One month and a few days...agh, the wait is killing me. Couldn't God or whatever being that created the universe hit the "Universe Evolution Fast Forward" button? Just a bit?
There is something I don't understand, maybe one of the tech-heads can clarify it. On the posted document, it references Power4 this...Power4 that, but not 970 (though it mentions 970 at the top). Does this mean that it is sort of generically for the 970, but not specifically? Or does it mean that the 970 is treated (by the compiler) as exactly the same as the Power4?
Internally, Power4 and 970 are virtually identical, so optimizing for Power4 is the same as optimizing for 970.
Apple is contributing POWER4/970 code to GCC, they must be doing that for a reason.
I don't see where Apple is contributing POWER4/970 code to GCC. In that thread I see an Apple employee talking about making GCC run faster (which Apple has been working on for years and has nothing to do with the 970). I see an IBM employee presenting benchmark results on a Power4, which doesn't seem to have much to do with anything either.
Anyway, do you still feel like you need more clues?
I don't see where Apple is contributing POWER4/970 code to GCC. In that thread I see an Apple employee talking about making GCC run faster (which Apple has been working on for years and has nothing to do with the 970). I see an IBM employee presenting benchmark results on a Power4, which doesn't seem to have much to do with anything either.
Anyway, do you still feel like you need more clues?
You're right it's picking at straws, but it's about GCC 3.3 on Power 4 and sent from the IBM guy to Matt Austen at Apple, so I would guess he's directly interested. Just a clue and not needed, as you imply, as it's clear the 970 is coming. Good it will have the next version of the compiler ready though, in time for Panther.
All this says is that GCC is readying to compile code for the 970 processor. Nothing more. Remember that GCC is used by all kinds of people and that IBM is a Linux vendor. It's just as likely that the GCC updates are so IBM can compile Linux and other sofware for the 970s.
One thing it does confirm is that IBM has the capability to compile AltiVec enabled source code for a Linux system. So we may see the IBM 970 blades capable of running SIMD apps of all kinds.
Actually, this says absolutely nothing about the next PowerMacs using 970s.
While I think there's good odds on the 970, there's nothing in this doc that says APPle is going to use it. As BRAD notes, IBM is using Linux, so they are going to want GCC to compile for their chips - the Power4 and the 970.
Anything other than that is extrapolation well beyond the facts.
While I think there's good odds on the 970, there's nothing in this doc that says APPle is going to use it. As BRAD notes, IBM is using Linux, so they are going to want GCC to compile for their chips - the Power4 and the 970.
Anything other than that is extrapolation well beyond the facts.
Anyone in Future Hardware criticising a hypothesis about Future Hardware should probably go to Apple Outsider.
Comments
Originally posted by Jupiter
Apple is contributing POWER4/970 code to GCC, they must be doing that for a reason.
Exiting times indeed.
Do I hear a CONFIRMED?
Originally posted by Programmer
At this stage people saying that Apple may not use the 970 strike me as akin to people on the beach who are ignoring a tsunami warning.
Don't take my comment to Brad the wrong way... I'm in the 'have known since last summer' camp that Apple will be using the gpul/970.
Dave
giga processor ultra lite?
and if so did it simply mean that the lite part was due to the fact that is was a jenny craig version of the power4?
Originally posted by Thai Moof
There is something I don't understand, maybe one of the tech-heads can clarify it. On the posted document, it references Power4 this...Power4 that, but not 970 (though it mentions 970 at the top). Does this mean that it is sort of generically for the 970, but not specifically? Or does it mean that the 970 is treated (by the compiler) as exactly the same as the Power4?
Internally, Power4 and 970 are virtually identical, so optimizing for Power4 is the same as optimizing for 970.
Originally posted by Jupiter
These clues pretty much sum up what's going on...
GCC Power4 and 970 specification
GCC mailing list with Apple employee contributing
Apple is contributing POWER4/970 code to GCC, they must be doing that for a reason.
I don't see where Apple is contributing POWER4/970 code to GCC. In that thread I see an Apple employee talking about making GCC run faster (which Apple has been working on for years and has nothing to do with the 970). I see an IBM employee presenting benchmark results on a Power4, which doesn't seem to have much to do with anything either.
Anyway, do you still feel like you need more clues?
Originally posted by Programmer
At this stage people saying that Apple may not use the 970 strike me as akin to people on the beach who are ignoring a tsunami warning.
Get on the the boat! The Island is sinking!
It's not sinking. Don't be ridiculous.
Yes, it's sinking!
Nonsense. Now everyone let's sing. Tee...tum...tee...tum...
Originally posted by wmf
I don't see where Apple is contributing POWER4/970 code to GCC. In that thread I see an Apple employee talking about making GCC run faster (which Apple has been working on for years and has nothing to do with the 970). I see an IBM employee presenting benchmark results on a Power4, which doesn't seem to have much to do with anything either.
Anyway, do you still feel like you need more clues?
You're right it's picking at straws, but it's about GCC 3.3 on Power 4 and sent from the IBM guy to Matt Austen at Apple, so I would guess he's directly interested. Just a clue and not needed, as you imply, as it's clear the 970 is coming. Good it will have the next version of the compiler ready though, in time for Panther.
Originally posted by Brad
All this says is that GCC is readying to compile code for the 970 processor. Nothing more. Remember that GCC is used by all kinds of people and that IBM is a Linux vendor. It's just as likely that the GCC updates are so IBM can compile Linux and other sofware for the 970s.
One thing it does confirm is that IBM has the capability to compile AltiVec enabled source code for a Linux system. So we may see the IBM 970 blades capable of running SIMD apps of all kinds.
MM
Originally posted by T'hain Esh Kelch
Haha, nice zig! And nice song!
It's not a song, it's take on the opening lines of Trainspotting
Originally posted by Brad
Actually, this says absolutely nothing about the next PowerMacs using 970s.
While I think there's good odds on the 970, there's nothing in this doc that says APPle is going to use it. As BRAD notes, IBM is using Linux, so they are going to want GCC to compile for their chips - the Power4 and the 970.
Anything other than that is extrapolation well beyond the facts.
Originally posted by Clive
Anything other than that is extrapolation well beyond the facts.
Well, in this case, there is indeed a spoon.
WWDC or bust?
Lemon Bon Bon
http://forums.appleinsider.com/showt...0&pagenumber=2
Originally posted by moki
Well, in this case, there is indeed a spoon.
1) The Tick: "Spoon!"
2) Trad Blues: "Some fill spoons full of diamonds..."
3) The Matrix:
Neo: "What is the truth?"
Boy: "There is no spoon."
Originally posted by Tomb of the Unknown
3) The Matrix:
Neo: "What is the truth?"
Boy: "There is no spoon."
Thanks, I was trying to figure out the spoon reference.
Originally posted by Clive
While I think there's good odds on the 970, there's nothing in this doc that says APPle is going to use it. As BRAD notes, IBM is using Linux, so they are going to want GCC to compile for their chips - the Power4 and the 970.
Anything other than that is extrapolation well beyond the facts.
Anyone in Future Hardware criticising a hypothesis about Future Hardware should probably go to Apple Outsider.