News for hackers. Stuff that matters.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by andromeduck on Saturday October 08, 2022 @06:10AM (#33131137)
Do you remember Itanium?
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @06:14AM (#33131152)
I do and I'm giving them credit that they won't make the same mistakes again with a product that fails.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by andromeduck on Saturday October 08, 2022 @06:38AM (#33131220)
But why not just jump and let momentum carry x86 like before? The problem with Itanium was largely that the ISA itself sucked to work with and SW/tools were not ready for prime time when 3p SW/HW vendors like Nvidia were expected to have ported to it.

It doesn't seem like ARM and maybe RISC-V will have the same issues, at leatin terms of magnitude so I don't really see why it matters.

I think long term they risk ending up like Canon/Nikon in the DSLR market when Sony came along with the new fangled mirrorless technology - sometimes you gotta disrupt yourself and skate to where the pick is going.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @06:50AM (#33131261)
Largely agree. I'd be astonished of they don't have high performance RISC-V designs under development.

Problem is that the endpoint is much less attractive (for them) than where they are now and the transition will be very, very messy. At a time when the business is under strain for other reasons it's a risky move.

Should have done it a few years ago - when they had process lead - but hindsight is a wonderful thing!

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by xorcist on Saturday October 08, 2022 @07:30AM (#33131445)
Mistake? They killed almost all their competition in one blow. Only IBM survived.

Ruthless, yes. But mistake?

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @07:36AM (#33131486)
You think Itanium was entirely successful? Really?
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by mepian on Saturday October 08, 2022 @09:07AM (#33131982)
It was indeed successful at killing PA-RISC, high-end MIPS, and Alpha, because their owners (HP, SGI, Compaq) abandoned them for Itanium.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @09:30AM (#33132131)
You think Itanium was entirely successful then?

My point is that it doesn't matter how good the architecture is or if a few firms follow you - you need it to be successful in the market.

If you think Itanium was a cunning plan to take a few competitors out of the market only to abandon a few years later then we'll have to disagree.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by xorcist on Sunday October 09, 2022 @05:41PM (#33145588)
Itanium was absolutely a plan to take out the competitors in the unix- and minicomputer market, where margins were much, much higher than in the generic x86 market. That worked. SGI, Digital, HP, Compaq/Tandem, they all fell.

It was not necessarily the plan to abandon the architecture, but once it was won, it also wasn't terribly important to keep going. Much like most corporate takeovers to this day.

Intel would have been happy to keep the market segmented for a few more years, but what happened instead was that the market vacuum was filled by Linux and x86 instead. That would likely have happened sooner or later anyway, but there you go.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Monday October 10, 2022 @07:12AM (#33149987)
Correlation is not causation.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @09:07AM (#33131984)
Itanium only failed because AMD exists, with patent agreements that allowed them to create AMD64.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by temac on Saturday October 08, 2022 @10:51AM (#33132766)
It was a poor architecture though and if not by AMD64 it would have been killed by something else more in line with traditionnal high perf superscalar. Maybe even just Arm.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @11:57AM (#33133367)
Underestimating Wintel mighty.

ARM was meaningless during those days.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by monocasa on Saturday October 08, 2022 @12:15PM (#33133539)
PowerPC probably would have eaten it's lunch. There was even a dual x86/PowerPC chip called the PowerPC 615.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @12:38PM (#33133748)
It was hardly successful outside Apple, and IBM systems, not sure.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by monocasa on Saturday October 08, 2022 @12:57PM (#33133926)
Even Microsoft directly shipped a PowerPC system running a modified NT kernel in the Xbox 360.

If Itanium continued to implode with no other alternatives, PowerPC would have been the most likely to pick up the slack. The main reason why it more or less failed was from a lack of volume to pay for leading edge R&D for the process side. Without AMD64, Intel's Itanium obsession combined with the mid aughts dennard scaling wall catching everyone with their pants down would have given a nice bit of breathing room for PowerPC to exceed x86-32.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @01:27PM (#33134204)
That Windows version was long gone when Itanium came to be, no way Microsoft would bring Windows NT for Power PC from the dead.

https://www.itprotoday.com/windows-78/windows-nt-powerpc-no-...

