I give up being a CPU enthusiast.

[LiuKangBakinPie]

And maybe they can design a cpu cooler like they design their cpus. Not trying to cut off two of your fingers installing it. I mean they can put billions of things on a tiny piece of silicon. Is it that hard to design a decent cpu cooler?

 

[SHAQ]

CPU space won't get interesting until they ditch silicon and move to graphene transistors. At best they can do is pack more and more cores, move to fusion, and 3D stacking.

If they lose the market there may not be a graphene. How much computing power do you need for a phone/tablet? Not much more than they have now.

 

[Idontcare]

Nowere in Hebrew scripture on genesis is the term days used . The hebrew word YORM is used which is an undefined period of time . The scribes had to change to make money threw religion . The Sabbath and the the seventh day of creation the DAY of rest are not the same . Fact is the sabath day in actual fact is a day of rest from task and has nothing to do with worship . and have nothing to do with GODS holy Day as your soon going to find out . Also the christ uses the word AEON

And you know this because? You read it in a book that you believe to be true while refuting one which you believe to be untrue.

Think about it. When was the last time you met and heard Christ use the word AEON? You take it on faith from someone who says so...which makes it no less true or credible than the dude who takes it on faith that they literally were days of the 24hr sort.

CPU space won't get interesting until they ditch silicon and move to graphene transistors. At best they can do is pack more and more cores, move to fusion, and 3D stacking.

Why does the industry need graphene to make any of that happen? (the moving beyond more cores and so on?)

Process tech (vacuum tubes, Si xtors, BiCMOS, graphene) is completely decoupled from microarchitecture (Nehalem, Pentium, Cortex A9, Niagara, Power7) is completely decoupled from ISA (x86, ARM, SPARC, POWER, IPF).

Cores and so on is a microarchitecture thing, has nothing to do with process tech like graphene.

 

[MobiusPizza]

Why does the industry need graphene to make any of that happen? (the moving beyond more cores and so on?)

Process tech (vacuum tubes, Si xtors, BiCMOS, graphene) is completely decoupled from microarchitecture (Nehalem, Pentium, Cortex A9, Niagara, Power7) is completely decoupled from ISA (x86, ARM, SPARC, POWER, IPF).

Cores and so on is a microarchitecture thing, has nothing to do with process tech like graphene.

First, you misinterpreted my sentence. I am replying to the initial topic, that CPU race in current form, more cores, SoC, are 'boring'. You don't need graphnene to do this as they are silicon tech.

To answer your 2nd question,power ceiling and TDPs. The performance inprovement of recent CPUs so far, is still achived by process shrink and achitectural change.
Process node cannot go much below 22nm, that's a physical constraint, before leakage cannot be contained effectively and silicon process is no longer viable. I.e. you can't shrink chips continuously without end to get better thermal and power efficiency. Once you get to a limit, say 18nm, you must rely on clever designs like power gating, FinFET etc. But you can only milk a cow so much, they eventually die.

If they lose the market there may not be a graphene. How much computing power do you need for a phone/tablet? Not much more than they have now.

That's like quoting Bill Gates being shortsighted, 640k is ought to be enough for everyone. No one is satisfied about computing power. If there is technology available someone will utilise it. 3D gaming is an obvious trend. If you want a phone which do realtime voice-text perfect language translation in the future, god knows how much computing power that needs. Technology can go the other way, graphene transistors not only runs faster, they use less power. Who doesn't want a phone that can charge itself just from some solar cells on the case and never need the mains.
Hack if in future a phone is a true PDA (personal digital assisitant) being a some AI on a quantum computer chip, I would rest in peace.

 

[lopri]

Yup. The CPU world is just incredibly not exciting for the PC. Sure, in the mobile/ultramobile space, it's neat and all, but at the high end, it's become a clusterfunk. Intel with two lines of CPUs with the lower end stuff getting the better uArchs first, Bulldozer disappointing, ARM basically holding a knife to AMD/Intel's throats and basically killing the desktop market, etc.

All that's left for us enthusiasts is new, badass graphics cards. These are the only thing that the f*cking phones/tablets couldn't hope to catch up with anytime soon. Let's hope AMD/NVIDIA's 28nm round are exciting.

Hey you're making me feel guilty. Posting on an iPad here. In a bed. ^^

Edit: Now I'm being "proper" and on my desktop. I got to learn about Bulldozer yesterday while watching a VOD on my phone (an ad popped up screaming "world's first 8-core desktop CPU") so I went onto read about it. That thing seems to be a "beast", to put it mildly.

But other than that I noticed nothing really has changed from a year ago.. still Thubans, Sandies, 6900s and 580s. It looks like I did not miss much. And I agree that desktop PCs are dying. Look no further than contents of this site, or take a walk to your local Bestbuys.

 

[Nemesis 1]

And you know this because? You read it in a book that you believe to be true while refuting one which you believe to be untrue.

Think about it. When was the last time you met and heard Christ use the word AEON? You take it on faith from someone who says so...which makes it no less true or credible than the dude who takes it on faith that they literally were days of the 24hr sort.



Why does the industry need graphene to make any of that happen? (the moving beyond more cores and so on?)

Process tech (vacuum tubes, Si xtors, BiCMOS, graphene) is completely decoupled from microarchitecture (Nehalem, Pentium, Cortex A9, Niagara, Power7) is completely decoupled from ISA (x86, ARM, SPARC, POWER, IPF).

Cores and so on is a microarchitecture thing, has nothing to do with process tech like graphene.

Take it to OT and will discuss it there . More dancing around by you.

