Paul!

Finally, a forum which shows a lot of potential, i will try to get them all to join, you know which peopl i mean...

Oh, BTW, have a great one!

Patrick

__________________
No question is to dumb if you are prepared for a technical answer...

Hey, I just wanted to say welcome, and I hope you all find this your new home. We will do our best to make this the best possible place to post.

__________________
ActiveTuning - Partner & Director Of Sales & Marketing

Thank you very much Dave!

If this forum will discuss questions like the bandwidth problems of the newer dual P4 boards and such issues, you can bet your butt that i will be a regular poster...

In fact, i will ask anyone who has more knowledge (or different opionions) regarding the bandwidth problem with the dual P4 boards to explain why they did not opt for a 4 channel board instead...

Wingz, PM, are you out there?

Patrick

__________________
No question is to dumb if you are prepared for a technical answer...

Patrick,

I am defintiely not in those leagues quite yet But just wanted to make a quick note for you and others. For this forum to become the "standard" for you hardcore techies, it needs to be done with your help. So tell the people you think belong here to come on over here, and start and add on to discussions. Tell them to bring the people they know, etc, etc. It can only be done with the communities help.

Looking forward to seeing you guys around.

---Dave---

__________________
ActiveTuning - Partner & Director Of Sales & Marketing

I don't know how long it takes for the typical chipset verification (and subsequently, most motherboard design, and verification with the chipset), but could this simply be an economic move?

If there were enough time between the introduction of the P3 w/i820, and Intel had enough time to redesign the i850 to be merely dual channel like the i840, then it might have been simply a political/face-saving view.

People hate rambus as a technology (often due to flamers), and even more hate rambus as a company. If (this is the if that I don't know) there were enough time to redesign the i850 to use only two channels, then they'd be able to do a few things:

A) keep the cost of the system to sane levels - Intel must have known shortly after the i820 that RDRAM might not be the "next big thing" (right away - it still has niche markets for sure, and may expand in the future), and might not become commonplace.

B) They really don't seem to need it. Of course more is better, but perhaps this was an internal move as well? While yes, the Itanium has better SPECfp scores, with increased bandwidth, might the P4 have scored too high? I don't know how the politics work within Intel, but it would be rather embarassing if Intel's "miserly anchient x86" architecture could beat their brand spanking new architecture.

Intel has made some slip-ups, for sure, but the additional costs associated with more RIMM's might detract from Intel's prestige more than they could take. Remember, the P4 is a desktop processor, not a backend server

Besides, ServerWorks is working on a 4-way interleaved DDR solution for 4-way SMP (one chipset). I think that a uniprocessor in such a system could score rather nicely in FP/bandwidth intensive benchmarks.

I understand your reasoning and perhaps you are correct, but i still think it is stupid...

If you are willing to buy two high performing (because of available bandwidth) CPU's from Intel and placing them on an motherboard with lower bandwidth per CPU, it sounds like a sucky deal to me...

Regarding the server works chipset, 4way ddr would equal 256pin solution (or rather it IS a 256 pin solution) which would requrie the same kind of shielding and technology that is used on the PCRES motherboard (unless they have a special solution) so to their tries for the future i can only say, good luck...

The higher bandwidth per available channel bit of RDRAM requires a high level of noice reduction, i know that, you know that, but simply adding PCB layers will solve that, so why not?

Patrick

__________________
No question is to dumb if you are prepared for a technical answer...

Well, obviously i did not read your post properly...

"The P4 is a destop CPU not intended for backhand servers" well, i can only say that i fully agree... except for one thing...

The ultimate bandwidth that comes from the PCRES RBDRAM board might be used with Intel cpu's instead...

But enough about that... what is your take on the hammer?

Patrick

__________________
No question is to dumb if you are prepared for a technical answer...

I dont' know a heck of a lot about the hammer family...then again, few people do, and those that do have NDA's . Sure there's the x86-64 documentation, but that's implementation agnostic.

Some more points about the whole bandwidth thing: sure you can add more layers to the PCB, but can't you do that anyway with SDR/DDR if you added additional channels?

I'm not arguing that SDR/DDR is better, I'm just saying that I don't think Intel wanted to simply add more layers, due to cost issues. While the P4 as a chip is cheap compared to other processor introductions of the past from Intel, it's still not cheap compared to AMD (god bless competition ).

Plus, having to install RIMM's in units of 4 eliminates the ability of the consumer to do much serious upgrading - in order to upgrade, they'd have to buy 4 RIMM's of the same type! Common', though I know the vast majority of corporate users don't do that, some do, and that'd be a major pain/expense. Especially if there were only 4 RIMM slots - you'd have to throw them all out! Otherwise, you could have 8 RIMM slots, but that just takes up a lot of space, increasing the size of the board - would Intel really want to do that?

You could add layers to reduce increased line noise with a multiple channel DDR solution, but the pin count problem is still there...

Maybe a new serial technology like DDR-II which is serial just like RBDRAM would be the answer, but there is not enogh info regarding this yet.



Good point!

Patrick

__________________
No question is to dumb if you are prepared for a technical answer...