Quad-Core AMD Opteron deliver world record web performance

[BaronMatrix]

Its for anything that seems like AMD bashing or those who have talked negative about AMD beofre or those like me who disagree with BM.

Where I disagree is the crap he spreads and thats all. Fun times.

Soon this site will be crap. It was so nice in the begining.

The problem started when AMD moved ahead. It got worse when Intel made a comeback. It was like people thought it was Gods divine will and that AMD should never have gotten ahead. That's a shame.

It's OK though, I just switched keyboards. My wireless is possessed by bad software.


BTW, what CRAP are you talking about?

 

[BaronMatrix]

Well I hate to go back to the Baron bashing, but surely you can believe he would use irrelevant benchmarks to prove his "point." That's kinda what he does 😉

Back on topic, Shangai is more than 'just' a die shrink. It triples the L3 cache and has some other enhancements which should improve IPC quite a bit in certain applications. Whether it will be enough to take on Nehalem is questionable, I lean towards "no way in hell," but we'll see

How is a comparison to Penryn as a baseline irrelevant? Extrapolate 20% for Penryn and you basically have Nehalem, according to Anand's tests. We don't know how scaling will be until MP chips come out next year.

 

[jimmysmitty]

My keyboard is possessed. I was using Penryn as a baseline. It's(Nehalem) supposedly 20% faster and not due to HyperThreading. See Anand. Things are totally different with K10 (K10.5) in server. We've already seen the proof with this very post. Intel is 500MHz ahead in clockspeed. Well, really 700MHz based on 2356(2.3GHz).

I'm not saying Intel doesn't have a chance but remember there will be no MP Nehalem this year. I heard Q209. Shanghai also upgrades to HT3 links which will really count in heavy multithreaded loads. AMD is being hush-hush because they are emphasizing Deneb improvements. Fudo thinks it will beat Yorkfield.

A 15-20% improvement (reported) over K10 just makes the wall for Nehalem higher, especially in server loads.

You are not even taking into account the reason why AMDs chips outperform Intels chips in that area. The IMC and HTT. Nehalem will have a IMC and QPI and QPI is said to be faster than HTT 3.0. So based on that (I'm going to assume like you) that means that Nehalem should be faster. I don't have any substantial proof and no real links that show Nehalems server performance but I am going to go based on the performance gains in the desktop market because we all know that the server and desktop market all get the same boost from a IMC.

Seriously. Stop trying to assume what it will do. There have yet to be any Nehalem server benchmarks. The only ones whe have are desktop. In the desktop market they are expecting a boost of 15-20% overall, not in the server market.

 

[ryanthesav]

I think this is a great way for AMD to come back to the desktop market. Go AMD!!!

 

[homerdog]

How is a comparison to Penryn as a baseline irrelevant? Extrapolate 20% for Penryn and you basically have Nehalem, according to Anand's tests. We don't know how scaling will be until MP chips come out next year.

It is irrelevant for the reasons stated above by jimmysmitty. K10 is good in supercomputing and server environments because of its IMC and HyperTransport link. Nehalem has a faster IMC (well it's just a faster processor overall) and CSI (or QPI, whatever you want to call it) which is faster than HT3.0. There is no "wall" for Nehalem to climb, it's the exact opposite. And with every post you further confirm my suspicion that you are a God's honest troll, albeit much more subtle than Thunderman 😉

 

[uguv]

From t

Off topic, but how does this rate +1? Someone must have too much time on his hands to go around and do a +1 on all Baron's posts and a -1 on all of Jimmy's.

 

[jimmysmitty]

^You will notice this very much now. It seems that someone is downing mine and a few others (like poor Grimmy) because we don't agree or something. And only in the BM posting threads.

Oh well.

 

[thunderman]

Intel just give you what you need....AMD give you what you really want

AMD4Life!

 

[The_Blood_Raven]

This is the most amazingly pathetic thread I've seen in a long time. It is like a battle straight out of one of the Lord of the Rings movies with Intel and AMD fanboys lining up their pike lines and catapults. I'm sure this will get rated down to about -15 soon, but hey this is the TRUTH.

First the AMD fanboys (they are first alphabetically), this is good news yes, but this does not in ANY way mean that AMD is making a comeback. That said they might still, no one really knows for sure. I have high hopes for AMD with Deneb and the new Optys, but I don't think we will be seeing a 48xx series type victory here, if any victory. The difference is minimal, and as it has been said the only way AMD can really make the sale with these chips, even if they are godly fast, is if they can prove they still have some life left in them.