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by monocasa on Saturday October 08, 2022 @02:47PM (#33134850)
They killed it at about the time they started working on AMD64 internally with AMD. Dave Cutler himself gave feedback on the pre silicon design; it was basically co designed with Microsoft. They kept PowerPC in public products until they had another way out. And to this day PowerPC support still exists internally in the NT kernel.

They also literally were shipping an NT derived kernel for Xbox 360 into the 2010s.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @02:56PM (#33134931)
Between 1997 and the release of XBox in 2001 there was enough years of code changes (two major NT based releases), also XBox NT kernel was basically that, a stripped down kernel without any relation to Windows 2000 userspace.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by monocasa on Saturday October 08, 2022 @03:13PM (#33135086)
And notice I said 360; they reimported the PowerPC support from mainline NT in 2005. And to this day they still have PowerPC support internally in mainline NT.

And that small bit doesn't address the core of what I'm saying, that PowerPC support would have seen even more support if the two options were that and Itanium.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @06:02PM (#33136445)
I guess we will have to agree to disagree that Itanium would have won anyway.

Which doesn't matter which of us is right, because AMD created AMD64 and killed Itaninum in the process.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by panick21_ on Saturday October 08, 2022 @01:05PM (#33134014)
I'm sure people said the same thing when they did Intel iAPX 432.

Surely they are not doing that again. Well ...

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @01:24PM (#33134177)
Indeed. I suspect the 'x86 always' mindset though led to the expensive failure in mobile - a huge miss. Easy to see this in hindsight though.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @09:06AM (#33131978)
Yes, it only failed because AMD came up with AMD64 thanks to their patent agreements.

Had it not been the case, and everyone would be using Itanium no matter what.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @09:27AM (#33132109)
> Had it not been the case, and everyone would be using Itanium no matter what.

That's quite a strong statement.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @09:32AM (#33132146)
It is the hard reality, no AMD64, Itanium would have been the only 64 bit CPU successor for x86 coming out from Intel.

Take it or leave it, and with Wintel going Itanium, that would be it.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @09:44AM (#33132218)
Today not 'everyone' is using x86 - by a long margin - so it's a bit of a stretch to say that in a hypothetical alternative history an architecture that failed in the market would be utterly dominant.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @09:52AM (#33132258)
85% of desktop users and about 90% cloud users certainly are.

Mobile phones and tablets are mostly consumer devices, Apple's ARM laptops are only relevant for about 10% of the desktop market.

Outside embedded devices and electronic appliances, every other CPU is a rounding error in what concerns the general public.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by temac on Saturday October 08, 2022 @10:54AM (#33132786)
If you exclude everything that does not go into your direction and make the hypothesis that a poor architecture would have risen and then would not have been replaced, you conclude that "everybody" would use it?
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @11:58AM (#33133381)
Without AMD64, Windows and UNIX would, that is what matters in this context.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by klelatti on Saturday October 08, 2022 @11:16AM (#33133000)
So not everyone, even in your chosen subset of the market.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @11:57AM (#33133374)
It is a matter of market share.
Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by panick21_ on Saturday October 08, 2022 @01:09PM (#33134050)
No people would just use 32 bit x86 and continue with that for many more years and move to SPARC/PowerPC for the few cases where you really need 64 bit.

People not just gone use really bad processors because they have no other options.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @01:30PM (#33134237)
Assuming Sun and IBM would sell them to Microsoft outside their own UNIX and mainframes usage.

Apparently people didn't have any problem using the bad 80x86 architecture like so many complain around here.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by panick21_ on Saturday October 08, 2022 @01:40PM (#33134303)
This was right in the time period of Windows Everywhere. Windows on MIPS, Alpha and so on. And for server workloads just using going to Unix is totally fine.

People would rather run server workloads on Unix rather then using windows with shitty expensive processors.

People that are unwilling to move to Unix very likely just stick around on 32bit instead.

Re:Intel, SiFive Demo High-Performance RISC-V on Intel 4 by pjmlp on Saturday October 08, 2022 @02:57PM (#33134942)
That version of Windows died with NT 4.0, several years before Itanium was a product.

We were running Windows 2000 in production, alongside Aix, HP-UX and Solaris workloads across all our customers back in 1999 - 2003, before we got hit in the first .com startup crysis.