TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

MIPS Becomes RISC-V

424 点作者 zimmerfrei大约 4 年前

30 条评论

zokier大约 4 年前
The progression of headlines is funny:<p>1) MIPS Strikes Back: 64-bit Warrior I6400 Arrives <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=8258092" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=8258092</a><p>We are still in the game<p>2) Linux-running MIPS CPU available for free to universities – full Verilog code <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=9444567" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=9444567</a><p>Okay, we are not doing so great, maybe we can get young kids hooked?<p>3) MIPS Goes Open Source <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=18701145" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=18701145</a><p>Open Source is so hip and pop these days, lets do that!<p>4) Can MIPS Leapfrog RISC-V? <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=19460470" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=19460470</a><p>Yeah, sure, that&#x27;ll happen<p>5) Is MIPS Dead? Lawsuit, Bankruptcy, Maintainers Leaving and More <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=22950848" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=22950848</a><p>Whoops<p>6) Loose Lips Sink MIPS <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=24402107" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=24402107</a><p>And there is the answer to the question from previous headline<p>And now we are here.
评论 #26390274 未加载
评论 #26390698 未加载
brucehoult大约 4 年前
In 2010 MIPS wanted $2 million from Berkeley to allow them to use the MIPS instruction encodings for processor cores that Berkeley would design entirely themselves. So they made up their own encodings instead.<p>The rest is history.<p>In many ways modern MIPS and RISC-V are pretty much just different binary encodings of the same ideas.<p>[this was already posted as a comment in a thread, but on reflection it probably deserves its own]
评论 #26394766 未加载
评论 #26395622 未加载
评论 #26396494 未加载
mortenjorck大约 4 年前
This is more or less analogous to Blackberry moving to Android, isn’t it? Storied, old-guard tech company loses most of its market share, trades in its first-party stack for a rising open-source alternative.<p>Is MIPS still a big enough name to make this much of a coup for RISC-V? Or is this the last-ditch effort of a fallen star of the semi market?
评论 #26389990 未加载
评论 #26390001 未加载
评论 #26391827 未加载
评论 #26390658 未加载
评论 #26391193 未加载
评论 #26395279 未加载
musicale大约 4 年前
There&#x27;s a lot to like about MIPS. It&#x27;s a perfectly usable RISC architecture that:<p>- is easy to implement<p>- is supported by Debian, gcc, etc..<p>- is virtualizable<p>- scales from embedded systems (e.g. compressed MIPS16 ISA) up to huge shared-memory multiprocessor systems with hundreds of CPUs<p>Like RISC-V, MIPS traces its lineage to the dawn of the RISC revolution in the 1980s, though on the Hennessy&#x2F;Stanford side rather than the Patterson&#x2F;Berkeley side.<p>And it was supposed to go open source: <a href="https:&#x2F;&#x2F;www.mips.com&#x2F;mipsopen&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.mips.com&#x2F;mipsopen&#x2F;</a><p>but that effort sadly seems to be dead: <a href="http:&#x2F;&#x2F;mipsopen.com" rel="nofollow">http:&#x2F;&#x2F;mipsopen.com</a><p>That&#x27;s really too bad. MIPS doesn&#x27;t deserve to die.<p>Fortunately as mentioned above it will live on as long as there are PS2 consoles or emulators around.