Intel fanboys, you are the most annoying so far I'm afraid. All those who just post to insult Thunderman and BM, STFU because that makes you WORSE than them, not better. Thunderman posted an article I would have likely never have found, and I thank him for it. That said we must move on to the Nehalem debate, ah here I come -15 score. Nehalem will NOT bring a huge performance boost, that I am SURE!!! This is because of the software bottleneck, and the lack of true multi threaded 64 bit applications. This will not be remedied for a long time. For all of you who keep saying Nehalem will be the Messiah CPU, please think about this. Nehalem will bring performance boosts in time, but what will really give the immediate performance boosts will be the DDR3 optimization, but that wont be worth the price for awhile to come.

This is how I think things will go, Deneb will compete with the high end Intel CPUs (Yorkfield) and will be priced reasonably. Nehalem will perform better than both, but not by much. The price of Nehalem will assure that it is not worth the trouble until around Q2 09. The real crown will go to the current Intel high end vs. Deneb. The server market will also be about tied between AMD and Intel.

I can not predict the future in anyway, but this is the most likely scenario, I think atleast. As usual only time will tell, but until then please stop hyping unproven products because you personally hope that they will be great, and please stop insulting others before they have done ANYTHING to merit your abuse, because that's just ridiculous, ACT YOUR AGE!!!!!

 

[jimmysmitty]

^I don't mind thunderman posting news.... its just his fanboy attitude that makes it hard to listen to him.

BM is different. I don't like BM because he doesn't post links, he post opinions. That urks me because it starts to spread false information which does not help anyone.

As for Nehalem, no its not the messiah of CPUs. But it does look to be of good design. And you are wrong about it ONLY being the DDR3 for the performance boost. Nehalem, contrary to prior belief, is not just Penryn with a IMC. It includes IPC improvements from just the core itself. The IMC will add a bit of a boost as will the 3 channel DDR3 but those will matter most in the server and MP market.

Fun times.

 

[homerdog]

This is the most amazingly pathetic thread I've seen in a long time. It is like a battle straight out of one of the Lord of the Rings movies with Intel and AMD fanboys lining up their pike lines and catapults. I'm sure this will get rated down to about -15 soon, but hey this is the TRUTH.

First the AMD fanboys (they are first alphabetically), this is good news yes, but this does not in ANY way mean that AMD is making a comeback. That said they might still, no one really knows for sure. I have high hopes for AMD with Deneb and the new Optys, but I don't think we will be seeing a 48xx series type victory here, if any victory. The difference is minimal, and as it has been said the only way AMD can really make the sale with these chips, even if they are godly fast, is if they can prove they still have some life left in them.

Intel fanboys, you are the most annoying so far I'm afraid. All those who just post to insult Thunderman and BM, STFU because that makes you WORSE than them, not better. Thunderman posted an article I would have likely never have found, and I thank him for it. That said we must move on to the Nehalem debate, ah here I come -15 score. Nehalem will NOT bring a huge performance boost, that I am SURE!!! This is because of the software bottleneck, and the lack of true multi threaded 64 bit applications. This will not be remedied for a long time. For all of you who keep saying Nehalem will be the Messiah CPU, please think about this. Nehalem will bring performance boosts in time, but what will really give the immediate performance boosts will be the DDR3 optimization, but that wont be worth the price for awhile to come.

This is how I think things will go, Deneb will compete with the high end Intel CPUs (Yorkfield) and will be priced reasonably. Nehalem will perform better than both, but not by much. The price of Nehalem will assure that it is not worth the trouble until around Q2 09. The real crown will go to the current Intel high end vs. Deneb. The server market will also be about tied between AMD and Intel.

I can not predict the future in anyway, but this is the most likely scenario, I think atleast. As usual only time will tell, but until then please stop hyping unproven products because you personally hope that they will be great, and please stop insulting others before they have done ANYTHING to merit your abuse, because that's just ridiculous, ACT YOUR AGE!!!!!

