I don't see how the gaming debate factors into a server chip in any way. But I tend to think its the hard core PC gamers that are becoming extinct, which is the driving factor for the decline in those games. I mean you can only make so many shoot-em-up games and still be original for plot and ways to kill people. Not being a hard core gamer, the genres that I do care about (racing and sports) are not being neglected on the PC level, although it would be nice to see some decent competition for EA Sports/EA Games as they've become a bit complacent with glitches. Anyway, I really dislike seeing threads hijacked so I'll refrain from commenting further on this subject.

Prices for the A-64 6000+ and Intel E-6420 are identical in AU as well. While the 6000 enjoys a slim lead in most benchmarks at stock speeds, the E-6420/E-6400 will overclock much higher and (as stated by Brandon DeCoppel) blow away the 6000 under those conditions. Even an overclocked 805D will give a 6000 a run for its money for performance. If multi-tasking is your thing, any Intel chip will out-perform any AMD chip.

As for Sander's article, the numbers that the HWA article refers to were presented by AMD themselves. The point of Sander's article seems to be that the numbers that AMD posted were slanted toward making their chip seem faster than it actually is at this point, and that AMD used marketing tactics that they are trying to sue Intel for using. Sander made no claim about how fast the chip is or isn't, just critisized the marketing ploy and justifiably so. How fast the Barcelona chip is at release has no bearing on HWA's article.

Efficiency remains to be seen. Most of the hype I've seen generated about efficiency shows that AMD chips run more efficiently at idle, pretty much across the board of chips. Since a server is hardly ever at idle, it would make little sense to brag about the efficiency of a server chip at idle. Under load, AMD and Intel both win some battles and lose some battles. So for now I see it as a toss up as to who actually makes a more energy efficient chip. When the final numbers of Penryn server chips' and Barcelona server chips' energy efficiency are made public I'm sure we'll have more than enough to read and argue about on the subject.

http://www.canadacomputers.com/index.php?do=ShowProduct&cm...id=CPU.907 X2 6000+ 273$

http://www.canadacomputers.com/index.php?do=ShowProduct&cm...cid=CPU.84 E660 - 262$

If my memory serves stock E660 can beat 6000+ in some benchmarks

Which would imply as of now in Canada it's better to just stick to intel in performance.

I just wanted to throw that out there because people from ENG/AU are saying it's = in price to the E6420

i thought i'd just through this in..... aka. better cdn prices

http://www.pccanada.com/viewitem.asp?id=4778
retail box E6600 $249cdn.
http://www.pccanada.com/viewitem.asp?id=6072
retail box amd 6000+ $279

Well the question was about E6400, not E6600. And yes the E6400/E6420 does beat the 6000+ in some benchmarks when both are at stock speeds. But the difference no matter what the benchmark is hardly worth mentioning. The E6600 beats the 6000+ consistently, but the E6400/E6420 will overclock to beat the overclocked E6600 in most benchmarks. So best bang for the buck is still the E6400/E6420 IMO.

Right now both (E6420 and 6000+) are listed at $230AUD, which is wholesale. I don't buy retail so am not sure what differences are there.

Actually, here is one example of the E6420 vs 6000+. But keep in mind this is a pretty lame review (IMO) because they state that 3.23GHz is the highest stable OC they could achieve on air, which I believe is due to the inferior motherboard. I've have my E6400 @ 3.68GHz stable on air, and not so stable at 3.72GHz, but I'm using a nice air system (Thermaltake Big Typhoon, fan replaced with Thermaltake Smart Fan II) and motherboard that I consider much better than the Asus Commando (Gigabyte GA-965P-DQ6 Rev 1).

http://www.hardwarezone.com.au/reviews/view.php?cid=2&id=2291&pg=4
http://www.hardwarezone.com.au/reviews/view.php?cid=2&id=2291&pg=5
http://www.hardwarezone.com.au/reviews/view.php?cid=2&id=2291&pg=10

My system, try this on a E6600. :

http://www.ultramaxcc.com.au/crucial6400.htm

I disagree. AMD is taking a much more difficult approach to a very complicated “Native” Quad-Core technology. AMD had the chance to simply glue two of there Athlon 64 X2’s, but they decided not too. They have the architecture to come out with a much more sophisticated design. Now all they have to do is execute these new CPU’s the right way or they will get into further trouble.

