IT'S OVER, X86 AND ARM ARE FINISHED.
MIPS HAS WON THE CPU WAR.
http://venturebeat.com/2015/09/03/chinas-loongson-makes-a-64-bit-mips-processor-that-can-run-x86-and-arm-code/
>>52296493
Who?
SLIDE IT GOYM
>MIPS
>2016
>can also execute x86 and ARM code using LoongBT, a binary translation technology for Linux.
It's literally a software side emulator. Basically QEMU.
>>52296519
Dragon chip
Linux distributions that work on Loongson:
Debian, specifically their mipsel port
Gentoo Linux, work in progress
Red Flag Linux
Mandriva, since September 2007
gNewSense, work in progress
CLFS, MIPS 32-bit,64-bit,Multilib
UTUTO, work in progress
EVEREST Linux
Non-Linux operating systems:
NetBSD, evbmips/gdium
OpenBSD, cvs repository work in progress
>>52296556
This isn't new either, it's been around since 2009.
http://www.computer.org/csdl/mags/mi/2009/02/mmi2009020017-abs.html
The reason x86 and ARM CPUs don't need instructions for aiding emulation of opposing ISAs is because they have market adoption.
IT'S MOOPS!
>>52296493
So will that MIPS programming I did in school be of some use now?
>>52296493
MIPSfags confirmed for being as delusional freetards.
>>52296788
No, because the x86 emulation was implemented by way of new instruction set module.
>>52296493
That's comparatively not very interesting. Now RISC-V, that's might become a real threat to ARM and mobile Atoms.
https://en.wikipedia.org/wiki/Elbrus_2000
Good give them a legit license, unless Iintel is skerrd. Another popular x86/arm CPU would mean big things for consumers.
>>52296819
This
>>52296996
>chinese
>getting a legit x86 license from Intel/US govt
lol
>He added, “You probably know the old line about dancing bears—they aren’t remarkable because they dance well, but because they can dance at all. I’ve seen so many ‘dancing bears’ in the CPU business that I no longer find them remarkable. Show me one that dances well, and I’ll feel differently.”
Give this guy a medal
>>52296996
Why do people think x86 is still patented, it's not. AMD's 64bit extension of x86 is actually a bigger deterrent to x86 market entry than anything Intel has.
>>52297132
Because intel and AMD are the people that grant licenses to make other 64bit cpus, y so pedantic
>>52297155
He didn't say AMD, he only said Intel. AMD is just as bad as Intel. If more players entered the x86 market AMD would be even more fucked than they are now. AMD needs the duopoly more than Intel does.
>>52297132
x86 alone doesn't get you very much because that leaves out the SIMD extensions. If you don't have access to those you're making something roughly equivalent to an i486-i586
>>52297233
MMX is expiring this year. 3DNow and SSE1 are going to expire in the next few years. AMD64 doesn't expire for another ~6 years.
>>52297374
no clue how that number got changed, was meant to reply to this post >>52297235
>>52296493
>x86
>finished
Hahahahaha.
>>52296996
Even such a long standing figure as AMD can't do it, what makes you think some irrelevant chinese company would be able to? Even if they could get a big chinese OEM like Lenovo to sign on, it would more than likely only see use in the entry level and most consumers would look at that shitty chinese branding and view it as a cheap, sub-par knockoff, which it probably actually would be.
Even in an era when there was tons of competition in the x86 market, Intel always at least maintained a rather comfortable plurality in the market, often an outright majority.
MIPS is pretty cool. I've got a CI20 and am getting a CI40, also want to pick up a Lemote Yeeloong if I could find one anywhere
>>52297459
Yeah, most of /g/ is too young to remember WinChip, NexGen and Cyrix.
>>52297604
I remember Cyrix but WinChip? Come on, nobody remembers them.
>>52297671
>be via
>own winchip
>own cyrix
I think i'll base my product line off winchip.
>>52296493
So meta. So trans.
Humm. This sounds familiar...
https://en.m.wikipedia.org/wiki/Transmeta_Crusoe
Did we time warp back fifteen years, or are they just hoping nobody notices this had totally been done before?
>>52297604
Don't forget
>VIA
>Rise
>Transmeta
>IBM
>NEC
>Numerous second-source vendors making Intel clones under license like Harris, TI and ST
>>52297805
>Don't forget
>>VIA
I've owned a laptop with one of their chips. It's been a long time. I thought long enough. But I still can't forget.
Why can't I forget that nightmare?
Why?