Yes m'am 😀

 

[The_Blood_Raven]

jimmysmitty, as far as I can tell Thunderman only posted a link and quoted parts of the article and stated the obvious... half the time I'm am sure I have read something similar to what BM posts and I find most of his posts to be exaggerated truths, not lies. That said, the insults towards BM before he even POSTED anything are what annoy me. I never said that DDR3 will be the ONLY performance boost, I just feel it will be the most immediate. Nehalem will be fast, that is for certain, but I do not think it will be fast enough to justify the price for awhile and I think a Q9770 or possibly a Deneb system will carry it's weight for a good few years to come. I suppose this thread angered me because it is becoming depressing watching Tom's degrade over time to a filthy pit of trolls, flame baiters, and fanboys.

homerdog that's "Yes Master" to you, your my b1tch 😀

 

[speedbird]

I can not predict the future in anyway, but this is the most likely scenario, I think atleast. As usual only time will tell, but until then please stop hyping unproven products because you personally hope that they will be great, and please stop insulting others before they have done ANYTHING to merit your abuse, because that's just ridiculous, ACT YOUR AGE!!!!!

How can you say that? Nehalem has to be great because it's Intel, Everything Intel rocks!

I'm kidding ofcourse :lol:

I completely agree as it's time for all fanboys to see the facts first before hyping a Product. I think it's laughable that there's people here who abuse pro AMD supporters and call them Fanboys, when really their no better themselves.

It's a sad world when someone gets upset when somebody says something negative about their favorite brand of Technology. It's just comical watching these people sling mud at each other.

 

[The_Blood_Raven]

speedbird, that brown stuff doesn't smell like mud... oh dear...

 

[jimmysmitty]

Lets all congratulate AMD on this latest Achievement....Intel left for dust here

Where this was pretty short compared to normal, what does this have to do with just AMD? Why couldn't he have just posted it with what it says? That AMD did great as always in the server arena.

As for BM, using Penryn as a performance scale for Nehalem, using Anands desktop review as Nehalems server performance, stating that Intel was pricing their CPUs lower than cost is to make them is exaggerating the truth? Sorry but BM thus far has been using his opinion in almost everything he has stated and has yet to back his comments up.

I for one don't care what CPU company is the best as I buy whatever is the best when I have the money. But don't try to tell me BM just stretches the truth. He has posted so much false information since he came back it makes thunderman look good.

 

[The_Blood_Raven]

I know what thunderman was talking about, overclocking a Phenom is much funner and fulfilling than an Intel quad, but the performance.....

jimmysmitty, I did say half the time....

 

[jimmysmitty]

I know. Its just getting annoying because wrong info gets thrown around and 90% of people are gulible idiots who believe everything that they see and or hear on the internet.

I just want to be able to discuss something without incorrect info and when they are proved wrong they don't fight back without proof.

 

[spud]

^You will notice this very much now. It seems that someone is downing mine and a few others (like poor Grimmy) because we don't agree or something. And only in the BM posting threads.

Oh well.

I have complained to Fredi aka Web Master about this issue, multiple accounts are being rectified, slowly but surely this issue will resolve itself.

Word, Playa.

 

[yomamafor1]

Links please. You continue to remind me of a seven year old who isn't getting his way.

This is funny. I've yet to see a person using this sort of logic to argue: I make a claim without proof, so if you cannot provide a proof, I'm right.

Where does the "K10.5 is 10~20% faster than K10" figure come from?

 

[jimmysmitty]

My keyboard is possessed. I was using Penryn as a baseline. It's(Nehalem) supposedly 20% faster and not due to HyperThreading. See Anand. Things are totally different with K10 (K10.5) in server. We've already seen the proof with this very post. Intel is 500MHz ahead in clockspeed. Well, really 700MHz based on 2356(2.3GHz).

I'm not saying Intel doesn't have a chance but remember there will be no MP Nehalem this year. I heard Q209. Shanghai also upgrades to HT3 links which will really count in heavy multithreaded loads. AMD is being hush-hush because they are emphasizing Deneb improvements. Fudo thinks it will beat Yorkfield.

A 15-20% improvement (reported) over K10 just makes the wall for Nehalem higher, especially in server loads.

How is a comparison to Penryn as a baseline irrelevant? Extrapolate 20% for Penryn and you basically have Nehalem, according to Anand's tests. We don't know how scaling will be until MP chips come out next year.

Thats just part of the crap you have been saying. Using Anands DESKTOP results for Nehalem as a baseline with Penryn when you know that with a IMC and very fast interconnect (QPI) makes a world of difference in the server market.

