英特尔isa架构_x86 isa和intel的游戏结束了吗

英特尔isa架构

Ever since Apple announced they are switching to their own Apple chips with the ARM instruction set I have spent time trying to develop some educated guesses about the future.

自从苹果公司宣布他们将使用ARM指令集转向他们自己的苹果芯片以来，我就花时间尝试对未来进行一些有根据的猜测。

The future is not looking good for x86 and intel, and that is not due to Apple alone, but rather a multiple development all happening in parallel conspiring to unseat intel and the x86 from the CPU throne.

x86和intel的前景并不乐观，这不仅是因为Apple本身，而是多种发展同时发生，从而使intel和x86失去了CPU的宝座。

Let us first look at the big picture.

让我们首先看大图。

CPU是一场游戏 (CPUs Are a Volume Game)

Those people who rant on about superior performance of intel and AMD chips and denounce ARM as weak, have simply not been paying attention to CPU history. In fact they have failed to grasp the very history that brought Intel on the top in the first place.

那些对Intel和AMD芯片的卓越性能大加赞赏并谴责ARM薄弱的人根本就没有关注CPU的历史。 实际上，他们未能掌握将英特尔排在首位的历史。

Back in the 90s we had NeXT, Sun Microsystems, Silicon Graphics and several other powerful RISC workstations. These command a hefty premium price, but that worked because they outperformed PCs and their intel CPUs by a big margin.

上世纪 90年代，我们拥有NeXT ， Sun Microsystems ， Silicon Graphics和其他几个功能强大的RISC 工作站 。 这些产品的价格很高，但之所以奏效，是因为它们在性能上大大超过了PC及其Intel CPU。

However one by one they folded, obliterated by intel and the PC juggernaut. Why was that?

然而，他们却一一折叠，被英特尔和PC巨头吞没了。 怎么会这样

Economics of scale. Doesn’t matter if you make shitty and puny hardware. However those who produce in large volumes will eventually gain a cost advantages and be able to outspend the competition.

规模经济 。 无论您制作的是劣质还是微弱的硬件都没关系。 但是，那些大批量生产的企业最终将获得成本优势，并且能够超越竞争对手。

Intel’s x86 architecture may not have been as good as their RISC counterparts. However due to higher volume, they could afford to throw more money at the problem and produce their chips at smaller nano-meters undoing a lot of their architectural deficiencies.

英特尔的x86架构可能不如RISC架构好。 但是，由于体积大，他们有能力花更多的钱解决问题，并以较小的纳米尺寸生产芯片，从而消除了许多架构缺陷。

This was no different from PCs themselves. As an Amiga user back in the 80s and 90s I could not help but shake my head about what a terrible hardware design the PC was. Of all computers you could buy, no architecture was probably worse than the PC. It was a shoddy product in every possible way. Do I need to remind anyone of the 640KB boundary, the many ways you had to tweak the DOS startup configuration files to get anything running?

这与PC本身没有什么不同。 作为80年代和90年代的Amiga用户，我不禁对PC的糟糕硬件设计摇了摇头。 在您可以购买的所有计算机中，没有任何一个架构比PC差。 这在所有可能的方面都是次充好产品。 我是否需要提醒任何人640KB的边界，您必须通过许多方法来调整DOS启动配置文件才能运行任何东西？

Or having fiddle with drivers for graphics cards and sound cards every single time you installed a game. Setting up interrupt lines (IRQ) and I/O addresses for graphics and sounds card was not my idea of fun.

或每次安装游戏时都花时间摆弄显卡和声卡的驱动程序。 设置图形和声卡的中断线(IRQ)和I / O地址并不是我的主意。

Despite it multitudes of flaws, the PC had something important going for it. It had an army of makers, competing to make it, which brought low prices and volume production. As PC volumes grew all the flaws of the PC architecture was simply hacked around by throwing money at the problem.

尽管存在许多缺陷，但PC仍具有重要意义。 它有一群制造商，竞争制造它，从而带来了低价格和批量生产。 随着PC数量的增长，PC架构的所有缺陷都只是通过花钱来解决。

Now ironically the PC itself and Intel have become the new RISC workstations. They may hold the performance crown for now.

