If you hit issues with DoubleDouble and/or generic math algorithms in
Julia's stdlib, please do file issues. We want to get to the point where
*everything* just works.

On Mon, Jun 20, 2016 at 4:31 PM, Chris Rackauckas <[email protected]>
wrote:

> I see. I surely can't wait for native bigs, but there are definitely more
> pressing issues to work on first. Tim's proof of principle results look
> really nice.
>
> I tried DoubleDouble and ran into some errors. I don't think it worked
> with all standard library math. I'll probably take another look at it in a
> month. That is definitely the right direction to go for now.
>
> On Monday, June 20, 2016 at 7:09:57 PM UTC+1, Stefan Karpinski wrote:
>>
>> I think it might actually be easier for BigFloat since BigFloats are
>> fixed-size, whereas BigInts are variable-size.
>>
>> Chris, there is a DoubleDouble package
>> <https://github.com/simonbyrne/DoubleDouble.jl>, which implements
>> efficient higher-precision floating-point arithmetic, albeit not IEEE
>> 128-bit floats. As soon as hardware and LLVM support 128-bit IEEE floats,
>> Julia can easily support them as well – as I'm sure you realize, much more
>> easily than any other system.
>>
>> Nobody wants BigFloats to be inefficient; the current state of the
>> compiler's ability to reuse them and eliminate allocations simply isn't as
>> good as it could potentially be. That doesn't mean that this won't be
>> improved in the future – it will be, although it's hard to say when since
>> there are a lot of competing priorities and a limited number of people who
>> can do the kind of compiler work necessary to improve this situation.
>> Fortunately, the problem is closely related to a number of other
>> performance issues that we also need to address (strings, array views).
>>
>

Reply via email to