Sat 11 Oct 2008

RSS Feed

Edited by Paul Hales

Published by Incisive Media Investments Ltd.

Terms and Conditions of use.

To advertise in Europe e-mail here

To advertise in Asia email here.

To advertise in North America email here.

Join the INQbot Mail List for a weekly guide to our news stories:

Subscribe

Barcelona B3s are just fine

Rumours to the contrary are wrong

THERE ARE A bunch of rumours floating about AMD's B3 stepping parts, saying they are have the same bug as the B2, IE the new step did not fix things. This is categorically not the case, it was, is, and never will be true.

When the rumours started to fly, I called AMD and talked to Pat Moorehead who answered all of my questions. The B3s are not bugged, period. The fix was tested on B2 silicon and verified (Don't ask, the elves who work at the Fraternal Order of Silicon Repair Mythical Creatures Local 17 get antsy if you reveal their secrets).

Basically there is no B4, there won't be a B4, and the next stepping is Cx. The Cx parts are the 45nm Shanghai cores, so the B3 will be the end of the line for 65nm. B3s are going to be sampling in the not so distant future, production in the end of Q1, and retail availability in Q2. That part is a slip.

In summary, B3 is fine. It is a little later than people expect, but what can you do? The fixed chips are on the way, sit tight, it is only a few weeks before this soap opera ends. µ

Comments

Hmmm...

Paul or Charlie?

Tough call, so I'll have to let my AMD fanboyism make me plump for Charlie.

Sure hope he's right. I don't want to have to spend the money for a socket 775 board to go quad core. Already got the AM2.
posted by : hoohoo, 11 January 2008

What about the new bugs found on QX9600 and Harpertown?

Accordings to the german http://www.heise.de/ct/08/02/018/ intel has updated it's CPU bug list twice during December 07. It says that AMDs TLB bug almost sounds harmless compared to the list of new bugs found on intels CPUs. Funnily enough there even is one affecting the TLB :-)
posted by : holger, 11 January 2008

Fixing Si

We do it all of the time. An ION mill ($3,000,000+) is used to cut and re layout metal on the top few layers of metal. If the problem is on the bottom layer or two, your burn a hole through the back side of the die and use the ION mill again. No magic needed, just $$$$$
posted by : Ken, 11 January 2008

amd

i have heard many good things recently about amd's B3 stepping, also heard that this erratum was holding clockspeed's down, bad news for those who bought the black edition phenoms because even tho they have unlocked multiplyer you wont be gettin those to 3ghz stable.
looks like higher speeds will be coming with this new stepping also :), the first steppings of phenom wasnt looking healthy but personaly if amd can get it rite me thinks k10 has HUGE potential lets see what happens in the coming weeks.
posted by : jason, 11 January 2008

who to believe

You guys at Inq can't even agree with each other? :) Anyways, let's hope AMD can get their stuff ready in time so they can compete against Intel.
posted by : az, 11 January 2008

B3 is last stepping??

Are you sure B3 is the last stepping? I thought AMD would push for more steppings to recover frequency lapses. If B3 is end of life for barcelona then Intel will be walking their way to glory. 45 nm core is a whole new project which will have its own set of issues and I imagine those wont be out for quite some time.
I am sure AMD will have more steppings on barcelona.
posted by : meckenna, 11 January 2008

Bug Off Buggers

What they promised was "Tick Tock."
The way that works is simple really:
They're all the Tock,
While'st you get Ticked Off.

I know. Impossibly corny due to excessive kernel size; but don't fault me for finding whiskey in a jar.

posted by : Karlsbad, 11 January 2008

But you have to keep in mind...

The rumors said that AMD would likely have to spin out newer silicon. The rumor suggested it was possible it may not receive a stepping above b4, that it could just be a second spinning of b3. The rumor also suggested a release of this new spin in March (end of Q1)... meaning retail in Q2. Which aligns perfectly with AMD's current statements. I'm not agreeing one way or the other, but one must admit it lines up...
posted by : Comdrpopnfresh, 11 January 2008

Read Between the Lines!

Yes the TLB bug may be fixed and while I'm sure that is technically accurate... doesn't it seem like B3 is behind what it was projected to be.

AMD ran into other issues and shoved them all under the giant TLB bug carpet. Did they mention the clocks of these B3 parts?

"B3's are Fine" = TLB bug fixed? Talk about lowering the expectations! Wasn't this stepping originally going to yield parts at 2.6GHz and beyond? Next time ask some tough questions and stop lobbing the softballs. I know AMD is having some tough times, but this is not a reason to go easy. Why are they releasing 1.8GHz quads for DESKTOP that supposedly 'people are demanding'? Who is demanding a chip that operates only marginally above Intel's IDLE speedstep speed?
posted by : joe, 11 January 2008

Eh...

No dancing in the aisles this time Charlie? Aww...
posted by : Pete, 12 January 2008

Time to stop the FUD then

Good to get a confirmation that the latest stepping is alive and healthy. I hope it will stop the Intel FUD, but I really don't think so. Whatever it takes to stifle the competition. Kick it while it's down, back at ya AMD right? I can't wait until AMD responds, Intel deserves a solid kick to the head rather sooner than later.
posted by : Demoan, 12 January 2008

Or Maybe They're Not Fine

Or, If you want a more objective opinion, go read Ed Stroligo's take on Overclockers

http://www.overclockers.com/tips01277/
posted by : TomH, 12 January 2008

Charlie Cheerleader

Charlie, when are you going to get out of your green and white AMD girlie cheerleader outfit and tell people (read: consumers) the real issues behind the dismal failure called Barcelona? The problem is transistor subthreshold leakage (Isub), indicative of the high thermals associated with design as whole. You know this. With your technical knowledge you’ve known it all along.

The TLB bug doesn’t affect the thermals which are off the wall, by the way, for Barcelona’s progressively higher clocks. Stepping B1, B2, and B3 didn’tt correct it because the overall SOI process is/was a failure from the onset. You know this, too.

No mater how much AMD tries to tweek this thing, it will bleed power as bad as or worse than Prescott, which is why B4 will never happen. It is a wasted effort.

Beauty, (elegant native quad design), is only skin deep, but Ugly (Isub threshhold leakage) goes to the bone. The TLB is a minor issue even if they fixed it; the steppings didn’t correct the real issue, the thermals.

SPARKS
posted by : SPARKS, 13 January 2008

Phenom

Tested Phenom 9500 on ASUS M3A32-MVP Deluxe. Runs like crazy, scales linearly. Will wait for Barcelonas. No Idiot inside.
posted by : Just 2cnts, 14 January 2008
IThound
Search for solutions, reports & analysis

Newsletter signup



 

Top INQ Stories