hopefully in 2.8 there is a button “do it” i mean we all live in 2017 and mankind landet on moon in 1969…“IT CAN NOT BE THAT DIFFICULT”
cheers! 
You can build 2.8 the sameway as juang3d builds the mantaflow build, you just check out a different branch, other than that it’s identical.
See! You just did it for the first time! 
hahaha
Thanks for the advice, regarding the libraries, cool, I found problems when trying to build with VS2015 and I was unable to configure it for 64bit, it was getting me just the 32bit version, so I found easier to go with the CMAKE gui, also I think is easier because you are able to understand what kind of Blender you are building 
And what about the extra parameter in cmake Sebbas told me that I needed, “-DWITH_CXX11=ON”, is not present in make.bat, isn´t it?
In the end I tried to make it as easy as possible and as understandable as possible for people that don´t code and don´t understand Visual Studio at all, so small and safe steps hehe but thanks for the advices, I´ll try them
@Piet
As LazyDodo said, instead of using the “fluid-mantaflow” branch you can use the “blender2.8” branch
Cheers!
EDIT: BTW, you can check the different branches here:
by default make.bat will just try to see what the best is it could build, it should have preferred x64 over x86, that being said, you can force a specific architecture by adding the visual studio version (2013/2015/2017) or x86 or x64 to the make.bat parameters
make release 2015 x64
would force msvc 2015 and a 64 bit build.
It’s not needed afaik for msvc, the compiler doesn’t allow you to turn off C++11/14 support like gcc does.
We have similar goals 
I’d like to get to a point where you just clone git, and run ‘make release’ and it’ll do all the manual steps you currently need to do automatically , but i’m not quite there yet.
the best mantaflow foam I ve seen.
what can be the foam settings??
That’s actually quite good. Glad to see some quality work from people.
Now that’s really impressive, given how painfully slow (and from what I can tell, single threaded) the current fluid system is, I’d really like to know how long it took to process the simulation for that.
11 hours 45 minutes said the translated tweet.
ahh, OK, so no matter what, fluid sim is still hard work, but I guess that’s to be expected no matter what software is used and of course, we have no idea what spec’s of PC he was using, plus mantaflow/blender is I guess very much in beta stage, not fully optimised, etc, etc.
Even so, just the fact it can be done and look that good bodes very well for the future.
Here, a test about surface tension factor.
It would be super nice if there was a dissolve factor for the velocities in the volume after the smoke leaves it… I get really amazing smooth looking turbulence in my sim until it does a full rotation and hits the velocities left by the smoke there before it
You are right. Angular velocity should be sensible to a damping produced by medium although vorticity is not null.
Fluid Guiding may help in some cases but it is not easy to control.
Because using it to damp angular velocities implies that it provides accelerations for linear velocities, too.
This is becoming a real pain lol
Ok, I see the problem but I don´t know how to reproduce it, can you explain a bit your setup and why do you think this is happening?
Cheers.
Well the velocities keep adding up on simulations that are in the same place or move around and then come back to a place a fluid was before… it basically gets so chaotic the entire smoke sim just has random noise from all the left over turbulence left by older smoke
It seems the attachments did not upload
Oh but they did and it posted xD but after a little while they died
Pretty cool renders!
graphicall.org have noting to do with BF its a independent group who offer blender builds. So here I am agree with Fweeb. Also compiling your self is very easy i found out.