> > In general POV-Ray compiles on 64bit systems but it won't be faster (in 
> > most cases even slower).  The main advantage would be to be able to 
> > address more than 4GB of memory.

> I couldn't believe this and did some testing with my own compilations of 
> POVRay.

  As noted in other articles of this thread, there's a slight confusion
of concepts here.

  64-bitness in itself does not bring any speed benefit to POV-Ray.
There's nothing special in having 64-bit pointers (and long integers,
which are not used almost anywhere in POV-Ray) which would bring
extra speed.
  However, making an AMD64-optimized binary can and most probably will
give speed benefits. However, this is not because AMD64 is a 64-bit
processor. It's because it has new features which the compiler can use
to make faster code and which the 32-bit processors don't have.

  In a 64-bit processor which has no extra speeding features compared
to its 32-bit equivalent (the UltraSparc being a good example of this)
you won't get any speed benefits from making a 64-bit binary.

-- 
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}//  - Warp -

Post a reply to this message

Warp schrieb:


>> > In general POV-Ray compiles on 64bit systems but it won't be faster (in 
>> > most cases even slower).  The main advantage would be to be able to 
>> > address more than 4GB of memory.
> 
>> I couldn't believe this and did some testing with my own compilations of 
>> POVRay.
> 
>   As noted in other articles of this thread, there's a slight confusion
> of concepts here.
> 
>   64-bitness in itself does not bring any speed benefit to POV-Ray.
> There's nothing special in having 64-bit pointers (and long integers,
> which are not used almost anywhere in POV-Ray) which would bring
> extra speed.
>   However, making an AMD64-optimized binary can and most probably will
> give speed benefits. However, this is not because AMD64 is a 64-bit
> processor. It's because it has new features which the compiler can use
> to make faster code and which the 32-bit processors don't have.

And thats the point. A 64Bit-compile on an AMD64 box will increase the 
performance of POVRay. If this increase of performance is caused by the 
64-bitness or the increased number of (SSE) registers is meaningless. It 
_will_ be faster. Thats the point. And so a version of POVRay for 
WindowsXP 64Bit edition will bring advantages against a 32 Bit Version.

Tonight I am going to make some tests with different versions of POVRay 
and the official POVRay benchmark file and also with a file with a lots 
of spheres to test the influence of the memory interface.

-- 

mat### [at] matweide
http://www.matwei.de

Post a reply to this message

> And thats the point. A 64Bit-compile on an AMD64 box will increase the 
> performance of POVRay. If this increase of performance is caused by the 
> 64-bitness or the increased number of (SSE) registers is meaningless. It 
> _will_ be faster. Thats the point.

  No, I don't think that was the point.

  The original question was: "Has anyone tried to compile 3.5 with 64bit
optimizations?"
  From my part I can answer: Yes, I have tried. It didn't give any
speed benefits. In fact, it was some percents slower than the 32-bit
version. (More specifically, I have tried it with an UltraSparc processor
which supports both 32-bit and 64-bit binaries.)

  That was the point.

  If the question had been "has anyone tried to compile an AMD64-optimized
version of POV-Ray" that would have been a different thing.

> And so a version of POVRay for 
> WindowsXP 64Bit edition will bring advantages against a 32 Bit Version.

  That depends on the CPU you are using. That statement of yours seems
to imply that a version compiled for a 64bit edition of WindowsXP will
automatically be faster, period, *regardless of the processor used*.
That's not true. The OS doesn't matter, it's the CPU which matters.

  Could we *please* stop widespreading false concepts? If you keep
claiming that "a 64-bit version of POV-Ray will be faster than
a 32-bit version" without any further details, you are just giving
misconceptions to people. People are going to think that there's
something magic in the 64-bitness which POV-Ray can benefit from
for extra speed.

  Instead, say "an AMD64-optimized version of POV-Ray will be faster
than a 32-bit version of POV-Ray run in the AMD64 processor".
  After all, it's not all that different from "a P4-optimized version
of POV-Ray will be faster than a version optimized for earlier processors
when run in a P4".

-- 
plane{-x+y,-1pigment{bozo color_map{[0rgb x][1rgb x+y]}turbulence 1}}
sphere{0,2pigment{rgbt 1}interior{media{emission 1density{spherical
density_map{[0rgb 0][.5rgb<1,.5>][1rgb 1]}turbulence.9}}}scale
<1,1,3>hollow}text{ttf"timrom""Warp".1,0translate<-1,-.1,2>}//  - Warp -

Post a reply to this message

Warp schrieb:


>> And thats the point. A 64Bit-compile on an AMD64 box will increase the 
>> performance of POVRay. If this increase of performance is caused by the 
>> 64-bitness or the increased number of (SSE) registers is meaningless. It 
>> _will_ be faster. Thats the point.
> 
>   No, I don't think that was the point.

So I see we understand the original posting in a different way.

>   The original question was: "Has anyone tried to compile 3.5 with 64bit
> optimizations?"

