[gentoo-user] i386 vs amd64

[gentoo-user] i386 vs amd64

[Sean]

I have a dual opteron here and I am thinking of putting Gentoo on it. I 
am trying to decide to go with either the amd64 or i386 version.

 From looking over the online portage database it does appear that at 
least openoffice-bin 1.1.5 lists as available and 2.0 in testing for 
amd64, one of my main apps I use often, beside firefox and thunderbird.

I have several Loki game titles around, are they able to run on amd64 
gentoo?

So I am asking some Gentoo amd64 users, are you happy with the version 
or would you have gained more with i386?
Do most applications work on amd64 or are there some important ones missing?

I know above is a little broad, but I am just trying to get an idea of 
the state of amd64. I have been looking over the list for several days 
now, but not much traffic.

I also posted this to the amd64 list directly.


                 Thanks
                 Sean

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64

[Peter Gordon]

>From my understanding, if you install Gentoo/AMD64, you still can run and
execute 32-bit ('x86') stuff natively, so that shouldn't be too much of an
issue.

I don't own any 64-bit hardware though, so I'm not certain about this.

--Peter
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64

[Bob Sanders]

On Thu, 20 Oct 2005 20:07:47 -0400
Sean <rsh.lists@comcast.net> wrote:
 
> I have several Loki game titles around, are they able to run on amd64 
> gentoo?
> 

Some do, with a bit of finding out when to wave the chicken.  Unreal Tournament
installs and runs without problem, once it's unmasked.  

Others I've run under amd64 -

I've had Railroad Tycoon2 running.

Half Life running under Crossover Office.

The Sims, Linux Edition runs, until asking for help, then it crashes.

Have never, ever, gotten Tropico running under Cedegra, on either x86 nor x86_64.
(And being a dictator is a critical need!)

Non-game - Both Comicworks and True-Basic Gold run under Cedegra.  Or used
			to, I haven't messed with them in awhile.

> So I am asking some Gentoo amd64 users, are you happy with the version 
> or would you have gained more with i386?

I've been running amd64 for over a year now.  About the only real problem I've stumbled
on has been some PVR issues, one with Abiword where it print.  But those are being worked
on, with the PVR issue resolved.

Pulling video from DV tapes via ieee1394 works well.   As does using a D-Link USB FM tuner.

> Do most applications work on amd64 or are there some important ones missing?
> 

Pure-FTPD doesn't work on amd64.  I've been unable to get lighttpd running.

My complaint at the moment is I have to run my monitor at 1280x1024 if I want 3D acceleration.
But that's Gfx card/driver interaction that'll get sorted out eventually.  Normally. I should
be running at 1600x1024.  At work, I've been running a 23" flat panel at 1920x1200 with
3D acceleration on an amd64 system, though I have problems with ut2004 on that system,
while UT runs great.  (Both systems use an nVidia 660GT.)

Bob
-- 
-  
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64

[Sean]

All,

	Thanks for your responses, I plan to try out the amd64 version.

			Thanks,
			Sean
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64

[Justin Patrin]

On 10/22/05, Sean <rsh.lists@comcast.net> wrote:
>
>
> All,
>
>         Thanks for your responses, I plan to try out the amd64 version.
>

Be very careful if you're doing any cross-compiling. The system
headers in Gentoo AMD64 are hacked to allow compiling for both 64 and
32-bit. If you try compiling for, say, ARM and it picks up the system
headers you get *nothing*....wonderful, eh?

To Gentoo devs:
Might I suggest that the system headers default to 32-bit?

--
Justin Patrin

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64

[Scott Tiret]

[-- Attachment #1: Type: text/plain, Size: 828 bytes --]

On Thu, 2005-10-20 at 20:07 -0400, Sean wrote:
> I have a dual opteron here and I am thinking of putting Gentoo on it. I 
> am trying to decide to go with either the amd64 or i386 version.
> So I am asking some Gentoo amd64 users, are you happy with the version 
> or would you have gained more with i386?
> Do most applications work on amd64 or are there some important ones missing?

I have been running an x86_64 (amd64) system for a few months now.  The
only thing I have been missing is a 64bit version of Macromedia
Shockwave plugin.  Apparently, there is no 64bit version for this
proprietary software.

Everything else is fine.  I have all I need on my desktop.  x86_64
version of Openoffice-bin (rc3) takes a long time to open, but is
promising.

