Multitasking, physical core vs virtual cores

[taltamir]

Your argument is that adding more cores has been hampering the architecture developments.

No it isn't.

What holds back the performance is total R&D required for the next step

I said that.

If you think multiple cores have no merits

That is not what I said.

Are certain applications faster with multiple cores or not?

Seriously have you read a single word I typed?

Where is the proof that more cores has been inhibiting the development of faster/more efficient architectures?

I never said more cores has inhibited the development of more efficient architecture. I said more cores are a simple solution, but are not the most efficient design. Due to the cost of designing a more efficient design it is more practical to simply add more cores until we hit a miniaturization wall. At which point performance increases will significantly slow down, but we can expect to see improved architectures rather then simply throwing more transistors at the problem.

Your problem is that you insist on attributing things to me which I have never said. Reread my posts, I am clearly not saying anything you say I am. I don't know why you find my words so difficult to comprehend.
And please don't retort with "so now you are changing your mind and contradicting yourself". Because I am not. I stand by everything I said, its just not what you claim I said.

 

[GaiaHunter]

I never said more cores has inhibited the development of more efficient architecture. I said more cores are a simple solution, but are not the most efficient design. Due to the cost of designing a more efficient design it is more practical to simply add more cores until we hit a miniaturization wall. At which point performance increases will significantly slow down, but we can expect to see improved architectures rather then simply throwing more transistors at the problem.

Your problem is that you insist on attributing things to me which I have never said. Reread my posts, I am clearly not saying anything you say I am. I don't know why you find my words so difficult to comprehend.
And please don't retort with "so now you are changing your mind and contradicting yourself". Because I am not. I stand by everything I said, its just not what you claim I said.

And isn't that what we have today?

Make new architecture to improve IPC, work on the silicon to increase speeds and add cores to improve performance for some applications and just general multi-tasking.

I disagree. I think it is significantly cheaper and faster to develop and design more slower cores then less faster cores. But I completely disagree that there is any physics effects that make more cores MORE efficient.
Remember that the THERMALS argument of 2 cores at 2.5ghz or 1 core at 3ghz is because of the non linear increase of power consumption with core clock speed and voltage. however, do remember that the dual core is nearly twice as many transistors.
A single core design with the same amount of transistors as the dual core @ 2.5 would be even more efficient overall... but requires significant changes from x86.
Performance: single core @2.5ghz @2B transistors > dual core @2.5ghz @2B transistors > single core @ 3ghz @ 1B transistors

Well you said this and the bold part isn't necessarily true - it depends on the workloads, It will be faster for single threaded applications but it needs to be much more efficient to beat the dual core in other kinds of workloads and multi-tasking. (yes, you stated changes to x86, but that will be a completely different paradigm).

An x86 core has multiple ALUs, multiple registers... that is "good parallelism". Multiple cores are "bad parallelism".

Curiously, AMD with BD seems to be intent in proving that it is better to have two integer cores with around the same amount of transistors.

And again, multiple cores have their uses and can do stuff a single core can't, "bad parallelism" or not.

I understand that multiple cores will never scale perfectly, because there is a point, even in workloads that are easy to split, where the work done by the multiple cores will have to be joined, but even what you define as "good parallelism" is hard to implement. Just increasing individual cores resources doesn't mean they will be put to good use or will show perfect scaling. That is why you there is stuff like HT. The moment your workloads needs to be split you will lose overall usage of resources efficiency, and that happens regardless if you have single cores or not - look at GPUs theoretical peeks and real ones.

 

[taltamir]

And isn't that what we have today?

Yes, it is what we have today. At this stage in time multiple cores approach is the most practical approach due to costs (including the cost in time) of development and market constraints (the need to maintain backwards compatibility).
However, they are not a perfect / ideal approach, and when (not if) the cycle of constantly improving process technology slows down we would likely move away from them to more efficient architecture that take longer to develop and implement.
Alternatively, we might see the rise of a a new type of already highly parallel architecture, such as what we have in GPUs right now (once it matures).

The original point of it all was that multi-core x86 is predominant due to being a simpler and easier (relatively) to design in order to achieve extra performance (by using up a lot of extra transistors in less than ideal efficiency). Not because there is an inherent superiority to multi-core design over a single core design of the same transistor count.

 

[GaiaHunter]

The original point of it all was that multi-core x86 is predominant due to being a simpler and easier (relatively) to design in order to achieve extra performance (by using up a lot of extra transistors in less than ideal efficiency). Not because there is an inherent superiority to multi-core design over a single core design of the same transistor count.