评论 #26393853 未加载
评论 #26393087 未加载
MaxBarraclough大约 4 年前
Everyone&#x27;s right to celebrate the success of RISC-V, but part of me thinks it&#x27;s a shame that there&#x27;s relatively little architectural diversity (<i>edit</i> I should have said <i>ISA diversity</i>) in modern CPUs. MIPS, Alpha, and Super-H, have all but faded away. Power&#x2F;PowerPC is still out there somewhere though. Apparently they&#x27;re still working on SPARC, too. [0]<p>At least we&#x27;ll always have the PS2. ...until the last one breaks, I guess.<p>[0] <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;SPARC" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;SPARC</a>
评论 #26391324 未加载
评论 #26390551 未加载
评论 #26390803 未加载
评论 #26393297 未加载
评论 #26391327 未加载
评论 #26393926 未加载
评论 #26395370 未加载
评论 #26390684 未加载
评论 #26394129 未加载
st_goliath大约 4 年前
I just read the official statement[1] that&#x27;s linked to in the article.<p>Just so I get this straight: Wave Computing, the company that bought the remains of MIPS, is now (after bancruptcy) spinning it off as a separate company, that is going to work under the name MIPS, holds the rights to the MIPS architecture, but is doing RISC-V?<p>[1] <a href="https:&#x2F;&#x2F;www.prnewswire.com&#x2F;news-releases&#x2F;wave-computing-and-mips-emerge-from-chapter-11-bankruptcy-301237051.html" rel="nofollow">https:&#x2F;&#x2F;www.prnewswire.com&#x2F;news-releases&#x2F;wave-computing-and-...</a>
评论 #26390801 未加载
评论 #26390152 未加载
nickysielicki大约 4 年前
This article from 2015 (&quot;The Death of Moore’s Law Will Spur Innovation: As transistors stop shrinking, open-source hardware will have its day&quot;) is getting better and better with age.<p><a href="https:&#x2F;&#x2F;spectrum.ieee.org&#x2F;semiconductors&#x2F;design&#x2F;the-death-of-moores-law-will-spur-innovation" rel="nofollow">https:&#x2F;&#x2F;spectrum.ieee.org&#x2F;semiconductors&#x2F;design&#x2F;the-death-of...</a>
评论 #26395290 未加载
gautamcgoel大约 4 年前
This is huge. It looks like the only architectures widely-deployed in ten years will be x86, ARM, Power, and RISC-V (maybe also SPARC64 in Japan, although that&#x27;s rare in the US).
评论 #26389732 未加载
评论 #26389951 未加载
评论 #26389741 未加载
评论 #26390530 未加载
评论 #26389884 未加载
评论 #26389945 未加载
评论 #26389680 未加载
评论 #26390534 未加载
tyingq大约 4 年前
The MIPS name was originally was an acronym for &quot;Microprocessor without Interlocked Pipeline Stages&quot;. The RISC-V docs that I&#x27;ve skimmed seem to show quite a lot more hardware pipeline interlocking than the last gen MIPS processors. So the name is a bit funny now.
评论 #26391837 未加载
notacoward大约 4 年前
A few jobs ago, the company I worked for based their own custom processor on a MIPS core. Why MIPS? The answer I got when I asked was that it was the only affordable option. ARM in particular was called out as beyond reach financially. Years later, long after that company was gone, RISC-V came in at an even lower price point. AFAICT there&#x27;s no need to look for other reasons behind this news.
Andrex大约 4 年前
My historical skepticism on the acceptance and proliferation of RISC-V looks more antiquated by the day. No real dog in the fight, but I would love to see this take off like ARM did.
ThinkBeat大约 4 年前
I started assembly programming on early x86 chips. Then I landed a job programming MIPS chips.<p>They were so much easier to work with in assembler than x86.<p>I was quite impressed.<p>Then came the SGI Indy for me. Good times.<p>Sorry to see them go.
p1mrx大约 4 年前
Mirror: <a href="https:&#x2F;&#x2F;archive.is&#x2F;S1s80" rel="nofollow">https:&#x2F;&#x2F;archive.is&#x2F;S1s80</a>
yaantc大约 4 年前
It&#x27;s very good move. First, it&#x27;s for embedded: if you&#x27;re not designing system on chip it won&#x27;t matter to you.<p>But for those who design SoC, and who needed embedded CPU of intermediate power, it&#x27;s very good news. ARM is expensive here (it&#x27;s considered cheap compared to Intel, the embedded world is different). The RISC V newcomers are interesting but... new, and it always get a bit of time to bring solutions to maturity, which matters in embedded. And if ARM owns the high-end (where a design must be co-optimized for the latest advanced nodes to really shine, which is very labor intensive and costly), for the mid-range it&#x27;s much more open.<p>MIPS had a good mid-range design with the I-7200. Their problem was that the old MIPS ISA was not dense enough (larger I cache, larger Flash footprint) compared to the competition, and their compact versions not good enough. So they designed for the I-7200 a new ISA, nanoMIPS, which has &quot;MIPS&quot; in its name but is completely different. And guess what: nobody cared. It became another proprietary ISA, only supported by MIPS own GCC version.<p>But still, the design was good, and in particular the LLC&#x2F;coherency support is much more mature than what many newcomers offer today. Which shouldn&#x27;t be a surprise, as it&#x27;s the result of a long line of (good) mid-range CPUs.<p>By evolving their design to RISC-V, MIPS will have one of (if not the) best mid-range CPU&#x2F;cluster IP in the market soon for quality vs price --- depending on how fast the competition move, they&#x27;re definitely not sitting still! And RISC V will solve the toolchain&#x2F;tools support nanoMIPS has. If there had been such a RISC I-7200 equivalent a few year ago, I may have used it.<p>So very nice, and I look forward for more competition in the embedded mid-range CPU IP market soon.
