TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Which Machines Do Computer Architects Admire? (2013)

174 pointsby dhotsonover 5 years ago

26 comments

xscottover 5 years ago
I&#x27;m not a computer architect (so my opinion shouldn&#x27;t count in this thread), but as someone who did a lot of numerical programming over the years, I really thought Itanium looked super promising. The idea that you can indicate a whole ton of instructions can be run in parallel seemed really scalable for FFTs and linear algebra. Instead of more cores, give me more ALUs. I know &quot;most&quot; software doesn&#x27;t have enough work between branches to fill up that kind of pipeline, but machine learning and signal processing can certainly use long branchless basic blocks if you can fit them in icache.<p>At the time, it seemed (to me at least) that it really only died because the backwards compatibility mode was slow. (I think some of the current perception of Itanium is revisionist history.) It&#x27;s tough to say what it could&#x27;ve become if AMD64 hadn&#x27;t eaten it&#x27;s lunch by running precompiled software better. It would&#x27;ve been interesting if Intel and compiler writers could&#x27;ve kept focus on it.<p>Nowdays, it&#x27;s obvious GPUs are the winners for horsepower, and it&#x27;s telling that we&#x27;re willing to use new languages and strategies to get that win. However, GPU programming really feels like you&#x27;re locked outside of the box - you shuffle the data back and forth to it. I like to imagine a C-like language (analogous to CUDA) that would pump a lot of instructions to the &quot;Explicitly Parallel&quot; architecture.<p>Now we&#x27;re all stuck with the AMD64 ISA for our compatibility processor, and it seems like another example where the computing world isn&#x27;t as good as it should be.
评论 #21988741 未加载
评论 #21988764 未加载
评论 #21988767 未加载
评论 #21988801 未加载
评论 #21988716 未加载
评论 #21988804 未加载
评论 #21991684 未加载
评论 #21988708 未加载
tachyonbeamover 5 years ago
I think one of the most influential designs of recent times has been the DEC Alpha lineage of 64-bit RISC processors[1]. Originally introduced in 1992, with a superscalar design, branch prediction, instruction and data caches, register renaming, speculative execution, etc. My understanding is that when these came out, they were way ahead of any other CPU out there, both in terms of innovative design and performance.<p>Looking at this chip, it seems to me that almost all the innovations Intel brought to the Pentium lines of CPU over many years were basically reimplementing features pioneered by the DEC Alpha, just over a decade later, and bringing these innovations to consumer-grade CPUs.<p>[1]: <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;DEC_Alpha" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;DEC_Alpha</a>
评论 #21988685 未加载
评论 #21988779 未加载
评论 #21988673 未加载
评论 #21995603 未加载
评论 #21988752 未加载
评论 #21988713 未加载
评论 #22002047 未加载
评论 #21991756 未加载
kragenover 5 years ago
There are some really great designers on the list, like Sophie Wilson and Gordon Bell, but the list of admirable machines comes up really short — and missing a lot of really significant and admirable machines.<p>Maybe these are the machines <i>bad</i> computer architects, like Alpert, admire. Alpert is notable mostly for leading the computer industry&#x27;s most expensive and embarrassing failure, the Itanic (formally known as the Itanium), despite the presence on his team of many of the world&#x27;s best CPU designers, who had just come from designing the HP-PA --- a niche CPU architecture nevertheless so successful that HP&#x27;s workstation competitors, such as NeXT, started using it. Earlier in his career he sunk the already-struggling 32000, the machine that by rights <i>should</i> have been the 68000. (And maybe if they&#x27;d funded GCC it could have been.)<p>What about the Tera MTA, with its massive hardware multithreading and its packet-switched RAM, which was gorgeous and prefigured significant features of the GPU explosion?<p>What about the DG Nova, with its bitslice ALU chips and horizontal-microcode instructions? What about the MuP21, with its radical on-chip dual circular stacks?<p>What about the HP 9100, with its dual stacks and PCB-inductance microcode, where the instruction set was the <i>user interface</i>?<p>What about the LGP-30, which managed to deliver a usable von Neumann computer with only 113 vacuum tubes (for amplification, inversion, and sequencing)?<p>What about the 26-bit ARM, with its conditional execution on every instruction, and packing the program status register into the program counter so it automatically gets restored by subroutine return, and, more importantly, interrupt return?<p>What about Thumb-2 with its unequaled code density?<p>What about the CM-1? Anyone can see that AVX-512 (or for that matter modern timing-attack-resistant AES implementations!) owe everything to the CM-1.<p>And the conspicuous omission of the Burroughs 5000 has already been noted by others.<p>I mean, there are some good designs on the list! But it hardly seems like a very comprehensive list of admirable designs.
