Whatever happened to Moore's law, amd??

[cbn]

If the xtors have problems getting smaller....maybe they can go faster instead?

How about 100 Ghz processing speeds ?

 

[alyarb]

100 GHz at ~260 nanometer. I wonder if it would be faster at 28 nanometer.

 

[Enigmatic]

The thing is, I remember that even when AMD had the lead (k8) intel processors were still not quite priced competitively. Intel doesn't really have to compete at all, their history and marketing are so good that big OEMs like dell and HP will continue to buy from them for years to come, and tbh, your average consumer really doesn't care what's in his/her computer.

Yeah, Intel has the financial clout to stay quite profitable even when they had the inferior products at higher price points. I remember before I was into PC hardware that the Pentium brand was such an established name I wouldn't even think of AMD. It took a while for me to be convinced otherwise (my first gaming rig was a boutique system and it had a... shudder... Pentium D).

Honestly, AMD has to be really glad they bought ATi. Their future essentially depends on these Fusion processors that they keep touting, so that they can power their GPU successes into the CPU market.

 

[Cogman]

no im not going to contribute anything else to this thread.

Because it seems like NO ONE CARES on the stuff i post.
So you guys can wait and see when gulftown does Tear Magony a new one.

Because i have seen it, oh... but its all inside information that i cant really leak out.

Let me ask you guys who do know me..

HOW MANY TIMES HAVE I BEEN WRONG?
And HOW many times have i made people go OOPS :hmm:


The guys who know me know not to go against me, unless i grossly messed up on something.

But IDC hasnt posted... neither has mark... we dont see Dmens..

Once again in a few months when things become offical, lets see who was wrong now shall we?

Do you have both sets of processors on hand? You may be right, but Its a bit premature to declare anything without having both chips to compare to each other.

(BTW, I was confusing you with AleleVanuatu early, my mistake. I don't think I said anything in here relations to a discussion I had with him about the merits of multithreading, but all the same, sorry for mixing you up with someone else.)

 

[EarthwormJim]

moors law is really more like "moors general observation that isn't entirely accurate".
AMD hasn't been able to keep up with it. Intel has been intentionally holding back for years and actually uses moors law as a marketing target.
We are getting into a point where advancing CPU tech is very difficult.
If you look at SSD, most companies have advanced far faster then moors law, with the overall SSD speed being doubled every 3 months early on, later 6 months... and now we continue to slow the trend, but are still advancing far faster than moors law.

Moore's law is not speed it's transistor density. AMD has been able to keep up with it pretty much.

SSD controllers have gotten faster, but the flash memory itself has not really doubled in speed every three months.

 

[taltamir]

Moore's law is not speed it's transistor density. AMD has been able to keep up with it pretty much.

SSD controllers have gotten faster, but the flash memory itself has not really doubled in speed every three months.

if you are only looking at transistor density, then this was a target for intel. with their tick tock cycles they have intentionally aimed at moving to a new processes every 18 months.

oh, and it definitely doesn't hold true for GPUs where nvidia and AMD (but mostly nvidia) have been packing transistors at breakneck speeds.

 

[IntelUser2000]

oh, and it definitely doesn't hold true for GPUs where nvidia and AMD (but mostly nvidia) have been packing transistors at breakneck speeds.

Well... not really, because die sizes have also been increasing tremendously in GPUs. Process technologies have been advancing similarly for them. Using more texture caches and such also increase transistor count without impacting power/die size much.

 

[taltamir]

Well... not really, because die sizes have also been increasing tremendously in GPUs. Process technologies have been advancing similarly for them. Using more texture caches and such also increase transistor count without impacting power/die size much.

so? the bottom line is that they have been increasing transistor counts faster then moors law. That is because unlike intel:
1. They have competition
2. They are not intentionally limiting themselves to moors law.

I read an interview with an intel exec who said they use moors law as a target. Amusingly, moors law is about marketing and economics, not about technology.

 

[IntelUser2000]

No, cause Moore's Law is about Transistor Density, not transistor counts. Think of how long it takes to go from one generation of process to another. It's approximately 2 years.

Plus, you cannot compare general purpose processor like a CPU that is designed to make every code faster vs a GPU which makes very limited number of specific code faster. It is easier to do the latter.

When you have bunch of identical circuits that you can just duplicate them all over the die(like SPs) it's easier to optimize for transistor density, which is partly the reason why GPUs have more transistors per die.

 

[piesquared]

no im not going to contribute anything else to this thread.

Because it seems like NO ONE CARES on the stuff i post.
So you guys can wait and see when gulftown does Tear Magony a new one.

Because i have seen it, oh... but its all inside information that i cant really leak out.

Let me ask you guys who do know me..

HOW MANY TIMES HAVE I BEEN WRONG?
And HOW many times have i made people go OOPS :hmm:


The guys who know me know not to go against me, unless i grossly messed up on something.

But IDC hasnt posted... neither has mark... we dont see Dmens..

Once again in a few months when things become offical, lets see who was wrong now shall we?

I hope you haven't seen MC numbers from Anandtech. Because if so, isn't that a violation of the NDA?

 

[taltamir]

No, cause Moore's Law is about Transistor Density, not transistor counts. Think of how long it takes to go from one generation of process to another. It's approximately 2 years.

ah that... well that IS a matter of pure marketing, not technology.

actually, if you want to know what he really said read this:
ftp://download.intel.com/museum/Moo.../Excepts_A_Conversation_with_Gordon_Moore.pdf
His original (later revised) publication:
ftp://download.intel.com/museum/Moores_Law/Articles-Press_Releases/Gordon_Moore_1965_Article.pdf

I got the links here:
http://en.wikipedia.org/wiki/Moore's_law#cite_note-0

 

[PCboy]

no im not going to contribute anything else to this thread.

Because it seems like NO ONE CARES on the stuff i post.

You're right. We don't. This topic was about the discussion regarding Moore's Law and if AMD would be able to follow it. Now what was the first thing you posted?

gets raped by gulftown.

Almost as bad as Rollo.

 

[IntelUser2000]

The thing is you can't bring forward product launch cycles much and 2 years is about the limit of how fast they can progress. "Moore's Law" was also one reason process technology cycle accelerated from 3 to 2 years, everyone needs something to lean or follow.

And no, you can't compare amount of transistors on a CPU and a DSP-like logic like GPUs.

 

[JFAMD]

Do you have both sets of processors on hand? You may be right, but Its a bit premature to declare anything without having both chips to compare to each other.

(BTW, I was confusing you with AleleVanuatu early, my mistake. I don't think I said anything in here relations to a discussion I had with him about the merits of multithreading, but all the same, sorry for mixing you up with someone else.)

I doubt they have both sets of processors in hand. I have one of the two in question and I am pretty confdent, based on what I have seen, that this person has neither.