Intel's Core Architecture vs AMD's Next Gen. Architecture

qcmadness

Distinguished
Aug 12, 2006
1,051
0
19,280
Source:
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2748
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2939

Here are some of the differences between Intel's Core Architecture and AMD's next generation architecture (NGA) called K8L or K10, with the reference of AMD K8 architecture.

1. Processor manufacturing technology:
Core Arch.: 65nm, 45nm in 2007 H2, 8 metal layers
K8: 130nm SOI, 90nm SOI, 65nm SOI, 9 metal layers
NGA: 65nm SOI, 45nm in mid-2008, 11 metal layers

2. Cache system
Core Arch:
L1 cache: 32KB, 8-way, latency: 3 cycles
L2 cache: 2-4MB shared, 16-way, 256-bit (64GB/s at 2GHz), latency: 11-20 cycles
L3 cache: absent
K8:
L1 cache: 64KB data+64KB instruction, 2-way, latency: 3 cycles
L2 cache: 512KB, 16-way, 128-bit (32GB/s at 2GHz), latency: 11-20 cycles (90nm version)
L3 cache: absent
NGA:
L1 cache: 64KB data+64KB instruction, 2-way, latency: 3 cycles
L2 cache: 512KB, 16-way, 128-bit (32GB/s at 2GHz), latency: unknown
L3 cache: 2MB shared, 32-way, unknown width and latency

3. x86 decoding ability
Core Arch:
x86 decoders: 3 simple + 1 complex (the complex decoder can decode 2 simple codes in a pass)
Out-of-order execution buffer: 96 instructions
K8:
x86 decoders: 3 complex
Out-of-order execution buffer: 72 instructions
NGA:
x86 decoders: 3 complex
Out-of-order execution buffer: 72 instructions with improvements

4. ALU, FPU and SSE units
Core Arch:
ALU units: 3
Maximum dual-precision (64-bit) FP per cycle: 4
SSE units: 3 units, 128-bit
K8:
ALU units: 3
Maximum dual-precision FP per cycle: 3
SSE units: 2 units, 64-bit
NGA:
ALU units: 3
Maximum dual-precision FP per cycle: 3
SSE units: 2 units, 128-bit

5. Pre-fetch and other tune-ups
Core Arch:
Out-of-order loads: Present
Stack manager: Present
Pre-fetchers: 2 data, 1 instruction (to core), 2 pre-fetchers (to L2 cache)
Instruction fetch width: 24 byte per cycle
K8:
Out-of-order loads: Absent
Stack manager: Absent
Pre-fetchers: 1 data, 1 instruction (to L2 cache)
Instruction fetch width: 16 byte per cycle
NGA:
Out-of-order loads: Present
Stack manager: Present
Pre-fetchers: 1 data, 1 instruction (to L1 cache), 1 DRAM pre-fetcher (to dedicated buffer)
Instruction fetch width: 32 byte per cycle

6. Memory controller
Core Arch: absent
K8: 1x128-bit memory controller (1 operation per cycle)
NGA: 2x64-bit memory controller with NGMA (max 2 operations per cycle)

7. Power management
Core Arch: EIST (min. x6 multiplier), switch off transistor when not in use
K8: Cool'n'Quiet (min. x5 multiplier)
NGA: improved C'n'Q, two separate power planes for crossbar and cores, separate clocks for each core
 

akahuddy

Distinguished
Oct 23, 2006
241
0
18,680
While I know a lot about PC's I didnt study nerdology so I will let someone translate all of this.

There's something to be said for holding your tongue and patiently waiting, but who wants to do that?

More importantly, nice complilation qcmadness.
 

dietzjack

Distinguished
Jun 10, 2006
81
0
18,630
From everything I have read, it sounds like the new AMD architecture could be good. We will know for sure come release time around September.
 

NightlySputnik

Distinguished
Mar 3, 2006
638
0
18,980
Add on-die memory controler to C2D and it looks like we'll have a fairly even competition starting next summer.

It's good to know we finally might have real choices to do when purchasing a new PC. When you think about it, until C2D, unless stupid or really into video encoding, you had to get a A64 cpu for good performances. Following C2D launch, unless you were a die-hard AMD fanboy or only upgrading, you had to get a C2D for top performances.

Now, if Barcelona perform up to expectation, we might finally be facing a dilemna when getting a new PC depending on the apps we use, price, availability, etc...

Good for us.
 

sabenfox

Distinguished
Mar 23, 2006
132
0
18,680
though i read the compiliation, i'm still at a loss as to what exactly it means...

perhaps one of our more technically advanced members could break it down for the lesser peons :)
 

wolverinero79

Distinguished
Jul 11, 2001
1,127
0
19,280
While I know a lot about PC's I didnt study nerdology so I will let someone translate all of this.

