Whatever happened non-Intel RISC workstation computers?

[Quartz]

@engineer_comp_geek

By the way, the nail for the coffin for the DEC ALPHA was when Microsoft decided to kill the NT ALPHA operating system. Once that died, the market for the ALPHA chip died with it. There were people using Unix operating systems, but not enough of them to keep the ALPHA viable.

Not true. PCs were never the main market for the Alpha. I’m not at work for another week, so I can’t get cites, but the DEC Alpha was developed for use in DEC’s VAX / VMS computers. Then DEC got taken over by Compaq (and then HP) which decided to use the Intel Itanium processors instead of developing the Alpha CPUs. Nevertheless enhancements to the Alpha CPU have happenned. VMS kit tends to be kept a long time, and some contracts specify 10 years or more of availability.

Here’s the Wikipedia article.

[RaftPeople]

@Derleth

RaftPeople: Performance is difficult to quantify, and it commonly doesn’t really matter anyway.

You can quote benchmarks all day and all night, but people really care how fast their applications run:
[ul]
[li]If none of their applications can be run on non-x86 hardware, game over. Their choice is made.[/li][li]If their applications are optimized for x86 hardware (they are written with a hand-tuned assembly kernel that does all the heavy lifting, their loops are arranged to fit into the Pentium’s L1 cache, etc.), benchmarks won’t make much of a difference to them, either, because they won’t reflect how their applications will actually perform.[/li][li]Benchmarks don’t tell the complete story. A benchmark that focuses on integer arithmetic and branch prediction won’t tell you jack about the speed of an application that’s floating-point heavy, or is I/O-bound.[/li][/ul]Application developers–or at least the ones who have to care about performance–know how their target hardware behaves, and can tune applications to take advantage of that behavior. Benchmarks give them that information, but benchmarks can’t say how fast a given machine will perform in truly real-world conditions.

We are in agreement on all points listed.
My point was that x86 did not overtake all other processors from a technical/performance perspective as others claimed.

From an economic perspective x86 is the clear winner.

I tend to follow this stuff pretty closely, and since the introduction of the Power5 I’ve only seen evidence supporting it as the performance leader. Is there evidence you see that shows x86 outperforms Power5 for workstations?

[engineer_comp_geek]

@Quartz

Not true. PCs were never the main market for the Alpha. I’m not at work for another week, so I can’t get cites, but the DEC Alpha was developed for use in DEC’s VAX / VMS computers. Then DEC got taken over by Compaq (and then HP) which decided to use the Intel Itanium processors instead of developing the Alpha CPUs. Nevertheless enhancements to the Alpha CPU have happenned. VMS kit tends to be kept a long time, and some contracts specify 10 years or more of availability.

I work for an industrial control system manufacturer that started with PDP-11’s and moved their way up to Vaxes and then Alphas. We always had a pretty close relationship with DEC and one of my cow-orkers used to work for them once upon a time. You are correct that the Alpha was developed for VMS, but they also has Unix and Windows in mind when they made it.

Compaq bought DEC mostly for their customers, I think. At that time, VMS was on its way out, and Windows was starting to take over the world. There was a big push to port most VMS stuff over to Windows, and there was a version of Windows NT that was built for the Alpha processors. There were also lots of handy dandy things developed for NT, like software that would convert VMS command files into NT executables, just to make it easier to port things over.

Compaq was still supporting VMS, but they were pushing all of their VMS customers (like us) towards NT. When Microsoft killed the NT Alpha project, Compaq pushed Unix operating systems for it for a bit, then gave up and pretty much put all of their VMS and Alpha stuff into support mode. They still sold it, but they weren’t pushing it at all (at least not that I could see). The quality of their technical support also went right into the dumper (IMHO).

We have since moved off of the Alpha platform and are now x86 based. I wasn’t aware that any significant development for the Alpha platform was still being done.

[Quartz]

@engineer_comp_geek

I work for an industrial control system manufacturer that started with PDP-11’s and moved their way up to Vaxes and then Alphas. We always had a pretty close relationship with DEC and one of my cow-orkers used to work for them once upon a time. You are correct that the Alpha was developed for VMS, but they also has Unix and Windows in mind when they made it.

Who wouldn’t? But AFAIAA neither were a priority.

Compaq bought DEC mostly for their customers, I think.

Yes

At that time, VMS was on its way out, and Windows was starting to take over the world. There was a big push to port most VMS stuff over to Windows,

Windows could never handle the big stuff.

and there was a version of Windows NT that was built for the Alpha processors.

My colleagues tell me it was only really aimed at the small stuff. NT couldn’t handle the loads that VMS on dedicated (but Alpha-based) kit could. It was only recently that one of my clients migrated off their VMS-hosted mail system.

Compaq was still supporting VMS, but they were pushing all of their VMS customers (like us) towards NT.

The same colleagues will tell you that this was a combination of arrogance and ignorance on Compaq’s part. As you say, Compaq bought DEC for its customers.

We have since moved off of the Alpha platform and are now x86 based. I wasn’t aware that any significant development for the Alpha platform was still being done.

It’s stopped now, with the move to Itanium, but VMS is still going strong.

It’s sad really, but that’s the free market. You’re free to screw up as much as you like.

[HMS_Irruncible]

Somewhat off-topic, but there is a bit of commotion in our office because we benchmarked a SunFire server against a Linux box running on a Dell (for compiling massive amounts of Java code), and the Dell completed the task about 5x faster (for about 1/10th the cost). The people who paid for the Sun boxes have some rather pointed questions about this phenomenon, and the people whose job security depends on the Suns are scrambling for answers. It’s a bit surprising to me since I always believed the RISC hype.

[gazpacho]

Derleth posed on another thread something that I think is relevant.

Either RISC won by becoming CISC on the outside, or CISC won by becoming RISC on the inside.