[Bf-committers] Jarred, please add me to your LinkedIn network

2013-08-08 Thread Dwight Hutto via LinkedIn
LinkedIn Dwight Hutto requested to add you as a connection on LinkedIn: -- Jarred, I'd like to add you to my professional network on LinkedIn. - Dwight Accept invitation from Dwight Hutto http://www.linkedin.com/e/-g

Re: [Bf-committers] Disassembled Blender

2012-10-09 Thread Dwight Hutto
On Tue, Oct 9, 2012 at 6:46 PM, Campbell Barton wrote: > ... to be clear, I meant optimizing C/C++ to get better assembler output. > > as I understood it, you wanted to check the number of CPU cycles each > assembler instruction does to see if the code can be tweaked to use > more-efficient/less i

Re: [Bf-committers] Blender command line 'cpulimit' in Linux

2012-10-09 Thread Dwight Hutto
On Tue, Oct 9, 2012 at 5:45 PM, Dwight Hutto wrote: > Has the command line usage of cpulimit been given any thought to in > regards to starting blender In terms of a switch to blender's usage. in a maximum mode for cpu utilization, or > any other command line optimizations

Re: [Bf-committers] Using 'const' on primitive function arguments passed by value (Please don't do this)

2012-10-09 Thread Dwight Hutto
On Tue, Oct 9, 2012 at 4:02 PM, Dahlia Trimble wrote: > I don't poke around much in blender source but I've found in c++ in general > that I kind of like being able to declare a function parameter as const, > for the simple reason that I don't see well and it helps me easily > differentiate foo(co

[Bf-committers] Blender command line 'cpulimit' in Linux

2012-10-09 Thread Dwight Hutto
Has the command line usage of cpulimit been given any thought to in regards to starting blender in a maximum mode for cpu utilization, or any other command line optimizations which could be utilized as the OS operates? -- Best Regards, David Hutto CEO: http://www.hitwebdevelopment.com ___

[Bf-committers] Disassembled Blender

2012-10-09 Thread Dwight Hutto
I had a question in an earlier thread about disassembled Blender. I'm not new to Assembly, but just a little naive. Has anyone ever tried to disassemble, and optimize instructions given by the compilers to the processors instruction set(cycle count)? If not, is this a good project to work on? --

Re: [Bf-committers] performance variation of BKE_add_object function

2012-10-07 Thread Dwight Hutto
> > Another option they ended up using was to have linked instances of the > object and 'make real' all at once. This implies threading, but still seems 'unrealistic' at the processor level. > -- Best Regards, David Hutto CEO: http://www.hitwebdevelopment.com _

Re: [Bf-committers] performance variation of BKE_add_object function

2012-10-07 Thread Dwight Hutto
On Sun, Oct 7, 2012 at 2:56 AM, Dwight Hutto wrote: > Now I'm not that great at Assembly, but it would seem that even more > time could be shaved by looking at the disassembled code, and removing > compiler specific instructions that may 'increase' time in the bin. I say

Re: [Bf-committers] performance variation of BKE_add_object function

2012-10-06 Thread Dwight Hutto
Now I'm not that great at Assembly, but it would seem that even more time could be shaved by looking at the disassembled code, and removing compiler specific instructions that may 'increase' time in the bin. -- Best Regards, David Hutto CEO: http://www.hitwebdevelopment.com _