Sign in with
Sign up | Sign in
Your question

CPU Questions: What we Should Be Asking (Barcelona vs. Harpertown)

Tags:
Last response: in CPUs
Share
September 23, 2007 1:51:50 AM

These are a few things I was curious about. Anyone have answers to these questions. Appreciate it.

1. Why was Harpertown given 16GB of RAM and Barcelona 8GB of RAM in the recent benchmarks?

2. Why didn't they use a mother board that supports HT 3.0 and dynamic power management for the individual cores? i.e Socket 1207+. I realise that a lot of people will go the route of flashing their bios and popping in the new Barcelona - BUT we have new features on this chip that some of us are curious to see in action just in case we are buying new and not upgrading.

3. Would it make sense for data center to upgrade Clovertown with Harpertown (15% performance gain) or Santa Rosa with Barcelona (80-100% performance gain)? Would a company with AMD servers go for Harpertown for that extra 10-20% performance gain on SINGLE CPU applications.

4. Why does Intel have the performance crown When we are comparing 2.0Ghz and 2.5Ghz Barcelona to 3.0 Clovertown and Harpertown? Why not compare 2.0Ghz Clovertown/Harpertown to 2.0Ghz Barcelona? Compare 3.0Ghz Barcelona with 3.0Ghz Harpertown in January/Febuary 2008.

5. Also are these same companies going to upgrade to Harpertown when a new socket and infrastructure is comin out in 2008 from Intel - yet in 2008,2009 you just keep dropping in new AMD CPU's and flashing bios? Wouldn't this be cheaper especially if AMD keeps Intels performance advantage at 10%.

6. Why would anyone care about a 4-5.0Ghz Harpertown that would consume more power than California? Isn't the real bread and butter in the 2.0-3.0Ghz range for both companies? In fact, aren't the lower clocked CPU's the best sellers since the strike the best balance?

7. Why didn't they show us a multi-CPU setup? 2 Harpertowns (8 cores) vs. 2 Barcelona's (8 cores)? 4 Harpertowns vs. 4 Barcelona's? I believe 8 Harpertown Cores behave like 6.6 Cores and 8 Barcelona Cores behave like 7.2 Cores.
- So at what point would AMD CPU's be untouchable?

8. Why hasn't anyone overclocked their Barcelona's - do they?

9. On the benchmarks done with the old 1207 mother boards where bios was flashed, doesn't memory for Barcelona run slower than if they used new boards???? That is - even though they used DDR 667 it was running at 533 on the old boards whereas on new boards it would run at 667. Is this correct? I believe it is something to do with independent power for memory controler on the CPU. Is this right?

10. Is AMD gunning for the 3.0Ghz mark on the production side - any Quad CPU that misses the 3.0Ghz mark by way of a faulty Core becomes a 3.0Ghz tri-core? Presumeably most problems involve one core not working when they push the envelope on the Quad - so Tri-cores are clocked pretty high. Anyone else heard this?
September 23, 2007 2:28:04 AM

people with AMD servers that go for Harpertown will aslo have to get FB-DIMMS
September 23, 2007 2:40:18 AM

I believe they have compared the cpu's at close to the same frequency as well.

Regardless of how the tests have been performed, I personally think that AMD is now competitive again in the performance arena.
Related resources
September 23, 2007 3:02:10 AM

enigma,

Sales are not the problem. ASP is the problem. They really NEED to raise CPU prices. I want AMD to raise prices. I want them to stay in business and to continue to make fast cpu's for me to buy.

The fact is, AMD has been brining up the rear. They have been losing a ton of money over the past couple of quarters, and they cannot continue to do so. They need to be able to raise CPU prices in order to compete in the long run, otherwise they will not have any money to compete with.
September 23, 2007 3:08:40 AM

sedaine said:
These are a few things I was curious about. Anyone have answers to these questions. Appreciate it.


Are you an AMD shill? :D 

sedaine said:
1. Why was Harpertown given 16GB of RAM and Barcelona 8GB of RAM in the recent benchmarks?


Perhaps they couldn't find bigger capacity DDR2 modules at such a short notice?

sedaine said:
2. Why didn't they use a mother board that supports HT 3.0 and dynamic power management for the individual cores?


Are they commercially available?

sedaine said:
Would a company with AMD servers go for Harpertown for that extra 10-20% performance gain on SINGLE CPU applications.


Depends on their application list.

sedaine said:
4. Why does Intel have the performance crown When we are comparing 2.0Ghz and 2.5Ghz Barcelona to 3.0 Clovertown and Harpertown?


