That article bodes well for the potential of Bulldozer, however I'm tired of hearing these fairy tales from AMD. I want more performance *now*. If they really can gain 30% in terms of performance through a scheduler and proper programming, they should be working with Microsoft to patch Windows 7 ASAP.

The real weakness of Bulldozer is that it has too many cores and they're weak. For most people, having more than four cores is going to do nothing to change their computing experience. For me, personally, to see Bulldozer doing *worse* in most of the benchmarks I care about compared to my Phenom II X4 does not bode well for the future of the architecture. The power consumption is icing on the cake.

I can really see the future gaming consoles having increased relevance. Having more performance on the PC is pointless aside from gaming, workstation apps, and servers. More cores will benefit those things. The issue is that it requires massive R&D investment on behalf of the software firms, and even with a ton of research, there's only so much they can gain by going with massively multi-threaded code.

FX is faster in gaming than your Ph II X4 and performance continues to scale with clockspeed.


Power consumption is great only for cold locations where you can game and heat your room at the same time
-- i am looking forward to FX QuadFire gaming next month




i think it will settle in at $200. Again, FX is not a high-end CPU; its performance and positioning is midrange - just as the 2900XT was positioned against the GTS, not the GTX.
--AMD conceded the high end - for now

AMD is not "losing" anything. Consider the alternative - throwing them away. They are offsetting their losses by selling them for what they think can get.

Both Bulldozer and 2900XT were high-end products in terms of cost with R&D and manufacturing.
Just because they were epic failures doesn't mean they weren't as expensive to make as high-end parts with proper performance. As such AMD's business model fell to pieces, because there was no way to get the return on their investment back with such low prices.



Yes, Captain Obvious.
The real problem is much bigger than that: They've been doing that since Barcelona failed...
Which resulted in not enough revenue to invest in R&D for Bulldozer (not to mention delays and downwatering of Fusion). So the downward spiral was already set in motion years ago. Bulldozer would have been an opportunity to escape from that, but it is not.
So again, selling CPUs with far too low profit margins, not enough R&D, next generation products not going to be competitive enough etc.

I'm not sure why you don't get this, to be honest. The theories I have aren't that favourable.

Your theories have been unfavorable but in practice, AMD is still here
- and evidently turning a profit as per last quarter

I was talking about the theories I have on why you just don't get it.

And yes, AMD is still here, but it hasn't looked good for years. Selling off GF and getting a billion from Intel are the main reasons why AMD is still around. What will they come up with next?

What is it that you think that i don't "get" ?


Less advanced AI computes that there is only one answer and only one correct way to arrive at it.

The first part of what you said (The real weakness of Bulldozer is that it has too many cores) is irrelevant, there is a 4 core BullDozer as well as the 8 core BullDozers, and it is still gutless.

The last part of your statement is accurate (and they're weak.) although understated. Each "core" in BullDozer is gutless, not even as powerful as a core in the preceding architecture, and on top of that it is forced to live in a module arrangement and share what precious few resourses it does have at its disposal with another gutless core.

And that is why BullDozer sucks.

Well, I think it will settle into the landfill site where all the Atari 2600 E. T. cartidges are buried.

Actually, I'd argue that AMD looks about the same as it has for 32 of its 40 years of existance. In hindsight the 8 years of K7/K8 success were nothing but an anomaly.

Out of stock at Amazon and at NewEgg

http://www.newegg.com/Product/Product.a ... 6819103960
http://www.amazon.com/s/?ie=UTF8&keywor ... 75ig03tb_b

Actually, it's Scali's quote - not mine

Most of the chips are being sold to Cray as opterons for their supercomputer. Only a few trickle through to the consumer as Zambezi, etc. So out of stock proves nothing (other than FanATIcs would purchase sh!t with sugar on it if you stuck an AMD/ATi logo on it somewhere).

Who is to say that the new CEO isn't spending some of his renumeration package buying them all up at to make it look like they are selling, probably got shareholders in on the act too.

Sorry about the misquote.

So if they are selling them - to whoever cares - then it is not the same as the "ET" cartridge flop
- your argument is shot down in flames - by you. Thanks!