SGI: Discussion

Prior to Silicon Graphics deciding on going down the Itanium route was there discussion on IRIX 6.6 or even a roadmap to 7.0?

The rumors I've heard were that the next release would have been IRIX 7, and was planned for release around 2000/2001. I've never seen any hints of the plans.

Adrenaline wrote: Prior to Silicon Graphics deciding on going down the Itanium route was there discussion on IRIX 6.6 or even a roadmap to 7.0?

Internally processor/OS/SW projects had codenames usually left to the whim of the PM in charge.

There must have been tons of "roadmaps" floating around during the roaring late 90s. But they were mostly vaporware, anything that had to do with what would have been Irix post 6.5 was abandoned probably around 1999 when Oracle dropped support for Irix and it was clear that any new MIPS arch was not going to materialize.

I assume most discussions among the Irix team around that time were about resume writing tips.

R-ten-K wrote:

Adrenaline wrote: Prior to Silicon Graphics deciding on going down the Itanium route was there discussion on IRIX 6.6 or even a roadmap to 7.0?

Internally processor/OS/SW projects had codenames usually left to the whim of the PM in charge.

There must have been tons of "roadmaps" floating around during the roaring late 90s. But they were mostly vaporware, anything that had to do with what would have been Irix post 6.5 was abandoned probably around 1999 when Oracle dropped support for Irix and it was clear that any new MIPS arch was not going to materialize.

I assume most discussions among the Irix team around that time were about resume writing tips.

Cool, that's what I thought. Hopefully some of those roadmaps get released into the wild, it would be neat to see what could have been.

Below is part of a post from a Slashdot user I saved back in 2002 or 2003.

"I haven't been following any IRIX roadmaps lately, but it seems that IRIX 7 will likely never happen. I went to SGI's Developer Forum in '97, when 6.5 was relatively new. Several of the sessions talked about all the great things that they were doing for IRIX 7, due out no later than the following summer. 5 years later and we're on something like 6.5.17. SGI continues to pump out minor maintenance and feature releases, but you certainly get the feeling that IRIX isn't going much further these days. 'Tis a shame, really. Back in the day, IRIX was the best Unix out there, bar none... "

I also remember another post on some site years back where a user talked of attending an SGI conference (or maybe he worked at SGI; don't remember for sure) where it was discussed that IRIX 7 would be a port of IRIX from MIPS to Intel Itanium
-- I thought I saved the above post or URL, but I guess I didn't.

Interesting that this got brought up. Imagine if Sgi still utilized MIPS , perhaps R20K cpus well over the 2GHZ mark.... (I never really liked a wintel box)
Irix 7 might have gone the same route as MacOS.. bubbled up, eye-candied up.. pwetty...
And if prices were competitive or equal to what Apple offered , then perhaps Sgi might still have the upper hand today in the graphics world.

Oh well..

Ryan Fox wrote: Interesting that this got brought up. Imagine if Sgi still utilized MIPS , perhaps R20K cpus well over the 2GHZ mark.... (I never really liked a wintel box)
Irix 7 might have gone the same route as MacOS.. bubbled up, eye-candied up.. pwetty...
And if prices were competitive or equal to what Apple offered , then perhaps Sgi might still have the upper hand today in the graphics world.

Oh well..

I really wish they were still the original Silicon Graphics I grew up hearing about from my dad and even as late as when I got into 3D Animation in the late 90s they were still king. It's sad, PCs put the nail in the coffin for MIPS. A large part of that could be attributed to Microsoft's dominance as a platform.

Growing up on Intel (and later switching to AMD in 2000), it was cool to spend a little money at a time here and there, especially when 3Dfx was around. Adding a Diamond Monster 3D (Voodoo 1) back in December 97 was probably the second biggest jump I ever saw in performance in gaming. Watching the blaster pistol's bullet in Quake II light a corridor as it traveled down it at 640x480 in OpenGL mode was amazing. Not sure if anyone else got into Glide programming back in 98, but that triggered me into getting serious about learning C/C++ because of the ease of use.

The problem I see now is there is so little competition now. You've got Intel taking hold of the Highend Desktop market and AMD taking the lower end. Graphics flipflop between nVidia and AMD/ATI. More and more people are using cloud environments like Rackspace, GoGrid etc, so the need for internal servers is slowing (at least from the companies I have seen in the Baltimore/DC areas). Before you know it your "desktop" will be connected to a private cloud with your cell phone as a primary interface. Slowly moving the home/small business desktops & servers into the larger companies' cloud network, essentially shutting out the small to medium sized companies need for their own internal servers.

The thing that could still save the idea of a MIPS cpu is the low power usage per node, but then having to have an emulation layer to convert MIPS -> x86/x86-64 would be costly. I'm not sure even the current MIPS implementation could perform better per watt than a low power Intel Atom or AMD C or E Series after taking that into account. Maybe a MIPS/x86-64 hybrid would be the answer, taking the best of both approach? AMD seems to be thinking of using ARM cpus in conjunction with their APUs so maybe (read up on their SeaMicro acquisition).

I could be completely wrong, but that's my thoughts.

You can thank Rick Belluzzo for the deaths of IRIX and MIPS.