Odd calculated deco "ripples" (was Re: RFC: color change for calculated deco)

[Miika Turkia]

On Mon, Jan 14, 2013 at 2:50 AM, Linus Torvalds
<torvalds at linux-foundation.org> wrote:
> On Sun, Jan 13, 2013 at 4:42 PM, Linus Torvalds
> <torvalds at linux-foundation.org> wrote:
>>
>> Hmm. Sadly, I don't have access to any machine with 128-bit FP "long
>> double". I wonder if a 32-bit version with the 80-bit "long double"
>> would make any difference. The Intel 80-bit FP format has 64 bits of
>> mantissa, iirc, so it would be an extra 12 bits, and if the ripple
>> goes away..
>
> Ugh. That looks like a disaster. Tons of compat libraries - all the
> gnome, cairo, pango stuff, and trying to build subsurface and
> libdivecomputer with -m32 and everything else too.. I don't think I
> have any native 32-bit machines left.
>
> Anybody who as a 32-bit build environment all set, and willing to just
> do a search-and-replace of "double" with "long double" in the deco
> code? Do the ripples go away?

I just created a 32-bit chroot environment to test this and did the
double to long double conversion. Same rippling effect as with the
"stock" subsurface.

miika