Well moving to a smaller die size is the same with the entire product execution. Intel has been on 32nm for what 2 years or so now? I believe the original Bulldozer has always been designed for 32nm, I might be wrong on that.
If this came out two years ago then it would have been on 45nm.
I don't see how making it hotter and slower would have made it any more competitive "in its day".
How did that work out for Phenom? What Phenom needed, and eventually got, was a shrink from 65nm to 45nm. It did not need to be released 2yrs earlier on 90nm.
Now if what you meant to say is that the BD we are going to see released this summer (i.e. the 32nm one) should have come out two years ago then yes, I totally agree. A 32nm-based BD in 2009 would have rocked.
Oops I was wrong, its original design in 2008 was for 45nm.
Of what has been said about it, the bulldozer design that was targeting 45nm release was nothing at all (in terms of microarchitecture) like the bulldozer they eventually designed with 32nm in mind.
They kept the project name and initial design team, but scrapped the initial bulldozer concept and started with a blank slate.
I could be wrong about that, I've never verified this to be true from anyone within AMD that would be in the know. But this is the current school of thought based on public comments made by various AMD employees over the years.
You are probably right and that's the failure of this company. Here we have a 4 year old cpu concept that has failed to make it to the market regardless of the changes. That's just mind boggling.
Well...Dirk was let go in rather hasty order right around the timeframe of those initial bulldozer results coming in hand internally.
We've all pretty much bought into the story that his dismissal had to do with his failing to have a strategic plan in the ultramobile markets, but maybe it has more to do with bulldozer and we are just now starting to see the signs of this reality?
Is that what Intel did with Prescott? Or did they bring it to market despite all its lackings?
I can't tell if you are being facetious or if you genuinely are unawares of the history of Prescott...
http://www.anandtech.com/show/1230
90nm Prescott was slower (less IPC and did not clock higher), hotter, and higher power-consuming than its older 130nm Northwood sibling.
And yet Intel brought it to market anyways, knowing all this to be true well in advance of launch day and they continued to sell them as their leading product until the launch of Core 2 Duo >2yrs later.
Depends. If architecture is ready, it takes at least 2 years. It not, it would be 3 years minimum. And I don't think either Intel or AMD can achieve that.Probably.
How long does it take a design team to design a cpu to tape out?
Depends. If you start from scratch to define architecture, it takes 4 years. If architecture is ready, it should take 2-3 years.Probably.
How long does it take a design team to design a cpu to tape out?
Thank you - some much needed common sense. I agree 100%. If BD wouldn't put AMD in the mid range market, they might as well scrap the whole thing. There's absolutely no money to be made off introducing a new cpu design in the low end market. Low end is and always will be flooded with yesteryears designs and down-binned or crippled mid end parts.Yeah, but Intel knew they could still just pay companies to use it.
The problem is, if Bulldozer was such a complete failure, it would make sense to scrap it and just go with 32nm Phenom, since they already put in the effort to do that. I would guess that 90nm Prescott cost less money to manufacture than 130nm Northwood, which would make the decision to go with it a lot easier.
Also, I thought that it clocked higher (not by a whole lot, but they went from like 3.4 to 3.8GHz)?
Thank you - some much needed common sense. I agree 100%. If BD wouldn't put AMD in the mid range market, they might as well scrap the whole thing. There's absolutely no money to be made off introducing a new cpu design in the low end market. Low end is and always will be flooded with yesteryears designs and down-binned or crippled mid end parts.
That's not the same as to say BD will wipe the floor with Intels top offerings. I personally hope it will, but that shouldn't be the criteria for success.
Thank you - some much needed common sense. I agree 100%. If BD wouldn't put AMD in the mid range market, they might as well scrap the whole thing. There's absolutely no money to be made off introducing a new cpu design in the low end market. Low end is and always will be flooded with yesteryears designs and down-binned or crippled mid end parts.
That's not the same as to say BD will wipe the floor with Intels top offerings. I personally hope it will, but that shouldn't be the criteria for success.
Beta hardware is beta and not gamma (production) for a reason.
If everything was sunny and rosy with bulldozer (including the platform) then it would be launched already.
There's a reason it ain't. And there's a reason a new stepping is involved...they don't spend money doing respins for the heck of it.
I think AMD has a proven track record. Going up against the 800-pound gorilla called Intel and staying in business for so many years is quite an accomplishment in my book. Intel has the luxury of being able to drop the ball and still make a ton of money (P6 vs K7) - AMD has not.You are assuming that AMD management is totally rational. History has shown that some of the brightest people in the way can act very irrationally sometimes.
Personally, I can't help but think that Llano is a sort of backup plan should Bulldozer suck. Why didn't they use Bulldozer cores in that to start off with? Probably because Bulldozer's concept is unproven.
Bulldozer is gonna be released in September. Guess that means I'll be buying an unlocked Core i5 in August. This thing has been hyped like there's no tomorrow and yet it keeps getting delayed... I wonder why.
The two bolded parts are not useful in this demonstration. They would need to be the same clock speed as the previous two (2.8GHz) to be feasible for what you're trying to show.Disregarding the fact that the site also tested with Cinebench R11.5, what's wrong with Super Pi as a measure of single threaded performance?
My old Pentium 4 @ 2.8 GHz did 1M in about 45 seconds
My Athlon 64 3500+ @ 2.8 GHz did 1M in about 32 seconds
My Core 2 Duo T8300 @ 2.4 GHz did 1M in about 21.5 seconds
My 2500K @ 5 GHz does 1M in around 7.5 seconds.
That shows a pretty good correlation between better single thread performance and super pi 1M times.
Back on topic though, I don't think SuperPi is great for testing new processor architectures. Nor is it good for testing the full performance a processor offers... Having said that, I hope that this ES is a flaky one cause it looks weak to me.
HardOCP claim to be organizing a event on July 16th in Texas it seems & bulldozer will be demoed & will be available for hands on experience & some tourneys & free giveaway e.t.c
http://hardforum.com/showthread.php?p=1037335089#post1037335089
Any news on what speed the BD will be running? Is it going to be one of those low speed ES CPUs or something closer to what can be expected from an actually released version?
I would think & hope AMD is smart enough & for their own good don't give fans & like wise interested people attending this event a hands down experience of Bulldozer on some broken Bulldozer ES , that will be a epic disaster lol.
Bulldozer is gonna be released in September. Guess that means I'll be buying an unlocked Core i5 in August. This thing has been hyped like there's no tomorrow and yet it keeps getting delayed... I wonder why.
I certainly hope that AMD's marketing dept. isn't being taken over by the panic of putting "something" out there to quiet the naysayers and in doing so plunk a 1.5 GHz crippled BD at this event just to show that "something" is there. AMD's reputation at least for high end silicon might not recover after that.