Performance is not everything, power savings is. And as I read it, Barcelona has both.


Yes, I agree, though I think the ATI buyout was the best thing for AMD. May not look like this right now, but ATI’s buyout was for future security. Now AMD has the opportunity to build & design stuff for other markets including the Mobile Phone industry etc.



Sorry, they may have newer CPU’s out, but they are far from being Quad-Core, they fit the class of Double Dual-Core. Two physical Dual-Core’s stuck together. AMD could have easily done this, but they did not. It’s uninnovative. No, really they have the technology to do what they are doing right now, “Native”




Your opinion, nobody knows what & how Barcelona will perform. Like I’ve said 1000’s of times, if AMD & Intel reversed ROLES, then AMD would have been out of business years ago. Think about that for a minute. I don’t need to spell it out for you.



I agree that K8 was designed to complete with Netburst, but I don’t think Barcelona is K8. They did a lot more then just enhancements. I think Barcelona will exceed 3.40 GHz at the least. Not overnight, but quite possible.

S_U_P_E_R X_P said:
The hardest route is rarely ever the best. Considering how much better Clovertowns are performing compared to the paper launched Barcelona's, you're entire argument is moot.

S_U_P_E_R X_P said:
If you can't live through today, why worry about tomorrow? You can't take your second step without taking your first. The acquisition of ATI could be the very reason why AMD goes under. They accumulated more debt than they could handle. You have to be profitable yourself before you can safely consider expanding. If you fail to do so, your business is bound to suffer a similar demise as the RMS Titanic.


S_U_P_E_R X_P said:
Do tell how this does not qualify as a quad-core:
http://img402.imageshack.us/img402/2421/quadcoresr3.png

You just blew the mercury bulb on the "I'm-An-Idiot-Meter."


S_U_P_E_R X_P said:
Actually, you do because what you said makes absolutely no sense.


S_U_P_E_R X_P said:
Show me documentation, published by AMD, which declares it as anything other than K8 then.

All they did was double the width of the SSE unit, do a little tweaking on the FPU, add an L3, and stick two more K8 cores on it for kicks. None of the above qualify for a major revision change. I've said that so many times before, I've lost count.

AMD has yet to produce a stock chip with a clock over 3.0 GHz, even single core, how could a chip four times more complex ever do it if the simple ones can't? AMD isn't going to get far with clock speeds until their next microarchitecture and hopefully they'll be doing a lot of talking with IBM to learn some tricks used in the POWER6 line of processors.

FordGT90Concept said:

You said it, Paper Launch. How will it perform when it is actually launched? We shall see when they come out soon enough.



I believe AMD was not planning on Intel releasing the faster Core 2 Duo design. AMD was already deep into the ATI buyout, so there was no turning back. It all happened too fast. AMD also made some bad decisions by removing Socket 939 too quickly, but anyway, AMD will come back with vengeances we all hope.



LOL, funny why not show us all an actual picture of the physical Double Dual-Core ey? Because that is what it is a Double Dual-Core. The picture of the CPU will show this LOL,




No, I make all the sense in the world. I believe you are confused, so let me elaborate for you what I mean. If AMD came out with the Pentium 4 & Intel came out with the Athlon 64, then AMD would have gone belly up. That is my point. Intel can afford to make mistakes which they’ve demonstrated several times in the past, where as AMD cannot. Understand? Dublin would agree with me, we had this discussion a year ago if my memory serves me well.



Here are some features among many others. I don’t know the details 100%, but I think it is classified as built from the ground up from what I’ve been reading.