Umm... why AMD had performance crown when it had Opteron at 2GHz versus Prescott at 3GHz? Or in other words, why frequency suddenly matters and it didn't when AMD was ahead?

sedaine said:
Compare 3.0Ghz Barcelona with 3.0Ghz Harpertown in January/Febuary 2008.


Judging by SkullTrail demo, Harpertown at 4GHz is not impossible in that timeframe so why bother waiting, AMD will be trounced anyway.

sedaine said:
yet in 2008,2009 you just keep dropping in new AMD CPU's and flashing bios?


You know that for sure? Also, Harpertown can be just plugged in into existing server boards if I heard correctly. It will surely work fine until the Nehalem comes out.

sedaine said:
Why would anyone care about a 4-5.0Ghz Harpertown that would consume more power than California?


Err... because Intel can sell it cheaper?

sedaine said:
2 Harpertowns (8 cores) vs. 2 Barcelona's (8 cores)?


8 cores you can see in SkullTrail demo, as for Barcelona, by looking at tri-core announcements it seems that AMD still hasn't produced two working quad-core chips yet, so they couldn't test :D 

sedaine said:
So at what point would AMD CPU's be untouchable?


I know this one it's simple... if they don't hit the shelves nobody could touch them :D 

sedaine said:
8. Why hasn't anyone overclocked their Barcelona's - do they?


You are simply expecting too much. It barely works at 2GHz :D 

sedaine said:
10. Is AMD gunning for the 3.0Ghz mark on the production side


They are but it is an evasive target if your gun is small :D 

/joke off

Seriously, what do you expect, Intel has just announced that they made first functional 32nm SRAM chip with 1.9 billion transistors -- every single one of them works, and AMD can't even manage to get 4 out of 4 cores working.
September 23, 2007 3:30:25 AM

"by looking at tri-core announcements it seems that AMD still hasn't produced two working quad-core chips yet, so they couldn't test"

lol. can't stop laughing! Comment of the month :-)
September 23, 2007 3:53:52 AM

levicki,

you accuse him of being an AMD shill, but you sound like an Intel shill.
September 23, 2007 3:54:10 AM

sedaine said:
These are a few things I was curious about. Anyone have answers to these questions. Appreciate it.

1. Why was Harpertown given 16GB of RAM and Barcelona 8GB of RAM in the recent benchmarks?

2. Why didn't they use a mother board that supports HT 3.0 and dynamic power management for the individual cores? i.e Socket 1207+. I realise that a lot of people will go the route of flashing their bios and popping in the new Barcelona - BUT we have new features on this chip that some of us are curious to see in action just in case we are buying new and not upgrading.

3. Would it make sense for data center to upgrade Clovertown with Harpertown (15% performance gain) or Santa Rosa with Barcelona (80-100% performance gain)? Would a company with AMD servers go for Harpertown for that extra 10-20% performance gain on SINGLE CPU applications.

4. Why does Intel have the performance crown When we are comparing 2.0Ghz and 2.5Ghz Barcelona to 3.0 Clovertown and Harpertown? Why not compare 2.0Ghz Clovertown/Harpertown to 2.0Ghz Barcelona? Compare 3.0Ghz Barcelona with 3.0Ghz Harpertown in January/Febuary 2008.

5. Also are these same companies going to upgrade to Harpertown when a new socket and infrastructure is comin out in 2008 from Intel - yet in 2008,2009 you just keep dropping in new AMD CPU's and flashing bios? Wouldn't this be cheaper especially if AMD keeps Intels performance advantage at 10%.

6. Why would anyone care about a 4-5.0Ghz Harpertown that would consume more power than California? Isn't the real bread and butter in the 2.0-3.0Ghz range for both companies? In fact, aren't the lower clocked CPU's the best sellers since the strike the best balance?

7. Why didn't they show us a multi-CPU setup? 2 Harpertowns (8 cores) vs. 2 Barcelona's (8 cores)? 4 Harpertowns vs. 4 Barcelona's? I believe 8 Harpertown Cores behave like 6.6 Cores and 8 Barcelona Cores behave like 7.2 Cores.
- So at what point would AMD CPU's be untouchable?

8. Why hasn't anyone overclocked their Barcelona's - do they?

9. On the benchmarks done with the old 1207 mother boards where bios was flashed, doesn't memory for Barcelona run slower than if they used new boards???? That is - even though they used DDR 667 it was running at 533 on the old boards whereas on new boards it would run at 667. Is this correct? I believe it is something to do with independent power for memory controler on the CPU. Is this right?

10. Is AMD gunning for the 3.0Ghz mark on the production side - any Quad CPU that misses the 3.0Ghz mark by way of a faulty Core becomes a 3.0Ghz tri-core? Presumeably most problems involve one core not working when they push the envelope on the Quad - so Tri-cores are clocked pretty high. Anyone else heard this?


1: Find one benchmark that needs even close to 8GB and we'll talk about the importance of 16GB.

2: Those board aren't available yet. The Barcelona were, so they hhad to go HT2.0

4: Because freakin 3.0GHz Harpertown is available now while only 2.5GHz Barcelona is on AMD side.

5: Would company switch to AMD if they are to loose 10% (which is more like 30 with 3.0GHz by the way) performance with them?????

6: Who's talking about 4.5GHz Harpertown? Oh yeah... you biased AMD fanboy!!!!!!!!

7: Who fuXXing need 8 core anyway for his server? I'll tell you, about 1% of the market as oppose to maybe close to 20% for 2 and 4 S system.

8: I didn't see much overclocking test on Harpertown side neither. Oh yeah, they are both server cpu on server board with very limited OC capability... Fanboy don't see this usually... :kaola: 

Didn't want to be rude, but you sound exactly like a biased fanboy, and that piss me off. Who care if Intel has the upper hand. Has long as AMD survive to get a better architecture out we don't have a problem, do we????
September 23, 2007 4:17:34 AM

NightlySputnik

"Who care if Intel has the upper hand. Has long as AMD survive to get a better architecture out we don't have a problem, do we????"

That is all I care about, AMD doesn't have to be faster all the time.... they just need to compete and survive.
September 23, 2007 5:55:23 AM

Oh hell no, not another Intel supporter vs AMD fanboy forum...
September 23, 2007 6:01:33 AM

Quote:
Why didn't they show us a multi-CPU setup? 2 Harpertowns (8 cores) vs. 2 Barcelona's (8 cores)? 4 Harpertowns vs. 4 Barcelona's? I believe 8 Harpertown Cores behave like 6.6 Cores and 8 Barcelona Cores behave like 7.2 Cores.
- So at what point would AMD CPU's be untouchable?


They don't become untouchable. AMD CPUs don't scale past 8P. Harpertown and Clovertown are for DP servers only. Tigerton is Intel's MP server product (up-to 32P).

What's with the stupid 2ghz vs. 2ghz comment? You expect people just to ignore the 2.33, 2.66 and 3GHz Clovertown CPUs that are out there now until AMD has faster CPUs? :sarcastic:  :lol: 

FWIW, Harpertown goes to 3.16GHz and 3.2Ghz at launch in November.
September 23, 2007 6:22:54 AM

sedaine said:
These are a few things I was curious about. Anyone have answers to these questions. Appreciate it.


Quote:
1. Why was Harpertown given 16GB of RAM and Barcelona 8GB of RAM in the recent benchmarks?

Because Anand is a Intel paid pumper. Anand has always been using hardwares in favor of Intel.

Quote:
2. Why didn't they use a mother board that supports HT 3.0 and dynamic power management for the individual cores? i.e Socket 1207+. I realise that a lot of people will go the route of flashing their bios and popping in the new Barcelona - BUT we have new features on this chip that some of us are curious to see in action just in case we are buying new and not upgrading.

Same answer as above. Although the boards were not commercially available, there got to be a prototype. Why Anand didn't use the board is pretty clear.

Quote:
3. Would it make sense for data center to upgrade Clovertown with Harpertown (15% performance gain) or Santa Rosa with Barcelona (80-100% performance gain)? Would a company with AMD servers go for Harpertown for that extra 10-20% performance gain on SINGLE CPU applications.

Because Intel's predatory pricing, as well as some under-table deals will "force" most company into abandoning AMD platforms. As Hector said, if Intel didn't use this immoral tatics, AMD should already have 50% of the market share.

Quote:
4. Why does Intel have the performance crown When we are comparing 2.0Ghz and 2.5Ghz Barcelona to 3.0 Clovertown and Harpertown? Why not compare 2.0Ghz Clovertown/Harpertown to 2.0Ghz Barcelona? Compare 3.0Ghz Barcelona with 3.0Ghz Harpertown in January/Febuary 2008.

That's because the test was unfair. As Anandtech's article showed, Barcelona is generally 300Mhz faster than identically clocked Clovertown. This means Barcelona has the IPC advantage.

Quote:
5. Also are these same companies going to upgrade to Harpertown when a new socket and infrastructure is comin out in 2008 from Intel - yet in 2008,2009 you just keep dropping in new AMD CPU's and flashing bios? Wouldn't this be cheaper especially if AMD keeps Intels performance advantage at 10%.

Exactly. This is why every companies should adopt AMD's platforms, as they are more future proof than Intel. On the other hand, with Nehalem debuting, a completely new socket and chipset need to be used, and that means tremendous extra cost.

Quote:
6. Why would anyone care about a 4-5.0Ghz Harpertown that would consume more power than California? Isn't the real bread and butter in the 2.0-3.0Ghz range for both companies? In fact, aren't the lower clocked CPU's the best sellers since the strike the best balance?

Definitely true. AMD pretty much has the performance/ watt advantage with their Barcelona. Harpertown is going to be a lot hotter, since they're clocked higher, and that needed to be cooled with stronger air-conditioning. This also means higher costs.

Quote:
7. Why didn't they show us a multi-CPU setup? 2 Harpertowns (8 cores) vs. 2 Barcelona's (8 cores)? 4 Harpertowns vs. 4 Barcelona's? I believe 8 Harpertown Cores behave like 6.6 Cores and 8 Barcelona Cores behave like 7.2 Cores.

I believe Harpertown cores will behave like 3 cores, while 8 Barcelona cores will behave like 8 core processors, due to native design.

Quote:
- So at what point would AMD CPU's be untouchable?

To put it simply, none.

Quote:
8. Why hasn't anyone overclocked their Barcelona's - do they?

They certainly do, I believe. As Fuad wrote, Barcelona can be easily overclocked to 3.0Ghz.

Quote:
9. On the benchmarks done with the old 1207 mother boards where bios was flashed, doesn't memory for Barcelona run slower than if they used new boards???? That is - even though they used DDR 667 it was running at 533 on the old boards whereas on new boards it would run at 667. Is this correct? I believe it is something to do with independent power for memory controler on the CPU. Is this right?

The memory was actually running at 667mhz. However, if Barcelona has a newer board, DDR2-800 can be used. This means a solid increase in its performance, as AMD's CPU always needs to be paired with decent performing RAM.

10. Is AMD gunning for the 3.0Ghz mark on the production side - any Quad CPU that misses the 3.0Ghz mark by way of a faulty Core becomes a 3.0Ghz tri-core? Presumeably most problems involve one core not working when they push the envelope on the Quad - so Tri-cores are clocked pretty high. Anyone else heard this? said:
10. Is AMD gunning for the 3.0Ghz mark on the production side - any Quad CPU that misses the 3.0Ghz mark by way of a faulty Core becomes a 3.0Ghz tri-core? Presumeably most problems involve one core not working when they push the envelope on the Quad - so Tri-cores are clocked pretty high. Anyone else heard this?


According to [H]ardOCP, AMD are pretty proud of their yields. AMD's 65nm process has always been known for its high yield, with low leakage on the manufactured die. I believe they're just holding the ace in their sleeve, to retard the reaction of Intel. If AMD suddenly releases a 3.0Ghz, Intel will need to spend a decent amount of time ramping the clockspeed up. Since Intel's CPU has diminished return as clockspeed increases, a 3.0Ghz Barcelona should be able to outperform a 3.4Ghz Harpertown by a sizeable margin.
September 23, 2007 6:37:59 AM

Mandrake,

Could you post info proving AMD doesn't scale well past 8p? I have always been under the impression that it's Intel's who don't.

wes
September 23, 2007 6:57:21 AM

found some info. How many HT links does K8 have? If it's less than 3, Barcy should scale better over 8p.
September 23, 2007 7:02:00 AM

Quote:
Because Anand is a Intel paid pumper. Anand has always been using hardwares in favor of Intel.


Proof?

Quote:

Same answer as above. Although the boards were not commercially available, there got to be a prototype. Why Anand didn't use the board is pretty clear.


You do know that Anandtech had about two days to run the tests, don't you? Where are they supposed to get a prototype ES motherboard with two days to run all these tests?

Quote:

Because Intel's predatory pricing, as well as some under-table deals will "force" most company into abandoning AMD platforms. As Hector said, if Intel didn't use this immoral tatics, AMD should already have 50% of the market share.


This is hilarious. One minute AMD is claiming that Intel's 'predatory pricing' (pricing CPUs below cost, while magically making profit!) is causing AMD to lose a ton of money. The next they claim that Intel is responsible for high CPU prices and milking the market for profit! :lol:  (60bn in monopoly profits press release anyone?!)

AMD saying they are 'proud' of their yields means NOTHING. It's not proof. See this for yields information. (When you look at the 'supporting document', scroll to Figure 3-9. The Barcelona die-size is 280mm squared.)

Finally, the performance. There are other reviews besides the Anandtech article.

I could go on, but I have better things to spend my weekend doing. :kaola: 
September 23, 2007 7:04:58 AM

weskurtz81 said:
Mandrake,

Could you post info proving AMD doesn't scale well past 8p? I have always been under the impression that it's Intel's who don't.

wes


It's a simple matter of not being able to scale past 8P. Current Opterons max out at 8 sockets. You can't go higher than that. Intel Tigerton supports up-to 32 sockets with the IBM X4 chipset. But the real high end with more than 32 sockets is all POWER, SPARC etc. where you'll never find a Xeon or Opteron.
September 23, 2007 8:00:40 AM

@Mandrake: Using a calculation method for RAM from 1996 isn't exactly accurate. Basicly you're saying that no improvements have been made in manufacturing since 1997(the year the repport came out)? I wonder how Intels processor production would perform using the same calculations...

About Intel, Where is my 10Ghz PC for 1500$? After all this claim is newer(from 2000), and from Intel themselves, so that predicion should be at least as accurate as your calculation method.
Link to Intel prediction http://www.theregister.co.uk/2000/12/11/intel_plans_150... what is the fastest Intel processor today and how are they doing on the 30 nm manufacturing?
September 23, 2007 8:02:56 AM

Mandrake_ said:

Proof?

http://sharikou.blogspot.com/2006/06/our-coverage-of-anand.html

Quote:

You do know that Anandtech had about two days to run the tests, don't you? Where are they supposed to get a prototype ES motherboard with two days to run all these tests?

AMD should sent them the prototype board along with the Barcelonas. Why didn't they use them?

Quote:

This is hilarious. One minute AMD is claiming that Intel's 'predatory pricing' (pricing CPUs below cost, while magically making profit!) is causing AMD to lose a ton of money. The next they claim that Intel is responsible for high CPU prices and milking the market for profit! :lol:  (60bn in monopoly profits press release anyone?!)

But if Intel didn't resort to low predatory pricing tactics, AMD should have 50% of the market right now, based on their innovation, and leadership.

Quote:
AMD saying they are 'proud' of their yields means NOTHING. It's not proof. See this for yields information. (When you look at the 'supporting document', scroll to Figure 3-9. The Barcelona die-size is 280mm squared.)

The document is from 1996, which is completely outdated. Do you have any information regarding 2007 AMD's yield? I guess not.


Quote:
Finally, the performance. There are other reviews besides the Anandtech article.

And all of them agree that clock-for-clock Barcelona is better. Should AMD pushed Barcelona's clock up to 3.0Ghz, it will be pretty much be like the old K8 days.

I could go on, but I have better things to spend my weekend doing. :kaola: said:
I could go on, but I have better things to spend my weekend doing. :kaola: 

or maybe you don't have much evidence supporting your arguments?
September 23, 2007 8:41:53 AM

So would I be right in assuming that if we use the right boad for Barcelona we should expect better performace/watt since dynamic power management kicks in? After all, isn't this one of Barcelona's main features?

With new boards that have HT 3.0 won't 2p servers benefit from faster link as opposed to the boards they used?

And what about 4p?

It seems to me that all these test setups are for 1p market - why not test a 2p and 4p server setup and give both 16GB of RAM.

Also, still wondering about the memory controler on AMD CPU - wouldn't it be independently powered on newer boards? I believe if that were the case it would operate at 1.8Ghz as opposed to 1.6Ghz on the old mother boards.
September 23, 2007 9:14:54 AM



Linking to Sharikou's blog is not proof. Sharikou also predicted that Intel will be bankrupt in under one year from now. :lol: 

Quote:

AMD should sent them the prototype board along with the Barcelonas. Why didn't they use them?


Should, but didn't. How is that Anand's fault?

Quote:
But if Intel didn't resort to low predatory pricing tactics, AMD should have 50% of the market right now, based on their innovation, and leadership.


If Intel sold CPUS below cost ("predatory pricing") how have Intel been posting profits every year? Oh, that's because they really sell CPUs at prices that are far too high to make "monopoly profits"! So which is it?!

We've all heard AMD complaining that they should have gained more market share throughout 2004 and 2005 when they had a superior product. It's their own problem for only having one 200mm wafer fab in operation throughout that time. They were selling all the CPUs they could make during that time. It's hardly Intel's fault that AMD didn't have enough CPUs to sell. By the time AMD's FAB36 was online Core was here with superior performance.

Quote:

The document is from 1996, which is completely outdated. Do you have any information regarding 2007 AMD's yield? I guess not.


Who cares how old the document is? The formula stated in the document that is used to calculate the yields has remained constant. The data from AMD is from July 2007 (i.e. quite current).


