|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> Chambers <ben### [at] pacificwebguycom> wrote:
> > If what you want to do is compare different 64 bit systems to each
> > other, then it's possible that this will suit your needs completely.
> > Especially considering that the major bottleneck in POV is the double
> > precision FPU, which is the same either in 32 or 64 bit mode.
>
> Except that the de-facto chip for floating point calculations in 64-bit
> intel-based processors is SSE, while in 32-bit systems it's the FPU.
Uh - "chip"? I guess this wording might (re-) trigger some discussion... :P
> This
> means that most compilers will usually automatically use SSE when compiling
> for an x86-64 target and FPU when compiling for an x86-32 target.
We're talking about POV-Ray here, the 32-bit release of which installs with SSE
support enabled on suitable systems. At least that's what the beta page claims.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 18 Mar 2009 04:21:22 -0400, Warp wrote:
> Jim Henderson <nos### [at] nospamcom> wrote:
>> Generally 64-bit and 32-bit performance are going to be very similar;
>> the advantage to 64-bit systems is the ability to address more memory.
>
> Except that the x86-64 architecture has more processor registers than
> the x86-32 architecture, which means that 64-bit executables can be
> optimized by the compiler better than 32-bit ones. This can cause a
> measurable difference in speed.
>
> (OTOH 64-bit programs will inherently consume more memory than 32-bit
> ones, which may also affect speed the other way around. Which one is
> more significant with POV-Ray, hard to say. I'd say the registers
> probably win, though.)
Interesting, now that's something I didn't know.
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 3/18/2009 1:21 AM, Warp wrote:
> Except that the x86-64 architecture has more processor registers than
> the x86-32 architecture, which means that 64-bit executables can be
> optimized by the compiler better than 32-bit ones. This can cause a
> measurable difference in speed.
>
> (OTOH 64-bit programs will inherently consume more memory than 32-bit
> ones, which may also affect speed the other way around. Which one is more
> significant with POV-Ray, hard to say. I'd say the registers probably win,
> though.)
Both true, which is why you can't compare a 32bit app on one system to a
64 bit app on another.
However, there's nothing wrong with comparing the same 32 bit app on two
different 64 bit systems.
--
...Chambers
www.pacificwebguy.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chambers <ben### [at] pacificwebguycom> wrote:
> However, there's nothing wrong with comparing the same 32 bit app on two
> different 64 bit systems.
OTOH then you will be comparing how efficient those systems are at
running 32-bit apps. Is that really the goal?
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dkanter wrote:
> There seem to be quite a few questions, and I'll respond to them all at
> once.
Don't; that's not how newsgroups work ;)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I'm currently compiling beta 31 for Win64. Could I have the POV-Teams
permission to release this build, since it's only available for Win32?
--
...Chambers
www.pacificwebguy.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"dkanter" <dka### [at] gmailcom> wrote:
> I'd really like to continue using POVRay to benchmark, as I think it adds value
> for my readers, and it certainly is valuable for POV Ray users.
Just to ensure that it gets said; yes it does, yes it is, and thankyou. :)
Charles
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dkanter wrote:
> However, I cannot wait till beta 32 comes out (since that will probably be the
> middle of April), so I'd like to get my hands on beta 30 for win64. I don't
> want support, since I'm sure the benchmark test will run without crashing.
> Could Chris or someone email me the needed files?
I'll try to get one built. The issue for me was the boost libraries, this isn't
difficult to fix, just fiddly.
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 3/24/2009 4:34 PM, Chris Cason wrote:
> dkanter wrote:
>> However, I cannot wait till beta 32 comes out (since that will probably be the
>> middle of April), so I'd like to get my hands on beta 30 for win64. I don't
>> want support, since I'm sure the benchmark test will run without crashing.
>> Could Chris or someone email me the needed files?
>
> I'll try to get one built. The issue for me was the boost libraries, this isn't
> difficult to fix, just fiddly.
>
> -- Chris
As I've noticed. Though I've managed to compile all POV files for 64
bits, I can't link the final .exe because I can't convince Boost to
build 64 bit libraries.
Grr...
--
...Chambers
www.pacificwebguy.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chambers <ben### [at] pacificwebguycom> wrote:
> On 3/24/2009 4:34 PM, Chris Cason wrote:
> > dkanter wrote:
> >> However, I cannot wait till beta 32 comes out (since that will probably be the
> >> middle of April), so I'd like to get my hands on beta 30 for win64. I don't
> >> want support, since I'm sure the benchmark test will run without crashing.
> >> Could Chris or someone email me the needed files?
> >
> > I'll try to get one built. The issue for me was the boost libraries, this isn't
> > difficult to fix, just fiddly.
> >
> > -- Chris
>
> As I've noticed. Though I've managed to compile all POV files for 64
> bits, I can't link the final .exe because I can't convince Boost to
> build 64 bit libraries.
>
> Grr...
>
> --
> ...Chambers
> www.pacificwebguy.com
Chambers <ben### [at] pacificwebguycom> wrote:
> On 3/24/2009 4:34 PM, Chris Cason wrote:
> > dkanter wrote:
> >> However, I cannot wait till beta 32 comes out (since that will probably be the
> >> middle of April), so I'd like to get my hands on beta 30 for win64. I don't
> >> want support, since I'm sure the benchmark test will run without crashing.
> >> Could Chris or someone email me the needed files?
> >
> > I'll try to get one built. The issue for me was the boost libraries, this isn't
> > difficult to fix, just fiddly.
> >
> > -- Chris
>
> As I've noticed. Though I've managed to compile all POV files for 64
> bits, I can't link the final .exe because I can't convince Boost to
> build 64 bit libraries.
>
> Grr...
>
> --
> ...Chambers
> www.pacificwebguy.com
Chris and Chambers,
Thanks a ton for looking at beta 31 for me. I was able to get a hold of beta 30
from a kindly user, so I'll be able to include POV-Ray results, although they
are slightly outdated I suppose.
Just to add a comment on 64b vs. 32b performance, the major differences:
1. 16 registers vs. 8
2. 64 bit pointers vs. 32b pointers
3. >4GB of memory
2 is a subtle point - it substantially increases the pressure on caches and
TLBs, since cache lines are fixed size (64B).
David
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |