On 23.03.2015, at 18:14, tim Rowledge <tim at rowledge.org> wrote:

> On 23-03-2015, at 10:06 AM, Tobias Pape <Das.Linux at gmx.de> wrote:
> 
>> 
>> On 23.03.2015, at 17:57, tim Rowledge <tim at rowledge.org> wrote:
>> 
>>> On 23-03-2015, at 8:53 AM, Tobias Pape <Das.Linux at gmx.de> wrote:
>>>> 
>>>> For the trunk I get:	'7,070 per second. 141 microseconds per run.'
>>>> For my version I get:	'2,630 per second. 381 microseconds per run.'
>>>> 
>>>> Yes, it is slower, but we are long away from noticeable effects on the UI, I'd say.
>>>> Are there still objections?
>>> 
>>> Not so much an objection as a reminder to test these things on slower hardware as well as “my nice new Mac Pro 12 core with a twazillion gigglebytes of ram”. Say, a Raspberry Pi model B, for example.
>> 
>> I shall do that.
> 
> Thank you; it’s long been a bit of a joke that developers should be forced to work on the slowest possible machines so they don’t get carried away by the speed of {insert latest toy here}. Now with things like the Pi it is so easy to have a slower device around for testing that there really isn’t a good excuse.
> 
> If you haven’t already, consider installing the xrdp package on the Pi and a suitable xrdp client on your main machine. I find it a lot more convenient that having the Pi (or Pis in my case) hooked up to separate screens.

Well, the NEC EA241WM has both a display port
and a HDMI in, so it is ok-ish to hook it up.