I guess this really is a Science/enthusiast vs engineer argument then.

Simpler and easier and achieve extra performance means higher efficiency for an engineer.

So while I understand what you are saying, being an engineer myself (although not related to computer science or computer hardware) my view is biased.

 

[taltamir]

you know, I just recalled something relevant to this.
Remember when AMD focused on unified monolithic designs, while intel always started out with a quick to market MCM, followed by a unified monolithic design at a later date?
A unified monolithic design is clearly more efficient, but it takes longer to develop (and might divert resources from developing other improvements, which might be better).

just because monolithic performs better than MCM doesn't monolithic is the ideal choice. You have limited budget, development time, manpower, etc. that you must decide how to allocate. And lets not forget manufacturing costs, where MCM means smaller 2 smaller die, thus improved yields.

Simpler and easier and achieve extra performance means higher efficiency for an engineer.

Well, it is certainly Higher efficiency in R&D.

And it is also higher efficiency product in the constraints of your budget (of time, money, manpower, manufacturing capacity, etc)

But some specific design decisions might including the implementation of a component that is less efficient then a known alternative, simply because the resources to improve them were better spent elsewhere.

 

[IntelUser2000]

Ok, whoever that said that "look at task manager" to support a theory that it would hypothetically be better on a many core CPU, look at how much % each process takes.

I have 49 processes running right now. CPU utilization at 1-2%. Do you know what the two highest CPU using apps are and what % they are at? It's Internet Explorer and Task Manager with 1-2% each. 47 other processes take zero CPU time.

You do not need to make 47 more cores in order to power those apps that take at most, 1% CPU time.

My view? Going from 1 to 2 cores were perfect. Major bottleneck on multitasking is opened up by 1 to 2 cores. I would stretch and say even 4 is fine. Beyond that? Consumer apps are NOT server apps. Looking at how many apps take forever to code for 4 cores, and exponentially increasing time required to program beyond that, what's the reason for going 8, or even 16, large general purpose cores?

Let's compare to number of lanes on a busy road. With 1 lane, the average utilization would be 100% most of the time, and the sole limitation is lack of parallelism. With 2 lanes the balance between utilization and parallelism is pretty good here. But you get to 8+ lanes then you start having under-utilized lanes. Will it make everything better by increased amount of lanes on a road that's already under utilized? Probably not.

Better make many smaller cores and add them with few bigger cores. We're back 25 years ago when specialized units aided the central microprocessor.

The key here is that there's no one way or easy way to gain performance anymore. To increase performance, one must improve on ALL categories: # of cores, instruction level parallelism, memory performance, specialized units, better caches, optimized programs.

 

[Scali]

Look, what Intel did (it wasn't proved in courts - there was a settlement between before that, although there are other cases in the court) wasn't exactly reducing price to the final consumers - they weren't selling their CPUs at a loss or something like that. What Intel is/was accused was selling their CPUs at a cheaper rate to retailers in exchange of not selling AMD CPUs and/or directly paying to prevent AMD sales.

I never denied that (so there's no need for you to reiterate it again, we all know what happened), but I don't think you understood my point.
The problem here is, anytime Intel gives discount, they again risk a lawsuit, which they probably will lose again (justice has little to do with it, they've been convicted before, and I'm sure AMD's legal team can spin it to make them look guilty again).
Why would Intel take that risk?

 

[GaiaHunter]

Ok, whoever that said that "look at task manager" to support a theory that it would hypothetically be better on a many core CPU, look at how much % each process takes.

I have 49 processes running right now. CPU utilization at 1-2%. Do you know what the two highest CPU using apps are and what % they are at? It's Internet Explorer and Task Manager with 1-2% each. 47 other processes take zero CPU time.

You do not need to make 47 more cores in order to power those apps that take at most, 1% CPU time.

But by the same token do you need a faster architecture for that?

 

[GaiaHunter]

The problem here is, anytime Intel gives discount, they again risk a lawsuit, which they probably will lose again (justice has little to do with it, they've been convicted before, and I'm sure AMD's legal team can spin it to make them look guilty again).
Why would Intel take that risk?

My point was that those previous discounts didn't translate into cheaper prices for the consumer.

In fact Intel processors were never as cheap as they are now. In fact as long as Intel isn't selling the CPUs at a loss no lawsuits will be issued - the lawsuits that were issued were about something else.

