backtop


Print 22 comment(s) - last by Desslok.. on Feb 15 at 1:59 PM

Intel's roadmap reveals more about the Viiv platform, and Conroe's VRM problem

Intel quietly released a list of Viiv specifications to its manufacturing partners last week.  Intel's Viiv platform is an attempt to unify PC components into a concise, easily supported platform for media center PCs.  Unfortunately, the project is already starting to show some missteps -- the Intel Conroe CPU, the workhorse of the Intel Viiv 1.5 and 1.6 platform, are not compatible with Viiv 1.0. 

Documents from Intel mentioned that the next generation Intel desktop processor, Conroe, will require a platform refresh on existing chipsets.  The upcoming Intel G965 and P965 chipsets will support Conroe, but any other Intel chipsets from 975X on down will require a modified VRM and BIOS updates.  The VRM, or voltage regulator module, is the component on the motherboard that adjusts the voltages to the CPU.  Even if today's chipsets are physically compatible with Conroe, no motherboard today will support the CPU if manufacturers followed the VRM guidelines set forth by Intel when those chipsets were designed.

The major software change to Viiv will be the operating system.  Viiv 1.5 will feature Microsoft Windows XP MCE, while 1.6 will feature Microsoft Vista Premium. Windows Vista is due out later this year after the Intel Conroe processor launch.  A footnote declares that the 1.5 Viiv software package will be available June or July of 2006.

Another new addition to the Viiv platform will be the addition of 802.11 a, b and g WiFi chipsets.  The specification details that the 3945ABG or the 3965ABG WiFi chipsets may be added to a Viiv device and still receive support from the platform software.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Confusion
By segagenesis on 2/14/2006 4:00:52 PM , Rating: 2
Is it just me or is Intel starting to get Alphabet Soup syndrome with all thier product line ups? Ever since they dropped the Pentium name I can hardly keep track of which processor or chipset is going to be what these days.




RE: Confusion
By Desslok on 2/14/2006 4:25:35 PM , Rating: 2
Intel has NOT dropped the Pentium name.


RE: Confusion
By Knish on 2/14/2006 4:28:59 PM , Rating: 1
you mean just because a "tabloid" says it is true doesn't mean it really is?


RE: Confusion
By shadowzz on 2/14/2006 4:31:01 PM , Rating: 1
RE: Confusion
By Desslok on 2/15/2006 1:59:30 PM , Rating: 2
Look at the above chart the Pentium name is still there.


VRM Problem
By Questar on 2/14/2006 4:05:29 PM , Rating: 2
Can you explain what the VRM problem is that is mentioned in the headline?




RE: VRM Problem
By kattanna on 2/14/2006 4:15:59 PM , Rating: 2
the new chip requires a new voltage that older boards cannot support..therefore they are useless and you need new stuff



RE: VRM Problem
By huges84 on 2/14/2006 4:18:19 PM , Rating: 2
VRM = voltage regualtion module

From the article it seems that the VRM, which is a physical component of the motherboard, on Viiv 1.0 boards will not be able to support Conroe chips if the motherboard manufacturer produced its board strictly according to Intel's specifications. This would be because the VRM would not be able to supply the proper voltage to the Conroe CPU.

Though the article seems to imply that if the manufacturer used a VRM with different specs, then the board may be able to handle Viiv 1.0 and Conroe.


Is this a surprise?
By TomZ on 2/14/2006 4:49:00 PM , Rating: 2
It was widely known late last year that 975X would be needed to support these next-generation processors. Why is anyone surprised by this "news"?!?




RE: Is this a surprise?
By KristopherKubicki (blog) on 2/14/2006 4:51:08 PM , Rating: 2
Maybe I didn't clarify enough. Existing 975 motherboards will not work either. You will need 975 "with Conroe support"

Kristopher


RE: Is this a surprise?
By JackPack on 2/14/2006 6:58:26 PM , Rating: 2
Well, Intel has stated from day-one on their roadmaps that 975X would work with Conroe only with a "platform refresh."


