The real competition is no longer Intel. Its AMD. Or if you're just buying Apple because you like them then there is no competition, you buy what they offer.
I like fanless systems so I'll be watching Apple closely.
I can't help but chuckle whenever I read these comments on HN about how doomed Intel is.
Intel just does this. Every now and then, they get so far ahead that the rest of the market just totally disintegrates, which allows them to screw around and juice up their margins while failing to actually innovate. Their brand is so strong it takes years for it to erode, even when they do suck, and when they have actual competition they've got plenty of cushion to keep selling old designs while they catch up.
The reality of semiconductor tech is that most people are not paying enough attention to understand when a company is doing well and when it's doing poorly. AMD stock was trading in the single digits well after the company was firing on all cylinders. Intel has encountered major bumps in the road since the early days of 10nm, yet research analysts and technologists alike were praising Intel all over the place until recently.
I wouldn't call it a KO until Intel 7nm chips come out. At that time, we'll see if this is a comeback story or just another IBM.
> Keep in mind you're trying to pin magical properties on a brand, and meanwhile people and technologies come and go.
Sure, things change! But Intel is huge, and it's got a track record of repeatedly weathering setbacks and missteps only to come back with market dominance.
Maybe it's really going to be all about Apple and AMD while Intel plays catch-up for generations to come - I just feel like it's a bit premature to come to that conclusion.
Delays aren't necessarily underperformance. Let me offer an example.
If Boeing had delayed the rollout of the MAX 8, or even simply reduced the production rate, it may have been able to identify and rectify the MCAS failure mode, thus preventing suspension of the MAX aircraft. In retrospect, they could have delivered more aircraft prior to the pandemic and avoided many of the order cancellations that it brought.
Sometimes, it's better to go slowly and get things right than to forge ahead at full steam. We won't know if these 7nm delays are good or bad for Intel, until 7nm actually rolls out.
It is an underperformance relative to their previously published roadmaps. There may be other reasons than just mismanagement for those delays, but they're still delays.
Intel has had problems with architectural cul-de-sacs before certainly. But they had other architectures on the back burner they could go back to to recover when things didn't work and they still had their process lead to keep them in the game. Recently, though, Intel's architecture work has been at fine but their process has been the thing having trouble. Maybe they'll be able to get 7nm working properly first try despite the problems at 10nm but I'm not sure given the attrition in their process team after the 14nm death march.
I don't know, this feels different. Intel's competition is bloodthirsty and attacking from all sides. Intel has nothing in the pipeline and just keeps dropping the ball (don't forget Intel's lack of penetration in the mobile market).
This feels more like Microsoft getting blindsided by Google, Apple and Amazon amongst others. Intel isn't going anywhere any time soon, but their reign as king of the mountain may very well be over.
Right, and I think some benchmarks I've seen this morning indicate that it throttles during more intensive tasks. The Pro benchmarks better, the only difference being the fan.
All CPUs throttle, all of the time. It's been years since anyone shipped a high-performance processor without a closed-loop dynamic thermal control system.
If your CPU always runs at its steady-state temperature that means it sucks and it leaving performance on the table. A CPU that can run at a steady 3 GHz (or whatever) should be capable of 5+ GHz momentarily given the right initial conditions.
You're choosing the argument first and then trying to justify it post hoc. That's the downside of having these conversations in thread format; it's easy to disagree, and tough to acknowledge that the other guy can be right about some things and wrong about others.
Sure, the critical path setup and hold time limit clock speeds, but that's not the reason for throttling a chip that can turbo at a higher clock. Even if it were, certain operations with a shorter critical path could run at faster clock even when hot.
If thermals weren't the dominant factor, you wouldn't need better cooling to overclock.
My perspective (correct me if wrong):
Hot semiconductors can damage themselves, and this becomes more important as the lithography shrinks. Binning is designed to identify which silicon can be pushed harder and which is not quite up to the task.
I agree with the other guy that if your CPU always runs at its steady-state temperature that means it is leaving performance on the table.
I like fanless systems so I'll be watching Apple closely.