I have to go with what intel had on Intel site. They talked about tri gate were getting tri gate . Now their talking quatium well . I stick with Intels info since they lead the pack

 

[mosox]

My 2c on the offtopic.

Christianity 101.

Old Testament - Jewish religion.

New Testament - Christian religion.

The OT is in the Christian Bible only for the historical context. It should NOT be taken literally BTW - ST Augustin (De Genesi ad literam).

The only still remaining God Law in the OT are the ten commandments. The rest (including a lot of stoning going on) is the Moses' Law. The Law of Moses has little to do with the Law of God and even contradicts it.

The Moses' Law is the Old Covenant (Mosaic Covenant).

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

The Christianity is the New Covenant

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

and it opposes the old covenant.

Some US neo-Protestant churches (evangelists) advocate a literal interpretation of the Bible. Those are fringe Christian or even fake Christian (all those who advocate the hate/violence).

 

[Nemesis 1]

Take it to OT section of forum and I will school you big time . and use the proper language allowed there to beat you down .

 

[bononos]

.......
I personally don't subscribe to their positions as they do not come across to me as self-consistent nor mentally satisfying. That said, fundamentally speaking, science is about endeavoring to refute or refine the existing understanding of nature and physics...if, as scientists, we were to assume that today's version of scientific understanding is "correct" then as scientists we would have no argument to justify the continued existence of our school of thought for there would be no further need of refining or refuting the science of today.
......

But aren't scientists already supposed to do what you are talking about, what with all the stuff about Popperian falsification?

You mentioned global warming, vaccination etc which should have more debate but the problem seems to be how scientific endeavor is being shaped more and more by moneyed interests which brings the question as to whether science can always be trusted. IMO there is enough inherent problems within the science of global warming, vaccination, GMO to allow for heterodox views - why should dissent be left to the tinfoil hatters?. It is a fact that there is alot of inertia and individual bias because of being very personally invested within the many fields above or to take a more closer example - people working in the tech industry seem less interested about dangers of non-ionizing radiation emnating from cellphones/wifi.

 

[Idontcare]

Take it to OT section of forum and I will school you big time . and use the proper language allowed there to beat you down .

Actually, by your own admission of needing to rely on the "language of beatdowns" to make your point, you kinda make a much larger point.

In the lingo of debate mechanics, one big difference between the technical forums and the OT forum is that in the technical forums you must refrain from degrading your side of the debate to that of the lowest levels of the hierarchy of disagreement (responding to tone, ad hominem, and name calling), the least productive/effective forms of disagreement in a debate.

The fact that you feel you need to access these lower levels of disagreement modes by receding to the OT (where DH0 through DH2 are allowed) tells us you likely don't have a well-reasoned and thought out rebuttal to refute our points; otherwise, you would skip straight up to DH5 (Refutation) or DH6 (Refuting the Central Point) and QED the thread.

How To Disagree

Most readers can tell the difference between mere name-calling and a carefully reasoned refutation, but I think it would help to put names on the intermediate stages. So here's an attempt at a disagreement hierarchy:
.
.
.
DH4. Counterargument.

At level 4 we reach the first form of convincing disagreement: counterargument. Forms up to this point can usually be ignored as proving nothing. Counterargument might prove something. The problem is, it's hard to say exactly what.

 

[Idontcare]

But aren't scientists already supposed to do what you are talking about, what with all the stuff about Popperian falsification?

You mentioned global warming, vaccination etc which should have more debate but the problem seems to be how scientific endeavor is being shaped more and more by moneyed interests which brings the question as to whether science can always be trusted. IMO there is enough inherent problems within the science of global warming, vaccination, GMO to allow for heterodox views - why should dissent be left to the tinfoil hatters?. It is a fact that there is alot of inertia and individual bias because of being very personally invested within the many fields above or to take a more closer example - people working in the tech industry seem less interested about dangers of non-ionizing radiation emnating from cellphones/wifi.

I think that you are, in essence, pointing out that science on an individual level has been "bought out" because of the conflict of interest between pursuing a salary to pay a monthly mortgage and that of science possibly generating an inconvenient truth or two.

We certainly see no shortage of examples of this, look at the labyrinth of red tape the FDA has been forced to institute surrounding the field of medicine and clinical trials because of the fact that they can count on companies (and supposedly independent academic researchers) to lie-by-omission in reporting their results if given the legal (but no question about the moral/ethical) grounds to commit such lies of omission.

The old model was that the pursuit of science was independent of corporate conflicts of interest because the majority of scientific discovery was privately funded (monarchy and philanthropist established institutions) or products of an academic environment (universities and so on).

This model has changed in the post-WWII era with corporations R&D investments vastly outstripping that of the government, university, and private-sector investments.

Not that this model is broken or in need of change, evidently the model reflects our priorities as a species and as a culture. It is what it is because we enabled it, allow for it, and support it to be so. We get the society we deserve.

(it is ironic you mention cellphones and EMF...the irony was not lost on me when I worked on cellphones at TI that I was a willing hand in the creation of a market that was resulting in billions of people sticking these things next to their heads, I had concerns of the EMF effects on children and do to this day, but it was not even a discussion held in whispers at TI, we neither refuted it nor studied it, it was simply one of those discussions that was voluntarily never broached nor discussed...not that it was taboo or something, we all just wanted our paycheck and we knew we were one key medical study away from having the entire program at TI being torpedoed by lawsuits, personal motivation trumped scientific curiosity and concern at the time, we ALL wanted to be willfully ignorant of the danger until someone else proved otherwise)