Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

When teaching computer architecture, why are universities using obscure CPUs?

academia.stackexchange.com

Post reply on HN
Print view View original

61–70 of 117 posts

Re: When teaching computer architecture, why are universities using obscure CPUs?

#61

Post by jumploops »

DeathArrow wrote: post #55
jumploops wrote: post #41

Less optimizations, simpler instruction sets, easier to implement. Going from transistors to C is like petrol to driving. Do you want to learn how a car works by building some 2024 safety and tech enabled car? Or would you rather start with something that resembles a 60s Beetle but with even less bells and whistles?

But aren't there real architectures that are simple enough? Z80, MOS 6502, Pic, AVR?

Real hardware handles real trade-offs. For example, there are hundreds of instructions in Z80 to support various quirks of the implementation (like the duplicate register file).

Sometimes less is more, and it can be freeing to know that your code compiles to a CPU with just a dozen or so instructions.

It’s the same reason most universities don’t teach git, UNIX, and other industry tools. Sometimes computer science is best seen in its purest form. Same with computer engineering.

Re: When teaching computer architecture, why are universities using obscure CPUs?

#62

Post by simonbarker87 »

Universities aren’t corporate training grounds.

The job here is to teach you the concepts involved in CPU architecture and using the most general purpose architectures possible that have been developed over millions of person hours to cover every task possible would be a very inefficient way to do this.

Re: When teaching computer architecture, why are universities using obscure CPUs?

#63

Post by moonchild »

inkyoto wrote: post #45
rdtsc wrote: post #31

Because my professor hated the x86 architecture. At least I appreciated the honesty. We looked at some popular RISC and CISC architectures, even looked at Itanium [1] as it was the hot new thing at the time. The professor didn't necessarily hate Intel, just the x86 architecture. [1] https://en.wikipedia.org/wiki/Itanium

x86 should be taught as a follow-up course in the following semester as an example of how not to design a CPU architecture, specifically from the long term evolution perspective. It would have been fine if x86 ceased to exist after 80486 or 80586, yet it has persevered and has continued to cause trouble with an incessant stream of extensions supplanting the previous generations, or expanding or «enhancing» them: MMX,…

I won't make the argument against variable-length vectors here, but I will note that, although some of the extensions you mention do change the vector length, for none of them is the vector length change the most interesting aspect; you could keep the vector length fixed, and they would all still be interesting. (The exception is avx10, but that is basically forwards-compatible and barely an 'extension', so.)

Contrariwise, although I think some of amd's choices in the initial design of amd64 have hurt its longevity—they could have made more breaks from 32-bit x86, although it is understandable why they didn't at the time—intel has been very careful and has done an extremely good job of playing the hand it's been dealt.

Re: When teaching computer architecture, why are universities using obscure CPUs?

#64

Post by moonchild »

gizmo686 wrote: post #60
vanviegen wrote: post #47

Earlier quoted context omitted.

Yeah, though the concept of segmented memory access can be confusing and so far hasn't been very useful this century. :-)

I think Linux uses it to implement thread-local storage.

No, not really. A couple of features of the architecture and instruction encoding which were historically used for segmentation remain on amd64, and these features are used to implement thread-local storage on every mainstream os targeting the arch, but amd64 is not a segmented architecture in any meaningful sense.

Re: When teaching computer architecture, why are universities using obscure CPUs?

#65

Post by Divver »

Hmm at Berkeley 10-14 years ago during undergrad we used MIPS in both our computer architecture class and compilers class and I wouldn’t call it obscure. Uncommon yes but not obscure.

It had a RISC (Reduced Instruction Set Computer) ISA (Instruction Set Architecture) which at the time was considered academic but not popular in real world computers since CISC (Complex Instruction Set Computer) ISA like amd64 was all the rage but since Apple Silicon and its RISC-based (ARM) architecture that notion has been turned on its head. And the upcoming Qualcomm ARM laptop chip is also following the trend on high performing RISC architecture chips.

The ARM comeback has made RISC popular again for uses beyond mobile phones and academia.

Re: When teaching computer architecture, why are universities using obscure CPUs?

#66

Post by ffgjgf1 »

Divver wrote: post #65

Hmm at Berkeley 10-14 years ago during undergrad we used MIPS in both our computer architecture class and compilers class and I wouldn’t call it obscure. Uncommon yes but not obscure. It had a RISC (Reduced Instruction Set Computer) ISA (Instruction Set Architecture) which at the time was considered academic but not popular in real world computers since CISC (Complex Instruction Set Computer) ISA like amd64 was all t…

Isn’t the RISC vs CISC distinction a but blurry these days with both ARM and x86 adopting a lot of stuff from the other side?

Re: When teaching computer architecture, why are universities using obscure CPUs?

#67

Post by creshal »

ffgjgf1 wrote: post #66
Divver wrote: post #65

Hmm at Berkeley 10-14 years ago during undergrad we used MIPS in both our computer architecture class and compilers class and I wouldn’t call it obscure. Uncommon yes but not obscure. It had a RISC (Reduced Instruction Set Computer) ISA (Instruction Set Architecture) which at the time was considered academic but not popular in real world computers since CISC (Complex Instruction Set Computer) ISA like amd64 was all t…

Isn’t the RISC vs CISC distinction a but blurry these days with both ARM and x86 adopting a lot of stuff from the other side?

It's been blurry for over 20 years at this point. ARM has had "cisc-y" complex instructions for almost as long as x86 has been splitting instructions into "risk-y" micro ops.

Re: When teaching computer architecture, why are universities using obscure CPUs?

#68

Post by onethought »

simonbarker87 wrote: post #62

Universities aren’t corporate training grounds. The job here is to teach you the concepts involved in CPU architecture and using the most general purpose architectures possible that have been developed over millions of person hours to cover every task possible would be a very inefficient way to do this.

> Universities aren’t corporate training grounds.

So the ridiculous fees that correlate to average wages of the profession you graduate into is just a weird coincidence?

Re: When teaching computer architecture, why are universities using obscure CPUs?

#69

Post by kalleboo »

Reminds me of the COMP-X machine and CAP-X assembly language that was invented for Japanese IT exams.

It was invented so that nobody taking the test would be at an advantage or disadvantage depending on platforms they had experience with, so the exams could just test pure understanding of the concepts.

But standardized exams being what they are, so many people started studied this specific ISA that eventually it made its way all the way to the US in the form of a few Tandy pocket computers sold at Radio Shack.

Exhaustive article about the history and the ISA here: http://oldvcr.blogspot.com/2024/02/cap-x-and-comp-x-how-tand...

Re: When teaching computer architecture, why are universities using obscure CPUs?

#70

Post by wjnc »

onethought wrote: post #68
simonbarker87 wrote: post #62

Universities aren’t corporate training grounds. The job here is to teach you the concepts involved in CPU architecture and using the most general purpose architectures possible that have been developed over millions of person hours to cover every task possible would be a very inefficient way to do this.

> Universities aren’t corporate training grounds. So the ridiculous fees that correlate to average wages of the profession you graduate into is just a weird coincidence?

Rent seeking causes that correlation (so there is actually a causal relationship): Some professions require certain university degrees. They pay high wages. Universities found out they can extract a rent on future earnings of potential entrants. So you could actually argue that by requiring the degree the firms not only create the rent extraction, but also de facto agree on the curriculum.
Post reply on HN