Some Features for Barcelona among many others:

SSE execution width = 128 bits wide

Instruction fetch bandwidth = 32 bytes/cycle

Data cache bandwidth = 2 x 128 bits loads/cycle

L2 cache/memory controller bandwidth = 128 bits/cycle

Floating-point scheduler depth = 36 dedicated x 128-bit ops

L3 cache = 2MB+
etc.

Since we are on the topic everybody:
AMD's quad-core 'Barcelona' to launch next month
http://news.cnet.co.uk/laptops/0,39029680,49291367,00.htm

SuperXP, why was 'double dual core' a great idea when AMD did it with the QuadFX but rubbish when Intel do it with their processor?

S_U_P_E_R X_P said:


Last time I checked, "Quad" derives from Latin which translates into "4" and only "4", the value. There is nothing in the meaning of "Quad" that suggests position, setup, or alignment of any such thing. So, a double dual-core processor is still a quad-core processor. Please don't give that BS about it being "not really quad-core" because they aligned it differently so that it was simpler to manufacture in the first place, something AMD doesn't seem to get, and is therefore on a downward spiral at this present moment. Hard to argue for a side that is currently sitting (in relation to a house) in the sewer below.

I'm a bit sick of the Super_XP bashing. While I may not agree with all of his views and everything he says, he's one of the nicest and most helpful members on this forum. I've never seen him flame someone else for their posts even when they're based on nothing but a pipe dream.

Super's only problem is that he's a fan of AMD to the point of fault. I have no problem with that.

None of these arguments are worth reading as it's all opinion based and not factual. Quad core means different things to different people. Who cares?! And no one knows yet what Barcelona will be like when it's released. No one.

Personally, I couldn't care less if a design is new or 20 years old. I just want what works. But that doesn't mean others won't think differently.

Diversity is strength. Anyone with a brain should be encouraging differing points of view, not flaming them.

Super, the difference there is that the P4 wasnt as far behind K8, as K8 is behind Core 2.

Sure, AMD dont have the resources of Intel, but then again, not only did Intel design the first ever uP, they made the design by which every other desktop CPU (PC) follows, the x86 architecture.

This was a design which they licensed for years, until they were forced not to license it anymore for being monopolistic.

All the other desktop CPU manufacturers (AMD, Cyrix etc) that followed Intels designs, had to pay to copy them.

Now, you cannot talk about Intel & AMD swapping positions for a section of their history, as that makes no sense. Intel has the funds & resources because its been there, done that, and basically created the market which they dominate.

If you swap the roles, AMD would be pretty much were Intel are at now.


Anyway, Barcelona at the moment doesnt look to good, but they are rvising the silicon (again, and again, and again lol) and with any bit of luck they will release a competitive product.

But one thing I disagree with you on is your definition of a quad core.

Quad core = 4 cores in the one package.

Core 2 Quad based processors have 4 core in the one package.

Thats a quad core, no matter how they implemented it.

S_U_P_E_R X_P said:
Those were AMD's paper launches, not Fudzilla, not The Inquirer, not CNet. No one but AMD. Therefore, the paper launches are extremely credible (2.6 GHz model performing just below an Intel Xeon 2.66 GHz and the highest launch speed announced at 2.0 GHz). Barcelona is beat before it even left the starting gate and AMD is well aware of this (hence more delays).

S_U_P_E_R X_P said:
The buyout was announced on July 24, 2006 and completed on October 25,2006.[1] The first Core 2 processors were launched on July 27, 2006.[2] AMD had plenty of time to call if off.

1. http://en.wikipedia.org/wiki/Advanced_Micro_Devices#Merger_with_ATI
2. http://en.wikipedia.org/wiki/Intel_Core_2#Conroe


S_U_P_E_R X_P said:
Quad is derived from the Latin prefix "quadri-," meaning four.[1] Core is derived from the Latin word "cor," meaning heart.[2] Therefore, "quadcore" or "quadricor" means "four hearts" which is true of both Kentsfield and Clovertown. The proof is in the amount of work it can do: almost double that of a dual core Core 2 Duo and almost four times that of a single core Core 2 Solo (coming soon). What you are regurgitating is merely AMD propaganda and therefore, has absolutely no leverage.

