Conroe shows performance flaw

Rob Updegrove 5 Jul, 2006 at 2:33pm ET Article published in Tech

Tests of Intel’s upcoming Conroe offerings are showing a disturbing fact: onboard RAID configurations aren’t performing up to spec when used in conjunction with test boards fitted with the new CPUs.

It seems to us that Intel’s Core 2 is suffering from a specific overhead when on-board RAID controllers are used. If you’re unfamiliar with current on-board RAID5 controllers, let’s just say that they really look like soft-modems and soft-sound cards, utilising power of the CPU for everyday work.

I only wonder how will a two-meg cache Conroe work on a fully integrated motherboard, with Gigabit Ethernet, software-based sound-card – ever popular AC’97 codec and so on. Probably not as advertised.

Can Intel solve this problem before Conroe is released into the wild? Or is an architecture flawed in some irreparable way?

Source: The Inquirer

Comments

5 Jul 2006 ~ 2:36pm Thrax Ho noez, RAID is broken. How else will I guarantee unrecoverable data?!?!?!1one
5 Jul 2006 ~ 2:38pm GHoosdum I think the key point is the extrapolation the author makes to other onboard (CPU dependent) hardware solutions.
5 Jul 2006 ~ 4:29pm Thrax For the general populace, yeah, that's something of a problem. I'm just glad I enjoy add-in sound cards and PCI WiFi cards.
5 Jul 2006 ~ 4:32pm GHoosdum

Thrax wrote:

For the general populace, yeah, that's something of a problem. I'm just glad I enjoy add-in sound cards and PCI WiFi cards.

Are you implying that you might buy an Intel when Conroe comes out?

(what happened to the :thrax: smilie that looked like the Intel logo?)
5 Jul 2006 ~ 4:53pm Thrax I've implied it several times on other forums. AMD is going to lose this round.. I want the best for my money, just like anyone.
5 Jul 2006 ~ 6:06pm RADA

Thrax wrote:

I've implied it several times on other forums. AMD is going to lose this round.. I want the best for my money, just like anyone.

I agree completely,

Intel, AMD

or

ATi or nVidia

..matters not to me.

Quality and performance matter when I'm buying new equipment.
5 Jul 2006 ~ 6:09pm airbornflght

RADA wrote:

I agree completely,

Intel, AMD

or

ATi or nVidia

..matters not to me.

Quality and performance matter when I'm buy new equipment.

yep. although, Im wondering if amd doesnt have a secret weapon. I'm sure they do, but I wonder who will be carrying the biggest stick this time?
5 Jul 2006 ~ 6:42pm rapture I'm also a bank statement fanboy but this is making me wonder. Hope it doesnt end up being a problem...
5 Jul 2006 ~ 7:07pm TheLostSwede We see this over and over again on every major release. One company spams out things to blame the other and vice versa. Yesterday we had the "patch" from Amd which is just utter nonsense. ONE guy have reported a 130% increase in CSS. Makes me wonder if that guy should't just have gotten a 360 instead. Note that this is Raid 5 and how many are using Raid 5 instead of Raid 0 of us here?
5 Jul 2006 ~ 7:21pm airbornflght not many.
6 Jul 2006 ~ 9:23am lemonlime I really don't think this has anything whatsoever to do with Conroe or the 'Core' architecture. This sounds like a chipset or mainboard flaw of some kind. I agree with Mack.. just fluff at release time to shake up competitors.
6 Jul 2006 ~ 9:46am rapture

This sounds like a chipset or mainboard flaw of some kind.

I think the Inq tend to agree, they were pointing fingers at the Southbridge. If thats the issue they by the time we see the new boards being released it may not even be a problem.

Some delays have been announced by the board manufacturers, wonder if they were in any way related...
6 Jul 2006 ~ 5:14pm Leonardo Apparently Woodcrest/systems are affected also.

US government unit throws Intel out over RAID problems

WoodCrest systems were delivered for an evaluation for one of the US government departments, and it all ended up in tears, with Woodcrest being rejected in the first round of trials. The reason was that there were severe problems when Woodcrest was paired with a 1E RAID field when using IBM ServeRAID controllers. The problems didn't occur just in benchmarking, it was the every-day usage model that produced unexpected errors.
7 Jul 2006 ~ 4:21am Gobbles I run a raid 5 in my server here in my house but I also use a dedicated raid controller. 3ware 7506-4lp 4channel raid controller running raid 5 with 4 250gig drives. Im glad i have it that way as 1 drive failed over the weekend. Popped in another 250 and 30min later raid was rebuilt and rocking away.
7 Jul 2006 ~ 9:24am GHoosdum Isn't the IBM controller that the Woodcrests failed with also a dedicated RAID controller?