Intel 840 Pentium X to cost $1000 - Page 2

The key word here is "WILL", which means who knows when. I wouldn't count on it. For example, just 6 months ago, MS told us that the new WinFS will be out in 2006 with the first LongHorn release. But now they are telling us that it will be out AT THE EARLIEST at the release of the first service pack, and likely will be later than that, which puts it in the timeline of 2008 range. Don't listen to all the "projections" that you hear, which amount to not much more than speculations, even if it comes from the developer.

enough with fanboism ffs! The price shouldn't be surprising as any time a new technology is released to the public it at premium price. Prices will deflate as time passes.

You need to realize is the reason for this kind of initial pricing, is that Intel will not have enough yields initially to supply all of the people who want to buy them at a lower price, such as 600 USD. They need to do this only due to their launch of this processor with very few, if any actual chip on hand, i.e. what is called a "paper launch". Any CPU/chipset/graphics makers have been doing this for years; and Intel does this nearly every time, when coming out with a new line of products.

This is not as innocent as you seem to think, not simply a case of supply and demand, but an effort to cover up what they don't want to tell you: that the yield of the processor is low and should not have been launched at this early date to begin with.

and to answer that question....just as many as use 64bit tech (not counting linux)

You're sounding very much like a run-of-the-mill AMD fanboi here. Most of your claims above are clearly speculation. Care to elaborate and prove me wrong?

At this stage of the game, I tend to believe that the winner will be the player with the strongest platform; not necessarlily the 'fastest' or 'best' CPU technology. There will always be a need for fast CPUs, sure; but the PC industry is changing -- dramatically -- and the days of buying a PC based solely on CPU performance are numbered, at best.

What we are seeing now is a fundamental shift in the way we view computing. The paradigm is shifting from a focus on single-core processors and single working thread applications, to multi-core processors and multiple working thread applications. We're also seeing steady convergence of the consumer electronics industry and the PC industry, and this alone is looking to usher in some interesting technology.

I, for one, am very much looking forward to Intel's virtualization technology becoming a reality. The ability to run multiple, isolated operating system instances on a single CPU has long been available on mainframes, but never before on comodity PC platforms; well, not natively in hardware, at least. Virtualization can be achieved using software such as Xen, but Xen is a virtualization monitor -- a software layer sandwiched between the OS and hardware. Xen requires an OS to be specially ported to the monitor before virtualization is possible.

True virtualization enables a user to run both Windows and Linux concurrently, for example. Both operatings systems are running on the CPU at the same time, but completely independent of each other; even if one OS suffers a critical fault and requires a reboot, the other will continue to run unabated. It also allows the user to switch between the two running operating systems at will, without requiring an OS shutdown and the resultant loss of state that entails. How cool is that!!?

I think perhaps the single most important feature virtualization brings to the table is its ability to allow the use of a single computer to serve the needs of an entire household. One could run one OS instance for home-entertainment needs, another OS instance for computing needs, and perhaps a third for back-end server tasks (i.e. internet/NAT/firewall etc) -- all on the one machine. This, I believe, is Intel's vision, and the motivation behind bringing virtualization to the PC platform. I alluded to this before when I mentioned the convergence of the PC and consumer electronics industries. Looking at it again from the virtualization perspective, it seems clear to me that this is one direction in which the industry is headed.

In regard to who will own who on the multi-core CPU front -- well, let's just wait and see, shall we? It's certainly going to be an interesting race, that's for sure; but I don't think that I, nor you for that matter, can predict the outcome of the race this far in advance. Truth is, the best we can do at this point in time is speculate. So, in future, if you do decide to add your two-cents, do us all the courtesy of backing up your claims, or at least provide us with some indication of intelligent thought behind your ravings.

i might as well just buy a lamborghini murcielago...

I guess you have never heard of AMD Pacifica or Presidia ??
http://www.theinquirer.net/?article=21607
http://www.vr-zone.com/?i=1842
http://www.amdboard.com/pacifica.html
http://www.engadget.com/entry/1234000340035583/
http://techrepublic.com.com/5102-10596-5600552.html
http://www.extremetech.com/article2/...1644513,00.asp
http://www.linuxbusinessweek.com/story/48591.htm?DE=1
http://www.linuxbusinessweek.com/story/48591.htm?DE=1

Now we know how WELL INFORMED you are, let's move onto serious matters.


Intel is certainly racing for the quick fix, simply linking two prescotts on the same die, and slap a packaging over it.

AMD K8 was designed from the ground up to be Dual Cores, and HTT, on die mem controller, and other advanced features such as cache coherence were part of the effort from the beginning to enhance Dual Core implementation (actually caused a delay in the original Opteron's debut).