By kilkennycat on 2/14/2006 9:47:04 PM , Rating: 2
The Conroe VRM issue has been known for a few weeks -- I came across it in a PC newsgroup. However, I am not at all surprised. Intel today cares more about maximizing their profits on new-product shipment from Dell and other PC manufacturers and cares little about supporting existing owners of Intel systems. The CPU provides the core lever for the chip-set and motherboard (of which, not surprisingly, Intel also make a huge quantity). So the more new motherboards that have to go with new CPUs is all the better for Intel's profit line. Why do you think that Intel recently invested $350million in upgrading their chip-production facility - not to have it sit idle.... The VRM issue follows a now well-known element of Intel strategy-planning -- with each new CPU family, make sure that a completely new motherboard is required, while soothing customers with marketing spin on all the wonderful new features. When are customers going to wise up enough to strip the Emperor naked ??

Here are a couple of the most glaring recent examples of Intel's version of planned-obsolescence :-

775LGA with 915 & 925 chipset -- first LGA 775 motherboards, but not dual-core compatible, chipsets discontinued after 6 months. Considering that CPU designs typically take well over a year to reach shipment, any reason whatsoever why the 915 & 925 chip-sets were not dual-core compatible ? All of the issues were obvious at the architecture stage. Presumably the CPU architects and the chip-architects at Intel do communicate ?

Yonah has exactly the same pin-count as the Pentium-M, but deliberately designed not to retrofit into any version of the Pentium-M motherboard.

And now we have the VRM game. Intel has known about this for a LONG time. The VRM specs are fundamentally tied to processor architecture coupled with process specs. The exact VRM requirements are captured in process-characterization and early simulation, long before any CPU chips are built. A VRM design capable of handling both the existing 775LGA P4-derivatives and Conroe could probably have been installed from day 1 on all 775-pin LGA boards and most certainly when the current 945/955/975 dual-core compatible 775LGA boards first became available.

Intel is no longer customer-centric, if they ever were in the last 5 years. Sales and short-term profits are now the only motive for all of their current product decisions in their PC-related business. The AMD lawsuit exposes some of the underbelly of the beast. Sad because Intel employs many talented and innovative people.

More power to AMD, nVidia and others in their battles with the arrogant giant. At least AMD has strong TECHNICAL reasons for shifting to the M2 socket (quad-core with DDR2 support etc, etc) and has endeavored to stretch the 939-pin backward-compatibility as much as possible. AMD still has not totally shut the door on further 939-pin versions of the desktop family beyond the FX-60.




By JackPack on 2/14/2006 11:28:46 PM , Rating: 2
The last VRD spec will be over 2 years old by the time Conroe is out. The new VRD spec calls for more steps in between and lower VID values to reflect Intel's NGA philosophy.

Sure, Intel could have called for an all encompassing VRD spec that lasts for 5 years. But it would be stupid to enforce such a spec since the vast majority of users don't upgrade the processor. Those that do, tend to replace the motherboard along with the processor. As technology progresses, it becomes less expensive to manufacture the same VRM, which means a long-term VRD spec incurs high costs for everyone when only a few benefit from it.


By DigitalFreak on 2/15/2006 9:09:33 AM , Rating: 2
Doesn't matter if it's the VRM or something else, Intel always finds a way to require a new systemboard with each processor refresh.


Nothing new
By Sunbird on 2/14/2006 5:49:54 PM , Rating: 2
Well, haven't Intel required a new chipset with whatever new processor they have come out with since after Prescott?

It might be news, but it ain't surprising. AMD FTW!!!




RE: Nothing new
By DigitalFreak on 2/15/2006 9:07:09 AM , Rating: 2
Yep. It's their way of bringing in more $$$ for their chipset business.


Typical huh
By Plasmoid on 2/14/2006 9:12:23 PM , Rating: 2
AMD stuff Dual Core Opterons, the latest FX chips, the lowest next to Semperon chips and everything in between thats come out through the years into basically any socket 939 board with and updated bios support.

