Re: [Flightgear-devel] Canvas rendering speed

[Erik H. <er...@eh...>]

On 12/16/2016 08:13 PM, James Turner wrote:
>
>> On 16 Dec 2016, at 14:31, Erik Hofman <er...@eh...> wrote:
>>
>> I've streamlined Canvas rendering using Single Instruction Multiple Data
>> (SIMD) instructions which is able to calculate on 4 floating poit
>> variables simultaneously (e.g 3 or 4 value vectors).
>
> Are you planning further work in this area? I ask because I am currently looking at making a Skia backend for the Canvas (to replace Shiva), and Skia already has quite intensive SSE/NEON optimisations, in addition to an optimised OpenGL render which I expect to be substantially more performant than the current OSG Node/Drawable hierarchy. I don’t want to duplicate / repeat work if you’re planning to look at this further.

By all means, SSE and NEON work best when considered in the design 
stage. Adding it afterwards is always suboptimal. I might still update 
it in a place or two but don't let that detract you from adding Skia.

Erik

-- 
http://www.adalin.com - High performance virtual reality audio software.