现在具有讽刺意味的是，PC本身和英特尔已经成为新的RISC工作站。 他们可能会暂时获得表演冠军。

However smart phones, tablets and all sorts of other embedded devices is where the volume is at. Chip makers such as TSMC, is in large part due to ARM getting such volume that they are able to outspend Intel.

但是，智能手机，平板电脑和各种其他嵌入式设备才是真正的市场。 台积电(TSMC)等芯片制造商，在很大程度上是因为ARM的销量如此之高，以至于他们可以超过英特尔。

Now intel are the ones trailing behind in the nano-meter race. A game they used to totally own, because they could outspend everybody else.

现在，在纳米竞赛中落后的是英特尔。 他们曾经完全拥有的游戏，因为他们可以超越其他所有人。

英特尔甚至具有性能优势吗？ (Does Intel Even Have a Performance Advantage?)

It has become a truthism that ARM chips are weak. Yet in the laptop space we saw that Apple’s iPad Pro when they came out beat most of their own intel based laptops on performance. That was insane as those ARM chips cost a fraction of the intel chips used in their laptops. Not to mention they where passively cooled.

ARM芯片较弱已经成为事实。 然而，在笔记本电脑领域，我们看到苹果的iPad Pro出现时在性能上击败了大多数自己的基于Intel的笔记本电脑 。 由于那些ARM芯片的成本仅为笔记本电脑中使用的英特尔芯片的一小部分，所以这太疯狂了。 更不用说它们在哪里被动冷却了。

That makes you wonder what performance you could get from an ARM chip with the same power/watt budget as an intel chip an an Apple laptop.

这使您想知道从ARM芯片获得的性能/功耗预算与Intel笔记本电脑和Intel笔记本电脑相同的功率/瓦预算。

I don’t think it is a stretch to imagine that Apple’s ARM laptops will outperform intel laptops. It depends a bit on what they aim for. Perhaps they only seek to match intel performance but give superior battery life time. Either way, intel laptops will likely have a hard time competing with ARM laptops in the future.

我不认为苹果的ARM笔记本电脑会胜过英特尔笔记本电脑。 这取决于他们的目标。 也许他们只是在寻求与英特尔性能相匹配的功能，但却可以延长电池的使用寿命。 无论哪种方式，英特尔笔记本电脑将来都可能很难与ARM笔记本电脑竞争。

So what you say “Intel gets most of their income from Cloud Data Centers. Apple laptops are a drop in the ocean!” Ok not totally, but a very large and growing portion of their revenue.

因此，您所说的“ 英特尔从云数据中心获得了大部分收入 。 苹果笔记本电脑是沧海一粟！” 不能完全确定，但是可以增加收入的很大一部分。

Except this intel stronghold is under siege by multiple ARM armies.

除了这个情报据点外，还有多个ARM军队围攻。

Amazon AWS has with the custom Graviton2 chips completely demolished intel on price/performance:

亚马逊AWS的定制Graviton2芯片在价格/性能上完全摧毁了英特尔 ：

If you’re an EC2 customer today, and unless you’re tied to x86 for whatever reason, you’d be stupid not to switch over to Graviton2 instances once they become available, as the cost savings will be significant.

如果您今天是EC2客户，并且除非出于任何原因与x86捆绑在一起，否则您会很愚蠢，因为一旦可用，就不要切换到Graviton2实例，因为这样可以节省大量成本。

Ampere is coming out with 128-core ARM beast for cloud computing, twice the number of cores of any x86 offerings.

Ampere推出了用于云计算的128核ARM野兽，是任何x86产品核心数量的两倍。

Yes they may not outperform x86 in every setting, but threat of ARM is at this point no longer a joke. x86 is being assaulted on home turf.

是的，它们可能不会在每种情况下都胜过x86，但是在这一点上，ARM的威胁不再是笑话。 x86正在家庭草坪上遭到袭击。

This process will begin to work in lockstep. As more people are seeing cost savings going with ARM cloud solutions, they are also going to want to have ARM laptops to develop on.