And goes on with: "I have a new PC with an AMD64 3200+ cpu (running 
Windows XP64)" So the processor we talk about has AMD64 extensions.

>   From my part I can answer: Yes, I have tried. It didn't give any
> speed benefits. In fact, it was some percents slower than the 32-bit
> version. (More specifically, I have tried it with an UltraSparc processor
> which supports both 32-bit and 64-bit binaries.)

The same for AMD64 here under Linux. Running 32Bit and 64Bit binaries.

>   If the question had been "has anyone tried to compile an AMD64-optimized
> version of POV-Ray" that would have been a different thing.

I understand the question in that way because the OP has telled us which 
hardware he is using.

>> And so a version of POVRay for 
>> WindowsXP 64Bit edition will bring advantages against a 32 Bit Version.
> 
>   That depends on the CPU you are using. That statement of yours seems
> to imply that a version compiled for a 64bit edition of WindowsXP will
> automatically be faster, period, *regardless of the processor used*.

Because WindowsXP 64Bit Edition supports only one processor-architecture 
(namely AMD64 or x86-64) there are no different processor architectures.

> That's not true. The OS doesn't matter, it's the CPU which matters.

Right.

>   Could we *please* stop widespreading false concepts? If you keep
> claiming that "a 64-bit version of POV-Ray will be faster than
> a 32-bit version" without any further details, you are just giving
> misconceptions to people. People are going to think that there's
> something magic in the 64-bitness which POV-Ray can benefit from
> for extra speed.

You are right here. Let me correct it:
A 64-bit version on AMD64 will be faster than a 32-bit version on AMD64.


-- 

mat### [at] matweide
http://www.matwei.de

Post a reply to this message

In article <4087bbac$1@news.povray.org>,


> And thats the point. A 64Bit-compile on an AMD64 box will increase the 
> performance of POVRay.

No, an AMD64 compile on an AMD64 will increase the performance. An Intel 
64 bit compile won't, if it will run at all. It's not due to the 
64-bitness, it's due to other enhancements the processor has available 
in 64 bit mode.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: <chr### [at] tagpovrayorg>
http://tag.povray.org/

Post a reply to this message

My apologies for starting such a controversial thread!  :)

Right now there are two versions of Windows XP 64bit, one for the Itanium 
and one for extended systems such as the Opeteron and AMD64.  I'm pretty 
sure I asked for a 64bit compile of POV for Windows XP 64bit for extended 
systems.  I didn't make my request specific enough for some of you, and 
again I apologize for that.

I'm really happy to see the reports of the Linux builds, and the apparent 
speed up that results.  I am thinking about replacing my Windows XP 64bit 
install with Mandrake 64bit, to take advantage of the successful compiles 
by Matthias, but there are even fewer drivers for my hardware available on 
Linux than there are on Windows XP 64bit at the moment.  <sigh>

I'm hoping someone will try a Windows XP 64bit for extended systems compile 
some time soon. I'll happily use my machine for testing new compiles!  :)

Rich Allen

Post a reply to this message

> A 64-bit version on AMD64 will be faster than a 32-bit version on AMD64.

  Given that the program was optimized for AMD64.
  Once again: 64-bitness does not give speed. AMD64 optimizations do.

-- 
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -

Post a reply to this message

Warp schrieb:


>> A 64-bit version on AMD64 will be faster than a 32-bit version on AMD64.
> 
>   Given that the program was optimized for AMD64.
>   Once again: 64-bitness does not give speed. AMD64 optimizations do.

Not the programm has to be optimiezed for AMD64. The compiler has to 
optimize the programm for the AMD64 architecture. I think nearly every 
programm will benefit from the increased number of registers.

-- 

mat### [at] matweide
http://www.matwei.de

Post a reply to this message

> >   Given that the program was optimized for AMD64.
> >   Once again: 64-bitness does not give speed. AMD64 optimizations do.

> Not the programm has to be optimiezed for AMD64. The compiler has to 
> optimize the programm for the AMD64 architecture. I think nearly every 
> programm will benefit from the increased number of registers.

  By "program" I was referring to the compiled binary.

-- 
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -

Post a reply to this message

On 18 Apr 2004 12:13:28 -0400, Rich <SrP### [at] ricoswebcom> wrote:

>What's being written 
>is that a program compiled to 32bit will run at simliar speeds on an AMD 
>XP and an AMD64 rated similarly.  When that same program is compiled to 
>64bit, using the optimizations intended for 64bit, the program will run 
>significantly faster on the 64bit processor (and not at all on the 32bit 
>processor).

I basically read this in the following way: the 32-bit core of the
AMD64 is identical to the Athlon series (XP, MP, Barton, Thoroughbred
...) and in addition the AMD64 has some 64-bit extensions like more
registers, enhanced instruction set (maybe), different memory
addressing, perhaps even a better memory caching scheme... Basically
in 32-bit mode it's just slightly better than the Athlon (higher bus
speed, larger and faster cache, better and more integrated core). I
may be entirely wrong here but this is the conclusion I am led to by
the statements above.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message