Anandtech confirms Bulldozer delay (and problems...)

[bamtan2]

Bulldozer's launch got pushed back by about a month

http://twitter.com/anandshimpi/status/74973660389183488

AMD originally wanted to launch Bulldozer at Computex but performance issues with its B0 and B1 stepping chips pushed back the launch. Now we're looking at a late July launch with B2 silicon, but performance today is a big unknown. Apparently the performance of B1 stepping silicon doesn't look too good.

Functionally the 9-series chipset is no different from the 8-series that it replaces; it'll simply be used on AM3+ boards exclusively.

http://www.anandtech.com/show/4365/c...8-motherboards

[xVeinx]

The way that this and other articles are written, it makes it seem as though some of these decisions are made last minute. For a component that has to spend more than a year in rigorous QC (specifically for server testing at least), it would seem to me that this should be old news by now. Granted, I have very limited knowledge of time to market with CPUs; I'm merely making an observation...

[-Boris-]

Sounds lika a Phenom launch.

[blindbox]

Awwww how disappointing. AMD better not delay their Llano notebooks as well.

[LexDiamonds]

If Bulldozer is targeting Sandy Bridge for overall performance, AMD has their work cut out for them...

[Swingline39]

This is frustrating...at least I'm not in desperate need of a new rig.

[CedricFP]

Wow. Not the best of developments for ol' AMD.

[TheBreezyBB]

This is bad.

[highoctane]

The way that this and other articles are written, it makes it seem as though some of these decisions are made last minute. For a component that has to spend more than a year in rigorous QC (specifically for server testing at least), it would seem to me that this should be old news by now. Granted, I have very limited knowledge of time to market with CPUs; I'm merely making an observation...

Well pushing your product launch date out a month at the last moment and not being able to launch at computex means something in the scheduling is getting fubar'd at the last minute considering how long things have been confirmed/planned.

We've heard this scenario with Phenom I, ATI 2900 & Nvidia's Fermi to name a few times where respins where needed/attempted at the last minute to attempt to fix issues which delayed product launches.

Sales & marketing is all set to go, engineering says we've got it under control and its going to work fine, last minute engineering calls management, things didn't work out the way they should have and we're going to miss the planned launch day, we need another respin to fix the issue, sheet hits the fan.

That's sort of my observation...

[Tenknics]

Well pushing your product launch date out a month at the last moment and not being able to launch at computex means something in the scheduling is getting fubar'd at the last minute considering how long things have been confirmed/planned.

We've heard this scenario with Phenom I, ATI 2900 & Nvidia's Fermi to name a few times where respins where needed/attempted at the last minute to attempt to fix issues which delayed product launches.

Sales & marketing is all set to go, engineering says we've got it under control and its going to work fine, last minute engineering calls management, things didn't work out the way they should have and we're going to miss the planned launch day, we need another respin to fix the issue, sheet hits the fan.

That's sort of my observation...

[Oliverda]

AMD originally wanted to launch Bulldozer at Computex

Since when? I've never seen the Computex as a launch event in any leaked presentation.

[kromosto]

i am waiting bd to upgrade but i think i will cancel this upgrade to next year at least i7 920 is not that bad.

[generics_user]

sounds really disappointing

[wez]

Hmm, we'll see

[iddqd]

delaying the launch by 1 month is pretty standard.

[Hornet331]

Wow, no "Anandtech paid intel shill herp derp" comments... I am suprised...

Well a month later really doesn't matter imho, the only thing that makes me a bit concerned are the "performance problems"

highoctane makes a good point... this sounds way to simmilar to phenom 1... Even without the TLB bug the first phenom itteration was only meh compared to the other cpus on the market...

[kuroikenshi]

Since when? I've never seen the Computex as a launch event in any leaked presentation.

I also tend to agree with this, I was never banking on seeing anything beyond llano platform at Computex

[Oliverda]

I also tend to agree with this, I was never banking on seeing anything beyond llano platform at Computex

I can see a chance that it was only in Anand's fertile imagination.

[kuroikenshi]

fertile or futile?

[Frontl1ne]

I think that's where the rumours about computex launch for BD came from.

[JumpingJack]

I think that's where the rumours about computex launch for BD came from.

Several reasons why people assumed computex. AMD (and Intel for that matter) typically align big product launches at large venue conventions to maximize press coverage. Then there was the very early slide deck that showed IP for BD in April leading some to speculate a late April or early May launch but lined up best with a computex timeframe. Then there is the official AMD positioning of early summer for desktop and late summer for server. Even this is sort of a computex or E3 alignment.

Launching end of July though is a mid summer not early summer timeframe so it looks like the rumors coming from the board makers in recent days is turning out to be true.

[FlanK3r]

Think, will see some oficially info from AMD side at Computex time. But think, it is no so bad as TLB bug at first Phenoms "Agena"

[kuroikenshi]

well here is the thing, is Cray already has them coming for their supercomputer, then the product must not bear any delays

[flyck]

well here is the thing, is Cray already has them coming for their supercomputer, then the product must not bear any delays

That product is not delayed, the desktop product is rumoured to be delayed (although we didn't have a launch time on which it would be released). Supposably they have problems scaling BD to higher frequencies. Which is less of a problem in the server space where clockspeeds are low but core counts are high.

[kuroikenshi]

I stand corrected, but not at attention