Another clueless post, Intel has been in chromebooks for over a year, look at your own post ...
•Review: Acer C720 Chromebook with Core i3, best value yet
Don't be condescending and you are not remotely keeping me honest rather I am just correcting you from your ignorance. No-one feeds me any beliefs, I do my own deep technical analysis often against the herd (i.e. I laughed in the face of all those who thought ARM and tablets would bring down Intel or that Itanium would slay Opteron) and invariably it is proved right over time as it has with AMD, NVDA, INTC etc and I was debating with Charlie D back (both agreeing and disagreeing at times) in the Aceshardware days (as mas) so my history and knowledge is long in these matters.
Right having told you what Denver is not i.e. a competitor to Core-M let me be positive and tell you what it could be. Now no-one looking at Denver would ever design such a cpu if you was going for just high performance i.e. in-order but it could just be an efficient design in performance/power in the ultra-mobility sphere although we will need results from final silicon to confirm.
If it can be positioned as a leading $30-40 chip for tablets and phones it could provide some very nice incremental revenue for NVDA that could become materially important over future years. It does need to stay ahead of ARM and Intel's future best efforts in that price range e.g. A57/Broxton but there is potential there especially when tied with a decent Nvidia gpu. It also needs to go to quad-core at 20nm as at 28nm it is not good enough to be a successful leading dual-core design like say Core-M. Anyways Nvidia is finally in the cpu designing business after all these years and that in itself is a useful diversification. Good luck with it.
Full transcript of the Crimean meeting of the Klingon High Council
My analysis of the situation now is that Putin will let the Nationalist elements in his country support the Donbass insurrection until its natural bloody conclusion, in effect letting them get it out of their system. Then he will make peace with Ukraine and the West in the aftermath. I also don't believe there will be any more sanctions as the EU is painfully finding out it needs Russia more than it needs Ukraine and will not fancy more retaliations this time against their car industries rather than just their farmers. So no more sanctions and only a localized conflict eventually petering out means world stocks like INTC are all set fair to coast along worrying just about economic factors rather than geopolitical ones.
p.s. the convey matter now seems agreed and will be handed over to the IRC to distribute.
Slide 14 of the following pdf shows 2.5 Ghz Denver being compared to a 1.4 GHz Haswell Celeron and 1.6 GHz Bay Trail Celeron as well as other ARM chips. Remember though these are Celerons with slugged clocks and caches for artificial sku stratification purposes not the top bins of both these chips that clock as high as 3.3 GHz in the Haswell-U, 2.9 GHz for Haswell-Y and 2.4 GHz for Bay Trail.
'Without it, they would be dead.'
Don't be absurd and alarmist, traditional PC processors are still selling in their billions of dollars every year and Nvidia depends on that ecosystem to make profits with its expensive gpus. If PCs die so does Nvidia's profits. In fact Intel would have made a billion more last quarter without its heavy investment in Atom and Modems.
'Don't count ARM for Windows out yet. '
I never counted it in as I knew and posted it would be a bust from the very beginning and have been proven completely right. No performance or power or price advantages over x86 with crippling legacy software incompatibility to boot is not a recipe for success and so it proved.
'Your post illustrates exactly why Microsoft needs to cultivate an Intel alternative if for no other reason than to keep Intel's pricing in-line.'
That's supposedly why AMD exists but Intel is now pricing their Atoms cheaper than them which shows you how irrelevant this line of thought is.
'Show me an unambiguous quote from a Microsoft executive saying ARM for Windows is dead and I will say that I am wrong. '
I don't need to as their actions speak louder than any words. On Surface Pro 3 release they also had a Snapdragon 800 Surface 3 ready to go too but canned it just before launch as I suspect they got tired of losing hundreds of millions on this aborted failed o/s. All the software work they have done on WinRT will eventually migrate to Windows Phone but for Windows on PCs and Tablets WinRT is DOA.
'I do not believe it will be with $125 Core-M's. '
It will be with $25 Cherry Trails or $40 Broadwell Celerons. Also don't get too hung up on Intel's list prices as the really big OEMs who buy in 6-7 figure bulk usually get around 50% discount.
