More Apple/970 Confirmation

[scem0]

well, yeah. G5 is a good name. But it seems like this is more
than a meer fifth generation processor . This is going to change
apple tremendously. It is a revolutionary product and so it needs
a revolutionary name. G5 just isn't sounding revolutionary .

 

[MacCoaster]

Re: G

Originally posted by Sol
...Apple has so far resisted re-branding any of Motorola's chips a G5, even with new technologies like Silicon On Insulator, more AltiVec units, etc...

That's because even with SOI, new Altivec, a G4 is a G4.

 

[nuckinfutz]

Re: Re: G

Originally posted by MacCoaster
That's because even with SOI, new Altivec, a G4 is a G4.

Typically CPU's will be renamed after a Core Change. Hence when Moto revamped the G4's Core to support 7 Piplines they could have called it a G5 if they had wanted to.

 

[ddtlm]

nuckinfutz:

Yeah, but what an embarrasing "G5" that would have been.

 

[nuckinfutz]

Originally posted by ddtlm
nuckinfutz:

Yeah, but what an embarrasing "G5" that would have been.

No Doubt.

ddtlm. How small do you think we'll be fabbing processor at. Is the next level to be .06um. I thought I read that Intel would be there in the next 3 years. Whaddya think?

 

[ddtlm]

nuckinfutz:

I bet Moto will keep making a 7455-type core for quite a while, since their seems to be little incentive for them to dramatically increase performance. They may more to RapidIO or something, but I bet the core will see 65nm. Years after Intel and IBM get there, or course.

 

[macrumors12345]

Originally posted by ddtlm
macrumors12345:

It is inconcievable to me that IBM would in fact claim that they do not intend to support the AtiVec unit in Linux userspace (at least) when in fact they would have to go out of there way to not support it, for no apparent benefit. On certain tasks AltiVec is unmatched and I'm sure that IBM knows that, and I'm sure that many prospective customers know that.

Once again, I agree that it is unlikely that IBM would actively prevent people from using Altivec. What I do not understand is why you interpret "no plans to implement it in their Linux servers, or AIX versions of the blade servers" as NECESSARILY meaning that they will (actively) not allow anyone to write to it in the userspace.

Like I said, I think this rumor is a fabrication, and I think the AlitVec chit-chat was thrown in to provide some meat and some excitement (purpose-built feature for Apple, etc).

Perhaps. I do generally come down on the side of skepticism, but the body of evidence out there is beginning to make it look like the 970 could be coming out pretty soon (say, within the next 6 months). And I would certainly be shocked if Apple chose not to use it. Well, we'll see soon enough (or not soon enough, depending on how patient you're feeling!).

 

[scem0]

Like I said, I think this rumor is a fabrication

hehe, fabrication..... I hope I'm not the only person that caught
that joke.

 

[ddtlm]

macrumors12345:

Perhaps. I do generally come down on the side of skepticism, but the body of evidence out there is beginning to make it look like the 970 could be coming out pretty soon (say, within the next 6 months). And I would certainly be shocked if Apple chose not to use it. Well, we'll see soon enough (or not soon enough, depending on how patient you're feeling!).

And I pretty much agree, just I think this particular rumor is bunk.

 

[Sun Baked]

Re: sinking the hopes of 970 powerbook sometime soon...

Apple updated the technical document page http://developer.apple.com/techpubs/hardware/Updates/updates.html

The PowerBook 17 is listed as using the same basic architecture of the PowerBook 12... This is a big change from using the same chipset as the XServe/Powermac in the top portables.

So the overiding question is why-o-why does the iMac and the PowerBook 12 not have FW 800?

The strangeness continues.

Since there is an all new chipset for the ALL the Powerbooks, and most likely iBooks when they get redone, don't expect any new architecture in the portable machines for 18-24 months.

Apple does like to get their money's worth out of chipsets.

 

[ddtlm]

Sun Baked:

Hmmm, I was under the impression that the system controller was pretty much the same one used in all DDR Macs, and if that is the case, then there is nothing to worry about cause its been spread around all over all ready.

 

[Sun Baked]

Originally posted by ddtlm
Sun Baked:

Hmmm, I was under the impression that the system controller was pretty much the same one used in all DDR Macs, and if that is the case, then there is nothing to worry about cause its been spread around all over all ready.

Slight to major change, PowerBooks used to use a PowerMac chipset -- now bumped in favor of the iMac single chip version.

If the DDR single chip controller is taking over in the portables and the consumer line, don't expect a change until Apple gets good use out of the controller. Especially if the DDR Intrepid controller is still being phased-in and displacing the old Pangea controller.

The DDR chipset in the PowerMac and XServe has been floating around a lot longer in machines.

Still odd that the PowerBook 12 and the iMac are not sporting FW800 if the Intrepid IC is FW800 capable... which it sure looks like it is looking at the PowerBook 17.