You know that what performance comes in a server app does not mean anything on the desktop. Server apps normally make use of the IMCs much higher memory bandwidth where as most desktop apps do not. This is especially true in MP servers and considering we have absolutely no server benchmarks for Nehalem I would expect you to not try to use desktop results.

I would post more of the useless info you have been spreading, much like the pricing BS where you got proven wrong so many times but I am too lazy to search for it all.

To top it all, the other AMD fans are saying Anand is biased towards Intel yet you are willing to use the results. This could be because they don't want to admit nehalem is going to do well in the desktop market or it could be a double standard. And you also use Fudo as if its proof. When you know they have spread wrong info and rumors.

 

[yomamafor1]

Apparently all good news for Intel = all bad news for AMD = fake news

Its funny that I've also seen non-analyst trying to argue that AMD is far from a financial crisis.

 

[amdfangirl]

I have complained to Fredi aka Web Master about this issue, multiple accounts are being rectified, slowly but surely this issue will resolve itself.

Word, Playa.

I think the whole rate system is being abused. People are continuously censoring other people's comments whether they be AMD'er or Intel Insider. I think I'll talk to Fredi about the whole rating fiasco. Maybe someone with enough sense to bluntly support one side will use a proxy (or ten). From what I know, the detection software is only capable of detecting multiple identities with the same recurring IP. My say is that we remove or disable it. Its simply not worth it to bother typing rant after flame. Rather we live in harmony?

It seems not a war but rather few angry posters on each side of the jousting fence poking each other. If you can't grow up and actually discuss the topic in a sensible manner...

Edit: This popped up

.....just ended.

Its stops, or I will stop it. The tactics of those who continue 'voting' as a game will include multiple logins.

My tactics will include thread locking, thread deletion and multiple bannings.

I will win.

Got it?

Opinion suppressed...

All hail Turpit...

Instead of completely trawling through the forums deleting and locking, could it be better to simply disable the feature if possible? Is that technically possible? Its fine if you mods/admins don't listen to me. I'm fine with it except you should at least reply or something (no offence tho...)...

 

[thunderman]

I don't even deny that AMD is financially crippled, we may even see some kind of takeover in the future. You have to ask yourself why that is? Is it because AMD sell terrible products? Any person in the know will understand this is not true...Is it because of Intel's Anti-competitive behavior? Well This is a huge factor in why AMD is in the situation it is.

AMD were destroying The so called 'Market leader' for a long time where performance was concerned..then Intel had a break with Core2Duo. Why did Intel still maintain a dominant position? Because of Intel Bribing and Bullying companies to use their products. Go into any typical PC store and you can tell they are Intel biased shoving the Intel Logo in your face and the salesman earning commission to sell Intel Machines to some novice buyer. It's a farce to be honest and I for one will not stand for this any longer...AMD have been denied there true right to be what they should be.

If AMD go under...enjoy your second rate Intel technology.

AMD4Life!!

 

[amdfangirl]

^ Thunderman, even if AMD dies, surely a company will need to replace or buyout AMD. Even if AMD dies and not a phoenix is in sight, Intel will still need sales to stay alive, right? There'll still be evolution, but it'll be slower and costlier. However Intel will realise few willing to buy marginal performance benefits... It all works out in the end, I hope...

 

[yomamafor1]

AMD were destroying The so called 'Market leader' for a long time where performance was concerned..then Intel had a break with Core2Duo. Why did Intel still maintain a dominant position? Because of Intel Bribing and Bullying companies to use their products. Go into any typical PC store and you can tell they are Intel biased shoving the Intel Logo in your face and the salesman earning commission to sell Intel Machines to some novice buyer. It's a farce to be honest and I for one will not stand for this any longer...AMD have been denied there true right to be what they should be.

That's incorrect. If I remember correctly, before AMD rolled out their K8 line, they enjoyed less than 10% of the market. By 2006, AMD had grown to over 20%, and close to 30% of the market. So in short, AMD had been given their rights to the market, IF they have a better product, and better image.

Here's where AMD's real disadvantages comes in. They have very limited resources, and most of the AMD executives didn't believe in marketing. So when Pentium 4 was endorsed by Blue Man Group, what did AMD do to spread their image? Most people don't even know what AMD is, and that's due to Intel's fault?

The only marketing AMD attempted to do was after Core 2's launch. AMD apparently rented a small screen in Time Square, and displayed the number of watts the world could've saved if AMD processor is used. How effective is that in spreading AMD's image?