' K1/64 changes the equation on what can be done with Windows. '
No it does not, it will barely live with a 1.4 GHz Haswell Celeron which is not groundbreaking.
'Architecturally, I would bet that Denver at 20nm is designed to produce equivalent performance at equivalent power to any 14nm out-of-order Core-M that Intel could conjure up (they key issue for Intel being 'out-of'order'). Further, Denver has first-to-market advantage by six months.'
You would bet wrong. The top-bin Core-M will have 2-3X the single-thread and 3-4X the multi-threaded performance/power of Denver in the same ~5W fanless class.
You are incredibly confused, which does not seem difficult or you. [Intel's] Mobile and Communications Group does NOT contain ANY CORE PROCESSORS (which includes the very low power ones too) in it as ALL COREs go into the PC GROUP, it only includes ATOM and MODEMS. See I spelled it out for you in big capital letters so you don't get confused again.
The PC group including all the low powers ones like Core-M is, along with Server group, what have made INTC touch very close to $35 this year so that strategy is working out perfectly both for share price and for subsidized Atom stealing marketshare over the ARMy hand over fist in cheap ultra-mobile slots while NVDA is left behind with expensive to make chips it can't sell in volume or at a high price.
'How is INTC going to charge oh, I don't know, $25-30 per device from their latest greatest 14nm process to be competitive in the mobile space? '
Why would they want to do that when they can easily charge $125 or $225 for the Core parts and sell more than say the equivalent Nvidia part in say the Surface models ? Listening to you you would think Intel had to aspire to make only a hundred million profit each quarter like Nvidia rather than the $3bn it does. The continuation of the Surface Pro tablet line as opposed to Windows ARM RT Surface already going EOL shows Intel does not need to charge only $25 like a puny ARM core to sell in mobility. Don't judge superior technology by inferior standards as this Core-M will turbo well over 2GHz at 4W annihilating Denver, A57 etc at the same power level.
'When the licensing agreement ends in '17 so will Intel's rights to use any of Nvidia's IP'
BS !! It's a lifetime license until expiration and unless Intel is stupid enough to want to use new Nvidia patents after 2017 Nvidia is SOL with this particular extortion racket ...
"4.1. Term. This Agreement and the rights and licenses granted hereunder shall become effective on the Effective Date, and shall continue in effect until the expiration of the last patent licensed hereunder to expire unless such rights and licenses are sooner terminated by a Party pursuant to Section 4.2."
' MS's statement was a pure manipulative, and baseless assertion with the agreement expiration being well over 2 years away. '
CLUE and LESS describe your 'knowledge' of this agreement. MS was completely right and only blind fanboy Nvidiots are denying this.
The licensing fees to Nvidia stop in 2017 and then NVDA share price collapses as they won't need to pay for any of those patents ever again in their lifetime as Morgan Stanley rudely pointed out to you recently and those license fees form a considerable part of Nvidia's profits every quarter so their absence will be immediately felt.
Intel has been in the ultra mobile segment a lot less than Nvidia and it has blown past it in market share pretty easily approaching 10% now while Nvidia languishes in the 1-2 per cent range. Also any advantage you think K1's Kepler gpu currently gives you will be taken away by 14nm Cherry Trail and 14nm Core-M. Intel will eventually dominate ultra-mobile by relentless outperforming and outpricing its competitors year on year.
4 years ago when I first posted on the NVDA board when Mosesmann's ridiculous $40 target was being touted by you guys I said NVDA was considerably overpriced and INTC considerably underpriced and that's how reality has turned out unlike all the circle-jerk fantasies all you clueless Nvidiaites had at the time.
'they will not be able to compete very effectively with already established market leaders in the mobile sector who do not have the very expensive cost of cutting edge fabs to support.'
PC processors support the cutting edge fabs and then some with billions of profits left over each quarter. Intel could afford to have no profits/losses on its ultra-mobile processors for literally decades and not worry. Alternately the ARM chips are potentially double the cost of intel's even at just wafer cost break-even because they have to pay a 45-50% of that wafer cost to a foundry for its profits which Intel does not as PC processors have already paid for the FABs. So you see it's the foundry players like Nvidia who have the 'very expensive cost'.
p.s. All these years you have been waiting for Project Denver and the best it could do at 2.5 GHz was be roughly in the same league as a 1.4 Ghz Haswell Celeron ! Much ado about nothing ;-).
