When will ARM enter the desktop processor market?

When will ARM enter the desktop processor market?

Other urls found in this thread:

howtogeek.com/309119/what-is-windows-10-on-arm-and-how-is-it-different-from-windows-rt/
en.wikipedia.org/wiki/Rosetta_(software)
twitter.com/SFWRedditGifs

Windows 10 on ARM Soon(tm).

When it becomes as poweful as a core 2 duo or core 2 quad.

Never. Do you expect people to start buying ARM cpus for their desktops and realize 99.9% of their software doesn't work?

Related question.

When someone runs linux on android, is it able to run all applications like gimp, pidgin etc or it is only for showing how big your e-penis is?

It already did? Everyone I know uses GNU/Rasperry Pi 3 as their main machine

x86 isn't going to last forever user.
They aint making it anytime this decade but they are getting pretty good market penetration in the datacenter and the software eco-system gets stronger by the day with Mobile technologies leading the way.

>is it able to run all applications like gimp, pidgin etc
Obviously. But android linux chroots usually don't have any hardware graphics acceleration, so it's still slow

server market is where ARM may well dominate

Hows life in Shenzhen?

>implying

Yeah no, EPYC delayed that a bit.

Desktop? meh never. Maybe they will move up to mobile seeing as they are already in tablets and phones, its not a big stretch to laptop

around 87'

ARM can't scale up anymore than x86 can scale down to meet it.

There's no reason for them to ever enter the desktop market.

Idk but my ARM will enter your MOM's pussy tomorrow. XD

Once Microsoft releases ARM builds of office. Then cheaper business workstations can be made with ARM versions of Windows with the applications that businesses rely on.

Sure LibreOffice and a lot of business software on Linux is open source and can easily be compiled for ARM. But we all know LibreOffice is shit and no non-tech businesses give normal employees Linux machines. At least in burgerland

Not until someone other than Apple builds a respectable ARM core.

Or maybe Apple will put it in their laptops and later their desktops.

>implying the desktop market will consist of more than a web browser + some basic office programs that could be emulated in a few years

Desktop is dying from that scene and is replaced by phones and tablets.

Desktop is enthusiast and office tool as of right now.

>ARM "core"
>Apple

That's why Microsoft wants to emulate X86. And because Intel trembles in the thought that x86 will be replaced with ARM, they threatened MS with suing if they do that.

Wait for RISC-V arm is doa on desktops and laptops

NEVAR
RISC V is next

RISC-MEME is a meme.

fucking animu posters

...

Yes but the google jew wants to shill shitty apps converted to desktop formfactor instead of using an entire existing linux ecosystem because they can't get their shitty revenue that way.

Xeon D killed the first ARM attempts. AMD will kill the second and with it the last attempt.

Hopefully never.
ARM ecosystem is proprietary cancer with no standards whatsoever.
Also x86/AMD64 is a very good architecture.

>AMD will kill the second and with it the last attempt.
AMD is making their own ARM CPU for the server market, dumbass

Apparently traditional software is less of a problem than older hardware drivers.

>While Windows 10 on ARM can emulate traditional desktop applications, it won’t be able to install hardware drivers written for traditional x86 or x64 Windows operating systems. It will need ARM versions of those hardware drivers to support different hardware devices.

howtogeek.com/309119/what-is-windows-10-on-arm-and-how-is-it-different-from-windows-rt/

Kind of a short term problem though.

Usually. The advantage Linux has is because everything is open source and uses a common set of libraries, running most software on a different architecture is just a recompile.

>Kind of a short term problem though.
Unless you want to install older hardware such as printers

>>x86 isn't going to last forever user.
It's lasted since the late 70s, and survived several attempts (including more than one by Intel themselves) to kill it off. Why can't it last forever?

>(including more than one by Intel themselves)
Fucking this. People seem to forget.

x86 will last forever

They are going to have to significantly change the way their licensing works if they are to compete with ARM.