Quote:
And all of them agree that clock-for-clock Barcelona is better. Should AMD pushed Barcelona's clock up to 3.0Ghz, it will be pretty much be like the old K8 days.


More BS from you? Looking at the TechReport data Xeon 5345 wins far more benchmarks than it loses vs. Opteron 2350. That's not even taking into account the faster 2.66 and 3Ghz Xeons that are available today, in volume. When is AMD coming out with a 3Ghz Barcelona? Randy Allen, VP of Servers and Workstation said that AMD plans to have a 2.5Ghz version out by the end of the year (Source: AMD's own Quad Core Opteron launch videos). Intel will certainly be ramping up the clockspeeds on the 45nm Harpertown. Plus Intel has the Nehalem architecture coming in the second half of 2008. The demonstration shows that they are on track to deliver Nehalem in their stated timeframe. Where is AMD's 45nm Shanghai CPU? Where is the tapeout announcement? Where are the demos?

Quote:

or maybe you don't have much evidence supporting your arguments?


It looks like you are the one spouting FUD. Linking to Sharikou as proof? That's pathetically low.

Quote:
It seems to me that all these test setups are for 1p market - why not test a 2p and 4p server setup and give both 16GB of RAM.


The TechReport test is a 2P test. Each system having 8 cores.
September 23, 2007 9:18:14 AM

Also in terms of comparing Oranges vs. Oranges:

Would we compare a Boeing 767 to Airbus A380?

Would we compare an S-Class 6.0L to a Audi A4 2.0L?

The answer is nobody has ever done that and probably never will. You should look at the prie of the CPU's and the determine which one gives you best performance at that point.

If I was an AMD shill then I would tell you this: It is much better for AMD to wait until Nov. 12 to see what Intel's pricing will be like for their new processors. If indeed AMD had a 2.5Ghz that was ready to go you will likely see it 2 weeks after Intel's release. It will likely be priced based on upcoming 2.6Ghz Harpertown, not the 3.0Ghz found in benchmark tests.

Also around this time if I was an AMD shill I would tell you that AMD now has the capacity to surprise Intel once more. Many people have suggested that AMD needs a 3.0Ghz. I would say they only need it after Intel releases new chips and prices them in November. This is not a short-haul business - some far-sightedness is needed.

In business, it is better to counter than be countered. It is also better to surprise than be surprised.

Then there is the issue of bad yeilds - let's assume that were the case - th question then becomes - If AMD attempts to make 10,000 Quad CPU's clocked at 3.0Ghz - how many would actually pan out? 20-30%? How many would clock higher - 3-5%? How many would have one faulty core (Tri-Core) - 50%? From those how many would be around the 3.0Ghz mark - 75%?

AMD will not officially release 3.0Ghz parts until they have sufficient quantities. But once that point is reached - Intel will need 3.6Ghz to compete with it.

All those things said - we must remember we are comparing Toyota (Intel) to Subaru in terms of production. Intel has been in business for years - and it does what it does like no one else can. Anything AMD makes, Intel can make. In fact AMD is in a position where they must be better to compete - Intel does not have to be better because of their production capacity vs. AMD's. AMD could never meet demand - evn if they had a CPU that was X2 faster than anything Intel had - they could never meet demand. So Intel is always going to have sales.

The good thing is AMD has kept Intel in check in terms of pricing. It's a buyers market because of AMD.
September 23, 2007 4:18:59 PM

Sedaine.... comparing a plane that flys to one that doesn't..... no, not a fair comparison.
September 23, 2007 4:25:22 PM

JustJC,

"Intel suggests that a 10GHz processor could power a universal translator as seen on Star Trek."

I love it.
September 23, 2007 4:27:09 PM

weskurtz81 said:
Sedaine.... comparing a plane that flys to one that doesn't..... no, not a fair comparison.




The point is different size planes are ideal for certain markets.

Slower, lower powerd CPU's like 1.7 AMD's and 1.8 Clovertowns are good because of power needs. Not everyone wants a 4.0Ghz chip - I wouldn't mind one in my desktop - but corporate America is not going to buy 5,000 4.0Ghz CPU's when they can get the 1.7Ghz and achieve the same goal - also save on power.
September 23, 2007 5:08:15 PM

sedaine said:
The point is different size planes are ideal for certain markets.

Slower, lower powerd CPU's like 1.7 AMD's and 1.8 Clovertowns are good because of power needs. Not everyone wants a 4.0Ghz chip - I wouldn't mind one in my desktop - but corporate America is not going to buy 5,000 4.0Ghz CPU's when they can get the 1.7Ghz and achieve the same goal - also save on power.

But they would rather buy 2000 servers with 4GHz CPUs which would save even more power and a huge amount of space as compared to 5000 servers with 1.7GHz CPUs while offering comparable performance.