For the love of God why?
nice hopefully these chinese companies develop cpu's to rival intel and sell them for 100$ each it will destroy both intel and amd
>>52297748
It's more like FX!32. It would be running a MIPS version of Linux. The emulation is only for running legacy software.
https://en.wikipedia.org/wiki/FX!32
The Crusoe is meant to be used as an x86 and the real instruction set is hidden.
>>52297748
I wonder if we'll ever see VLIW CPUs in the wild. They seem so cool
>>52296552
Fuck I love this show but I forgot the name.
>>52298103
They've come and gone.
>>52297837
I want a VIA just for the sake of having one, but they're ridiculously expensive for what it is here in aus.
>>52299189
Can buy a full VIA board with a 1 GHz C3 off of ebay for like $20 right now, dunno how the shipping to the auslard kingdom would be though
>>52299115
I remain hopeful that we'll see more of them. When things get too complex for RISC we'll probably see a return of them or something like them.
>>52299323
I can see something like the Itanium coming back if they ever figure out how to properly implement the advanced compilers that the architecture hinged on.
Based Loongson making my Comp. Architecture classes relevant.
>>52299091
Mission Hill. The entire series is available on YT, Hulu, and I think Netflix.
>>52299643
Thanks dude. I was completely blanking there
>>52299292
1.2ghz via eden with no sata and ddr2 ram.
63 dollars with shipping.
http://pages.ebay.com/link/?nav=item.view&id=281901549581&alt=web
>>52299753
No problem. It's a pretty great show, too bad MTV dropped their animation dept. There was some good stuff a while back.
>>52297076
srsly how hard can it be to reverse engineer x86?
>>52299867
For people scared of lawsuits against a tech giant like Intel, I'd say it's pretty hard.
>>52297805
>Transmeta
my god I can't think of a more worthless cpu line.
>>52299905
>China
>afraid
ok kid
>>52299795
http://www.ebay.com/itm/VIA-EPIA-CL10000-LVDS-C3-1-Ghz-CPU-Dual-Ethernet-LAN-Quad-RS232-Mini-ITX-MOBO-/391354072233?hash=item5b1e853ca9:g:bckAAOSwnDxUkLu0
This was the one I was looking at, didn't notice the free shipping either.
If all you want is just an example of a VIA chip this might be fun to pick up and play with, this one looks like it's for low-power server applications judging by the dual NICs, and it would probably still be nice in that role.
But some FSF distro on it or something as an ultimate botnet-free communication system, maybe. Or just a seedbox.
>>52299867
Any retard can make an x86 compatible chip, making a fast one is another thing.
>>52299848
Holy shit I was thinking about this show the other day. My FAVORITE scene is when they go bowling and he says "Now that kid, he should be ashamed." Then the kid with the orange hair rolls his bowling ball right into the robotic pin collector, sets off the smoke alarm, enables the sprinkler, then it shows his face all confused. Lol don't know why that always used to kill me.
>>52299905
kek china has been ripping off multi-billion dollar companies for years intel is nothing to them
>>52299941
If they want to actually export it they will be. Shit, even if they could get around it it would more than likely be a failure.
Not even counting the massive amounts of capital you'd need to throw into a business to get it remotely ready to produce the chips, you'd have to find ways around whatever bewildering amounts of patents Intel and AMD put on pretty much every modern instruction set/feature addition to the x86 line.
In the off-chance to you get around even that obstacle course, you'd be spending years catching up to Intel in performance, something not even AMD has thus been able to do. In the meantime, you're stuck servicing the low-end markets that AMD has practically locked down. And selling that shit to the west? Fucking forget it, even if you could get a big OEM to sign off on that, no consumer is going to trust some shitty chinese no-name chip that will likely be even more expensive than comparable AMD and maybe even Intel options due to their low volume output.
>>52300022
The main thing is, if chinkbrand electronics rips off x86 CPUs their products will be banned from the US. Like Huawei
>>52300109
even if they are china is a huge market. if intel loses china they will have to lower prices to compete which is a win/win for us.
>>52300147
But they won't, not unless they manage to produce something actually competitive.
I still want a 4 GHz Alpha. :^(
>>52299189
You know the allwinner A20 boards you get shall as shit from china now? They will run rings around said via board. Admittedly they're newer, but they're lower clocked, and designed for an embedded market instead of the laptop. Htpc market like via was.
That via machine I had was the only thing sonce 2002 I had graphics driver issues with in Linux. And it wasn't even the 3d portions that sucked (that did such though). The basic 2D operations were even bad on it. A first gen EEEPC laptop blazed past it.