Intel keeps demanding new motherboards, or at least modified ones, every single core revision....

Course it wouldent suprise me if those NForce 4 X16's will be able to handle those fine... (course no viiv... yet another reason to ditch it) and maybe the Intel boards will capable of running the cpu's a little overvolted.

Well i presume they will be capable of... the real decider will be if Intel is brave enough to put out a bios that would allow instant warranty voiding to take place.




RE: Typical huh
By Questar on 2/15/2006 1:32:57 PM , Rating: 2
quote:
AMD stuff Dual Core Opterons, the latest FX chips, the lowest next to Semperon chips and everything in between thats come out through the years into basically any socket 939 board with and updated bios support.


About half of A64 owners were unable to upgrade to dual core becuase their motherboard vendors never issued bios updates for various reasons.


.
By hans007 on 2/15/2006 9:04:09 AM , Rating: 2
i know a lot of you think amd is just way better off witht his, but really they are just abotu as bad.

socket am2 cant even use the heatsinks from the older one even though they probably could make it work.

socket 754 and socket 939 do not work together.

and socket 940 chips cant work in the other 2 sockets either.

not to mention when socket A came out all slot A users were screwed and there was no way to get an adapter. intel had slotkets that would actually work.

not to mention a TON of motheboards did not work with tbred chips and could only work with palaminos because of a similar vrm type issue.

yes, i am painting amd bad here, but i think a lot of you amd fan boys are well a little overzealous to bash intel at every single step of the way.




RE: .
By Griswold on 2/15/2006 10:31:55 AM , Rating: 2
quote:
socket am2 cant even use the heatsinks from the older one even though they probably could make it work.


They actually improved the design from what I saw. Made it more stable. Whats wrong with that? Besides that, if I was to "upgrade" from my current X2 system to a AM2 box, the HSF would be the last thing i'd move over to the new box - I would want that old system to remain operational, as I have to buy a new mobo, CPU and RAM anyway. The point here is, you have to buy a new CPU and Mobo due to DDR2 support, not because you need a new HSF.

quote:
socket 754 and socket 939 do not work together.


I hope you know why that is.. hint: dual channel memory.

quote:
and socket 940 chips cant work in the other 2 sockets either.


You do know that the current 940 socket CPU's require registered memory and therefore are not really want you want on the desktop in most cases? I could agree with you on the initial S940 A64's for the desktop. I didnt understand why AMD made that move, however, it was only a shortlived accident.

I wont comment on the yapping after this, as that is years old stuff. However, we're talking about Intel changing sockets and chipsets like others change their underwear over the last 6 to 12 months and another 6 months in the future.


C'mon
By nordicpc on 2/15/2006 9:49:03 AM , Rating: 3
Honestly guys,
Conroe is such a huge departure from the P4 how can you be upset by this? We're going to an entirely new architecture, miles apart from both the Netburst and Pentium-M stuff we've seen in the past 5 years. I'm actually really impressed that Intel is even going to make the new boards backwards compatable with the P4. That's gotta be an engineering nightmare.

I personally give Intel some props for keeping the P4 alive for a little longer, even though it's a pretty raunchy architecture and has been since it's birth ~6 years ago. Atleast you guys with 775 P4's will be able to get a motherboard, and then get a processor. I've gotta buy a motherboard, processor, and memory next time I upgrade.




What the hell?
By Runiteshark on 2/14/2006 11:34:23 PM , Rating: 2
What the hell? They have had atleast 6 new motherboard reqs from the last time I messed with an Intel based processor.

Hot damn, this feels like the old AMD days when there were so many sockets and slots.. And Intels, cyrixs and all that other good stuff. I don't even know (right now) what processor goes with what chipset.

Hell, I need to get a little crib sheet to tell me what goes with what, because this crap is getting so old.




"If you can find a PS3 anywhere in North America that's been on shelves for more than five minutes, I'll give you 1,200 bucks for it." -- SCEA President Jack Tretton











botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki