You have built a time machine to 1980. IBM is hard at work with

>team up with Motorola and co-develop the 68k instead of using the 8088
>work with Xerox PARC, develop a GUI for Version 7 UNIX
>game port is standard on all IBM compatibles

>>60289644
No headphone jack

>>60289644
I keep going back to meet Jesus

First of all, I wouldn't ask Microsoft for the OS.

>>60290294
Seems reasonable

>>60289644
Use Motorola 68k
OS based on BSD or other UNIXes

File: 1494277057575.png (93KB, 400x394px) Image search: [Google]

93KB, 400x394px

Before going, I would write down a list of the most common complaints I hear from other people.
I would then go and hand them the list, saying "this is what people of 2017 hate about computing".
I would then tell them to develop a high-speed peripheral standard that's capable of upgrade and to bribe other companies to adopt that standard, in addition to forming a committee of companies trying to make standards.

>>60290385
>I would then tell them to develop a high-speed peripheral standard
And how would they test high speed of the peripheral standard? Even megabyte was not a small value in 1980.

Use an openfirmware like interface instead of BIOS
yes to 68k like everyone else has said
Get them to use QNX instead of Microsoft

>>60290410
Would USB be a reasonable idea to implement in 1980?

>>60290466
Thunderbolt is the way

>>60290385
Well fuck if we have time to prepare I'm loading kindles with all of the info I need taking 10 chargers with me and going down in history

>>60289644
there's not really much I would change about it, maybe proper winchester disk options and a better monochrome option

all of the retards ITT suggesting high-end 68k parts and graphical UNIX interfaces seem to have forgotten the last 2-3 IBM microcomputers before he 5150 that were utterly overpriced failures that barely moved any numbers, businesses wanted workhorses, not workstations

>>60290545
>high end 68k
The 68000 was pretty cheap
>graphical UNIX interfaces
Well Windows is a thing, that was my alternative to it.

File: dFMquxU.png (741KB, 1688x518px) Image search: [Google]

741KB, 1688x518px

>>60290585
>windows
>1980

>>60290597
I never said it existed in 1980.

>>60290545
Would modern computers be better if they were based on a risc architecture?

>>60290585
IIRC, Windows wasn't a thing at launch of the IBM PC.

Still, I'd have put UNIX on it and later offered a GUI software package for it as an option.

>>60290606
Modern computers are based on a RISC architecture.

Daily reminder that x86 has been RISC since Pentium Pro

>>60290616
But x86 is Cisc isn't it

File: AMD K6.png (172KB, 1387x958px) Image search: [Google]

172KB, 1387x958px

>>60290610
I didn't say it was. I'm time traveling from the future, remember?

>>60290625
See >>60290621
The x86 instruction set you'd write assembly for is CISC but the actual architecture is RISC.

>>60290625
>But x86 is Cisc isn't it

IIRC, for a long time, they are internally RISC with extra machinery to interpret old code for backwards compatibility.

>>60290648
So when I type "clang++ main.cpp" then the resulting a.out will have RISC86 instructions in the .text segment?

>>60289644
If I brought a cpu from today to that period, would they be able to reverse engineer it, causing a huge leap in technology advancement?

>>60290648
>>60290675
No, the processor takes CISC instructions and converts them to RISC ones

>>60290585
not in 1980-81, the 68k was a decently high-end chip that required more expensive supporting components and more complicated implementations than an 8088 while ultimately not offering too much of an advantage given the target market, IBM wanted to compete with Z80 CP/M systems (that PCs utterly blew into the weeds already), not high-end workstation vendors

and don't forget that Windows was absolutely irrelevant until OEMs shipped 3.0 en masse, early GUIs were shit with few exceptions
>>60290606
they would be no different, ISAs are meaningless to the end user and what paradigm you picked maybe mattered for 5 or so years at best, once the 386 came out it was just a matter of time and market focus, as soon as Intel turned their eyes to the high-end the shitty, overpriced and already stagnating RISC vendors were as good as done while x86 vendors took all the good ideas for themselves

>>60290610
they did have UNIX available at least by 1982-83 in the form of PC/IX

nobody bought it, it was trash

>>60290687
I'm not sure we'd be able to build 12nm transistors cheaply in 1980

>>60290697
Isn't that kinda... Backwards? Because then the processor has to do the optimization on the fly. Wouldn't it be more prudent to give the compiler direct control over optimization?

>>60290687
maybe a few minor enhancements, but by and large most of the "advancements" in personal computers today were already on the mainframe and supercomputer decades before

>>60290687
Figure out how it works, probably. Re-create it... probably not. A lot of the advances in CPUs have involved advances in manufacturing technique, so they'd still need to figure that out the hard way.

File: 1450634889168.png (233KB, 501x648px) Image search: [Google]

233KB, 501x648px

"Ditch that ISA bus and stick to MCA".

>>60289644
>Hi in the future we have 20 core cpus and TBs of ram!
>We have graphics cards that push 4k!
>Okay how do you build them?
>I dont know I just shitpost all day on an anime website.

>>60290749
This. Taking back advanced manufacturing techniques and materials science information would be more valuable.

>>60289644
That's a stupid comic. Electron drift is actually incredibly slow to the point where it can effectively be ignored. Electrons aren't just flowing through wires like a tube. It's equally correct to pretend you're dealing with positive or negative charges since all you're really dealing with is charge propagation.

>>60289644
Please use Rust.

>>60289644
According to IBM's in-house history the real reason they went with Intel rather than Motorola was that Intel's dev tools were better.
The IBM board had given them only one year to bring their 3rd try at a PC to market. Therefore speed was of the essence.
The better architecture and faster/wider 68000 CPUs didn't matter as much as a shorter time to market.
The feeling was that the board wanted the project to fail; there were a lot of people who thought it wasn't the sort of product that IBM should be involved with.
They were proved right when IBM failed to make dollar one with the PC, and lost control of it to Microsoft.

>>60291329
Yup. And even then it'd take a little while for them to build the tools to build the tools to build the tools to build chips up to the level we've got.

>>60291367
And honestly, the reason IBM-compatible PC caught on as the de facto standard isn't that it's better than the other things around at the time or even actually any good. It's because it was more-or-less good enough, and possible to clone since it was thrown together with entirely off-the-shelf bits.

>>60289644
Just tell them to scratch everything and use Blit.
>>60289936
there is already a very good gui for version 8 unix.

>>60289644
Copyleft everything.

>>60290728
>Isn't that kinda... Backwards? Because then the processor has to do the optimization on the fly. Wouldn't it be more prudent to give the compiler direct control over optimization?
Intel tried something pretty similar to this with Itanium, they couldn't get their compilers to be good enough soon enough
By the time the compilers were somewhat ready the AMD64 parts were faster, without the need for new code

id take back the arm documentation as well as c++17 standard.

>>60289644
Tell them to add a longer bipeline.

Convince Linus Torvalds to get in my van (he'd be 10 at the time) so that Linux the kernel never happens, therefor the GNU project has to finish its kernel on its own. It might take a few years longer than it did in our timeline, but not only will we still have a fully free OS, we will have one completely maintained by people who actually care about freedom.
The GPL will be enforced such that nonfree device drivers will be brought into compliance instead of tolerated by a Swede who cares more about adoption than freedom, when GPLv3 is released (I won't warn them in advance about the problems it fixes; I'm already messing with the past too much) the GNU system will fully embrace it to fight against tivoization, software patents, et al. There might not even develop an open source "movement" to water down the issues of freedom so that morals don't come into it. "Open source" has lead the software culture to look at the practical issues *instead* of ethical, moral issues. There's "open source" software which has advantages X over "closed source" software, but if the latter still manages to deliver a technically superior product, even slightly, they will choose it instead. Without "open source," there will be the evil nonfree software subjugators who trample on their users' freedoms, and the only ethical developers who free their software. Framed like that, people will consider their freedom important, and we'll see a very different 2017.

>>60289644
GNU PUBLIC LICENSE EVERYTHING. Free software for a free future!

>>60294035
Linus Torvalds is from Finland

File: cetun.jpg (52KB, 620x407px) Image search: [Google]

52KB, 620x407px

Make it ternary

https://en.wikipedia.org/wiki/Ternary_computer

>>60294169
I was hoping you'd say "Linus Torvalds is Finnish" so I could reply "Yes, and bankrupt, too."

I know you think you're being really smart with that keyboard controller but cut that shit out.

>>60290728
Let the compilers optimize CISC. They get to use all the features.
In the processor, the assembly gets broken down to RISC micro instructions. That way, instead of supporting a ton of CISC instructions (and remember, the processor needs to contain a controller, i.e. state machine to handle all of those), you reduce the state-space by breaking up complex instructions into simple ones.
It's easier to optimize and predict branching that way as well, and it's more friendly towards out-of-order-execution.

>>60294240
It's bad and you should feel bad.
There's a reason it failed and you can read a bit more than the wiki article about it and apply some engineering judgment to understand why.

>>60289644
I go back in time and tell them that EPIC is the future. How badly did I fuck up computer development?

>>60289644
I'll just bring Silberschatz' book on operating systems, mention the vital importance of security in systems, and leave them to it.

>>60290294
T H I S
H H
I.....I
S......S

>>60294035
>Framed like that, people will consider their freedom important
people will never give a shit as long as the majority see their computers as tools instead of lifestyles as you do

>>60297787
>this hammer comes with a shitton of restrictions on use
>but that's okay because it's just a tool :^)

File: 1485804630642.png (250KB, 700x490px) Image search: [Google]

250KB, 700x490px

>>60298361
>b-but what if we were talking about a hammer instead
lmao

>>60289644
Bring an Android Phone all the way back to 1980 and let them study it.

>>60298838
IBM didn't make toys back then.

>>60298915
Hilarious.

>>60299112
90% of the shit your facebook machine does was already implemented in mainframes, what held engineers back from bringing those features to desktops was the cost of doing so with older manufacturing processes, not because they were a bunch of shit-chucking apes like you think they were.

I'd tell IBM to go fuck themselves, and then go to Sun's founders and tell them to make SPARC and Solaris not fail. I'd also tell them to never invent Java.

>>60290466
Yes

>>60298828
>computers are tools
>okay, let's compare the situation to other tools
>N-NO

i would just show up naked with a horse mask and pony tail butt plug and run around screaming PLUG AND PLAY USB DONGLES, then right before i get arrested i whisper instal gentoo in one guys ear, then slip him a copy of gentoo on a plug and play usb dongle.

>>60299821
not all tools are identical in their usage and expectations
your shitty analogy is like buying a hammer and whining that you can't use your nanoforge to freely copy/distribute their product and optionally replace the handle with a dildo, the majority of people wouldn't care for this because they buy hammers to nail things and there are already dildo-hammers available on the market if they want those instead

>>60289955
I emitted a hearty guffaw

File: 1490534091414.jpg (553KB, 1536x2048px) Image search: [Google]

553KB, 1536x2048px

>>60289644
Give me an ibn 5100

>>60299695
Is a cell phone really supposed to be anything other than a telephone + a facebook machine? Get over yourself.

>>60300159
No, that's why it's funny when phonetards jerk off to them like they're some kind of profoundly revolutionary device.

>>60300093
Nobody will stop you from using your hammer however you are capable. They won't threaten to sue you for replacing the handle with a dildo, and if you find a way to replicate hammers near 1:1 perfection, no one will stop you from doing that, either.
The analogy is shitty because no other tools have such retarded restrictions on them, so saying people accept them because they're tools is completely inaccurate. They accept them because they don't know any better.

>>60289936
fpbp

>>60300176
>The analogy is shitty because no other tools have such retarded restrictions on them
a computer is no other tool, but even then, only restricted by what is computable
what's restricted are certain ideas and implementations of prepared instructions you can compute with, that you are not obligated to utilize if you wish to maintain or modify them yourself
>They won't threaten to sue you for replacing the handle with a dildo, and if you find a way to replicate hammers near 1:1 perfection, no one will stop you from doing that, either.
this line boils down to "muh feelings"
you are too incompetent to build your own hammer, and too meek to influence its architects, so you demand absolute control over it yourself
>They accept them because they don't know any better.
and this pseudo-intellectual smugness is why you will continue to fail, nobody wants what you preach because what you preach is software that is more concerned with your feelings and sticking it to your bogeymen than being good at its genuine purpose, and you blame your failure to compete on lesser beings who "just don't know better" (even though thousands upon thousands of companies know of free software yet continue to intermingle it with proprietary software as well, because their job is their job, not jerking off to trivial ethics)

your ideal reformed GNU would be no more relevant than GNU/Linux is today, because the human hivemind tends to give credit where credit is due, not where it is desired

>>60289644
32bit protected mode interface for switching the video mode.

>>60300436
I'm trying to form a response, but what you said either is nonsense or just not relevant to what you replied to.
>a computer is only restricted by what is computable (and certain ideas and implementations but it's only restricted by what's computable)
>it's an opinion that people won't sue you for using a hammer without restriction
>"this pseudo-intellectual smugness is why you will continue to fail" doesn't boil down to "muh feelings"
???
Unless you can make yourself more clear, I can't reply to any of your points.

>>60300570
>I'm retarded so I'll just call it nonsense
and that's that, my work is done
>>a computer is only restricted by what is computable (and certain ideas and implementations but it's only restricted by what's computable)
you can't read
>>it's an opinion that people won't sue you for using a hammer without restriction
you can't read
>>"this pseudo-intellectual smugness is why you will continue to fail" doesn't boil down to "muh feelings"
it's not a matter of people rejecting you because you're a smug pseudo-intellectual, it's your being one that causes you to formulate shitty answers that don't solve the problem, while telling yourself that those who don't accept them simply "don't know better", thus ensuring that, like Stallman, you will continue to force the same shitty ideas and concepts forward and fail while those like Thorvalds choose to recognize reality for what it is and work within that framework to succeed

>>60300692
>y-you read it wrong

>>60300738
you still can't read

>>60290728
CPUs already do all kinds of trickery and out-of-order execution anyway in the name of speed. It's quite complicated but it's also pretty much a standard feature on any high-performance CPU.

>>60289644
"Make sure normies can't use them"