Hacker News Reader

Comment by mhh__ | Hacker News Reader

> basement

I was looking for an itanium server for a while and literally couldn't find one that wasn't silly money

Back in the day, we reluctantly abandoned PowerPC, UltraSPARC, PA-RISC, Alpha AXP, MIPS... because the bright future was Intel Itanium. Bye, bye, SUN, Silicon Graphics, Acorn, DEC/Compaq, Hewlett Packard, Sony NEWS.

That future never materialized.

About the only real "success" of Itanium was wiping out the competing architectures. ARM only survived because its low power consumption was perfect for mobile and embedded use. Once Acorn was gone, we didn't see a mainstream ARM-based personal computer again until the M1 Mac.

> About the only real "success" of Itanium was wiping out the competing architectures.

Arguably x86-64 would have wiped those out anyway, with or without Itanium.

Most likely, yes. But without Itanium, the competing architectures would have had several more years of development before x86-64 was released. While the end result would probably be similar, it is interesting to think about what improvements may have been made in competing architectures in those interim years.

Yeah, the big question I have is whether PowerPC or maybe Alpha would have done better. Both had competitive chips but a lot of companies were waiting for Itanium since the analysts were saying that was the future. PPC survived for a while on Macs but it’s interesting to consider hope many extra millions shipped per year it would have taken to make it a profitable and more competitive business. Microsoft had NT running on the hardware so a switch could have been relatively quick.

Worth noting - PA-RISC was Hewlett Packard's own baby, which they used for their higher-end systems. And, at least in Wikipedia's account of things ( https://en.wikipedia.org/wiki/Itanium#Inception:_1989%E2%80%... ), Itanium was born when HP approached Intel about collaborating on a new architecture, and Intel happily accepted a load of fundamental design ideas from HP.

Sure, but 64-bit x86 did materialize and killed Itanium, there is little reason to think it wouldn't have killed those archs too.

Maybe...but Itanium was a join announcement by A-listers Intel and HP in June '94. Vs. C-lister AMD didn't announce x86-64 until 1999, didn't release the full spec. until Aug'00, and didn't ship an actual product until Apr'03.

Without Itanium, those other architectures could easily have had an extra decade or more, and things might have turned out quite differently.

Maybe, maybe not.

I'd argue what really killed Itanium was the end of dennard scaling. The whole arch was based on clocking faster and faster rather than being a slower, but smarter core. The end of dennard scaling put a ~5GHz cap on CPU speeds, is it's whole value prop didn't make sense against smarter cores that clocked just as high.

IMO, on the technical merits, a lot of those dead archs are just as valid ISAs for slower, smarter cores as x64 is and would have been able to compete. PowerPC and AArch64 are pretty damn comparable.

I got one from ebay for basically the price of shipping (£58, that thing weighs a lot). They are less available now than they used to be though. The era of decommissioning data centers with Itanic in them is over. https://rwmj.wordpress.com/2014/09/08/raise-the-itanic/

It was difficult hardware to deal with because it had noisy high speed fans that couldn't be disabled. Really, disabling them caused the machine to automatically shut off after 5 minutes even if it was in no danger of overheating. You couldn't be in the same room when it was running.

I expect there are people out there who were in the right place at the right time when the company they work for was decommissioning some IA64 kit, and got it for nothing or close to. So there will be at least some hobbyists who haven't paid silly money. Because of the niche it ended up in, and how small that niche was, you either get the kit for very little that way or a lot by buying it from someone who got it that way, there is not much in between.

Reply on news.ycombinator.com