So would I be right in assuming that if we use the right boad for Barcelona we should expect better performace/watt since dynamic power management kicks in? After all, isn't this one of Barcelona's main features? said:
So would I be right in assuming that if we use the right boad for Barcelona we should expect better performace/watt since dynamic power management kicks in? After all, isn't this one of Barcelona's main features?

The Anandtech test already used a split-plane Socket 1207+ MB.


With new boards that have HT 3.0 won't 2p servers benefit from faster link as opposed to the boards they used? said:

With new boards that have HT 3.0 won't 2p servers benefit from faster link as opposed to the boards they used?

The HT 3.0 S1207 boards don't exist and Opterons are not scheduled to support it till Shanghai.


And what about 4p? said:

And what about 4p?

I'd expect Tigerton to beat 4S Barcelona in enterprise applications quite handily.
September 23, 2007 5:50:57 PM

sedaine said:
The point is different size planes are ideal for certain markets.

Slower, lower powerd CPU's like 1.7 AMD's and 1.8 Clovertowns are good because of power needs. Not everyone wants a 4.0Ghz chip - I wouldn't mind one in my desktop - but corporate America is not going to buy 5,000 4.0Ghz CPU's when they can get the 1.7Ghz and achieve the same goal - also save on power.


[extreme intel fanboyism]

The problem is, what about the companies that require high performance servers? As Accord99 pointed out, it would be more economical if you have 3k servers running at 4.0Ghz, than 5k servers running at 1.8Ghz. It will be a lot cheaper too.

It all comes down to the crtieria a company is looking for in a server. Within a scientific community, where FP performance is crucial, Barcelona will undoubtly standout. But for companies that require superior integer performance, Intel will be a wiser choice.

[/extreme intel fanboyism]
September 23, 2007 8:28:39 PM

sedaine,

I was poking fun at AB. That's all.
September 23, 2007 8:51:57 PM

enigma067 said:
geez weskurtz81,
You make it sound like AMD is pickin' up the rear.

http://www.news.com/AMD-surpasses-Intel-in-U.S.-retail-...
You funny. Did you bother to look at the date on that article. Let me help you.
Quote:
Published: November 8, 2005, 6:48 AM PST


Now don't you feel foolish :pfff: 
September 24, 2007 10:32:33 AM

weskurtz81 said:
levicki,

you accuse him of being an AMD shill, but you sound like an Intel shill.


Can't you see:

1. Difference between accusation and question (with a smiley)
2. /joke off at the end of the post

???

yomamafor1 said:
Because Anand is a Intel paid pumper. Anand has always been using hardwares in favor of Intel.


You mean he has been using all hardware capabilities to the maximum in his benchmarks as opposed to others who used legacy software like SuperPI for testing?

yomamafor1 said:
Although the boards were not commercially available, there got to be a prototype. Why Anand didn't use the board is pretty clear.


Yes it is clear, because prototype when AMD is concerned always worked much worse than finished product available in retail channel for some time.

yomamafor1 said:
Because Intel's predatory pricing


Make up your mind already -- are Intel's prices too low or too high?

yomamafor1 said:
Exactly. This is why every companies should adopt AMD's platforms, as they are more future proof than Intel.


Futureproof like in bankruptcy? :D 

yomamafor1 said:
Harpertown is going to be a lot hotter


Heh, 80W for 3GHz quad-core part. I am waiting for AMD to release something like that.

yomamafor1 said:
I believe Harpertown cores will behave like 3 cores, while 8 Barcelona cores will behave like 8 core processors, due to native design.


Yeah, yeah, yeah... you know what? 8 Barcelona cores will actually behave like 9 cores because if you want 8 cores you will have to buy three of those tri-core rejects due to their naive design. :D 

yomamafor1 said:
Barcelona can be easily overclocked to 3.0Ghz.


First, do you seriously think that someone would overclock CPU in a SERVER? Second, we all saw Harpertown running at 4GHz on air cooling (Zalman) on IDF. Now what?

yomamafor1 said:
If AMD suddenly releases a 3.0Ghz, Intel will need to spend a decent amount of time ramping the clockspeed up


You really need to check your facts... and your head.

yomamafor1 said:
AMD should sent them the prototype board along with the Barcelonas. Why didn't they use them?


"Should sent" (btw, it is "should have sent") is not the same as "sent". Without knowing for sure whether AMD did that or not, you can't ask "why didn't they use them".

yomamafor1 said:
Should AMD pushed Barcelona's clock up to 3.0Ghz, it will be pretty much be like the old K8 days.


