Dual Processor vs Dual Core

[Hitman_Jim]

It has always been a frequent question "Will I benefit from multiple processors?" With the growing popularity of dual core processors, the topic is more important than ever! Will multiple processors or a dual core processor be beneficial, and what are the differences between them? These are the questions this article will attempt to lay to rest.

http://www.frameworkx.com/Frameworkx/blog.aspx?blog=56&id=66

[TrinireturnofGamez]

 dual core >> dual processors , its generally cheaper , uses less power , puts out less heat = more overclockable .. and if you have 2 dual cores in a dual socket system.. you get quad cores  :p

[richjob]

I'll post this email from the Pc__Support list out of Florida.

Not gamers, but men working in enterprise environments.

First off, since when does L2 cache have _anything_ to do with
performance?  If that was the case, then Intel's 2-4MB L2 cache on the
P4 processors should have been be smacking the A64 silly!

Secondly, people _forget_ that AMD processors have *4*TIMES* the L1
cache of Intel processors!  L1 is far, far more of a consideration --
especially in how AMD's non-exclusive cache design works compared to
Intel's exclusive cache design.

I.e., In an Intel processor, it _must_ load data into L1 cache before
the registers.  In an AMD processor, it can _directly_ load from L2
cache into registers.  With a _small_ L1 cache, the Intel really hurts
from this.

People forget about AMD's L1 cache _advantage_.  It's a major reason why
AMD is able to "keep up" with Intel's fabrication lead.  In fact, Intel
often has to "waste transistors" in adding more L2 to "make up" for its
small L1.

> > Blah, its already testing out poorly.

No, not "poorly."  AMD HyperTransport (A64/T64) processors do _smack_
Intel NetBurst (P4) processors silly.

> > I'll wait for the Core 2 Duo for the laptop.

Intel has finally _chucked_ the _inefficient_ Netburst architecture.
The reason why Netburst existed is because it was a "stop gap" design to
merely bridge from IA-32 P6/Pentium Pro (ignoring the original Pentium,
it had too many bugs) to IA-64 Itanium.  So it did a quick, 18 month
"refit" by extending the pipes and doing _no_ redesign.

Why?  7 years ago, Intel thought we would all be Itanium by now.
Unfortunately, that wasn't reality (don't get me started on IA-64 --
it's a "computer science" "paper ideal" that almost _every_ single one
of us "electrical engineers" said would _fail_, _utterly_, and it
did  ;-) .

So Intel _finally_ decided to _really_ rev the P6/Pentium Pro with a
full 36-48 month re-design circa 2002-2003.  That is now the Core
architecture.  It's very efficient, just like the P6 was -- totally the
opposite of the Netburst (P4).

Understand that this is Intel's _first_ IA-32[e] redesign since the
Pentium Pro of 1994.  So now, with the introduction of the Core design,
Intel has a 5-6 year lead on AMD -- in addition to their 12-18 month
fabrication lead.

AMD's last redesign was the Athlon (yes, 32-bit) in 1999.  The
A64/Opteron (including Turion64) is the _same_ architecture.  The
original Athlon was actually a 40-bit platform -- based on the 64-bit
Digital Alpha 21x64 -- and easily extended to 64-bit registers.  That's
why AMD could do it without a major redesign.

The _only_ major advantage that AMD has over Intel is in the 2+ way
space, especially 4+ way.  It had already built a non-shared, _truly_
switched platform interconnect in the Athlon based on the Digital EV6.
They merely moved it from a switch crossbar to a partial mesh.  In fact,
as I understand it, AMD's multicore is the EV6 Xbar internally, meaning
they could easily go to 13-core with_out_ any redesign (EV6 is 16-way
minus 3, 2 for the dual DDR channels plus 1 for HyperTransport).

The switched/mesh design _forces_ AMD to put an I/O MMU on the CPU.
This took a _years_ for AMD to mature, which it did in the original
Athlon MP (yes, 32-bit).  That's because of memory coherency for I/O --
with a "shared bus" Intel just relies on the chipset.  With AMD, it has
to maintain _full_ I/O coherency on _each_ processor.  Now that has
turned into a _major_ advantage -- as Opterons maintain memory mapped
I/O affinity to _each_ processor.