 

[Scali]

My point was that those previous discounts didn't translate into cheaper prices for the consumer.

Is that Intel's fault though? I think the OEMs decided that. But not in all cases anyway.

In fact Intel processors were never as cheap as they are now.

Yea? I don't think so. There's absolutely no pressure on the faster Core ix models.
I think that during the Pentium D era, Intel had the cheapest CPUs ever. They undercut AMD's prices. And yes, the consumer benefited from that.

In fact as long as Intel isn't selling the CPUs at a loss no lawsuits will be issued - the lawsuits that were issued were about something else.

Intel has never sold CPUs at a loss. As I have explained many times before, Intel's production costs are lower than AMD's for various reasons (more efficient architecture, smaller dies for the same performance level, economy-of-scale advantages, 32 nm technology etc). So Intel can completely crush AMD on price if they wanted, and still not sell at a loss.
But do you think AMD won't sue? I don't. I think they'll just try to paint Intel as a criminal again, to try and cover up for the fact that AMD is failing to deliver proper technology to the end users, and is losing the competition on a fair basis.

 

[GaiaHunter]

Yea? I don't think so. There's absolutely no pressure on the faster Core ix models.
I think that during the Pentium D era, Intel had the cheapest CPUs ever. They undercut AMD's prices. And yes, the consumer benefited from that.

And during Athlon 64/FX and Athlon X2 days, with pressure of the P4 EE and PD EE were still priced at $1000 like the i7 980 and just like the AMD FX products then.

For the situation being similar AMD would have to be pricing their X6 parts at $1000 too (sure AMD lead wasn't as big as Intel lead now is, but was still pretty clear).

And don't the Athlon II X3/X4 and Phenom II X4 limit the price of i3 and i5? How about the X6 parts - don't you have to think exactly what are your uses before going one way or the other?

But do you think AMD won't sue? I don't. I think they'll just try to paint Intel as a criminal again, to try and cover up for the fact that AMD is failing to deliver proper technology to the end users, and is losing the competition on a fair basis.

Sorry but this is a non sequitur and a straw-man argument.

http://en.wikipedia.org/wiki/AMD_v._Intel

AMD launched the lawsuit against its rival Intel, the world's leading microprocessor manufacturer. AMD has claimed that Intel engaged in unfair competition by offering rebates to Japanese PC manufacturers who agreed to eliminate or limit purchases of microprocessors made by AMD or a smaller manufacturer, Transmeta.[1]

AMD didn't sue Intel on therms of Intel selling at a loss - so I don't understand what are the merits of repeating that Intel doesn't drop prices because of AMD - Intel doesn't drop prices because it is not advantageous for them. Period.

Second, Intel was fined before, hence considered guilty of anti-competition practices.

In 1991, AMD filed an antitrust lawsuit against Intel claiming that they were trying to secure and maintain a monopoly,[citation needed] and one year later, a court ruled against Intel, awarding AMD US$10 million "plus a royalty-free license to any Intel patents used in AMD's own 386-style processor".[5]

From the same source.

http://news.bbc.co.uk/2/hi/8047546.stm

Computer chipmaker Intel has been fined a record 1.06bn euros ($1.45bn; £948m) by the European Commission for anti-competitive practices.
[...]
The Commission found that between 2002 and 2007, Intel had paid manufacturers and a retailer to favour its chips over those of Advanced Micro Devices (AMD).

And of course you have the settlement between AMD and Intel (yeah this one doesn't prove guilty but doesn't prove innocence either).

http://www.washingtonpost.com/wp-dyn/content/article/2009/11/12/AR2009111210555.html

The world's two biggest chipmakers, whose Silicon Valley headquarters are separated only by a few miles, have put an end to one of the high-tech industry's longest and costliest legal battles.

But Intel's agreement to pay Advance Micro Devices $1.25 billion to drop multiple lawsuits Thursday won't stop investigations by U.S. and European antitrust watchdogs. The regulators are investigating allegations Intel used bribes and bullying tactics to maintain its dominance in the highly concentrated chip industry.

And there are more. Just google for it.

Nowhere was this related to Intel dropping their processors prices to the consumers.

 

[taltamir]

Intel has never sold CPUs at a loss. As I have explained many times before, Intel's production costs are lower than AMD's for various reasons (more efficient architecture, smaller dies for the same performance level, economy-of-scale advantages, 32 nm technology etc). So Intel can completely crush AMD on price if they wanted, and still not sell at a loss.
But do you think AMD won't sue? I don't. I think they'll just try to paint Intel as a criminal again, to try and cover up for the fact that AMD is failing to deliver proper technology to the end users, and is losing the competition on a fair basis.

Intel might be ABLE to do so, but instead they use exclusivity contracts, and reward sellers for not stocking AMD. Rewarding sellers for not selling your competition, and punishing them if they do, is anti competitive business practices and is illegal. If intel just sells all its products cheaper then they are breaking no laws. But its cheaper for them to put the squeeze on sellers than it is to discount their prices

 

[Scali]

Intel might be ABLE to do so, but instead they use exclusivity contracts, and reward sellers for not stocking AMD. Rewarding sellers for not selling your competition, and punishing them if they do, is anti competitive business practices and is illegal. If intel just sells all its products cheaper then they are breaking no laws. But its cheaper for them to put the squeeze on sellers than it is to discount their prices

This has nothing to do with what I said though. As I said: we know what happened (or at least, what they were convicted for).
Let me put that in less subtle words: I am interested in technology, and I don't really want to have legal discussions. Especially since they are not related to the point that I was making. I wish people would just stop bringing it up over and over again. I have no interest in discussing it whatsoever.

 

[Scali]

And during Athlon 64/FX and Athlon X2 days, with pressure of the P4 EE and PD EE were still priced at $1000 like the i7 980 and just like the AMD FX products then.

That's why they are Extreme models (and performance was close enough to justify the prices). I was talking about the cheap Pentium Ds, like the 805 for example. Intel had a bunch of Pentium Ds that had better price/performance than the Athlon X2 at the time.

For the situation being similar AMD would have to be pricing their X6 parts at $1000 too (sure AMD lead wasn't as big as Intel lead now is, but was still pretty clear).

Nonsense. But your part in brackets shows that deep down you already knew that. The performance lead determines the price difference. The EE processors were pretty competitive.

And don't the Athlon II X3/X4 and Phenom II X4 limit the price of i3 and i5? How about the X6 parts - don't you have to think exactly what are your uses before going one way or the other?

Only up to about the 860. The rest has no competition whatsoever.
And it's pretty clear that Intel's CPUs are technically a lot cheaper than AMD's, especially the X6. But Intel doesn't bother to undercut AMD's prices.

Sorry but this is a non sequitur and a straw-man argument.

No it's not. I'm saying that I think AMD will sue, regardless of whether they have an actual case or not. So they'll 'paint Intel as a criminal', as they did before.

AMD didn't sue Intel on therms of Intel selling at a loss - so I don't understand what are the merits of repeating that Intel doesn't drop prices because of AMD - Intel doesn't drop prices because it is not advantageous for them. Period.

So we have a difference of opinion.

Nowhere was this related to Intel dropping their processors prices to the consumers.

That's not the point. Point is that AMD has sued because it was the only way to survive, and they'll do it again.

 

[taltamir]

This has nothing to do with what I said though. As I said: we know what happened (or at least, what they were convicted for).
Let me put that in less subtle words: I am interested in technology, and I don't really want to have legal discussions. Especially since they are not related to the point that I was making. I wish people would just stop bringing it up over and over again. I have no interest in discussing it whatsoever.

fair enough.

 

[GaiaHunter]

This has nothing to do with what I said though. As I said: we know what happened (or at least, what they were convicted for).

First you said

*Because* of AMD, Intel is not lowering prices as far as they could.
Obviously, without AMD there is no need to lower prices into a competitive range.

Then you said

The problem here is, anytime Intel gives discount, they again risk a lawsuit, which they probably will lose again (justice has little to do with it, they've been convicted before, and I'm sure AMD's legal team can spin it to make them look guilty again).
Why would Intel take that risk?

But the question here is that lowering the prices of CPUs is not the same than giving discounts to retailers based on what those retailers do with the products of Intel competition.

This isn't the same as, for example, McDonald's getting better prices for being coca-cola exclusive.

 

[Scali]

But the question here is that lowering the prices of CPUs is not the same than giving discounts to retailers based on what those retailers do with the products of Intel competition.

I never said that they were sued BECAUSE they lowered prices for consumers... I said that BECAUSE they were sued by AMD many times before (basically AMD just threw all the shit they could find at the wall, and was hoping that some of it would stick), Intel is careful with competitive pricing.

 

[GaiaHunter]

That's why they are Extreme models (and performance was close enough to justify the prices). I was talking about the cheap Pentium Ds, like the 805 for example. Intel had a bunch of Pentium Ds that had better price/performance than the Athlon X2 at the time.

And so does AMD now.

Nonsense. But your part in brackets shows that deep down you already knew that. The performance lead determines the price difference. The EE processors were pretty competitive.

It is nonsense because we are talking about AMD and not Intel. AMD simply doesn't have the muscle Intel had to force their product on the market by via of removing the competition from the same market.

Basically Intel wasn't trying to compete with AMD, they were trying to prevent AMD to establish a bigger base in both marketshare and mindshare.

Only up to about the 860. The rest has no competition whatsoever.
And it's pretty clear that Intel's CPUs are technically a lot cheaper than AMD's, especially the X6. But Intel doesn't bother to undercut AMD's prices.

We agree, but you stated before Intel doesn't do it because of AMD threats of lawsuits.

No it's not. I'm saying that I think AMD will sue, regardless of whether they have an actual case or not. So they'll 'paint Intel as a criminal', as they did before.

And what will a lawsuit do? Take years to be reach somewhere and when it reaches somewhere the paradigms is completely different of what it was and what it could have been?

So we have a difference of opinion.

How can we have a difference of opinion over a fact?
"The sky is blue." "No, I beg to differ."?

It is the same here - anyone can read the lawsuits and see they are about pressures over retailers by Intel related to AMD products, not about Intel taking a loss over their products.

That's not the point. Point is that AMD has sued because it was the only way to survive, and they'll do it again.

Hmm, so AMD should just stay quiet over practices that are consider illegal by the market regulation and hamper their business?

I never said that they were sued BECAUSE they lowered prices for consumers... I said that BECAUSE they were sued by AMD many times before (basically AMD just threw all the shit they could find at the wall, and was hoping that some of it would stick), Intel is careful with competitive pricing.

So gross margins have nothing to do with Intel pricing scheme?

And are you saying that Intel pricing isn't competitive?

And how many times did AMD sued Intel over the fact Intel products are price competitive and how many times did it win? Links?

Last, are you forgetting that the agreement Intel and AMD reached include an agreement over business practices?

EDIT: This is the history of Intel and AMD legal disputes I found - feel free to add information.

http://news.cnet.com/Intel-and-AMD-A-long-history-in-court/2100-1014_3-5767146.html?tag=nw.20

1982--Intel and AMD sign a technology exchange agreement making AMD a second supplier. The deal gives AMD access to Intel's so-called second-generation "286" chip technology.

1984--Intel seeks to go it alone with its third-generation "386" chips using tactics that AMD asserts were part of a "secret plan" to create a PC chip monopoly.

1987--AMD files legal papers to settle the 386 chip dispute.

1991--AMD files an antitrust complaint in Northern California claiming that Intel engaged in unlawful acts designed to secure and maintain a monopoly.
1992--A court rules against Intel and awards AMD $10 million plus a royalty-free license to any Intel patents used in AMD's own 386-style processor.

2000--AMD complains to the European Commission that Intel is violating European anti-competition laws through "abusive" marketing programs. AMD uses legal means to try to get access to documents produced in another Intel antitrust case, this one filed by Intergraph. The Intergraph case is eventually settled.


2004--Japan's Fair Trade Commission (JFTC) raids Intel offices in Japan searching for documents. Intel cooperates with the investigation but does not agree with the outcome. JFTC officials find that Intel's Japan unit stifled competition by offering rebates to five Japanese PC makers--Fujitsu, Hitachi, NEC, Sony and Toshiba--which agreed not to buy or to limit their purchases of chips made by AMD and Transmeta.

2005--AMD files an antitrust suit against Intel in U.S. District Court in Delaware. The 48-page complaint alleges in detail that Intel has unlawfully maintained its monopoly in the x86 microprocessor market by coercing customers worldwide from dealing with AMD.

 

[Scali]

And so does AMD now.

Erm, duh? Haven't I already said that Intel *never* had price/performance advantage over AMD? Pentium D is pretty much the only exception (and other 'accidents' where Intel would introduce a new architecture, like Core2 Duo and Core i7, where performance was so much higher, that price/performance was completely redefined, and AMD didn't respond with price cuts quick enough... so Intel temporarily took price/performance until AMD restored order).

We agree, but you stated before Intel doesn't do it because of AMD threats of lawsuits.

No, the way you state it, it sounds like it is THE reason.
I said that I think it is a factor (and I have also mentioned other factors, so clearly it is not THE reason).

How can we have a difference of opinion over a fact?
"The sky is blue." "No, I beg to differ."?

It's not the facts, but they way they are interpreted.

Hmm, so AMD should just stay quiet over practices that are consider illegal by the market regulation and hamper their business?

That's not what I'm saying. I'm saying that Intel is extra careful not to take actions that could somehow be interpreted as illegal.

I think I'm about done with you now. Your style of arguing is very one-dimensional. If I say something about Intel, you assume the opposite goes for AMD, or vice versa.
Other than that, you are very creative at 'reinterpreting' things that I say, trying to connect things that have no connection.
It's a waste of energy (I need to start pretty much every post with "That's not what I said"... perhaps you need to take a step back, take a deep breath and assess what you're doing here, it may give you some personal insights). Apparently we don't agree. Bye now.

 

[GaiaHunter]

I think I'm about done with you now. Your style of arguing is very one-dimensional. If I say something about Intel, you assume the opposite goes for AMD, or vice versa.
Other than that, you are very creative at 'reinterpreting' things that I say, trying to connect things that have no connection.
It's a waste of energy. Apparently we don't agree. Bye now.

The problem is simple.

You have opinions and turn them into facts without providing proof.

I point to information/data that contradicts/doesn't support/expands on your opinions.

Instead of trying then to support/expand/clarify your arguments you either move forward or backwards on your arguments.

You said that is your opinion AMD will sue Intel if Intel drop prices.

I asked why and you say it is because it is the modus operandi of AMD.

Then I ask for you to support that modus operandi of AMD and you don't, because it simply there is no evidence to support it.

And basically most of the opinions you have regarding any subject involving AMD just goes like this - it either lacks support, is contradicted or there is a "catch" that makes your arguments less compelling/strong.

 

[Scali]

See, there you go again (see my edit, I clarified a bit)...
You're arguing like a boxer, trying to push me into a corner and keep pounding on and on. Even now... I decide to drop the discussion because it's getting nowhere, and you continue, getting personal too.
Forget it, I'm not interested. I'd also like to point out that you're the only one having a go at me pretty much. From that I assume that other people are perfectly happy with my arguments, and frankly I don't care what you think.

 

[waffleironhead]

See, there you go again (see my edit, I clarified a bit)...
You're arguing like a boxer, trying to push me into a corner and keep pounding on and on. Even now... I decide to drop the discussion because it's getting nowhere, and you continue, getting personal too.
Forget it, I'm not interested. I'd also like to point out that you're the only one having a go at me pretty much. From that I assume that other people are perfectly happy with my arguments, and frankly I don't care what you think.

No, I think most on here have realized that it is pointless to discuss anything with you.

 

[GaiaHunter]

Even now... I decide to drop the discussion because it's getting nowhere, and you continue, getting personal too.

Personal? How?

It is very simple - you show information that shows AMD suing Intel because Intel products simply are more competitive and your point is proven.

Otherwise, your opinions are just that, opinions, and I've no problem that you have whatever opinions you want to have.

My only problem is that you present your opinions as facts, and you do a good job at that (if your arguments were plain stupid I wouldn't lose my time) and even give some interesting technical information , until someone looks closer and do some research.

 

[Scali]

No, I think most on here have realized that it is pointless to discuss anything with you.

It's pointless because I know what I'm talking about and can back up my arguments better than anyone else.
Problem with Gaiahunter as I pointed out is that he doesn't attack my arguments, but pulls my words out of context and makes me defend completely other things... or just attacks me personally.
So it's pointless discussing with him.

 

[GaiaHunter]

It's pointless because I know what I'm talking about and can back up my arguments better than anyone else.
Problem with Gaiahunter as I pointed out is that he doesn't attack my arguments, but pulls my words out of context and makes me defend completely other things... or just attacks me personally.
So it's pointless discussing with him.

So is it AMD fault that doesn't drops the prices or not?

Because I'll be damned if this

Yes, that's what I thought aswell.
AMD's lawsuits eventually resulted in a billion+ fine. Intel will probably think twice before they sell their CPUs at low prices again.
So in an ironic twist of fate, AMD has killed competition.

doesn't mean that AMD is the responsible for us consumers not having cheaper prices for Intel CPUs due to use of lawsuits!

If you said AMD CPUs simple don't perform well enough to bring Intel CPU's prices down I wouldn't have said a thing.

And of course you attributed the lawsuit to cheap Intel prices instead of pressures over retailers.