评论 #21988715 未加载
评论 #21988640 未加载
评论 #21988698 未加载
评论 #21988688 未加载
tlbover 5 years ago
It&#x27;s disappointing that most machines today suck so badly. How did that become the state of the industry, with so many smart people working so hard and nobody likes their latest designs?<p>The last high-performance design I actually liked was the DEC Alpha. You could write a useful JIT compiler in a couple hundred lines.<p>I suspect that nVidia&#x27;s recent GPUs are wonderfully clever inside, but they don&#x27;t publish their ISA and the drivers are super-clunky. So I can&#x27;t admire them.<p>I appreciate the performance of intel Core chips, but there&#x27;s so much to dislike. The ISA is literally too big to fully document. The kernel needs 1000s of workarounds for CPU weirdnesses. You have to apply security patches to microcode, FFS.<p>RISC-V would be great if we had fast servers and laptops.
评论 #21988860 未加载
评论 #21988831 未加载
评论 #21995822 未加载
baneover 5 years ago
Surprised nobody picked the Atari 400&#x2F;800 and Amiga 500 computers (which are the 8-bit and 16-bit spiritual parent&#x2F;child machines by the same people).<p>On the other end, pure CPU only machines are kind of interesting as a study in economy, like the ZX Spectrum, a horrible, limited architecture that managed to hit the market at an unreasonably cheap price, make money, and end up with tens of thousands of games.
评论 #21988818 未加载
评论 #21988549 未加载
评论 #21995870 未加载
CalChrisover 5 years ago
Interesting that the B5000 didn&#x27;t make this list. Berkeley CS252 has been reading the <i>Design of the B5000 System</i> paper for years. The lecture slides don&#x27;t criticize it but <i>Computer Organization and Design</i> sorta does:<p><i>The Burroughs B5000 was the commercial fountainhead of this philosophy (High-Level-Language Computer Architectures), but today there is no significant commercial descendant of this 1960s radical.</i>
评论 #21988521 未加载
oddityover 5 years ago
The list seems biased towards pre-2001, so I’ll toss one in: Cell. I hold that it was so ahead of its time, it dragged game devs, kicking and screaming, into the future ahead of schedule when they were forced to support the PS3 for the extended console cycle. :)<p>Larrabee was cute, but to this day I still have no idea what their target workload was.
评论 #21988647 未加载
DonHopkinsover 5 years ago
I always thought of the 6809 as the Chrysler Cordoba of 8 bit microprocessors, with soft Corinthian Leather upholstery and a luxurious automatic multiply instruction.<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=Vsg97bxuJnc" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=Vsg97bxuJnc</a>
erosenbe0over 5 years ago
The CDC-6000 and Cray-1 designed by Seymour Cray are the most admired, hands down.<p>It is also notable that quite a bit of R&amp;D was done in Chippewa Falls, WI, which is just a regular old town in America&#x27;s Dairyland.
评论 #21988621 未加载
评论 #21988677 未加载
gumbyover 5 years ago
Surprised the PDP-6&#x2F;10 didn’t make the list as it was the dominant research architecture for a certain period. Another Gordon Bell jewel.