 

[ddtlm]

Sun Baked:

Slight to major change, PowerBooks used to use a PowerMac chipset -- now bumped in favor of the iMac single chip version.

Yeah the PMac DDR chipset is single-chip too as far as I know. I'm under the impression that all of the Mac DDR system controllers are essentially the same chip.

 

[Sun Baked]

Originally posted by ddtlm
Sun Baked:

Yeah the PMac DDR chipset is single-chip too as far as I know. I'm under the impression that all of the Mac DDR system controllers are essentially the same chip.

Since when?

.pdf comparison...

The DDR multichip version -- XServe and DDR PowerMac -- using UniNorth2.x & KeyLargo ICs for the controller/IO.

The DDR singlechip version -- PowerBook 12 and PowerBook 17 which uses the Intrepid IC. iMac pdf not update yet.

Look at Architecture - Block Diagrams

Unless Apple is wrong there are two different DDR chipsets currently being used.

And 2 non-DDR chipsets the old non-DDR UniNorth and the old Pagea -- which are still in the eMac (UniNorth) and the iBook (Pangea).

For Apple four chipsets are a lot, and the transition isn't complete.

 

[ddtlm]

Sun Baked:

Unless Apple is wrong there are two different DDR chipsets currently being used.

Well I certainly wouldn't claim that they are wrong.

 

[astrocity20]

hmmm XGrid...... anything pertaining to Mac and it providing more game support is of interst to me.

 

[Tim Flynn]

Re: Re: G

Originally posted by MacCoaster
That's because even with SOI, new Altivec, a G4 is a G4.

I think ... that because the processors are PowerPC, They should change to P5 !

I'll duck now

 

[areyouwishing]

and the new Mac will be called.... H1

reasons...

1. H is after G in the Alphabet
2. H is on the periodic table as Hydrogen (happens to be the 1st)
3. 1 being the best, first, newest (all those monikers)

"Introducing the all new Macintosh H1, the worlds first 64-bit desktop."

I think marketing could really make that burn.

 

[Remus]

Re: Re: Re: G

Originally posted by Tim Flynn
I think ... that because the processors are PowerPC, They should change to P5 !

I'll duck now

Quack Quack

 

[ffakr]

Originally posted by Sun Baked
Strange to put in a vector unit and then not make use of it in their machines, even if it's just an available library and compile time option -- it would speed up those apps that can make use of the code.

If Linux is going to be extended to include Altivec and IBM is shipping Linux boxes, there seems to be a disconnect in the information here.

Might not quite fit in with their compile once and run on a bunch of boxes strategy, but some people do end up tweaking code to cut run time at the expense of portability.

Linux has supported Altivec for ages. Yellow dog linux and LinuxPPC brought altivec enhancements to gcc. Apple has continued this.

yellow dog

 

[silvergunuk]

stating the obvious

come on lads enough of the specualtion we all know the 970s are coming..steve jobs is known to go through rumor sites and feeds off the mindless ranting we do..come on steve show us mac owners your cards so we can save up for a nice new computer. You know you wanna, our hearts will beat faster and maybe even cause some of us to climax you know you love it....ahem

 

[ffakr]

Originally posted by jettredmont
Umm, no.
A quick search for "linux altivec" on Google didn't turn up anything more recent except for a note from a developer working on Linux-PPC64's Altivec support late last year which indicated Altivec support was being developed but not quite "there" yet.

As I said before, IBM is not likely to be shipping the cutting-edge developer build of Linux on their blades. If Altivec support isn't time-proven yet, it likely will not make it into the IBM Linux-PPC distribution when it ships.

Do you have any more recent link indicating that Altivec support is in place and working?

That must have been a really quick search.. try yellow dog linux
Yellow Dog has been providing Altivec support for a LONG time and they sell third party altivec development tools for linux through their site.

I posted this on page one, but it seems to have been removed for some reason. ?????

DOH!! stupid ffakr!! I didnt' see the other post on the first tow pages and I thought it was removed. I could have swore I posted it on the first page!
My bad.

 

[scem0]

How about the G970 or is that too big of a leap?

 

[Sun Baked]

In the middle of year 2000 an effort was started at IBM in Rochester MN to create a new Linux port for ppc64. This new port would be a high performance, scaleable version of the Linux kernel to run on 64-bit PowerPC processors.

http://linuxppc64.org/

there is a link on the page back to the IBM Linux support page and the Linux PPC32 site.

Didn't really look, nor do I know, what type of support/structure IBM has for the Linux project. Obviously the Unix org does not like the amount of material IBM has ported.

 

[aasmund]

*cough* bull**** *cough*

This whole thing smells worse than the fish salesman in Asterix. And I can't imagine it would be called G5. Perhaps G64 or simply 64 or something else entirely. Also they may have decided to use it, they may not. However, if I was apple I would be very sceptical to have to rely on a sole supplier.

Regards