[TheReg] Intel Management Engine: Is this why we have seen slower speed growth?

[ehume]

Here. With the ME running under the OS, it has to be slowing the OS down. Maybe the growth in performance has gone partly to this.

Edit: thanks to zir blazer for this introduction to the IME

 

[ShintaiDK]

No....

 

[SarahKerrigan]

Here. With the ME running under the OS, it has to be slowing the OS down. Maybe the growth in performance has gone partly to this.

You realize the ME is a separate processor, right? And one with a negligible amount of die space at that?

 

[Magic Carpet]

Interesting bits on ME.

 

[TeknoBug]

Milking the market.

 

[dark zero]

So... in order to please the government of US and Israel, they screw up the improvement of the processors? Seems interesting.

 

[ehume]

You realize the ME is a separate processor, right? And one with a negligible amount of die space at that?

According to the presentation, it interposes between the CPU and every operation. There has to be a slowing because of that detour. Ain't nothin free. My speculation is that we have seen less improvement than we expect in the last several generations. This may be the culprit. I'm willing to be educated.

 

[ShintaiDK]

It doesn't. The ME engine is in the PCH, not the CPU.

 

[crashtech]

It's not tough to make the intuitive leap to realizing a system designed to help sysadmins with out-of-band management could become something that can potentially be used for malicious purposes. But I think the performance hit would be negligible, and anyway far from the main concern.

 

[ShintaiDK]

The single biggest (stealth) danger in a modern system is the UEFI.

 

[Jovec]

Anyone who thinks Intel is purposely slowing CPU performance gains is free to design a faster CPU and rake in all those billions and billions in profit.* Modern CPU design is hard, and doubly so when backwards compatibility is required.


*Beyond the obvious like different clocks and core counts.

 

[Arachnotronic]

Anyone who thinks Intel is purposely slowing CPU performance gains is free to design a faster CPU and rake in all those billions and billions in profit.* Modern CPU design is hard, and doubly so when backwards compatibility is required.


*Beyond the obvious like different clocks and core counts.

This.

 

[Fjodor2001]

Anyone who thinks Intel is purposely slowing CPU performance gains is free to design a faster CPU and rake in all those billions and billions in profit.

Barriers to entry.

Intel knows this and is making use of it to maximize profit. Their primary interest is just that, maximizing profit. This does not necessarily mean maximizing performance improvement per year. Especially in a monopoly-like environment.

 

[Azuma Hazuki]

And this, among many other reasons, is why I have said the cardinal human sin is greed, for a very long time now...

 

[mrmt]

Barriers to entry.

Intel knows this and is making use of it to maximize profit. Their primary interest is just that, maximizing profit. This does not necessarily mean maximizing performance improvement per year. Especially in a monopoly-like environment.

There is a small underdog company that has no barriers to the entry and it seems that despite Intel slowing down they are not making much progress in shrinking the huge performance gap they have when compared to Intel. Maybe they are being bribed by Intel to not develop faster processors too.

 

[III-V]

Maybe they are being bribed by Intel to not develop faster processors too.

Sounds like something Fjodor would say.

 

[ehume]

No, I'm sure they have good reasons for making this device, but my hypothesis is that ME is why we haven't seen larger gains in processor capability.

It must be worth it to them to build this in. But if someone ever gets inside enough to hack it, will we on the outside even know when we have been hacked, much less fix it?

 

[III-V]

No, I'm sure they have good reasons for making this device, but my hypothesis is that ME is why we haven't seen larger gains in processor capability.

It must be worth it to them to build this in. But if someone ever gets inside enough to hack it, will we on the outside even know when we have been hacked, much less fix it?

Ask the NSA.

 

[mrmt]

Ask the NSA.

What about Mediatek SoCs? Do they have somerhing fundamentally different?

 

[ShintaiDK]

No, I'm sure they have good reasons for making this device, but my hypothesis is that ME is why we haven't seen larger gains in processor capability.

It must be worth it to them to build this in. But if someone ever gets inside enough to hack it, will we on the outside even know when we have been hacked, much less fix it?

But its still on the PCH and not the CPU. So what's next, not making faster chipsets due to the ME?

 

[eternalone]

Does this affect AMD cpu's as well?

 

[DrMrLordX]

Ask the NSA.

Nothing to see here, please move along.

Does this affect AMD cpu's as well?

That would depends on whether or not they've integrated Intel's ME spec into their PCH/FCH. I'm not aware that they have done so.

 

[III-V]

What about Mediatek SoCs? Do they have somerhing fundamentally different?

I don't see why they'd be immune to a little bribery. Heck, I think they'd be even more prone to it.

 

[ShintaiDK]

Does this affect AMD cpu's as well?

AMD got their own security processor. So yes. But the issue is nonexistent.

 

[DrMrLordX]

Here's AMD apparent equivalent:

http://www.amd.com/en-us/innovations/software-technologies/security