Switch vm_vec_scale2 back to ALU
fixmuldiv uses higher precision now
This commit is contained in:
parent
a8784bc486
commit
0b18b4140e
|
@ -113,7 +113,7 @@ void vm_vec_scale_add2(vms_vector &dest,const vms_vector &src,fix k)
|
||||||
//dest *= n/d
|
//dest *= n/d
|
||||||
void vm_vec_scale2(vms_vector &dest,fix n,fix d)
|
void vm_vec_scale2(vms_vector &dest,fix n,fix d)
|
||||||
{
|
{
|
||||||
#if 1 // DPH: Kludge: this was overflowing a lot, so I made it use the FPU.
|
#if 0 // DPH: Kludge: this was overflowing a lot, so I made it use the FPU.
|
||||||
float nd;
|
float nd;
|
||||||
nd = f2fl(n) / f2fl(d);
|
nd = f2fl(n) / f2fl(d);
|
||||||
dest.x = fl2f( f2fl(dest.x) * nd);
|
dest.x = fl2f( f2fl(dest.x) * nd);
|
||||||
|
|
Loading…
Reference in a new issue