此过程将开始同步进行。 随着越来越多的人看到使用ARM云解决方案可以节省成本，他们也希望使用ARM笔记本电脑进行开发。

Running and testing on the same hardware platform locally as you deploy on is always an advantage. To quote Linus Torvalds:

部署时在本地相同的硬件平台上运行和测试始终是一个优势。 引用Linus Torvald的话：

That’s bull***t. If you develop on x86, then you’re going to want to deploy on x86, because you’ll be able to run what you test “at home” (and by “at home” I don’t mean literally in your home, but in your work environment).

真是牛逼。 如果您在x86上进行开发，那么您将要在x86上进行部署，因为您将能够在“家中”(而在“家中”进行测试)上运行您所测试的内容，但在您的工作环境中)。

With ARM based laptops Apple provide that final piece of the puzzle that threatens the x86 server dominance. As Linus points out, once you got viable ARM laptops, ARM in the server room starts to make sense:

苹果使用基于ARM的笔记本电脑提供了威胁x86服务器主导地位的最后难题。 正如Linus所指出的 ，一旦有了可行的ARM笔记本电脑，服务器机房中的ARM就变得有意义了：

my argument wasn’t that ‘ARM cannot make it in the server space’ like some people seem to have read it. My argument was that ‘in order for ARM to make it in the server space, I think they need to have development machines.’

我的观点不是说“ ARM无法在服务器空间中实现它”就像某些人似乎已经读过它一样。 我的观点是“为了使ARM在服务器领域实现这一目标，我认为他们需要拥有开发机器。”

This is where the advantage of ARM based Mac laptops really start to come together. Let us look at that in more detail.

这正是基于ARM的Mac笔记本电脑的优势真正开始融合的地方。 让我们更详细地看一下。

面向专业人士的ARM Mac笔记本电脑的三重优势 (The Triple Advantage of ARM Mac Laptops for Professionals)

When I worked as a consulted doing mobile app development, we often used Mac for the simple reason that it meant that we had one computer which could be used for Java server development, Android and iOS development.

当我担任移动应用程序开发顾问时，我们经常使用Mac，原因很简单，因为这意味着我们有一台计算机可以用于Java服务器开发，Android和iOS开发。

This advantage will simply be amplified with ARM. An ARM based Mac will be able to run both Android and iOS applications natively. There is no need for emulators, simulators or whatever. Mobile App developers are going to like that.

使用ARM可以简单地放大这一优势。 基于ARM的Mac将能够本地运行Android和iOS应用程序。 不需要模拟器，模拟器或其他任何东西。 移动应用程序开发人员会喜欢这样。

When you can throw in that the same laptop can also be used to test solutions for the AWS solution with best bang for the buck, the proposition only gets stronger.

如果您还可以使用同一台笔记本电脑测试AWS解决方案的最佳解决方案，那么这一主张只会变得更强大。

But we are not done yet. These laptops will also likely have the best battery life.

但是我们还没有完成。 这些笔记本电脑也可能具有最佳的电池寿命。

Why on earth would you then go with a PC laptop to run Windows? Windows for ARM will most likely be available anyway. Sure it may not run every Windows app, but so what? The money for developers is not in making Windows desktop apps. The money is in cloud deployment and mobile Apps. Both areas where ARM will dominate.

您到底为什么要带着PC笔记本电脑运行Windows？ 无论如何，ARM的Windows很可能将可用。 当然，它可能不会运行每个Windows应用程序，但是那又如何呢？ 开发人员的钱不用于制作Windows桌面应用程序。 这笔钱用于云部署和移动应用程序。 ARM将主导这两个领域。

苹果将​​如何推动 (How Apple Will Get the Ball Rolling)

This is the point where you should realize that laughing at the importance of Apple’s 9–10% marketshare impact is premature.