Good luck,

-- 
Scott Tiret <stiret@oneredshoe.net>

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] i386 vs amd64

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 710 bytes --]

On Fri, 21 Oct 2005 08:54:55 -0400, Scott Tiret wrote:

> I have been running an x86_64 (amd64) system for a few months now.  The
> only thing I have been missing is a 64bit version of Macromedia
> Shockwave plugin.  Apparently, there is no 64bit version for this
> proprietary software.

There's an ebuild on the forums to make the 32 bit nsplugins work with 64
bit Konqueror, but not for KDE 3.5 yet.

> Everything else is fine.  I have all I need on my desktop.  x86_64
> version of Openoffice-bin (rc3) takes a long time to open, but is
> promising.

Turn off Java in OOo prefs, it makes a huge difference to start up times.


-- 
Neil Bothwick

Only an idiot actually READS taglines.

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-user] i386 vs amd64

[Rob]

Scott Tiret wrote:
> On Thu, 2005-10-20 at 20:07 -0400, Sean wrote:
> 
>>I have a dual opteron here and I am thinking of putting Gentoo on it. I 
>>am trying to decide to go with either the amd64 or i386 version.
>>So I am asking some Gentoo amd64 users, are you happy with the version 
>>or would you have gained more with i386?
>>Do most applications work on amd64 or are there some important ones missing?
> 
> 
> I have been running an x86_64 (amd64) system for a few months now.  The
> only thing I have been missing is a 64bit version of Macromedia
> Shockwave plugin.  Apparently, there is no 64bit version for this
> proprietary software.
> 
> Everything else is fine.  I have all I need on my desktop.  x86_64
> version of Openoffice-bin (rc3) takes a long time to open, but is
> promising.
> 
> Good luck,
> 
I thought the email might be a good place to ask for some ideas:

I don't want to start a 64bit vs 32 bit war, or a Windows versus *nix
war, but it has been my experience so far that the fastest benchmarks
for a highly computation intensive program written in Numeric Python
came on my 3.5Ghz P4 laptop with hyperthreading- on Windows.  Also,
running the same program on an AMD Opteron gave me a slower speed no
matter what OS I was using.  I performed the experiments when the
Opteron was first introduced.  I paid a high price for the fastest chip
I could find- I don't remember the exact speed.  I haven't tried the
test lately though.  Maybe it has gotten much better.

Do not ask me why it happened, I have no idea. But even now, Windows+P4
has consistently been 3x faster in execution time than any Python on 32
bit *nix systems.  The specific program is a Numeric Python port of the
NEC2 EM Simulator program which calculates the Norton-Summerfield ground
coefficients under an antenna.  It makes much use of Complex-64
variables.  I ported it from FORTRAN so I could more easily see how the
program worked.

I am baffled by the behavior.  The only thing I can figure might be
occuring would be that the *nix 64 bit toolchains are much younger than
the 32 bit ones.  But as the 32 bit Numeric Python on Windows is still
3x faster than the *nix equivalents, I have asked Activestate, the
Windows Python provider, if they do anything special when compiling the
code and they say no.  I think they said that they use some ordinary MS
comiler.

Any ideas would help me to put to rest the problem.  I say it is a
problem as I really don't want to boot into Windows XP to run scientific
programs in Numeric Python.

Thanks,

Rob.
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64

[Richard Fish]

Rob wrote:

>I don't want to start a 64bit vs 32 bit war, or a Windows versus *nix
>war, but it has been my experience so far that the fastest benchmarks
>for a highly computation intensive program written in Numeric Python
>came on my 3.5Ghz P4 laptop with hyperthreading- on Windows.  Also,
>running the same program on an AMD Opteron gave me a slower speed no
>matter what OS I was using.  
>

Did you recompile or install a 64-bit version of python for the 
Opteron?  If not, you are comparing a 32-bit processor doing 64-bit 
computations using 32-bit instructions to a 64-bit processor doing 
64-bit computations using 32-bit instructions, which is probably not 
what you intended.