The Klingon Emperor is promising new nuclear weapons, how nice ...
The Russian Enigma: Winston Churchill
1st October 1939
I cannot forecast to you the action of Russia. It is a riddle wrapped in a mystery inside an enigma; but perhaps there is a key. That key is Russian national interest. It cannot be in accordance with the interest of the safety of Russia that Germany should plant itself upon the shores of the Black Sea, or that it should overrun the Balkan States and subjugate the Slavonic peoples of south eastern Europe, That would be contrary to the historic life-interests of Russia.
Thus, my friends, at some risk of being proved wrong by events, I will proclaim tonight my conviction that the second great fact of the first month of the war is that Hitler, and all that Hitler stands for, have been and are being warned off the east and the southeast of Europe.
Here I am in the same post as I was 25 years ago. Rough times lie ahead; but how different is the scene from that of October, 1914! Then Russia had been laid low at Tannenberg; then the whole might of the Austro-Hungarian Empire was in the battle against us; then the brave, warlike Turks were about to join our enemies. Then we had to be ready night and day to fight a decisive sea battle with a formidable German fleet almost, in many respects, the equal of our own. We faced those adverse conditions then; we have nothing worse to face tonight. We may be sure that the world will roll forward into broader destinies.
We may remember the words of old John Bright, after the American Civil War was over, when he said to an audience of English working folk: "At last after the smoke of the battlefield had cleared away, the horrid shape which had cast its shadow over the whole continent had vanished and was gone forever".
The first generation MMS-100 is half that weight and a third that price and the actual MMS sensor itself is a fraction of that weight/size, think palm-size.
Some of it looks legit as revealed by a US reporter for FT who is there with it.
With today's resignation of Strelkov/Girkin and the soft mood music coming from Putin in Crimea about Peace and reducing bloodshed I do hope the major countries can all get back to DefCon5 from DefCon4 and start thinking about how they can out-trade each other rather than out-fight each other because I really doubt the difference between EU/Ukraine/Russia are worth one life never mind the thousands that have happened.
'A legitimate comparison for device intended for the same target devices.'
but deceptive in that there are higher clocked Bay Trails and Haswells that are actually lower power than the Celerons they chose.
'this is where the ARM ecosystem (or at least this iteration) exposes intel despite all the noise about process'
There is no exposing being done as shown by Intel's continually high margins and profits despite years of ARM tablets being arrayed against them and despite wild claims by many over here that x86 and Intel would collapse under a cheap ARM assault when in fact Intel is pricing Atom even cheaper than ARM without any effect on Core profits and margins. Basically Intel is charging a premium price for its premium Core performance but that does allow fair room for bottom feeders like AMD and ARM to exist and make some money.
So you have dropped all pretense of being a fake Long today lol ?! Project Denver is not a threat to x86, it's going to be hanging with ARM A57 and Apple A8 in that performance neighborhood and there should be a niche for it in the ARM world. NVDA was up because of its earlier earnings.
remember also Nvidia is starting from a low volume base so it will not take much interest to start making a material difference. This chip is like Nvidia's version of Itanium but higher clocked with lower performance and power.
There is more potential for volume when it's made into a quad at 20nm because at the moment it's not going to be much faster than a A57 at the single thread core level and Samsung/Qualcomm will be churning out loads of those 20nm A57 quads next year. Still it does not look a dud assuming there are no glass jaws with the particular in-order concept they have chosen and even the 28nm dual-core version should sell some if the A15 K1 is managing to sell.
Apple 20nm will still be a dual but higher clocked so performance should be similar/greater than Denver depending on how much extra clock they squeeze out. At only 1.3-1.4 Ghz that is where the performance growth potential lies for Cyclone while Denver at 2.5 Ghz already looks maxed out on clock to me so will probably go quad at 20nm.