Bulldozer failed to beat out Intel.

[Doylez]

.... the i5-760 beat it? At 2.8GHz? What?!

I think the hilarious thing is how close the old Phenom II X6/X4 are. A Phenom II X4 9XX is also less than half the price, so I really see no reason for anyone to buy a Bulldozer for gaming.

[Xuvial]

Bulldozer was supposed to be (and is) the competition for Sandybridge.

But what generation of Sandy Bridge? I kept reading constant marketing stuff of AMD claiming to be anywhere from 30-50% faster than "Sandy Bridge" processors, but I can't help but think that 9 months after the SECOND generation of SB chips they were still comparing Bulldozer's performance to the FIRST generation chips. It makes good sense too, looking at the final results. What say you?

[DeltrusDisc]

But what generation of Sandy Bridge? I kept reading constant marketing stuff of AMD claiming to be anywhere from 30-50% faster than "Sandy Bridge" processors, but I can't help but think that 9 months after the SECOND generation of SB chips they were still comparing Bulldozer's performance to the FIRST generation chips. It makes good sense too, looking at the final results. What say you?

I'm saying second gen Intel Core, which is the first Sandybridge I know of, the i5-2500K/i7-2600K. That's what these were meant to compete with, so unless you're saying that old i5s or i7s are Sandy, then I'm confused. Obviously they aren't going to compare to the i7-3960x and well... yeah.

[Xuvial]

I'm saying second gen Intel Core, which is the first Sandybridge I know of, the i5-2500K/i7-2600K. That's what these were meant to compete with, so unless you're saying that old i5s or i7s are Sandy, then I'm confused. Obviously they aren't going to compare to the i7-3960x and well... yeah.

Indeed, it appears I've confused myself as well -_- thinking the older chips were SB.

Some interesting quotes from an AMD ex-employee

" On paper bulldozer is a lovely chip. Bulldozer was on the drawing board (people were even working on it) even back when I was there. All I can say is that by the time you see silicon for sale, it will be a lot less impressive, both in its own terms and when compared to what Intel will be offering. (Because I have no faith AMD knows how to actually design chips anymore). I don’t really want to reveal what I know about Bulldozer from my time at AMD.

What did happen is that management decided there SHOULD BE such cross-engineering ,which meant we had to stop hand-crafting our CPU designs and switch to an SoC design style. This results in giving up a lot of performance, chip area, and efficiency. The reason DEC Alphas were always much faster than anything else is they designed each transistor by hand. Intel and AMD had always done so at least for the critical parts of the chip. That changed before I left – they started to rely on synthesis tools, automatic place and route tools, etc. I had been in charge of our design flow in the years before I left, and I had tested these tools by asking the companies who sold them to design blocks (adders, multipliers, etc.) using their tools. I let them take as long as they wanted. They always came back to me with designs that were 20% bigger, and 20% slower than our hand-crafted designs, and which suffered from electromigration and other problems.

That is now how AMD designs chips. I’m sure it will turn out well for them [/sarcasm]

BTW, you ask how AMD could have competed? Well, for one thing, the could have leveraged K8 and the K8 team’s success and design techniques instead of wasting years of time on a project that eventually got cancelled using people that had never achieved any success. It took Intel years to come out with Nehalem, and AMD could have been so far ahead by that point that they’d have enough money in the bank that they wouldn’t have to accept a low-ball settlement offer in the antitrust suit and they wouldn’t have to sell off their fabs."

source: http://www.insideris.com/amd-spreads...ee-speaks-out/

[DeltrusDisc]

Well that's certainly an interesting read.... puts a twist on stuff. Sigh, I wish I had started studying hardcore computer hardware ages ago. =/

[Xuvial]

Moar interesting shit:





(note: WTF is an i7 2500K? Lets assume they meant i5)

source: http://hardocp.com/article/2011/10/1...mance_review/1

[Asera]

Speaking of the DEC Alpha, has anyone had the privilege of seeing one in operation?

