[libre-riscv-dev] evaluation of bit-witdh scoreboards

[Luke Kenneth Casson Leighton]

On Monday, December 17, 2018, Jacob Lifshay <programmerjake at gmail.com>
wrote:

> On Sun, Dec 16, 2018, 21:48 Luke Kenneth Casson Leighton <lkcl at lkcl.net
> wrote:
>
> > Good about ROM. Yes i meant 565 profile. That needs different ROM and
> > separate special pixel fns right? Only 16 bit so 2x data rate output,
> just
> > not 2x less input data rate as quads are still fp16 x4 right?
> >
> R5G6B5 does not have sRGB correction, so it's just a linear transformation.
>
>
Ah ok.

> The inputs are 4x 32-bit floats
>
>
Still seems like it is worth an op, to do 1 FPx4 to RGB565 per cycle.

Hm can we use FP16 on the pixel colours? Might be able to get 2x speed that
way.

>
> > On Monday, December 17, 2018, Jacob Lifshay <programmerjake at gmail.com>
> > wrote:
> >
> > >
> > > libre-riscv-dev at lists.libre-riscv.org
> > > http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev
> > >
> >
> >
> > --
> > ---
> > crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
> > _______________________________________________
> > libre-riscv-dev mailing list
> > libre-riscv-dev at lists.libre-riscv.org
> > http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev
> >
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev at lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev
>


-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68