The result is that Opteron scales much, much better than Intel Xeon
"shared bus" design.  Even it's forthcoming, split bus is still not the
same.  Although it will allow Xeon to scale a little better in
combination with IBM's X3 architecture (long story).

But it's clear that with Intel's return to the _efficient_ Pentium Pro
base in the Core processor, and the _end_ of Netburst (P4), Intel is
back in the lead as far as ALU performance.  Combined with their
proliferation of "lossy math" SSE (whereas AMD does SSE with its full,
3-issue FPU for greater precision -- great for scientific/engineering
apps, but who cares about video/games?), Intel is in a pretty position.


> > My wife has a Turion chip in her laptop, and it runs well.  However, the
> > cache (or lack thereof) on the Turion X2 is a dealbreaker for me.

Huh?  Why does the L2 cache matter?

For _me_, the L1 cache size in Intel's processors and the requirement
that _all_ L2 go through that _tiny_ L1 is the "deal breaker."  ;->

Seriously now, buying something for L2 cache size is like saying you're
not going to buy a car with a smaller gas tank.  Yeah, it can't feed the
engine and go as far without stopping for gas than a car with a bigger
gas tank.

But the L2 cache has _nothing_ to do with the L1 cache like the gas tank
has _nothing_ to do with the displacement of the engine.  Seriously!
Yeah, my car with a smaller gas tank will "stall" before the larger one
unless I get gas -- but with 4x the engine displacement like L1 cache, I
_could_ be much farther down the road when I do stop!  ;->

Com'mon Kyle!  You've been listening to marketing.  ;->

Buy a Core 2 Duo because of the _performance_.  The L2 cache has
_nothing_ to do with that.  There's no more proof in that than the
massive and often _useless_ L2 cache sizes of the Netburst (P4)
architecture.  You could put 32MB of L2 cache on a Netburst and it
_still_ would _not_ beat an AMD A64-based design.

Again, the Core 2 Duo is a return to and an improvement over the
original P6 (Pentium Pro) core.  That's why it's very, very competitive
with AMD MHz for MHz -- _unlike_ Intel Netburst (P4).  And because Intel
has a 12-18 month lead in fabrication over AMD (because fabs cost
_billions_ of dollars), Intel can offer higher frequencies for the same
series/price.

Hence why Core 2 Duo wins.  It has _little_ to do with L2 cache size,
especially given AMD's _superior_ L1 cache feed directly to the
registers.

[TriniXaeno]

lol, great read Rich, but what does that have to do with the question, Dual Core or Dual Processors?

on a side note, I'm very interested in real world Single Core vs Dual core comparisons.

Like from a layman's point of view.

Any peeps on Dual Core that could offer this?

[richjob]

Dread.  Like I post that in the wrong discussion.  LOL.

Like everything else, it depends on your applications.  Linux stuff can be coded for multiprocessing, (with the right kernel) so that isn't a problem there.  And it does improve performance.  Man, when yuh see the two penguins showing yuh that yuh two processors up and working...

Dual COre IS faster.  Not double speed, though.  I don't think that games are as yet being explicitly coded for dualcore stuff.  It does come in handy when you are trying to encode video while running a virusscan, though...

Or if yuh wukking with huge graphics in Photoshop.

[W1nTry]

Well I can first hand tell you Baego that dual core makes multitasking MUCH easier (no duh right) but I talking running a VM on a separate IP scheme so man could ping himself when testing 2 sides of a VPN link. Trust me when yuh have 1 machine atm it's a GODsend. I usually set the VM affinity to processor 1 and let it be. Also encoding has benifited. I reenoded some anime from avi to rmvb in about 2/3 the time on meh T2300 vs. meh A64 3500. Also, like man say, running AV scan/S (note the S as in AV, spyware, AND registry clean). Now the HD speed comes into play there but it's great not to totally lock up and be at a standstill. Also running those awkwardly coded java apps no longer bring me to a halt save when they begin to eat up more than 50% of meh memory... I jes wish I had a GPU to match, but I liming strong with meh X1400. Dual core/processor is the way to go man. I wouldn't have made it the past few months without the extra power for multiple VMs.