>I am baffled by the behavior.  The only thing I can figure might be
>occuring would be that the *nix 64 bit toolchains are much younger than
>the 32 bit ones.  But as the 32 bit Numeric Python on Windows is still
>3x faster than the *nix equivalents, I have asked Activestate, the
>Windows Python provider, if they do anything special when compiling the
>code and they say no.  I think they said that they use some ordinary MS
>comiler.
>  
>

Well, MS makes _very_ good compilers, from a speed standpoint.  It's 
difficult to find an objective comparision between the Visual C++ 
compiler and GCC, but it would not surprise me at all if the VC++ 
produced code that was 10-30% faster for many cases.

For example, VC++.NET can use "whole program optimization", where much 
of the optimization is delayed until the linking step, when data from 
all comilation units (.o files) can be used to make decisions.  This 
results in more inline functions, more unreachable code being deleted, 
better function ordering, and so on.

As for being 3x faster on Windows, that seems a bit strange to me.  Were 
the "*nix" versions of python compiled specifically for the processor?  
Running code 'optimized' for a 386 on a modern processor would account 
for this difference in performance.

Note that the OS should make very little difference here.  You could 
probably do a similar comparison using the ActiveState python on Windows 
vs the cygwin version of python.

-Richard

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] i386 vs amd64/ forget my Python comparisons

[Rob]

Richard Fish wrote:
> Rob wrote:
> 
>> I don't want to start a 64bit vs 32 bit war, or a Windows versus *nix
>> war, but it has been my experience so far that the fastest benchmarks
>> for a highly computation intensive program written in Numeric Python
>> came on my 3.5Ghz P4 laptop with hyperthreading- on Windows.  Also,
>> running the same program on an AMD Opteron gave me a slower speed no
>> matter what OS I was using. 
> 
> 
> Did you recompile or install a 64-bit version of python for the 
> Opteron?  If not, you are comparing a 32-bit processor doing 64-bit 
> computations using 32-bit instructions to a 64-bit processor doing 
> 64-bit computations using 32-bit instructions, which is probably not 
> what you intended.
> 
>> I am baffled by the behavior.  The only thing I can figure might be
>> occuring would be that the *nix 64 bit toolchains are much younger than
>> the 32 bit ones.  But as the 32 bit Numeric Python on Windows is still
>> 3x faster than the *nix equivalents, I have asked Activestate, the
>> Windows Python provider, if they do anything special when compiling the
>> code and they say no.  I think they said that they use some ordinary MS
>> comiler.
>>  
>>
> 
> Well, MS makes _very_ good compilers, from a speed standpoint.  It's 
> difficult to find an objective comparision between the Visual C++ 
> compiler and GCC, but it would not surprise me at all if the VC++ 
> produced code that was 10-30% faster for many cases.
> 
> For example, VC++.NET can use "whole program optimization", where much 
> of the optimization is delayed until the linking step, when data from 
> all comilation units (.o files) can be used to make decisions.  This 
> results in more inline functions, more unreachable code being deleted, 
> better function ordering, and so on.
> 
> As for being 3x faster on Windows, that seems a bit strange to me.  Were 
> the "*nix" versions of python compiled specifically for the processor?  
> Running code 'optimized' for a 386 on a modern processor would account 
> for this difference in performance.
> 
> Note that the OS should make very little difference here.  You could 
> probably do a similar comparison using the ActiveState python on Windows 
> vs the cygwin version of python.
> 
> -Richard
> 
Hi Richard,

After going to Tom's Hardware and looking at the dizzying array of 
processors, chipsets, etc, and all of the benchmarks, I think my 
comparisons need to be more highly defined just on that basis alone.  So 
forget my email, haha.

In the end, what matters to me the most is time spent emerging Gentoo 
ports and perhaps trying to get Python working faster.  I used a stage 3 
tarball when setting up this computer.  Maybe I will try upgrading 
Python.  I saw that there is a utility for doing so,

However, in any case I think this Laptop is damned fast.   I believe  I 
compiled Gentoo's Xorg x11-base in about 10 minutes.   The laptop uses 
an Intel 3.4Ghz P4 4-E Prescott with 1Gig of DDR-2 memory.  I also 
didn't know that it supported AMD64/EM64T.  I got that from SiSoft Sandra.

This laptop is huge, weighs alot, and emits large amounts of hot air 
when compiling.  I bought it as a desktop replacement as sitting on a 
chair in front of a monitor is problematic for me due to back problems. 
  I believe this unit is one of those where some of HP's batteries 
melted. I saw that in the news.  So far mine has not.

Sincerely,  Rob
-- 
gentoo-user@gentoo.org mailing list