They destroyed their forum on the BD hype. Completely different thing.
BD hype?
Yes. AMD shock troop hyped Bulldozer to death there. To them, Bulldozer was going to conquer the world. The forum suffered with that, lots of good people banned by the designs of the troop. After Bulldozer flopped, the credibility of the shock troop suffered and the guys who were banned and a lot of them very good people, didn't come back. The forum is now what it is.
In that regard I suppose Intel should be thankful those millions of investors don't think as poorly of Intel's future prospects as they do AMD's.
Haswell has always been about GPU and low power. I think Desktop Haswell might suck balls tbh. i have a feeling this is bad news for desktop.
I really want to see some benches and not all the conjecture that intel is spouting currently.
For the desktop, I have a feeling you are going to be right about that.
IB was the first processor I have ever bought that failed to OC higher than my processor from the previous node.
And all the talk about Haswell has been directly focused where you'd expect it to be - the power-efficiency in the low power end of the spectrum.
So what are we going to see on the high end of the power spectrum? Probably more of what we've already seen - clocks that come shy of what we can already hit with 32nm SB but at slightly lower power at least.
So the 100% die space committed to a CPU at 130W TDP aside, you're saying the thermal spreader on IB and Haswell chips is inferior to past iterations, including SB? Is this what the soldered vs. non-soldered thing is about? While on the subject, there was a post in another thread saying that Haswell will be going back to soldered, but in response, another poster claimed the opposite.Okay, Intel obviously didn't want TDPs to skyrocket to the moon. But they've made 130W chips before, so why can't they again? What would a Haswell chip with a proper thermal spreader, no integrated GPU and a 130W TDP look like?
My guess is it would look like something everyone on this forum would want to buy unless it was outrageously expensive.
That topic actually also has been just one generation away from happening for the last 4 years~Haswell IMO is compelling in the laptop space where it finally brings enough GPU grunt to make gaming reasonable, but only if Intel can get their drivers to work correctly! That is a big if.
That topic actually also has been just one generation away from happening for the last 4 years~
The biggest issue with Desktops I have is actually with the Celerons, Pentiums and Core i3s. While the highend stuff moved upwards, the slower parts have all but stagnated in performance and features for 5 years now. I'm not surprised that revenue is tanking big time.
So the 100% die space committed to a CPU at 130W TDP aside, you're saying the thermal spreader on IB and Haswell chips is inferior to past iterations, including SB? Is this what the soldered vs. non-soldered thing is about?
To make myself feel better though, I just tell myself that today's CPUs are dozens of times faster than ones from just a few years back, all while running cooler and requiring less power. That's still progress.
It's just IB, SB is still soldered. But Haswell is supposed to be like IB. I guess we'll see.
Okay, but is that really true? Is a top-of-the-line IB CPU really "dozens of times" faster than a 2006-era Conroe on typical workloads (i.e. not cheating by using stuff that's easily multithreaded)? I don't know, but I'm highly skeptical. My guess is that it's twice as fast, maybe a bit more than that, and maybe even less than that.
Consider the above "Idontcare bait".
Shouldnt that 3770K be a 3960X? You are comparing 1000$ CPUs with a 300$ one.
It's just IB, SB is still soldered. But Haswell is supposed to be like IB. I guess we'll see.
Okay, but is that really true? Is a top-of-the-line IB CPU really "dozens of times" faster than a 2006-era Conroe on typical workloads (i.e. not cheating by using stuff that's easily multithreaded)? I don't know, but I'm highly skeptical. My guess is that it's twice as fast, maybe a bit more than that, and maybe even less than that.
Consider the above "Idontcare bait".
I have much a much more optimistic outlook on Haswell. Let's face it; it's no Sandy Bridge. But it's a tock, which essentially guarantees that it will be a bigger jump than Ivy Bridge was, and given that this is an Intel product, there's little chance for a Phenom or Bulldozeresque flop.For the desktop, I have a feeling you are going to be right about that.
IB was the first processor I have ever bought that failed to OC higher than my processor from the previous node.
And all the talk about Haswell has been directly focused where you'd expect it to be - the power-efficiency in the low power end of the spectrum.
So what are we going to see on the high end of the power spectrum? Probably more of what we've already seen - clocks that come shy of what we can already hit with 32nm SB but at slightly lower power at least.
Okay, but can't they actually give us those 125W CPUs?
I knew you'd come through. :thumbsup:
31% increase in IPC over seven years is a pretty good indication that the microarchitecture tweakers at Intel have pretty much run out of tricks. Evolution is getting pretty slow; we're going to need something revolutionary to take the next step up.
The idea is that they're increasing IPC in a very power efficient way.
If power wasn't a problem, you could see huge IPC boosts. But in this world, we like our battery life.
well, there has been some gains,
the worse part was the fact that Intel eliminated overclock, so a older gen Pentium/Celeron/i3 overclocked can still probably beat the ivy bridge parts, but in reality it's irrelevant for most buyers sure,
2007/08 vs 2011/12 $60 CPU
http://www.anandtech.com/bench/Product/406?vs=67
GMA 3100 or 4500 compared to HD 2000 is even better I think,