CPU thread. How much of this stuff do you CompSci folks learn in school? Seems interesting.
Networking/Security crew checking in, so obviously this isn't a part of my mainstay; but it sure is neat.
What would you CompSci folks recommend as far as books and such go?
https://www.youtube.com/watch?v=cNN_tTXABUA
http://www.visual6502.org/
Bormp.
>>51936078
CPUs are rather electrical engineering. Comsci students only care learn about programming and shit.
>>51936375
Well that's not true. There is barely any programming courses in CompSci, some don't have any.
CPU die shot thread?
>>51936424
What kind of Comp. Sci are you taking OP? Are you sure that you don't mean Computer Engineering?
>>51937444
>>51937463
>>51937344
I don't take Comp. Sci. at all but I'm not sure what Comp. Sci. programs you are thinking of that have a big concentration on programming?
>>51937344
He's probably the same guy who previously argued the exact same thing and to whom a programming course means it's a course in learning a new language or API, not actually programming something new.
Comp. Eng. myself, but at my university a description of the CPU similar to the one in the video posted by the OP is brought up in the first intro course. After that there's an undergrad course in microcomputer hardware that goes even more in debt with stuff like registers,caches, superscalarity, slightly more advanced binary math, types of memory (SRAM, DRAM, EPROM, EEPROM, etc.) with a lab where you have to program an Atmel ATMega based microcontroller in assembly language for the CPU, but I'm not sure if it's mandatory for CS (for Comp. Eng. it is).
>>51936375
Daily reminder that comsci is not a degree in programming. Computer organization is very much a comsci subject.
>>51937512
The ones I'm (not the guy you're replying to) thinking about have quite a lot of it. At my university on the undergrad level there's two basic programming courses (one in Python, one in Java), two courses in data structures and algorithms where both have you implement the stuff (both in Java), two courses in systems design (both with programming in Java), a course in programming language paradigms (with programming in Prolog and Haskell), and an operating systems course (with programming in C).
So under the duration of the undergrad curriculum at my university you get to at least get to do some basic programming in Java, Python, C, Prolog and Haskell.
>>51937493
>>51937803
more?
Sparc m7
>>51938774
>>51936078
>How much of this stuff do you CompSci folks learn in school?
The focus of my CS degree was more on algorithms, programming language theory and software engineering practices, but I did take some extra undergrad courses in digital logic and close-to-metal development. Those covered many areas of CPU design, too.
I wish there would have been more courses like that, but my faculty seemed to prefer handholding people with no clue about computers or writing simple programs.
>>51936078
>How much of this stuff do you CompSci folks learn in school?
Depends on the school.
Are you talking about Akron Community or MIT?
>>51936375
>>51936424
>>51937344
>>51937544
>>51937580
>>51939395
>>51940673
I honestly don't think that public 4yr universities would want to actually sit down and exlain to wouldbe code monkeys how the underlying archetecture of a processor works. In todays realm of cybersecurity, once you can exploit things at the harsware level, then you could potentially cuck everyone's system.
So yeah to reiterate my question, why would public universities want this info to be easilly accessable? They just tell you it is "magic"
>>51941278
You're wrong. I'm at a university in the 40s (e.g. not a great one) for rank and to get into the major, they send us through discrete math and computer design. This weeds out all the maek gaymes and code monkeys. Year 2-3 is compilers and operating system theory. We do MIPS stuff though for the cpu class, so it's just theory about pipelines, and constructing a cpu from alus and multiplexors.
>>51941278
We also talk about malicious software and the implication of memory overwrites on a cpu and how people who can write past a buffer can essentially do stuff at a hardware level.
Basically, you don't know what you're talking about, and comp sci is not code monkey school
>>51936078
Read this OP
https://drive.google.com/file/d/0B7doMTp4mbo3cm81RUNuSlpDMEE/view
Wat if your brain is actually a cpu created by god
>>51941278
>cuck
let's see if I can do this
>>51941363
You are talking about bufferover. The 28 backspace meme is an example of this
>>51941338
So what can you tell us about processor architecture? Does the discrete maths tell you any technical details other than to convet to binary/hex/decimal? Ok agree that csci weeds people out within the second year, and algorithms will. My point is that there is no emphasis on understanding processors. No body will teach you how to bitbang stuff, or write/interperate assembl.
>>51943192
Discrete math is more about logic proofs and structural induction. We did learn MIPS, quite a bit of it, although I don't know why they didn't teach us x86.
As far as processor architecture, I can tell you a lot about MIPS. What do you want to know? I can tell you how it pipelines, and where each bit of each wire goes and what it's used for.
>>51936078
>How much of this stuff do you CompSci folks learn in school?
A standard CS curriculum would have one or two courses on processor design, often rolled together with a quick and dirty on ASM programming because it doesn't make a lot of sense otherwise. Honestly there are a lot more interesting problems in CS than processor design, compiler optimizations which are closely related are a lot more fun. If hardware is really what floats your boat and you don't enjoy higher maths then EE or CE would both be better picks (that that they're particularly math-light but a sense for the mirth of formal systems goes a long way in (good) CS)
>>51943300
MIPS is a clean architecture. X86 is the worst architecture to use that's commonly used today.
>>51943402
I agree strongly with this, but, can you have a 5820k using MIPS?
>>51943300
Saus plz
time to git gud
Anything outside of the realm of google?
>>51936078
Would you like Java or fries with that?
- all CS grads since 2001
SELECT ALL IMAGES WITH FRENCH FRIES!
>>51941278
are you indian? the point of comp sci is not to be a code monkey...
some go into that because it's a lucrative field but thats all. and most programmers arent even code monkeys anyway.
we know how processors work in great detail. well to be fair i wouldn't be sure about the architecture of the one in the pic, since we did mostly old processors and that one seems to be a multi-core.
i could know even more about them but i went into software concentration for the last year of the major
If the modern cpu's produced in nano scale, is it nano thechnology ?
>>51939437
>>51944705
well yea you could call it that m8
but i think nano technology refers more to the possibility of having nano sized actuators, ie robots that could be more useful due to that kind of size. the processor only "thinks"
>>51936078
thats hardware. CS is more theoretical math, programming languages and operating systems
go study EE or CE u peasent
>>51943706
I recommend reading computer organisation and design, and discrete mathematics and its applications. They're both able to be torrented
