Just get some easy names
Intel have:
i3 for just Web folk
i5 for people who run office
i7 for people who need full power
AMD have a series of numbers than mean nothing to "normal" folk
Industry execs are paid to put on a brave face when outlining company financials, and AMD’s CEO Lisa Su kept up her end of the bargain last night when the broken business reported a net loss of $497m for 2016. This was progress of sorts: in the prior year ailing AMD filed losses of $660m. Despite the latest numbers, the stock …
'AMD have a series of numbers than mean nothing to "normal" folk'
At the moment Intel has no less than:
- 13 different Kaby Lake i7 processors;
- 17 different i5s;
- 11 different i3s;
- 5 Pentiums
- 3 Celerons
- 2 Xeons
- 1 Core m3
The processors are divided into 5 different, easy to remember power usage categories: H,K,T,U, and Y. Throw in the mix 7 different kinds of weak 2D GPUs.
Is AMD processor naming really more complex? I thought their numbering scheme is rather simple: bigger number means faster.
Intel have:
i3 for just Web folk
i5 for people who run office
i7 for people who need full power
That's rather too simplistic ... and what about "Pentium" and Celeron and Atom?
It's nice to know, though, that "i3" means two cores and Hyperthreading, "i5" means four cores, and "i7" means four cores and Hyperthreading ... except in the case of mobile processors which all have two cores and Hyperthreading ... except when they don't!
I agree that AMD need some cuddly names -- what happened to "Sempron", "Athlon" and "Opteron" ... meaningless, but at least they kept it easy by only having three ... oh, and "Duron" and "Turion" and "Phenom" ... maybe it wasn't so simple after all?
Maybe there are just too many CPU types?
ARM, anyone?
"If they "lost" $497m but paid off debts of nearly $800m, that actually sounds like a pretty successful year. Revenue is up, profit is up (or at least less negative), market share is up, and debts are down. It really doesn't sound like such a horrible 2016 at all."
^ This. Put aside the one-off write offs and look at the underlying trading situation. It's certainly not disastrous by any means although they do need to do better. I'd certainly want AMD to survive to serve as counterweight to Intel especially since the third competitor, Cyrix, has long gone.
There's so many cards stacked against them, even if everything goes unrealistically right.
With a little honest scepticism borne from that fact there's still not yet public comparison available, it would be fantastic for AMD to be back in the performance space. If we assume a hypothetical (and it does seem to be being taken as a given) that this will be true, AMD's problem is still that on empirical grounds I, at least won't actually feel directly compelled to upgrade from a haswell; and won't, for as long as possible.
I'm not sufficiently 'ryzentful' to buy one just because I can bemuse myself it's socially responsible.
They'll have to create the space for that, because intel sure haven't been. It doesn't mean they can't; but I have to wonder if even 50% per core faster* would even be a game changer for those not calculating their FLOPS per rack; the market has changed since 5-10yr ago (draw your own line) when they were last a contender.
It's the latter that seems to me like their best hope of an end customer. With mobile everything on the business end, any performance desktop part - like my melted 21264 - seems almost boutique. How much can AMD make from this? And from consumers?
I'm sure the tier 1's OEM's will still just cram them in dog-food grade product, and that's their other access to real volume. And I'd not be surprised if intel schedule a release the day after/before and upstage them.
But keeping those happy thoughts.
*I'd be impressed if 5% * 8 cores, but again, that doesn't buy /me/ a lot more frames a second.
More cores is nice, but I rarely need more than a couple (although I'm sure having four rather than two makes everything a bit smoother). Most of my home use is gaming, so single thread performance is important. Mind you, it looks like AMD won't be including GPUs in their higher end chips, where as Intel are still stuffing GPUs into i5 chips, that's a whole hunk of silicon I don't need and would rather not have had to pay for.
there is still time to buy £4000 of shares
https://uk.finance.yahoo.com/quote/AMD?ltr=1
currently $12.12, they spike to $60 - > $90 everytime AMD releases a new product, and those products are ryzen and vega at the end of the month, you would make a sweet and easy $22,000+ profit
all the more better then ryzen and out do a i7 and the vega can out do the nividia 1080, then you have vega and cuda code convertor to use stream processor for floating point math with low level api's and all the other new arch for parallel computing
No, that's a fantasy version of share dealing. After some extraordinary growth AMD's shares have only just climbed back to their level of ten years ago - literally ten years ago - and that seems to be based on wishful thinking rather than a genuine turnaround. The company still doesn't pay a dividend. Their shares won't be spiking to $60 any time soon.
you can scroll through the past back to 2000 on https://www.google.co.uk/finance?q=NASDAQ:AMD
shares can spike upto $90 whenever the release a new product
Biting the hand that feeds IT © 1998–2022