Unofficial news/rumors say that turning off every second core, ie. the Bulldozer's version of HT will increase the processor's speed in single threaded tests by 20-40% making it actually faster than Phenoms and not suck quite as badly for people who already bought very expensive 990FX boards.

Now that's something that really proves the branch predictor is far too small and/or inefficient.

[haxartus]

Unofficial news/rumors say that turning off every second core, ie. the Bulldozer's version of HT will increase the processor's speed in single threaded tests by 20-40% making it actually faster than Phenoms and not suck quite as badly for people who already bought very expensive 990FX boards.

I wonder if they could make the quad core model with 4 modules and only one core per module.

[DeltrusDisc]

I wonder if they could make the quad core model with 4 modules and only one core per module.

They probably feel that would be going backwards and not forwards.... even though it would be the smart thing to do....

[vesseblah]

I wonder if they could make the quad core model with 4 modules and only one core per module.

Fixed the link in my post afterwards, but here's the original testing/benchmarking with disabled cores -> http://www.xtremesystems.org/forums/...-Threaded-Perf.

[haxartus]

Fixed the link in my post afterwards, but here's the original testing/benchmarking with disabled cores -> http://www.xtremesystems.org/forums/...-Threaded-Perf.

This confirms my theory that the architecture is still mostly unfinished, but AMD had to release it because they couldn't delay the CPU anymore.

[Deleted]

Unofficial news/rumors say that turning off every second core, ie. the Bulldozer's version of HT will increase the processor's speed in single threaded tests by 20-40% making it actually faster than Phenoms and not suck quite as badly for people who already bought very expensive 990FX boards.

Been trying to get someone to do that... neat if it works. Neat if the results are correct.

[DeltrusDisc]

Been trying to get someone to do that... neat if it works. Neat if the results are correct.

I stand by that you shouldn't have to do that to get any better results to make it out-perform its predecessors. They are marketing it as this "KILLER AWESOME ZOMG 8 CORE CPU WOW WORLD'S FIRST!!! SUPER DUPER FAST LOLINTEL IS SO SLOWWWWW" crap.... so it should be used at those standards. =/

[haxartus]

If they make the 4.2Ghz quad core (FX-4170 i think) with 4 modules/4 cores instead of 2 modules/4 cores, it will be a nice entry level gaming chip, especially if we consider a potential 5Ghz overclock.
Edit: Apparently the water cooling that will be sold with the FX-8150 is pretty good.

[Deleted]

(note: WTF is an i7 2500K? Lets assume they meant i5)

We all know what they probably mean but it's kinda hard to take seriously when they keep making that same mistake.

[DarkBlade6]

Moar interesting shit:





(note: WTF is an i7 2500K? Lets assume they meant i5)

source: http://hardocp.com/article/2011/10/1...mance_review/1

GPU bottleneck.

If they make the 4.2Ghz quad core (FX-4170 i think) with 4 modules/4 cores instead of 2 modules/4 cores, it will be a nice entry level gaming chip, especially if we consider a potential 5Ghz overclock.
Edit: Apparently the water cooling that will be sold with the FX-8150 is pretty good.

1 module = 2cores .. Thats how this CPU works, if u tell AMD hey dont make module, just make core well then this will be not BD.

[Deleted]

GPU bottleneck.



1 module = 2cores .. Thats how this CPU works, if u tell AMD hey dont make module, just make core well then this will be not BD.

The GPU is not bottlenecking an essentially worse version of the Phenoms, an 1100T is beating the damn thing in many benches. This is an utter failure on AMD's side of things, I've already canceled my reservation on a chip. I5 2500k here I come.

[Masterec]

Since no thread is complete without a random youtube video:

2.55 cracke me up so bad every time <3
Jeez almost fell off my chair the first time.

[Meleti]

It did have some remarkable performance in other areas though.

Yeah, it's a great processor if your workload mostly consists of using 7 Zip.

[haxartus]

Yeah, it's a great processor if your workload mostly consists of using 7 Zip.

Or... encryption.

There are types of calculations where the Bulldozer crushes everything short of Gulftown. And Gulftown is a worse overclocker.