The kicker is that, the current intel chipsets won't be able to work with dual core CPUs, for Smithfield or Montecito. But AMD CPUs will be compatible with a large number of current S940 and S939 boards, partially thanks for the on die mem controller.

http://www.pcper.com/article.php?aid=122

Not too mention that the dual core Opteron and A64s will have a lot lower TDP than the makeshift dual core Smithfield, which will be a temendous advantage in terms of implementation in Servers and DTR notebooks.

I don't recall saying that AMD didn't have plans for virtualization. That is a presupposition on your part; don't put words in my mouth.

That aside, have you actually *read* the article you linked to from the Inquirer? If not, I suggest you read it. If so, read it again; then come back and discuss this with me. With gems like:

There were some things that came out, the first of which was that there will be a Pacifica spec that comes out 'before the end of the month'. Hopefully, they mean March, but it wasn't specified as such.

and

The next bit was that it isn't totally compatible with VT, or at least that is the strong impression I get. When you ask the question 'is it compatible', and you get the answer 'well, isn't the weather nice today, how long are you in SF for?', it becomes clear that they don't want to say no in public.

AMD certainly don't sound too confident of their offering; at least in the near term.

In any case, like I said before: let's wait and see. It is my opinion that Intel is leading the virtualization charge with Vanderpool; it has been heavily involved with projects such as Xen; it has a distinct vision of where it wants to take the technology; and it has a definitive timeline for when VT is going to be available on the market.

This sits in stark contrast to that of the vague handwaving and claims from AMD of 'ours will be better, when it arrives'.

Well, sorry for the presumption on my part. But I certainly got the impression that you suggested that AMD didn't have a comparable offering.

If you look at the date on the article, you will relize that the date is last sunday, and that AMD will be offering details about it soon. I have certainly read more than just these articles, and can say that AMD is just as confident about their technology as Intel is about theirs. Just because the details havn't been released, doesn't mean that a tech is inferior.

I agree, let's wait and see. But the presumption is yours here, knowing nothing about Pacifica or Presidia, making the bold assumption that Intel's technology is superior
Knowing only one side of the story certainly help you to make a fair and balanced case Any way, the timeline is there for Pacifica as well, we know that the first processors supporting such technology are the Windsor and Orlean dual cores (2MB and 1MB L2, respectively), and both will be out in Q1 of 06, probably by late February.

Please open your eyes wide once, and realize that there is more than one chip maker out there. That will broaden your horizons and make you breathe easier as well. As you can see in my signiture, I can happily buy Intel products as well, if they have what I need; and I am certainly no fanboy of any company. But the most important thing is to keep a good competition between intel and AMD, between Nvidia and ATI, between MS and ?? :sigh:, anywayys, competition benefits us, the consumers; and we buy all means, should encourage competition and give every company a fair shake. If you would rather have a large monopoly, than I think I could say that you are in select company on this site or anywhere else, for all who follow this business.

Nonsense - it is simply R&D investment recovery. You see it in every other industry why so surprised to see it in tech. Look at clothing; Sping fashions at a premium close to the end of Winter then they are on clearance in a few weeks. Same thing will happen with these new fancy chips - same as always. New = more expensive for a while until the novelty wears off or until the next new thing.

R&D investment recovery??
Now that certainly explains the the price difference of $500 between the 840 and 840XE, running at exactly the same frequency, both of which required the investment in the exact same R&D, and the only difference between the two are the cache and 1066FSB, both are residual from the last generation of prescotts. And I'm sure the makeshift Smithfield of slapping two Prescotts together with an arbitration unit in between is certainly R&D intensive.

The raising of price between the two chips with select few architectural differences is deliberate, and certainly cannot only be attributed to market forces or R&D or anything comparable.

I agree, no, it doesn't. But at the same time, the only means we have of evaluating a technology, or its potential, lie in the details of the technology that have been disclosed. Intel have been making noise about VT for quite some time; I remember first hearing about it way back when HT was the big 'new-thing'. When Xen made it's existance known to the mainstream public late last year, Intel was already there, and IDF this year was very much a public showcase for muti-core platforms and VT.

To be honest, I just haven't seen enough of AMDs offering to be convinced of it's merits. Once again: wait and see.

Yeah, I'm aware of this. I've owned both AMD and Intel systems in the past, and with little to no preference for one over the other. The reason I got involved in this debate was due to the post I quoted initially; white noise posts like that just rub me the wrong way... I should probably show greater restraint.

Yeah, I didn't mean any of it against you personally, I totally agree that we should wait and see; and choose the better of the two products when it comes out, at least what's better for each of us. I just wanted to point out that both companies have comparable technologies, and point that out to those who are reading the thread, and who may not follow the industry as closely as we do.

Legit Reviews has the first review I've seen of the Intel Dual Core: Pentium EE 840 up.

They're popping up all over the place now. Check this thread:

http://notebookforums.com/showthread.php?t=73788