Shoulda, woulda, coulda...

sedaine said:
It is much better for AMD to wait until Nov. 12 to see what Intel's pricing will be like for their new processors.


AMD is going to suffer a lot at Nov, 12th because it is going to be hit hard from two different sides:

Side #1 -- Intel Harpertown
Side #2 -- NVIDIA (link)

Kick in the nuts from the NVIDIA could have been avoided by not absorbing already sick ATI.
That said, I would sell their shares now if I had any while they are still standing.

sedaine said:
Anything AMD makes, Intel can make.


You forgot to add "only cheaper and with higher material quality" at the end of that sentence.

sedaine said:
Intel does not have to be better


That is not true. They have to because of their stockholders, not because of AMD. True, Intel can afford to lose money, but why would they do that if they can do better?

sedaine said:
It's a buyers market because of AMD.


Agreed.
September 24, 2007 1:10:31 PM

levicki,

As long as Intel is meeting earnings estimates they really don't need to be better than AMD. They have such superior marketing they can survive with a inferior chip for years.
September 27, 2007 9:42:17 AM

weskurtz81 said:
levicki,

As long as Intel is meeting earnings estimates they really don't need to be better than AMD. They have such superior marketing they can survive with a inferior chip for years.


Yeah, the only problem is that they have superior chip :D 
a b à CPUs
September 27, 2007 1:28:15 PM

Journal of Pervasive 64 bit Computing ... If they target Anand like that then what have they got to say bout the reviews at THG ... I'll take a stroll.

Anyone remember the dual core shootout where the Intel box died twice during the test ? Whey still managed to try to make the Netbust POS sound good.

THG have the most pro-Intel reviews on the web fullstop.

The suspect benchies here from the last of the Netbust processors in comparison the the A64's is testament alone.

Plus even the low end reviews are carefully exploited to compare Celery and Sempy systems that are mismatched.

You have to remember Intel pays sites an aweful lot ... must be hard to resist eh?


a b à CPUs
September 27, 2007 1:30:05 PM

Intel's grip on this industry is like a cancer.
September 27, 2007 2:06:45 PM

Intel has the money to "persuade" many companies, and....has done so successfully. This is not merely supposition but documented over the course of many years but all is fair in love and processor manufacturing right?

I had always been curious as to why Intel maintained so much market share while AMD chips were clearly superior not only in performance but in price/performance. Now that Intel dominates in performance, but not in price/performance, the beat goes on. Ahhhhh yes, let the extremists form to the right.
September 27, 2007 5:49:09 PM

@Levicki

Intel does not have to be better to make sales. They have manufacturing capacity and economies of scale work better for Intel because of that. Plus of course they are a big marketing machine, Tom, Dick and Harry will buy Intel P4 over Athalon 64 because of marketing.

On the other hand AMD must be competitive or better to make sales and these sale are usually to enthusiests (read: people who know what they are buying) and can see beyond marketing. That is not to suggest that Intel does not have the most dominant desktop CPU right now, because they do - but looking at Phenom it looks like it will be very competitive.

Also, if Phenom X3 goes ahead in being priced to compete against Core 2 Duo, then it would be Intel marketing vs. AMD superior performance - seeing that it is likely that Phenom X2 will match Core 2 Duo based on early Barcelona figures.
September 27, 2007 5:59:03 PM

Reynod said:
Journal of Pervasive 64 bit Computing ... If they target Anand like that then what have they got to say bout the reviews at THG ... I'll take a stroll.

Anyone remember the dual core shootout where the Intel box died twice during the test ? Whey still managed to try to make the Netbust POS sound good.

THG have the most pro-Intel reviews on the web fullstop.

The suspect benchies here from the last of the Netbust processors in comparison the the A64's is testament alone.

Plus even the low end reviews are carefully exploited to compare Celery and Sempy systems that are mismatched.

You have to remember Intel pays sites an aweful lot ... must be hard to resist eh?

Proof?

THG has been pretty consistent back in the Netburst vs. K8 days (I don't know about now). The article you quoted had a conclusion in favors of AMD. Are you telling me that Intel paid them to favor AMD in that particular test?

Nonsense.
a b à CPUs
September 29, 2007 2:47:14 PM

I am simply saying that THG didn't bag the hell out of the Intel space heater, or when it blew up twice ... they simly made little comment.

I worked in marketing ... I can spin a story in any direction too.

Check out the gaming benchies for the last of the Netburst vs the K8 ... ???

October 2, 2007 6:58:04 PM

Reynod said:
I worked in marketing ... I can spin a story in any direction too.


Marketing is the worst and most fatal brain disease of the 21st century.
!