SELL SELL SELL
That intel stock is going to be worth a lot less this time next year.
Put your money in AMD instead.
Despite record-breaking earnings, Intel's shares took a modest dip on Thursday when the semiconductor behemoth reveal its financial results for the second quarter of this year. Revenues for the three months to June 30 rose strongly, year on year, and profits were up 78 per cent. However, it's clear that the markets are …
@ Tom 64
I did that 15 years ago when AMD's Opteron was the best server chip by miles, so I bought AMD and sold Intel. AMD stock went down and Intel's went up. 2 -3 years later when the first P3 Xeons came out and delivered better Price/performance that Opteron the stocks reversed and I lost again.
Either Wall Street knows nothing or they know everything, they just tell you or I, ever.
if you think China and the U.S. President have anything to do with Intel stocks, then you're greatly mistaken. Either that, or you merely WANT it to be so because the alternative disagrees with your world view or something. Whatever.
Intel's current problems are based on a few things:
a) Meltdown and Spectre, and their somewhat 'lackluster' responses to it all
b) Effectively hitching their wagons to Win-10-nic to sell new computers. THAT right there is a BIG part of it.
c) Moore's Law not doing what it was doing 10-15 years ago any more; i.e. computers aren't perceived as being 30 percent faster each subsequent year. [so people are more likely to hang onto what they have]
d) Features like 'Management Engine' or whatever it's called, and the inability to canonically DISABLE it.
None of these things are helping their bottom line. Getting 10nm processes going is just the latest excuse, I say.
To fix the problem, they need to do the following:
a) properly fix Meltdown+Spectre, including for older CPUs [when possible to do microcode updates]
b) properly document (and support) SHUTTING OFF the 'management engine'
c) Promote other operating systems, including Linux and Mac, _ESPECIALLY_ including marketing for new affordable non-windows computers, software ports of existing applications to Linux and Mac, and other stuff necessary to get people to switch to a non-windows OS so they don't have to hitch their wagon to Micro-shaft's poor decisionmaking.
d) Promote and develop multi-thread algorithms that actually MAKE A DIFFERENCE in the perceived speed of applications. This COULD include a client/serer model for the GUI (you know, like X11) which inherently could take advantage of multi-core.
But yeah, moving away from relying on Win-10-nic to sell computer chips is a BIG one. They should get serious about that.
"Intel's current problems are based on a few things:"
No. Intel's real problem is not Spectre, AMT or whatever you wrote. Or the President.
The problem ahead is lack of 10nm or smaller lithography because Intel bet their lithography process on non-EUV process, and couldn't make it work.
Now they're late to bid for those EUV machines from ASML - which just happens to be the only manufacturer in the world. And every other big silicon company is opting for the same EUV machines as well. The assembly and testing before go-ahead will take time and Intel will need several machines for each product line they wish to manufacture.
"Promote other operating systems, including Linux and Mac"
So the poor little Apple needs outside help promoting their proprietary computers and operating systems? I'll gladly let the Macolytes to spread their gospel but asking Intel to promote Macs when Apple looks set to change the camp to their own ARM based silicon? What are you smoking man?
And from what I've read in these forums, many pro-Linux people hate the complexity of x86 and claim that their Raspberries are fine and suitable to desktop and server duties. Promoting operating systems that are mostly used on other platforms doesn't seem very clever to me.
AFAICS, Intel has invested in Linux kernel and compiler optimization greatly.
The elephant in the room is a physics.
Intel get where it is by making a crap design run on smaller and denser transistors.
The capital required to get much smaller is no longer economic, at least for a horizontal integrated company, making and selling its own chips.
Intel will be wiped out by a company that cracks large scale cores, large caches, running at 3Ghzish.
And facebook's shares have plummeted 20%, shrinking their cap by a headache inducing $120bn - the largest such drop in corporate world history. Perhaps worth a mention, el Reg?
And with poetic timing this coincides with the release of the false and misleading "Vote Leave" ads run on the site by AIQ, including some rather sneaky phising methods. Perhaps worth a mention, el Reg?
Or are you just bored talking about them?
I do have to make one point. Intel and AMD have different ways of measuring their feature size. What AMD has dubbed their 7nm node is pretty comparable to Intels 10 nm products.
Intels was banking very heavily on ASML getting EUV working 2 or 3 years ago which in reality is only getting ready for volume manufacture now. When it became obvious that EUV wouldn't be ready when planned Intel was sent scrambling to come up with the needed techniques to do it on immersion DUV litho systems giving competitors time to catch up. First missing their traditional Tick-Tock schedule and inserting an extra tock, and possibly now needing even more time. It remains to be seen if Global Foundries/AMD has managed to fully catch up or if they will still be a step behind. I know GF is also looking heavily at EUV for the next node, so we'll see whats going to happen.
With that hugely increased profit you'd think they could put some extra into R&D and maybe helped solve their 10nm, spectre, meltdown, management engine, and nic issue. I recognize that the last two aren't "errors", although I think many consider them problems.
This post has been deleted by its author
Biting the hand that feeds IT © 1998–2021