There's something to be said for holding your tongue and patiently waiting, but who wants to do that?

More importantly, nice complilation qcmadness.

Come on, his name is based on the AMD X2 family (possibly...I'm assuming here) and his username was created in Aug 2006. Maybe he's just bitter about Core2 Duo :-D

And yes, good compilation.
 

Heyyou27

Splendid
Jan 4, 2006
5,164
0
25,780
While I know a lot about PC's I didnt study nerdology so I will let someone translate all of this.

There's something to be said for holding your tongue and patiently waiting, but who wants to do that?

More importantly, nice complilation qcmadness.

Come on, his name is based on the AMD X2 family (possibly...I'm assuming here) and his username was created in Aug 2006. Maybe he's just bitter about Core2 Duo :-D

And yes, good compilation.Actually, he has a Core 2 Duo E6400 overclocked to 3.4GHz with dual Geforce 8800GTXs. :?
 

dragonsprayer

Splendid
Jan 3, 2007
3,809
0
22,780
sounds like amd dual core will be the same not as good - but the quad core will benefit from being unified 4 core

[

While I know a lot about PC's I didnt study nerdology so I will let someone translate all of this.

dude you should skip this post - try "will my dual core chip show 2 charts in the task manager" or "will this 350w psu power my p4 systems with 8800gtx's in sli"

i could not help it! :twisted:
 

slim142

Distinguished
Jan 29, 2006
2,704
0
20,780
hey, Im in (almost) the same case as your friend. I wont upgrade my CPU in a looooooong time. I can OC to 333mhz and get the perfomance of a E6650 or whatever is the equivalent to the E6600.

When I said anything with Core arquitecture was good, I never said in any way Dont consider other options, but it would be better to buy something cheap like the E4300 which performs nice for its price dont you think?
Now, if you use the pc for like 10 mins a day, then find something cheaper which will do what u need.
 

turpit

Splendid
Feb 12, 2006
6,373
0
25,780
I am a nerdologist understudy :wink:

Basically what he is saying here is the chunky peanut butter is relative to non chunky rounded up and multiplied by2; Then increased by a magnitude of strawberry jam and divided by sliced bananas.

So in the end you can make a peanutbutter jelly and banana sandwich in half the time with either chunky or smooth peanut butter.


:lol: :lol: :lol:


uhheemmm


:lol: :lol: :lol:

So what your saying is this is the new AMD

pbj.jpg
 

qcmadness

Distinguished
Aug 12, 2006
1,051
0
19,280
Wouldn't it make more sense to compare Barcelona to Nehalem, which by all accounts will be its competitor?

We should know that the NGA by AMD will be present in server / desktop areas. :wink:

Why I did not include Nehalem:
1. I know little from Nehalem / CSI.
2. Whether Nehalem is a "major" update of Core Architecture is a problem.
 

qcmadness

Distinguished
Aug 12, 2006
1,051
0
19,280
I think once csi is out, if intel can get a top notch mem controller like barcelona's, amd won't have much else to turn to, though th etotal 12mb+ of cache on the penryn cpus is all waste of transitors, amd went for the mem controller since it couldn't afford the keep up in the cache race
May be or not may be :D
By the time, the second generation of direct connect architecture and the third generation of memory controller is present in AMD's processors :wink:
 

qcmadness

Distinguished
Aug 12, 2006
1,051
0
19,280
the problem, it has turned into a flame war of fanboys, otherwise this could have been a very nice sticky, and I believe it should be one, just with all the clutter taken

You may try to help me by sending PM to mods. :wink:
 

piesquared

Distinguished
Oct 25, 2006
376
0
18,780
I think once csi is out, if intel can get a top notch mem controller like barcelona's, amd won't have much else to turn to, though th etotal 12mb+ of cache on the penryn cpus is all waste of transitors, amd went for the mem controller since it couldn't afford the keep up in the cache race
May be or not may be :D
By the time, the second generation of direct connect architecture and the third generation of memory controller is present in AMD's processors :wink:


Yes, you have to believe that AMD are masters of the memory subsystem and interconnects.
 

slim142

Distinguished
Jan 29, 2006
2,704
0
20,780
(btw, hows the speed difference from the maxtor to the raptor, I have always wondered really how much bias is there when people claim the speed differences)

Dont have the raptor yet :( sorry

You know the reasons, I explained them to you already.

I thought I was gonna get my Corsair RMA yesterday, when I asked if any UPS package got here, and I heard a NO, I was like F********CK! now gotta wait til monday! :(:(:(