Could AMD be about to do a P4-era Intel and ditch Bulldozer altogether? http://www.bit-tech.net/news/hardware/2012/08/02/amd-jim-keller/1
AMD need to throw away synthetic designing and go back to hand crafting , the `man in the loop` who can see what will work and what wont, where a CAD programme cant tell the difference.
This pretty much. But will they ditch bulldozer? Nope, not if they follow their own roadmap. They've got a few more iterations of it. And truth be told, it isn't nearly as bad as other say it is. Outside of the relatively bad misprediction penalty and the abnormal voltage leakage spike (perhaps the Gate first issues) it's not a bad chip. Just a sub-par one in comparison to intel's offerings.
Intel need competition fast in the enthusiast market segment and AMD wont manage that if they stick their current course. A new direction is needed and the sooner the better. That said, AMD's APU's are still very good.
"AMD has rehired the man largely responsible for the design of the Athlon 64 and Opteron 64 processors - the first chips to feature a native x86-64 architecture - suggesting that the company is rethinking the whole Bulldozer approach to chip design." Nothing of the sort. It suggests they want the man who built Apple it's SoCs to build their own new SoCs.
What makes you so sure that he is hired for a new desktop cpu? And not for a mobile one? Seems more likely. ;-)
I'm sticking with the prediction I made in Micro Mart's Bulldozer feature: just as Intel did with Netburst and the P4s, AMD will eventually discard Bulldozer (on the desktop, at least - servers are another matter) for an upgraded version of the K8 architecture. It took Intel six years to make that decision; it just remains to be seen how long it takes AMD to come to the same conclusion.
seems stupid to have had a policy to let good engineers be poached. Surely the best engineers are worth 1,000 times their salary to amd, why didnt they just double / treble his salary at the time so he would stay. Do this with the other engineers too and push for the next product to keep moving instead of stalling and losing billions. *I only run a small business in order to make money. The above comment assumes amd have the same goal.
Even though it would be awesome to see AMD trying to keep up its desktop business. Its a bit contrary to what they announced lately. Like not keeping up the race with intel, planned revisions of bulldozer (piledriver being the first) and the announcement to expand into other platforms (mobile). Again, i hope you are right, though.
i think that amd will en up between a k8 and bulldozer chip. like a k8 whit CMT. if amds had marketed there 8 core for a 4 core whit mulittreading then it would have been seen al a much better chip. the gpu will become the FPU in the future and the modular design is the idea behind bulldozer not the high clocks. also the flex plex fpu is a nice idea they will probably make it wider like 512bit and split-able to 64 or 128 bit. (maybe they break the module apart so you could use all the fpu for one core or share it between all of them. i havn't had computer architectures yet so i don't know if that is possible). bulldozer is also a higher clocked design. maybe they hoped that it would be possible whit the new processes. but it isn't completely like netburst . in the netburst architecture the alus where double pumped so if the P4 ran at 3 ghz the alu ran at 6 ghz. that's why it used a lot of power.
It's entirely possible that their short-term plan is unchanged, it'll take a while before new ideas filter into bulldozer, so Piledriver likely isn't going to change much, off what's already been announced, consecutive versions, however, may be much better. It does mean, however, that two years down the line, or more, AMD might finally get their arse into gear and start producing chips worth half a damn.
If/when I upgrade in the next 6 months it'll be to Intel (again), but I hope next time after that there'll be a credible AMD alternative, if only to drive competition and push prices down for everyone.
What I'm wondering is how the whole node process will go down. As we know it, 7nm is about the limit. If AMD can hold on long enough, I have an inkling that once the process node wars end, (and things stop shrinking so quickly) then there will be some more interesting instruction set and architecture revolutions. Moore's law may be gone by then, but that's when things will get very interesting.