Pixar Beige Box... 970 or G5?
Forgive me if I ramble. I have been a lurker on these boards for a while and an occasional poster. Over the holidays I ran into an old highschool friend who, unbeknownst to me now works at Pixar. He is not a Mac guy but is in more of an IS or IT position at Pixar. We got to talking about his job and I of course asked about Macs. He said that they do use some macs there and that they are using a test machine in a beige box that is supposedly very fast. It is being tested with Shake compositing. He said he doesn't know what is in it and really doesn't care because it is sealed somehow and he is a PC guy. He said that the folks that using it are calling it the G5.
I have not seen this guy for a long time(almost 10 yrs) and he could be full of crap but this was pretty interesting to me. I don't run into much inside information in South Carolina(not exactly a Mac hotbed)
I have not seen this guy for a long time(almost 10 yrs) and he could be full of crap but this was pretty interesting to me. I don't run into much inside information in South Carolina(not exactly a Mac hotbed)
Comments
He might expect the Mac to be a sloth and is surprised at the speed of what may be a dual 1.4GHz. And the G5 is just an inside joke (as in, there is no G5, there will never be a chip dubbed the G5...)
Or it may indeed be a prototype 970.
The only really interesting bit from this -- if true -- is that Steve is using Pixar as a kind-of-in-house hardware testing ground.
Screed
[ 12-30-2002: Message edited by: sCreeD ]</p>
<strong>
He might expect the Mac to be a sloth and is surprised at the speed of what may be a dual 1.4GHz. </strong><hr></blockquote>
What I find promising is that it's the folks using the Macs (presumeably at least part time Mac users) that think it's fast. That's a good sign.
If it's true (chilleymac's story sounds true enough, but his source might be full of do-do of course.)
However, the truth is otherwise.
Well, as for G5, we can only hope.
[QB]"it runs very fast" from a PC guy can be very... oh what's the word? tentative? ambiguous? splunge?
<hr></blockquote>
Very nice use of the word splunge!
<strong>Very nice use of the word splunge!</strong><hr></blockquote>
Umm, splunge is not a word...maybe a cross between splurge and plunge, which may fit Apple right to do both at Macworld next month.
Hard to imagine that MPFC is 30+ years ago.
So anyway another point this brings up is, Are pre-970 PowerMacs good enough for Shake?
Screed... A shroe, a shroe, my kingdom for a shroe.
<strong>So anyway another point this brings up is, Are pre-970 PowerMacs good enough for Shake?</strong><hr></blockquote>
Well a 550 MHz PIII is, so I would guess that the answer is Yes!
<strong><a href="http://www.graphicszone.net/monty_python/scripts/Series_1/43.htm" target="_blank">splunge</a></strong><hr></blockquote>
Thank you for refreshing my memory on the Pythonish-twist on the English vocab...
<a href="http://www-3.ibm.com/chips/news/2002/1216_xilinx.html" target="_blank">Here</a>, we have IBM and Xilinx announcing 90nm production in 2nd half of 2003. IBM has stated in the past that the 970 will be initially released as a 130nm product. So if we have 90nm in 2nd half of 2003, who's to say we won't have 130nm product (and new Powermacs, yum!) much sooner, say 1Q 2003?
I only wish <img src="graemlins/oyvey.gif" border="0" alt="[No]" /> since i need to by a new tower in the next 3 months.
This rather contradicts IBM's own statement about any 970 coming out until Q3.
Besides, the 970 isn't mentioned. Just IBM's "new product line." This vaguery could go either way, for or against a sooner-than-expected-by-two-quarters-970-release.
I want to believe; I do. But we're splitting hairs and pitting press releases from the same company against each other. No good can come of this.
Screed Fintimlinbinwhinbimlim Bus Stop Poontang Poontang Ole Biscuit-Barrel
[ 12-30-2002: Message edited by: sCreeD ]</p>
i think of the end of the motorola agreement, the "we like options" statement, and the IBM 970. what's going on? well, in the motorola agreement Apple must have been left out with no choice, so now they are free to choose(Moto's contract ends now); well they probably went with ibm some time ago, probably closing an exclusive deal where performance had a pretty central part. is performance a threat to IBM? didn't think so. is it crucial to Apple? thought so. do IBM believe in SIMD? well, they didn't use to, but do now.. so as long as ibm keeps Apple in the performance lead, they got the whole pie. to secure both ends, apple is keeping a close watch at amd 64bit processors.
SPECULATING
i believe we will see new processor naming on all the chips from IBM, who is soon to be sole PPC provider. they probably will adapt a vector unit to the G3 and rename it too...
BYE-BYE G'this, G'that... welcome <insertnewname>6GB+ throughput.
this will happen in 03, but allready;
if the new deal runs from when the old ends.. IBM is bound to ship all the chips. and Steve to live up to the " **** you Motorola" he must have proclaimed at the time of the 350GHZ G4 (What a bummer..).
all chips after mwsf will be IBM, they will also get new names.
Schedules may have changed slightly, but maybe not by much, he seemed very confident. That's all I know.
Yeah. It...that is pretty weird. If the Moto' deal is at an end for any new chips...then...Moto' may give, or be obligated or Apple may only want current speeds.
For any new chips, say a 1 gig iMac flat then the next chip maybe the new G3 Macosrumors (yeah, I know they're a 'rumour' site... )/Thinksecret are talking about. G3 plus altivec plus 200mhz bus? Thinksecret says these chips going even higher re: mhz.
Yeah, I KNOW a g3 at 1 gig wouldn't be quite a G4 at 1 gig re: floating point.
BUT if there is an exclusive IBM deal in the offing then a 1 gig G3 on a 200 mhz bus with altivec would surely perform better and even cooler than an 800mhz G4? IBM seem to like the G3 chip and seem to leasurely proving a point to Motorola and Apple about whos chip tech' is best. IBM can hardly be said to improving the G3 at breakneck speeds of development. Take away the altivec and the fact the G3 is on a what? 100mhz bus? It aint mauled by a G4. The 100mhz bus masks little performance difference between the two chips bar altivec? Afterall the 'power'Mac is on a 167mhz bus. With DDR mem' (heh...) and the Powerbook has a faster bus? And higher mhz. Has anybody drawn a 800mhz G3 vs 800mhz G4 test bar altivec?
In short (I'm trying...), Moto' could be out on their ear after the current Mac specs run out? Maybe only supplying the Quark and Edu' '9' boot Macs?
"well they probably went with ibm some time ago, probably closing an exclusive deal where performance had a pretty central part. is performance a threat to IBM? didn't think so. is it crucial to Apple? thought so. do IBM believe in SIMD? well, they didn't use to, but do now.. "
Stick it on a G3 and clock them higher, say 1 gig-1.2 on a 200mhz bus...do we have the next iMac/emac revisions? There's been an insinuation about a 'new'/temporary cpu solution on a few rumour sites. Could this revised G3 be it?
"so as long as ibm keeps Apple in the performance lead, they got the whole pie."
That's what I've come round to thinking of late. Motorola's priorities lie elsewhere. IF Apple offers a guaranteed contract of all their four million sales on a return contract of enhanced performance....it would be hard not to see IBM and Apple going for it.
Apple will well and truly learned their lesson..?
IBM get the chip, the G3 and 970 (which they were going to develop anyway...), R&D payed for?
After the beating Motorola have given the 'power'Mac sales, I'd be surprised if Apple don't turf Motorola out on their ear at some point during 2003. If the agreement (puh...) came to an end in January...then a 2003 early bird surprise would be nice.
Steve Jobs inviting Moto exec' on to San Fran' stage. Giant washing machine crashes onto his head. Flattens him. On the side of the washing machine. 'G4'. Or bend over, 'Take this, Moto', this is the 970...can you feel the performance, huh, HUH?'
"...to secure both ends, apple is keeping a close watch at amd 64bit processors."
Maybe that is part of the 'Marklar' strategy every damn rumour site is covering. There's something in the air. Hypertransport? Perhaps. I won't rule out Sledgehammer either. If only to keep IBM on its toes? Or part of something more radical?
"SPECULATING
i believe we will see new processor naming on all the chips from IBM, who is soon to be sole PPC provider. they probably will adapt a vector unit to the G3 and rename it too..."
Yes. I could see this happening. The 'options' statement. A new naming scheme? Why not. It would put the G4 years behind us.
"BYE-BYE G'this, G'that... welcome <insertnewname>6GB+ throughput."
I'm dreamily watching the San Fran keynote in my head. Sigh. (whimsically...)
"this will happen in 03, but allready;
if the new deal runs from when the old ends.. IBM is bound to ship all the chips."
It sounds plausible? At some point?
If the Motorola deal 'ran out' in Jan'. Does that really mean an abrupt end to all Motorola/Apple ties?
Apple have the Powerbook, the 'power'Mac, the eMac, the iMac flat. The X-serve. That's a few lines there.
Would it be a phase out of Motorola during 2003 with the G4 chip gradually phased out in preference of the g3+/970?
"and Steve to live up to the " **** you Motorola" he must have proclaimed at the time of the 350GHZ G4 (What a bummer..).
all chips after mwsf will be IBM, they will also get new names."
Personally, I'd like to see it happen. Some Moto' payback.
Hey, I'm not bitter, I just remember Steve's blind Ati fury!
Lemon Bon Bon
<img src="graemlins/lol.gif" border="0" alt="[Laughing]" />
Sure wouldn't mind being surprised, however.
<strong>Sure wouldn't mind being surprised, however. </strong><hr></blockquote>
I think that all of us wouldn't mind being surprised.
I'm still waiting for KidRed to let loose with some more info about the inside knowledge he claimed he had about new hdw around MWSF. C'mon Kid... put the porn down and fess up! <img src="graemlins/smokin.gif" border="0" alt="[Chilling]" />