Chrome OS Switches To "Freon" Graphics Stack To Replace X11

[Luke]

Can Chrome OS be used without a Google account?

I just bought both: a cheap Windows laptop and a Chromebook. The Chromebook is way, way nicer for my needs. The Windows laptop is a $300 ASUS Windows 8.1 laptop with dual core celeron, 4GB RAM, 500GB HDD for my daughter for MS Office school work + web surfing. She also has an iPhone which she does most leisure/social computing with. The Chromebook is a $330 Toshiba Chromebook 2 with a 1080p display, 4GB RAM for my whole family to share.

The Windows 8 laptop can run Chrome just like the Chromebook as well as a ton of desktop software that the Chromebook simply can not run... But the Chromebook is smaller, lighter, faster, more responsive, suspends/resumes faster, quieter, has better battery life, and uses everyone's existing google login rather than requiring a device specific login. My kids, wife, and myself use the chromebook much more frequently and enthusiastically than the Windows laptop.

Chromebooks run Chrome OS based on Linux over a hacked Coreboot, and support reflashing with upstream Coreboot. My question is this: can you use Chrome OS as delivered by an OEM without ever having a Google account? There are those like myself that would no more log in with Google than with the NSA, in fact I consider Google a privatized NSA and the more dangerous of the two.

Glad Chromebooks exist as a source of hardware compatable with both Linux and Coreboot, but that does not mean I would want a Google account to use the vendor-provided OS.

[uid313]

asdads

Chromebooks run Chrome OS based on Linux over a hacked Coreboot, and support reflashing with upstream Coreboot. My question is this: can you use Chrome OS as delivered by an OEM without ever having a Google account? There are those like myself that would no more log in with Google than with the NSA, in fact I consider Google a privatized NSA and the more dangerous of the two.
asd
Glad Chromebooks exist as a source of hardware compatable with both Linux and Coreboot, but that does not mean I would want a sdGoogle account to use the vendor-provided OS.

asd
You can login to a guest account.

[Luke]

Can this be done new from the OEM?

asd
You can login to a guest account.

Does anything block someone who just bought a Chromebook from running it using ONLY "guest account" logins, without ever having a Google Account?

[uid313]

Does anything block someone who just bought a Chromebook from running it using ONLY "guest account" logins, without ever having a Google Account?

That I do not know.
I know you don't have to logout from the guest account, and if you logout from the guest account all your history gets wiped.

But I don't know if you can setup the computer without ever entering any Google account since I never performed the initial setup of any Chromebook, I have just used one.

[Zan Lynx]

Actually, asm.js* is now already faster than native in certain cases, look at the top graph:

http://arewefastyet.com/#machine=28&view=breakdown&suite=asmjs-ubench

* remember asm.js is: C++ compiled to LLVM bytecode and transpiled to a Javascript subset. And then JIT-or AOT-compiled to native in the browser

Yes, it doesn't make a lot of sense, but the world doesn't make a lot of sense. Deal with it. :-)

Whenever I see this happen (native code being slower) it can usually be fixed by compiling the code with Link Time Optimization and/or Profile Feedback. It may also help to use Whole Program, with GCC.

JIT compilers have the advantage of seeing the entire program (LTO) and seeing it while running (Profile Feedback). Give those advantages to GCC or clang and the native code will be as fast or faster.

[Luke_Wolf]

Whenever I see this happen (native code being slower) it can usually be fixed by compiling the code with Link Time Optimization and/or Profile Feedback. It may also help to use Whole Program, with GCC.

JIT compilers have the advantage of seeing the entire program (LTO) and seeing it while running (Profile Feedback). Give those advantages to GCC or clang and the native code will be as fast or faster.

Frankly I would love to know what exactly they're defining as "Native C++" (also for that matter the optimization level given that -O0 gives about 1/2 performance) vs this Javascript code, but I see no links to that. Meanwhile the Debian Language Shootout doesn't support this idea. http://benchmarksgame.alioth.debian....g=v8&lang2=gpp And unlike this other site, we know the algorithms behind the benchmarks for the Debian Language Shootout are competently written.