Barend Gehrels

Thanks for trying! My recent fixes should be in 1.70 Beta 1, I thought. So maybe it is not yet fixed for VS2017 Let's first wait on the regression matrix...

> I am also using tutorial's workflow that is why I never encounter to that issue. I am not against though. @barendgehrels what do you think? I don’t have that...

> Right, so this is different than what is described in the wiki tutorial where Geometry resides inside modular Boost. The tutorial is very good, and I followed it in...

@tinko92 , it is OK to close it? You can add it into your generic `.gitignore` file anyway.

INT_MAX is not just a large value, it is the max value. Internally calculations are made (e.g. determinate) multiplying values. We use larger types for this, but apparently this is...

Hi @tinko92 , thanks for this, it is very interesting. Can you also compare the (speed) performance? Side is called a lot in our code, so we have to know...

> @barendgehrels I know this benchmark is completely artificial. Do you have a large (as in computationally expensive), real-world problem for an Boost.Geometry algorithm that uses a side-predicate? Maybe that...

> here are my results: Very good! OK so it is not only better but also a bit faster, that is really good news! Thanks for your tests! So we...

@awulkiew thanks for heads up - normally I keep an eye on them but at the moment they appear at a rate that I cannot follow all of it. Because...

@awulkiew I shortly checked it in the rescaling branch, but that one is not yet prepared for this (does not compile yet). Should be fixed of course but cannot do...