Guess who's gone all 64bit!!!!!

Wayne wrote:
I wonder how much difference the similarly clocked Athlon 64 would actually make over the Athlon 2800+ I'm running now?  Haven't really seen a site that adequately explains the difference in real-world terms.

Wayne

Well, I never thought I'd hear myself syaing this, but 64bit windows with the latest 64bit radeon drivers feels just as responsive on this Athlon 64 3200+ with 1MB L2 cache as MorphOS on my Peg G3.
 :-o

[bloodline]

mdma wrote:

Wayne wrote:
I wonder how much difference the similarly clocked Athlon 64 would actually make over the Athlon 2800+ I'm running now?  Haven't really seen a site that adequately explains the difference in real-world terms.

Wayne

Well, I never thought I'd hear myself syaing this, but 64bit windows with the latest 64bit radeon drivers feels just as responsive on this Athlon 64 3200+ with 1MB L2 cache as MorphOS on my Peg G3.
 :-o

BURN THE HERETIC!!!!

Well, I never thought I'd hear myself syaing this, but 64bit windows with the latest 64bit radeon drivers feels just as responsive on this Athlon 64 3200+ with 1MB L2 cache as MorphOS on my Peg G3.

Perhaps, but the Windows machine is at least usable for something aside from "having one" and "hobbyitis".  (For the record, I'd say the same about the AmigaOne/OS4).

[Karlos]

mdma wrote:

Well, I never thought I'd hear myself syaing this, but 64bit windows with the latest 64bit radeon drivers feels just as responsive on this Athlon 64 3200+ with 1MB L2 cache as MorphOS on my Peg G3.
 :-o

I don't really know what to say in response. Do we congratulate them for acheiving this level of performance?

Or do we ridicule them severely for requiring such a hardware base to achieve it compared to the comparatively pathetic G3 Peg. Especially if we assume 64-bit recompiled version of the OS brushes aside all the previous x86 legacy excuses.

:-?

[iamaboringperson]

Karlos wrote:

T_Bone wrote:
I want a **TRUE** 64 bit CPU!!!

64 bit ASSEMBLY OPCODES baby!! YEA!
(with that many opcodes, assembly would pretty much be a high level language)

Well, suppose you made a super RISC load/store architecture where you had, eg 65536 integer and 65536 floating point registers, you could easily have 64-bit opcodes. 16-bits for the instruction and up to 16-bits each for each register definition :lol:

Ew! Just imagine the length of the context switches between tasks!

Karlos wrote:

mdma wrote:

Well, I never thought I'd hear myself syaing this, but 64bit windows with the latest 64bit radeon drivers feels just as responsive on this Athlon 64 3200+ with 1MB L2 cache as MorphOS on my Peg G3.
 :-o

I don't really know what to say in response. Do we congratulate them for acheiving this level of performance?

Or do we ridicule them severely for requiring such a hardware base to achieve it compared to the comparatively pathetic G3 Peg. Especially if we assume 64-bit recompiled version of the OS brushes aside all the previous x86 legacy excuses.

:-?

Oh, I have no doubt that once it's been installed for a few weeks and the registry starts getting bloated with crap, that it'll be just as slow as 32bit windows currently is on the same laptop.

Windows and Linux (other than gentoo or a self compiled distro), both run in the mixed 64bit/32bit mode anyway, so they can use the x86 legacy excuse for a few more years.

BURN THE HERETIC!!!!

If I don't die, am I a witch? ;-)

[the_leander]

mdma wrote:

BURN THE HERETIC!!!!

If I don't die, am I a witch? ;-)

Don't worry about it too much mdma, once you've got the burnt taste out of your mouth and some fresh clothes you'll be good to go. You'll get used to it after they've tried it a few times ;-)

@ Wayne

I went from a 2Ghz AthlonXP to a 2Ghz Athlon64 and the speed difference was as great if not greater then going from a 1.3Ghz duron to the AthlonXP @ 2Ghz, they really are that fast, as for ray tracing, I've been playing with POV over the past few days and of the intro scenes that come with the initial package, I've not come up against one that doesn't complete with all the bells and whistles at 1280x1024 in under a minute. This is with an Athlon64 - 3000+ (2Ghz) running BeOS R5 with 512Mb of ram, your milage may vary with windows.