pabs3大约 4 年前
There are a couple of vendors of multi-architecture CPUs where the native architecture is MIPSish and the other supported arches are MIPS, ARM, RISC-V &amp; x86. Tachyum is one of them, here is another:<p><a href="https:&#x2F;&#x2F;www.zhihu.com&#x2F;question&#x2F;414069789" rel="nofollow">https:&#x2F;&#x2F;www.zhihu.com&#x2F;question&#x2F;414069789</a>
评论 #26406741 未加载
thr0w__4w4y大约 4 年前
I consulted with MIPS and ARM many years ago - pre Y2K (background is microprocessor design and firmware). ARM was a little formal but easy to work with, cooperative and supportive (they opened doors and provided everything I asked for). Working with MIPS was a nightmare, every step of the way. Weren&#x27;t responsive, difficult to work with, engineers were stubborn, etc.<p>I realize these are generalizations and I&#x27;m a sample size of one. But I was plugged in at a pretty high technical level with both companies, and I remember telling my wife at the time that I thought ARM would skyrocket and MIPS would be unable to get out of its own way.<p>Glad I bought a lot of ARM stock before most people knew about them.
throwaway81523大约 4 年前
Actual source: <a href="https:&#x2F;&#x2F;www.eejournal.com&#x2F;article&#x2F;wait-what-mips-becomes-risc-v&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.eejournal.com&#x2F;article&#x2F;wait-what-mips-becomes-ris...</a>
评论 #26390938 未加载
pacman2大约 4 年前
Loongson still does MIPS. They are a little bit vaporware, even hard to get in China.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Loongson" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Loongson</a>
评论 #26393450 未加载
lallysingh大约 4 年前
I guess that they&#x27;re going to ship RISC-V CPUs? Makes sense. Do they still have design chops for making fast implementations?
ChuckMcM大约 4 年前
It was unclear if &quot;v8&quot; of the MIPS architecture is a re-branded RISC-V or if v8 MIPS is a combined RISC-V + MIPS or what.
评论 #26393180 未加载
评论 #26404693 未加载
cable2600大约 4 年前
I would like to see AmigaOS ported to a RISC-V platform for new RISC based Amigas, it would be cheaper than the PowerPC Amigas.
synergy20大约 4 年前
It could preempt RISC-V long time ago by doing this. I hope it&#x27;s not too late.<p>MIPS is still used in routers and set-top-boxes, but the steam is running out quickly, nearly all routers&#x2F;set-top-box new design are now using ARMs already. There is a last hope though.
buescher大约 4 年前
&quot;Development of the MIPS processor architecture has now stopped&quot;<p>Is anyone still developing SPARC?
评论 #26389882 未加载
评论 #26390540 未加载
评论 #26389880 未加载
phab大约 4 年前
I spent a summer interning with MIPS hardware when they were owned by Imagination. It&#x27;s kinda odd to see something you helped make die (be killed?) like this.<p>Strong &quot;not with a bang but a whimper&quot; vibes in my office today.
einpoklum大约 4 年前
Does this mean that design features from MIPS can now be adopted &#x2F; adapted in RISC-V? Or is it just a no-longer-used ISA and chip designs that become freely usable?
dingosity大约 4 年前
meh. there&#x27;s more to a CPU architecture than its instruction set. this isn&#x27;t the &quot;death of MIPS.&quot; far from it, it&#x27;s the remaining MIPS people applying their own understanding of design to implement a CPU with the RV64 instruction set (plus extensions, presumably.)<p>that being said... MIPS is certainly smaller than it once was. i hope there are still people around who remember how to build CPUs.
Lucretia9大约 4 年前
Hopefully, they&#x27;ll release the cores as OpenMIPS like OpenSPARC did.
moonbug大约 4 年前
If synthesisable MIPS cores from R3k to R20k enter the public domain, that would be something special. based on how previous such &quot;announcements&quot; have turned out, it&#x27;s not gonna happen this time either.
abrowne大约 4 年前
If you can&#x27;t beat &#x27;em, join &#x27;em?
评论 #26389735 未加载
joshgoldman大约 4 年前
If you did Nintendo 64 assembly, then you know about these guys<p>Disclaimer: I knew someone who worked at Nintendo
评论 #26395975 未加载