After my last reinstall of Windows a few weeks ago, I haven't done any 
coding.  However, I decided to try installing MS's latest (free) 
version, and give the 3.7 source a try.

Following the notes in the included readme, I was able to compile Boost 
properly, which I then installed on my system.

The POV solution file converted without a problem, and the project 
compiled and linked fine.

-- 
...Ben Chambers
www.pacificwebguy.com

Post a reply to this message

Chambers wrote:
> After my last reinstall of Windows a few weeks ago, I haven't done any 
> coding.  However, I decided to try installing MS's latest (free) 
> version, and give the 3.7 source a try.
> 
> Following the notes in the included readme, I was able to compile Boost 
> properly, which I then installed on my system.
> 
> The POV solution file converted without a problem, and the project 
> compiled and linked fine.
> 

Neat, thanks for letting us know. You may want to try some benchmarking vs
the official beta binary (nothing significant has changed in the source
between beta 25a and the source 25b, so it should be a valid comparison).

-- Chris

Post a reply to this message

Chris Cason wrote:
> Neat, thanks for letting us know. You may want to try some benchmarking vs
> the official beta binary (nothing significant has changed in the source
> between beta 25a and the source 25b, so it should be a valid comparison).
> 
> -- Chris

Official: 4m 14s
VCe08: 4m 11s

Nothing to get too excited about, ~1% difference.

-- 
...Ben Chambers
www.pacificwebguy.com

Post a reply to this message

Chambers wrote:
> Chris Cason wrote:
>> Neat, thanks for letting us know. You may want to try some 
>> benchmarking vs
>> the official beta binary (nothing significant has changed in the source
>> between beta 25a and the source 25b, so it should be a valid comparison).
>>
>> -- Chris
> 
> Official: 4m 14s
> VCe08: 4m 11s
> 
> Nothing to get too excited about, ~1% difference.
> 

I meant to say, this was running the benchmark on both cores, on a C2D 
@2.4gHz.

-- 
...Ben Chambers
www.pacificwebguy.com

Post a reply to this message

Chambers wrote:
> Nothing to get too excited about, ~1% difference.

if you're interested, you may want to try using the PGO feature (IIRC it's
a pain to set up in the VS2005 version, VC8 may be better). if you decide
to do it, use the benchmark and a few other random scenes as the guide that
you render when generating the PGO data.

at least with the Intel compiler (back when we could still use it), the PGO
version could shave about 10% from rendering times.

-- Chris

Post a reply to this message