Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Intel forms 'Apple' Group
- Thread starter MacRumors
- Start date
- Sort by reaction score
ZorPrime
macrumors regular
TrenchMouth said:
I completely agree. I think this is a very promising and significant announcement for a couple of reasons:
1. Like already stated; it signifies a material commitment to Apple, which is good for Mac users. In terms of research and development, it also could signify a major strategic initiative between Apple and Intel.
2. I noticed the reference to “Skunkworks”, to me, that’s a really significant insinuation; Lockheed’s Skunkworks is centered on the development of ultra-highly advanced research and technology. To specifically reference Skunkworks strongly suggests this announcement is research and development related… maybe we’re looking at a new AIM type relationship in which Apple and Intel will develop an entirely new breed of CPU. 😎
I for one will be looking forward to the 3rd and beyond generations Intel based Mac…
hayesk said:
Yes, this presumably was a continuation of the OPENSTEP environment that ran on Windows NT that I mentioned in a post above.
Apple/Intel Skunkworks
As ZorPrime suggested, I think an Applegroup at Intel is extremely promising. There have been many discussions of how Intel has been frustrated with the combination of Microsoft and OEMs like Dell refusing to market CPU innovations (from MS's POV, CPU innovations are more support headaches, and from Dells POV unless it makes their computers cheaper or better at running Windows CPU innovations are worthless), and that Intel was courting Apple primarily as a vendor that will showcase their new technologies.
I expect Yonah iMacs and 'books at MWSF in a couple of months, to be followed shortly by Memrom-based machines, as these chips are already in the pipeline and they will keep Apple's machines competitive for the next couple of years. However, I expect we will see some very exciting new hardware emerging in the form of the 2008 MacIntels.
Interesting times indeed.
Cheers
As ZorPrime suggested, I think an Applegroup at Intel is extremely promising. There have been many discussions of how Intel has been frustrated with the combination of Microsoft and OEMs like Dell refusing to market CPU innovations (from MS's POV, CPU innovations are more support headaches, and from Dells POV unless it makes their computers cheaper or better at running Windows CPU innovations are worthless), and that Intel was courting Apple primarily as a vendor that will showcase their new technologies.
I expect Yonah iMacs and 'books at MWSF in a couple of months, to be followed shortly by Memrom-based machines, as these chips are already in the pipeline and they will keep Apple's machines competitive for the next couple of years. However, I expect we will see some very exciting new hardware emerging in the form of the 2008 MacIntels.
Interesting times indeed.
Cheers
rlwimi said:
Has anyone recently checked Intel's announcement of Yonah/Merom power consumption with their various power-usage versions? As of yesterday?
rlwimi said:
Note the news, this morning, that Rosetta will now include Altivec support.
rlwimi said:
Depends on the system running Rosetta, which is significantly better performing than traditional emulation. Rosetta isn't emulating a G3/G4, by the way.
rlwimi said:
Depends on the power draw of the rest of the motherboard; if you assume that the CPU is the sole criterion of power consumption, you're wrong.
rlwimi said:
Which lies? Serious charge, Ace.
rlwimi said:
Try the decaf for a change.
AidenShaw
macrumors P6
Altivec support is not that important
BTW, support for Altivec is interesting, but not that important.
It would be fairly easy for QuickTransit to decode Altivec instructions to its intermediate language - since most Altivec instructions can be simply expressed as "do this operation on sets of 4 to 16 items that are side-by-side in memory".
In essence, that would simply convert a G4 or G5 instruction stream to a G3 instruction stream. Useful to run programs that *require* Altivec, but not necessarily any faster.
What would be noteworthy would be for QuickTransit to *generate* SSE2 instructions in its optimizing kernel. That would mean recognizing that the intermediate language describes parallel operations, and outputing SSE2 instructions to do them in parallel.
Something especially interesting about this, however, is that if the code generator could do that - it could generate SSE2 instructions for programs that didn't use Altivec originally!
In essence, it could turn a G3 into a G4 and at times run even faster than the original.
(QuickTransit technical brief)
Link, please.... Not even Apple has posted this news.sehix said:
BTW, support for Altivec is interesting, but not that important.
It would be fairly easy for QuickTransit to decode Altivec instructions to its intermediate language - since most Altivec instructions can be simply expressed as "do this operation on sets of 4 to 16 items that are side-by-side in memory".