[T_Bone]

mdma wrote:

BURN THE HERETIC!!!!

If I don't die, am I a witch? ;-)

Possibly a duck!  :python:

[bloodline]

I've been reading Microsofts's Technical docs... apparently the x86's FPU (the x87) registers are not saved during a context switch in XP64... this means that you can't use the crappy x86 FPU anymore!!! It took 20 years, but finally they managed to get rid of that peice of junk :-D

This also means that the MMX and 3DNow! instruction sets no longer work.

In future all Floating point work is to be done by the Vector units, SSE and SSE2 :-)

This is good news all around!!! This also explains why we are seeing speed ups in 64bit version of windows when running 64bit apps.

Forgive me.  I don't keep up with Windows a lot, but I didn't know the 64-bit version of Windows was even publicly available.

Wayne

[Karlos]

bloodline wrote:
I've been reading Microsofts's Technical docs... apparently the x86's FPU (the x87) registers are not saved during a context switch in XP64... this means that you can't use the crappy x86 FPU anymore!!! It took 20 years, but finally they managed to get rid of that peice of junk :-D

You mean a task context switch? How frequent is that compared to the number of instructions executed between switches? I know the x86 FPU was a {bleep} to code for properly, but register save on context switch is not that big hit. On any x86 from the last 10 years the chances that the registers weren't just moved to/from L1 cache are almost nil.

So now, instead of backing up a handful of x86 FPU registers when a task switch occurs, you instead to back up the vector units register file? These units are far bigger than the old FPU register file - how does that make it faster?

This also means that the MMX and 3DNow! instruction sets no longer work.

In future all Floating point work is to be done by the Vector units, SSE and SSE2 :-)

This is good news all around!!! This also explains why we are seeing speed ups in 64bit version of windows when running 64bit apps.

Buh... surely FPU code designed for sequential execution is not going to run faster in the vector unit, just becasue it is a vector unit.

[bloodline]

Karlos wrote:

bloodline wrote:
I've been reading Microsofts's Technical docs... apparently the x86's FPU (the x87) registers are not saved during a context switch in XP64... this means that you can't use the crappy x86 FPU anymore!!! It took 20 years, but finally they managed to get rid of that peice of junk :-D

You mean a task context switch? How frequent is that compared to the number of instructions executed between switches? I know the x86 FPU was a {bleep} to code for properly, but register save on context switch is not that big hit. On any x86 from the last 10 years the chances that the registers weren't just moved to/from L1 cache are almost nil.

So now, instead of backing up a handful of x86 FPU registers when a task switch occurs, you instead to back up the vector units register file? These units are far bigger than the old FPU register file - how does that make it faster?

My Bad phrasing... The FPU regs are not saved, but the SSE regs are, so there is no real speed increase there at all.

This also means that the MMX and 3DNow! instruction sets no longer work.

In future all Floating point work is to be done by the Vector units, SSE and SSE2 :-)

This is good news all around!!! This also explains why we are seeing speed ups in 64bit version of windows when running 64bit apps.

Buh... surely FPU code designed for sequential execution is not going to run faster in the vector unit, just becasue it is a vector unit.

The SSE instruction set has been designed to allow better branch prediction and pipelining etc... C compilers are able to pump out SSE code that runs faster than it's x87 equivilent... if there is no x87 on XP64 all fp operations are going to run faster, as all XP64 systems have SSE, and no x87 code is needed.

Wayne wrote:
Forgive me.  I don't keep up with Windows a lot, but I didn't know the 64-bit version of Windows was even publicly available.

Wayne

The beta is free to download for all from microsoft.com.

See here to get your own copy. :-D

It times out after a year.

[the_leander]

My only concern is that the cost of buying the final finished product - the current XP pro edition is something like £150 atm and was well over £200 when it first came out, but that said, I am very curious about what a difference in performance this would have over win2k pro.

[bloodline]

the_leander wrote:
My only concern is that the cost of buying the final finished product - the current XP pro edition is something like £150 atm and was well over £200 when it first came out, but that said, I am very curious about what a difference in performance this would have over win2k pro.

I'm sure there will be solutions! :-D