>>52300333
He wants one for collector's purposes. Same reason a person might want to own a WinChip.
>>52300401
>Same reason a person might want to own a WinChip
I've never heard the euphemism, "collector's purposes", used for painful masturbation before.
Jokes aside, I may have a working box with a winchip & trident graphics somewhere.... Mostly I used it to make myself feel better about only being able to afford the AMD socket 7 processor. It was a reminder of how much worse things could be.
>>52300499
the WinChip was kind of cool to be quite desu with you
make it simple as possible, clock it fast as possible, made a pretty decent chip for low-end business applications as long as you weren't doing anything big on floating point calculations
VIA's x86 offerings are still shit out by the same design team
https://en.wikipedia.org/wiki/OpenPOWER_Consortium
desktop devices when?
>>52296597
Of course it runs netBSD
You know, I'd actually like to try out one of the loongson CPUs, but the problem is you can't buy them. The only consumer device that uses them is the lemote yeeloong laptop, and those are basically impossible to acquire now.
>>52300685
You wish.
(me too ;_;)
>>52301327
Doesn't Lord Stallman use some sort of MIPS shitbook?
>>52301711
He had a lemote at one time. I think it got stolen by a BR or something. Now he uses a thinkpad with libreboot for the bios
>>52297080
I thought dancing bears were remarkable because their dicks were hanging out
>>52296493
>>52301619
>>52300685
Hope for Sonoma.
>>52296556
>ITANIUM
>>52296493
PERFECT N64 Emulation Incoming
Its Happening
>>52296493
was there ever really any doubt?
mips has been leading the charge w/ efficiency AND compatibility since sgi took over the designs,
i'm sure it's been whispered around /g/ how god-tier sgi was in its glory days.
mips only stood to get even better since then
>>52305032
MIPS is owned by PowerVR and Apple is a major shareholder in PowerVR so it's automatically shit.
>>52305068
mips is overseen by mips technologies, a division of imagine technologies you dunce.
sure powervr is also owned by imagine technologies, & apple does indeed own shares in powervr, but powervr produces graphics chips, not cpu's.
besides, why would apple invest in mips when they've already spent so much trying to go w/ arm architectures for ios?
>>52305125
besides, apple's hardware interests have never had anything to do w/ how good the hardware actually is.
how could anybody be so ignorant?
>>52305032
desktop MIPS under SGI like every other non-embedded RISC was stagnant as fuck after the dotcom bubble blew up though, any prospect that arch had in the high-end went down with them desu
>alien technology-tier workstations will never come back
>>52305249
i will admit, that is fair.
i'll stand by my statement that mips still is pretty much badass though.
& arm is pretty nice
>>52305296
yeah the arch seems pretty neat, and judging by the amount of SGIfags still daily driving their systems they seem to have aged way better than most other RISC workstation shit
>>52296667
IT'S MIPS. THERE'S NO MOOPS.
>>52305341
yea, & besides intel has run into a physics barrier, their cpu's literally cannot get any smaller or go any faster, it is physically impossible.
& if they want to keep shelling out progressively better cpu's, they will eventually have to abandon cisc, & if they were to do that, i'd recommend that they adopt mips
>>52305384
as an sgi fan, i must say that i deeply regret not owning an sgi machine.
that will have to change w/in the next 5 years
Anyone got info on HD mini vs display port?
>>52305511
Hdmi*
the question is why though? architecture is only a problem for plebs who use proprietary software or shit code with inline assembly
>>52305568
>t. some guy who's only used x64 and shitty arm sbcs that spoonfeed him pre-compiled packages and binaries
>>52305581
actually nevermind I misinterpreted you're funpost desu, I'm thinking more of the shit you have to do when compiling software between different operating systems, don't think I've ever had to fight with compiling from source on any of my itanic/sparc shitboxes as long as I run linux on them
>>52296597
Tell me when it runs Faildows7 64-bit or Ubuntu. Otherwise, fuck off , china.
>>52305607
what? targeting different systems ain't hard either
>>52305612
@ this point, it is conceivable.
that's probably what they're trying to do anyway
>>52305635
i mean, i'm guessing all they have to do now is reprogram the bios to see which architecture the system needs to run & reconfigure itself, then work out some bugs, & then they're golden
>>52305249
>high-end
Well, there was a PS2 cluster built after that.
>>52296597
Is that chip fast enough to be used for a home desktop?
>>52296493
Can it run Crysis?
>>52306920
Crysis is mips.