Code:
For example,
int a[4], b[4], c[4];
c = altivec_add (a,b)
is the same as
int a[4], b[4], c[4], i;
for (i=0; i<4; i=i+1)
c[i] = a[i] + b[i];
=========================================================
and
int a[4096], b[4096], c[4096], i;
for (i=0; i<4096; i=i+4)
c[i] = altivec_add (a[i], b[i])
is the same as
int a[4096], b[4096], c[4096], i, j;
for (j=0; j<4096; j=j+4)
{
for (i=j; i<j+4; i=i+1)
c[i] = a[i] + b[i];
}
which is also the same as
int a[4096], b[4096], c[4096], i;
for (i=0; i<4096; i=i+1)
c[i] = a[i] + b[i];What would be noteworthy would be for QuickTransit to *generate* SSE2 instructions in its optimizing kernel. That would mean recognizing that the intermediate language describes parallel operations, and outputing SSE2 instructions to do them in parallel.
Something especially interesting about this, however, is that if the code generator could do that - it could generate SSE2 instructions for programs that didn't use Altivec originally!
In essence, it could turn a G3 into a G4 and at times run even faster than the original.
(QuickTransit technical brief)
MacsRgr8
macrumors G3
AidenShaw said:
http://www.osx86project.org/index.php?option=com_content&task=view&id=68&Itemid=2
Still a rumor
Multimedia
macrumors 603
Who Cares About Intel Graphics On The Box
Apple never puts graphics of any sort on their computers any more. Only the prominent Apple logo.
You have got to be kidding. Faster is better no matter what graphics they put on the box.😛Piarco75 said:
Apple never puts graphics of any sort on their computers any more. Only the prominent Apple logo.
Multimedia
macrumors 603
Tuesday January 10, 2006 PowerBooks
That will be Tuesday January 10, 2006. Yes. 😀 😛 🙂rxse7en said:
I was under the impression that the latest Intel chips, like Yonah, were made power hungery. I saw someone mention earlier that there are new press releases on those chips, insinuating that they are doing better in that department. Where do I go to find this?
Since I have not seen this above mentioned press release I have to go on what I have read on Cnet and Ars. From those places it is really disconcerting to see that the first chips since Intel's public commitment to performance per wattage gains have come to be far from impressive. At the same time this happens, IBM comes out with the 970FX which apparently does a better job, though I have not seen any hard data, in that department than Apple lead us to believe.
I am not trying to flame the new Intel Mac arrangement, just showing concern about the fact that Intel's results have not yet matched Apple and their promises.
Since I have not seen this above mentioned press release I have to go on what I have read on Cnet and Ars. From those places it is really disconcerting to see that the first chips since Intel's public commitment to performance per wattage gains have come to be far from impressive. At the same time this happens, IBM comes out with the 970FX which apparently does a better job, though I have not seen any hard data, in that department than Apple lead us to believe.
I am not trying to flame the new Intel Mac arrangement, just showing concern about the fact that Intel's results have not yet matched Apple and their promises.
ezekielrage_99
macrumors 68040
rlwimi said:
It's a well know industry fact that Intel's first release of any consumer product is always rushed so there's always a few major bugs in it..... heat, speed, stability issues and insane overcloakability comes to mind 🙂
Remember the rushed release of the Celeron 300A??? The Intel guys got that one wrong because that processor could be clocked to 450Mhz no problems and it could easily outperform the 400Mhz Pentium equivelant.
Still I kind of feel that we are going to have one of those cool wow Apple moments in January....
Chundles
macrumors G5
No no no, you guys have it all wrong.
The Apple group formed by Intel are currently going through the sets of Buffy The Vampire Slayer looking through all the old books for some sort of incantation that will save their company from the Motorola Curse.
Intel is a big company and will resort to anything, including the occult to ensure that the processors they provide will not be forced down in speed, get stuck in development for nearly a year and then peter out at the top end of the scale with no possibility of improvement.
Hellboy is rumoured to be a member.
The Apple group formed by Intel are currently going through the sets of Buffy The Vampire Slayer looking through all the old books for some sort of incantation that will save their company from the Motorola Curse.
Intel is a big company and will resort to anything, including the occult to ensure that the processors they provide will not be forced down in speed, get stuck in development for nearly a year and then peter out at the top end of the scale with no possibility of improvement.
Hellboy is rumoured to be a member.