They've strong-held the market for this long because the sunk cost and barrier to entry is so high and they've created restrictive patents and technologies to deter startups.
Now that some of the biggest foundries in the world can can make you an custom ARM chip that performs well x86 is becoming more irrelevant by the day
.

It's not going to happen over night, i say 10-20 years and the tides may well turn against the current x86 model.

>Never. Do you expect people to start buying Intel Macs for their desktops and realize 99.9% of their software doesn't work?

go be a dumb cunt elsewhere

>Now that some of the biggest foundries in the world can can make you an custom ARM chip that performs well x86 is becoming more irrelevant by the day
But ARM still doesn't perform well relative to x86 and most presumably never will. As soon as you get into double digit TDP territory, the performance increase drops off a cliff with current ARM designs. This is also ignoring the chip specific quirks of each and every ARM chip basically necessitating a whole lot more software trickery than with x86.

Even if that falls into place, we're still looking at BIOS/UEFI issues as there really doesn't seem to be much in the way of cooperation there. We also haven't touched on peripherals, which are still an issue if ARM is to move into the laptop market never mind the desktop and workstation markets.

Yeah, all this is possible, it's just not feasible. It would require an incredible effort across the entire industry, which just won't happen.

What even is wrong with x86 that it needs to die?

Ignoring the diminutive third party software library available for Apple computers at the time, I'd also like to point out
en.wikipedia.org/wiki/Rosetta_(software)
I'm not even that guy but you're not in a position to insult the intelligence of anyone else.

Mainly that it offends nerds' sense of elegance. It's a sprawling, messy instruction set that's been extended many times and features a lot of legacy cruft. And Intel is well-known for sharp-elbowed business practices, so a lot of people are rooting against them.

Are you for real?

Are you telling me, that small, pre-iPhone Apple of 2006 was able to make the transition by offering a half assed emulation layer that run in *userland* and outright didn't support most pro appications like Photoshop, Final Cut, etc., while stronghanding 3rd parties to develop native software, but for some reason huge Apple of today can't?


When the transition happens, nobody will bat an eye, just like when Rosetta was dropped in Snow Leopard.

The new greatest and latest will be developed natively for ARM, big software vendors will rush to capture the marketsare by writing (more like optimising a bit + recompiling) for ARM, and an emulation layer will be offered for compatibility and peripherals.

ARM's potential is not even in the same league as x86. After barely 10 years of serious developent, it caughtup with i3s and i5s of yesteryear. All this in a passively cooled, thermally restrticted design, that of an iPad.

x86 overstayed its welcome and this happened bacause of Intel monopoly. That's all there is to it.

You don't know many people then

Apple could just up and switch architectures like that because they're a consumer company. Microsoft and the PC world can't, because consumers are a sideline these days, it's business where they make their money. Businesses do not like upgrading software. And they aren't at all swayed by any arguments of a cleaner instruction set. They'll ask "Why should I change? What would all the risk and disruption get me?" The answer is not much - ARM is power-efficient because it's slower. It's advantages go away when you try to push it up to speeds comparable to desktop x86. The folks in the datacenter also have a bunch of x86 software that they don't want to change and can't just recompile. And a whole new architecture means one thing to them: Complexity. These servers aren't like those servers, they're totally different and run different software. For business to swallow a switch to ARM it's not good enough for ARM to be as good as x86. It has to be better. It has to be A LOT better, because changing has big costs. Right now ARM is better in one niche (low power - that is, mobile devices and IoT crap) It's not better in any of the places desktops and servers play in.

The Year of the ARM Desktop will have arrived when I can go to any online parts dealer and buy an ARM motherboard in a standard form factor from any of several manufacturers. Into which I can plug ordinary off-the-shelf peripherals and expect them to Just Werk with no emulation layer, and onto which I can install any OS stock, no special versions, no workarounds, no hackery specific to this board or that board.

That day not only is not here, it's not even on the horizon.

>RISC V
>not OpenPOWER
yall niggas are delusional