1. http://en.wiktionary.org/wiki/quadri-
2. http://en.wiktionary.org/wiki/core


S_U_P_E_R X_P said:
Pentium 4 processor sales far exceeded Athlon 64 sales. Even if they switched, AMD would have been fine because they would still be making money. The Pentium 4 line is at least profitable and competitive (the proof is in the price where P4 EE processors were almost always the exact same price as Athlon FX processors).


S_U_P_E_R X_P said:
You basically regurgitated what I said and provided no evidence fabricated by AMD that declares Barcelona as anything other than a K8.


dark41 said:
It only means different things if you are Super XP and believe AMD is always right beyond a shadow of a doubt. The defintion of "quadcore" is above.

DublinGunner said:

My point in all of this was to explain that Intel can afford to make mistakes, where as AMD cannot.

So, look at it this way O.K. AMD & Intel DON’T switch roles O.K.

AMD comes out with the Pentium 4 design, Intel comes out with the Athlon 64 design. AMD would be in big trouble.
This is why AMD needs to innovate. Slower clocked CPU’s etc. performing very competitively against Intel CPU’s. Intel can keep pumping out the age old FSB slapping it with enhancements over & over again, and they will probably succeed. The market wants something that is competitive, even though it’s un-innovative & old.

AMD’s target was Price/Performance, and they’ve won this in the years of the Athlon XP & almost dominated with market share gains with the Athlon 64.

Anything Intel comes out with that uses a FSB is an underachievement in the computer industry. We all know this, but they seem to always make them competitive. I call it un-innovative, but we are talking about Intel, they can afford to do stuff like this & get away with it.

AMD does this & they are screwed Big Time. This is why it is very important for AMD to successfully execute the Barcelona & to prove to everybody that it is in fact 40% faster then anything Intel has out right now. If they can get that speed bug fixed fast, then Barcelona can start hitting its designed speed of 3.0 GHz +.

DublinGunner, tell me which design do you prefer?
Intel’s implementation of there so called Quad-Core?
AMD’s implementation of there “Native” Quad-Core?

I call AMD's Quad-Core intelligent

We are talking common sense here right?

New Intel quad-core chips are really double duo-cores
http://www.zdnetasia.com/news/hardware/0,39042972,39315888,00.htm

Quote:
Quote:

Sure Intel has a Quad-Core out right now, they've used the Dual Dual-Core approach. AMD's on the other hand is a Native Quad-Core.

AMD Phenom Launch Schedule & Roadmap
Page Title: AMD Phenom Launch Schedule & Roadmap
Category: Processors
Type: News
Posted By: Visionary
Date: June 21, 2007, 1:55 pm

http://www.vr-zone.com/?i=5078
http://images.dailytech.com/nimage/4622_large_stars_list.png

And here, AMD kicked in Damage Control
AMD s Better by design mobile logos out
http://www.fudzilla.com/index.php?option=com_content&task=...;Itemid=35

I dont prefer either, they're both quad cores, implementation means jack, its performance I've always been after.

And I really dont see the point in trying to undermine the FSB, when for instance, mine runs at 1700Mhz, and I have a chips faster than ANY offering from AMD, even overclocked.

Its performance, no implementation that rules supreme, that has, and always will be what wins the day.

BTW, in the quote from your article there, note the words

in theory

DublinGunner said:
I agree. I couldn't care less if Barcelona has wings. If it isn't faster, it doesn't mean s**t.

It’s the principle of this Quad-Core situation.

Which do you prefer?
Intel’s Double Dual-Core = Gas Guzzler but Very Fast
OR
AMD’s Native Quad-Core = Great on Gas & Fast

To reiterate:

DublinGunner said: