Breaking - AMD Warns! - Page 5

Last message on previous page:

Monopolies are not illegal in the US. Anti-competitive behavior is. The two are by no means the same thing.

I guess i should have put in one more word god knows how i feel the need to be accepted by you



http://www.usdoj.gov/atr/contact/newcase.htm

The Sherman Act also makes it a crime to monopolize any part of interstate commerce. An unlawful monopoly exists when only one firm controls the market for a product or service, and it has obtained that market power, not because its product or service is superior to others, but by suppressing competition with anticompetitive conduct.

The Act, however, is not violated simply when one firm's vigorous competition and lower prices take sales from its less efficient competitors; in that case, competition is working properly.

http://www.ftc.gov/bc/compguide/xintro.htm

To ensure consumer choice, the antitrust laws set two basic requirements: companies cannot agree to limit competition in ways that hurt consumers; and a single company cannot monopolize or try to monopolize an industry through unfair practices.

http://en.wikipedia.org/wiki/Monopolization

Under long-established precedent, the offense of monopolization under Section 2 has two elements. First, that the defendant possesses monopoly power in a properly-defined market and second that the defendant obtained or maintained that power through conduct deemed unlawfully exclusionary. The mere fact that conduct disadvantages rivals does not, without more, constitute the sort of exclusionary conduct that satisfies this second element. Instead, such conduct must exclude rivals on some basis other than efficiency.

Being a monopoly is not illegal in the US Vern. Only monopolization via unfair business practices.

You could proofread the article or even send it to Tanker for proofing. I am sure you could do as well as what we see for articles here.

Oh, I see now. Sorry about getting peeved about it, but it's such a common misconception that US anti-trust laws make monopolies per se illegal, when in fact they don't, that I have a tendancy to pounce on it.

Personally I'd of gone with avant-garde, but I guess that could be considered mildly flattering. Your definition of politically incorrect clearly needs updating, the only thing considered incorrect in the political sense now a days is being republican. Though you wouldn't know that from watching fox news.

Dude, I followed you up to the first sentence of your last post.

I would agree with you the feds are keeping the economy going right now just look at the M1 and M2 money supply numbers.

Again the Vista debacle is readily apparent MS really dropped the ball again like XP I would give it a year and wait for service pack 2 before you actually expect it to work.

As for AMD I don't expect to see any recovery until July and at that point Q2 results are due in. Q2 is going to be worse than Q1 with the price cuts and new Allendale processors from Intel putting additional pressure on AMD prices and volumes.

I put my recovery estimate out to Q3 mid to late qtr. I think for any extensive gains 07 was shot down by unprioritized planning dating back to 04/05.

The ball was dropped when AMD decided to scale 90nm instead of follow intels lead to 65nm in 05 dec.They should have responded with k8 am2 65nm in june 06 or there abouts but were overly confident in k8 and not giving Intel credit for closing the gap with core mobile chips in dec o5.

So as far as amazing gains are concerned 07 is shot outside of a miracle market turnaround. There are too many bad elements to consider before AMD will rebound after a stock dilution.

GFX will sell but it wont pick up until k10 is released and there after so I expect that mid Q3 will be much better depending on dilution ratios.

I should restate that I don't expect the stock slide to stop until July recovery would follow the release of Barcelona and is likely to be very short lived before Penryn nails AMD again.

Penryn knocks them off the top in performance and eliminates any control they could have on prices which means the bleeding by Intel continues until theere is no more. Intel is nothing if not predatory and when Intel smells blood in the balance sheets they are going for the kill. They want to squeeze AMD as hard as possible for as long as possible and they hope long enough before they end up filling up AMD's coffers with cash from that legal problem AGAIN.

Perhaps I am missing something here, but I don't believe SSE4 will amount to much, at least for a while. From what I understand SSE4 consists of *new* instructions. Will the apps/games/benchmarks used to test K10 and penryn actually have support for SSE4 instructions compiled in? Furthermore, isn't it mostly multimedia applications (i.e. encoders) that will benefit from SIMD type instructions?

Edit: I do however agree with you that AMDs numbers are suspect and we will have to wait and see which architecture is better.

******************! I like amd. Intel sounds tooo.... wierd. Amd sounds like man on an everlasting source of horse steriods. but of course amd is better than tiny men on roids.

It would be smart for AMD to do the same as ATI and release their full line all at once. Quad, Dual, Single and maybe in 3 months we'll be crying for Intel because noones' board can run DDR3.

As I recall it was roughly the same as the shrink from 90nm to 65nm. I belive the term is zilch.

Major what? and you mean DDR3 over DDR2 don't you?

So far there have been no performance gains from AMD in process shrink or memory types which holds concern for a shift from DDR2 to DDR3. The problem could be poor design or it may be inherent in the AMD IMC which means we will see a repeat when moving to DDR3. Without comments from an AMD insider on what the problem was going to DDR2 we can only guess if they will have the same probelm going to DDR3.

Amen to that. They need to put something worthwhile out soon AMD's offerings are becoming like the Intel releases of Heatburst.

Here you go.

Actually it lost by a small bit <.2%; their architecture relies on low latency not high bandwidth so much.

Somebody is.
The main difference between northwood a,b,and c was the mem speed.
Intel chips rely on good prefetch. This means they have the data, before it's called. It's not efficent use of bandwidth, so they need more.

BTW Jack, the reason Amd benefits from LL, is a percentages game.
System latency is low, so any change is a higher percent.
Say thier sys latency is 50, while Intel's is 100. Increasing latency by 4, is only a 4% cxhange for the Intel system, but an 8% change for the AMD system.

Somebody is.
The main difference between northwood a,b,and c was the mem speed.
Intel chips rely on good prefetch. This means they have the data, before it's called. It's not efficent use of bandwidth, so they need more.

BTW Jack, the reason Amd benefits from LL, is a percentages game.
System latency is low, so any change is a higher percent.
Say thier sys latency is 50, while Intel's is 100. Increasing latency by 4, is only a 4% cxhange for the Intel system, but an 8% change for the AMD system.

So basically its the same cos 50*8%=54 and 100*4%=104. Isnt this the same
benefit. Or am i wrong? Yes, that's wrong. Both add 4, but the first (50+4) is still an 8% improvement, the second(100+4) is only a 4% improvement.