评论 #21988809 未加载
PeterStuerover 5 years ago
As far as processors are concerned I loved the Zilog Z80 and the Motorola 68000. Oddly enough I really disliked the MT 6502 and the Intel 8086.<p>As total systems I loved the HP 41CX, the Sinclair ZX Spectrum, the Symbolics Lisp Machine and the Apple Mac IIcx (or really just any Mac before the PowerPC debacle).<p>After that era, I just started home-building x86 machines, and while there was the odd preferred component, it never went beyond the &#x27;A is better than B&#x27; stage.
评论 #21998045 未加载
foucover 5 years ago
Anyone admire forthchips? Such as the 144-core chip from <a href="http:&#x2F;&#x2F;www.greenarraychips.com" rel="nofollow">http:&#x2F;&#x2F;www.greenarraychips.com</a>
Merrillover 5 years ago
&gt;Processor design pitfalls - Designing a high-level ISA to support a specific language of language domain<p>Is there an equivalent pitfall in designing the ISA to support a specific Virtual Machine?<p>For example, wouldn&#x27;t the performance of a server processor when running the Java Virtual Machine be a key factor in determining its commercial success? I&#x27;ve always wondered whether the failure of Itanium wasn&#x27;t at least partly caused by the shift from binary executables to bytecode with the contemporary success of the Java language. Even when JIT compilers were used, they were probably too simple to take advantage of the VLIW architecture.
评论 #21995931 未加载
mtreis86over 5 years ago
The machines I most admire are mechanical computers like the ones used in WWII era battle ships for targeting their long guns. Those machines performed differentiation and curve matching using cams and gears.
评论 #22001835 未加载
cptnapalmover 5 years ago
I have a small IBM 390 which I haven&#x27;t been able to find out much, but I did spot while searching that my 1999 S&#x2F;390 has a 256 byte cache line. That&#x27;s 4x over a 2020 i7.
bshanksover 5 years ago
The ones listed by 4 or more people (not including Bell) were:<p>- CDC-6600 and 7600 - listed by Fisher, Sites, Smith, Worley<p>- Cray-1 - listed by Hill, Patterson, Sites, Smith, Sohi, Wallach (also Bell, sorta)<p>- IBM S&#x2F;360 and S&#x2F;370 - listed by Alpert, Hill, Patterson, Sites (also Bell)<p>- MIPS - listed by Alpert, Hill, Patterson, Sohi, Worley<p>Special mention:<p>- 6502 - only listed by Wilson, but she was the chief architect of ARM so i think her choice is important to note<p>- Itanium - mentioned in the top-ranked comment in this HN discussion<p>- DEC Alpha - mentioned in the second-ranked comment in this HN discussion
cameldrvover 5 years ago
Pentium Pro should be on the list. The out of order execution, especially with the micro-op translation was a huge breakthrough.
评论 #21988806 未加载
squarefootover 5 years ago
M68k and Z80 IMO deserved to be in that list much more than x86.
评论 #22002213 未加载
ChuckMcMover 5 years ago
I was always partial to the DEC-10 architecture. That said my first exposure to a machine that had been really well thought out was the IBM 360.
dillonmckayover 5 years ago
<a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;VAX-11" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;VAX-11</a><p>32 bit system from the late 1970s.
评论 #21995494 未加载
rootbearover 5 years ago
I&#x27;m tempted to suggest Babbage&#x27;s Analytical Engine, on the basis of shear audacity alone. Babbage was just amazingly ahead of his time.
bathtub365over 5 years ago
Well, pretty much anything will run AutoCAD these days.
评论 #21988689 未加载
tyingqover 5 years ago
Are there any notable&#x2F;not-just-academic &quot;clean sheet&quot; CPU architecture efforts other than what Mill computing is doing?
ragerinoover 5 years ago
I admire AMDs Zen architecture.
评论 #22002007 未加载
MaxBarracloughover 5 years ago
No mention of the SuperH.
vi-modeover 5 years ago
I&#x27;m into computers for decades. The first time after ages a computer blew my mind again was when I got deep into k8s.<p>If you haven&#x27;t yet do it now, check k8s out, get knee-deep into it and not from a devops perspective but as one who admire computers.
评论 #22002152 未加载