在这一点上，您应该意识到，嘲笑苹果公司9％至10％的市场份额影响的重要性为时过早。

They have a big part of the profitable premium segment. It is like their phones. They have a relative small part of the market but has 66% of the profit of the smartphone market. Apple gets 60% of the PC hardware profits, despite only selling 7.5% of computers. It is easy to forget just how big Apple is by just looking at marketshare.

他们在可盈利的保费市场中占很大一部分。 就像他们的电话一样。 他们在市场中所占份额相对较小，但在智能手机市场中却占有66％的利润 。 尽管只销售了7.5％的计算机，但Apple获得了60％的PC硬件利润 。 仅查看市场份额就很容易忘记苹果到底有多大。

With these advantages Apple will be able to crush the remaining competition in the premium segment and steal ever more of the total profits in the laptop market.

凭借这些优势，苹果将能够压制高端细分市场中剩余的竞争，并从笔记本电脑市场中窃取更多的总利润。

Not to mention anybody who want to run a Windows for ARM on higher performance hardware will have to get a Mac, as Microsoft’s ARM offerings have been pathetic.

更不用说要在高性能硬件上运行Windows for ARM的任何人都必须购买Mac，因为Microsoft的ARM产品令人遗憾。

Once this trend develops PC makers and Microsoft are going to start throwing bucket loads of cash at an ARM transition to make sure they are not experiencing an iPod or iPhone moment for the second/third time.

一旦这种趋势发展起来，个人电脑制造商和微软将开始在ARM过渡中投入大量现金，以确保他们第二次/第三次没有经历iPod或iPhone的时刻。

This will only further increase the ARM spending advantage over x86.

与x86相比，这只会进一步增加ARM的支出优势。

ARM is doing to x86 what PC did to everybody else in the 90s. By being an open platform with multiple chip makers you got fierce competition which will drive prices down and boost innovation.

ARM在x86上所做的就像PC在90年代对其他所有人所做的那样。 通过成为拥有多个芯片制造商的开放平台，您将获得激烈的竞争，这将压低价格并促进创新。

x86被超越 (x86 is Outgunned)

ARM is coming at x86 from all corners. There is a wave of different companies throwing money at the problems. x86 is getting outgunned and like RISC workstations in the 1990s there simply is no quick fix or silver bullet to get out of this predicament.

ARM正在四面八方出现在x86上。 涌现出许多不同的公司向这些问题投钱。 x86越来越强大，就像1990年代的RISC工作站一样，根本没有快速解决方案或灵丹妙药来摆脱这种困境。

If they try to become another ARM maker they will have to sacrifice way too much profit. Hence they face the same predicament NOKIA faced when deciding on whether to become an Android maker or not.

如果他们试图成为另一家ARM制造商，他们将不得不牺牲太多的利润。 因此，他们在决定是否成为Android制造商时面临着同样的困境。

They didn’t want to take the hit, but the end result was that the whole company collapsed. Intel may risk the same if they double down on x86 instead of doing a strategic withdrawal and regrouping.

他们不想遭受打击，但最终结果是整个公司倒闭了。 如果英特尔加大对x86的投入，而不是进行战略性撤资和重组，英特尔可能会承担同样的风险。

那AMD呢？ (What About AMD?)

AMD enjoys a certain niche dominance on game consoles and gaming rigs. I don’t see any ARM guys coming after that market any time soon.

AMD在游戏机和游戏平台上占有一定的优势。 我看不到任何ARM家伙会在不久的将来进入这个市场。

Also AMD has the benefit of not manufacturing their own chips. They can utilize the big chip makers.

此外，AMD还具有不制造自己的芯片的优势。 他们可以利用大型芯片制造商。

Thus in many ways AMD may survive longer than Intel. But it is a tough call to make. Intel sits on a large pile of cash and can sustain quite a lot of losses for some time.

因此，在许多方面，AMD的生存时间可能比英特尔更长。 但这是一个艰难的决定。 英特尔拥有大量现金，可以在一段时间内承受大量亏损。

翻译自: https://medium.com/swlh/is-it-game-over-for-the-x86-isa-and-intel-5ce1b00fbd1

英特尔isa架构