On Thu, Nov 27, 2014 at 7:18 AM, Alex Hornung <<a href="mailto:alex@alexhornung.com">alex@alexhornung.com</a>><br>
>> wrote:<br>
>> ><br>
>> > On 26/11/14 21:55, Carsten Mattner wrote:<br>
>> >> On Wed, Nov 26, 2014 at 5:01 PM, Cory Smelosky <<a href="mailto:b4@gewt.net">b4@gewt.net</a>> wrote:<br>
>> >>> On Wed, 26 Nov 2014, Tim Darby wrote:<br>
>> >>><br>
>> >>>> I just want to mention that the only "DragonflyBSD-certified"<br>
>> >>>> chromebook<br>
>> >>>> is<br>
>> >>>> selling for $179.99 on Amazon now:<br>
>> >>>><br>
>> >>>><br>
>> >>>><br>
>> >>>> <a href="http://www.amazon.com/Acer-C720-Chromebook-11-6-Inch-2GB/dp/B00FNPD1VW/ref=cm_cr_pr_product_top" target="_blank">http://www.amazon.com/Acer-C720-Chromebook-11-6-Inch-2GB/dp/B00FNPD1VW/ref=cm_cr_pr_product_top</a><br>
>> >>>><br>
>> >>>> Xmas present to myself? :)<br>
>> >>>><br>
>> >>>> Tim<br>
>> >>>><br>
>> >>> Not bad...aside from the 1366x768 resolution...that'd make it a no for<br>
>> >>> me.<br>
>> >> And with i386 dropped in 4.0 how useful are 2 gigs of ram?<br>
>> >><br>
>> > About as useful as before it was dropped.<br>
>><br>
>> Right :). My point is that without i386 in 4.0 you'll have to use older<br>
>> versions<br>
>> of Dragonfly or have less memory available in x86_64 mode. I'm not<br>
>> complaining i386 was dropped. Quite the opposite as dropping i386 might<br>
>> be an elegant way to solve the time_t limitation other systems incorporate<br>
>> non trivial solutions for because they keep i386 around.<br>
><br>
><br>
</blockquote></div><br></div>