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.
No Keynote at Apple Expo Paris
- Thread starter Chundles
- Start date
- Sort by reaction score
I for one am very pleased that there is no keynote....
... think of it as this way, any products Apple releases now are merely going to be excuses to hold off the releases of the wonderful upcoming MacIntel machines.
So what if Apple announces a, oh, 1.7Ghz G4 PowerBook!11!11111!, in MWP? So it is 70Mhz faster!!! Are you going to buy it?
Well you can go on and buy it.. and be happy with that "bargain!" I for one am happy that Steve releases just how lacklustre their lineup currently is and put more heat on the R&D department to speed things up a tad.
... think of it as this way, any products Apple releases now are merely going to be excuses to hold off the releases of the wonderful upcoming MacIntel machines.
So what if Apple announces a, oh, 1.7Ghz G4 PowerBook!11!11111!, in MWP? So it is 70Mhz faster!!! Are you going to buy it?
Well you can go on and buy it.. and be happy with that "bargain!" I for one am happy that Steve releases just how lacklustre their lineup currently is and put more heat on the R&D department to speed things up a tad.
FF_productions said:I had to come home to this? Does this mean no Dual-Cores??
That's already been addressed:
https://forums.macrumors.com/posts/1715401/
psycho bob said:
One problem, many pro apps use Alti-vec which is not supported by Rosetta. So now users have fancy Intel based Macs that won't run the software they need for business.
Yes, maybe. If it has a 7448 G4 it'll run cooler and faster than any recent PowerBook. Plus, it'll run all my current software without the need for emulation or upgrades. It'll run for several years until the Intel mess is smoothed out.generik said:
I can't understand why so many people are anxious for Intel processors to arrive in Macs. It's just going to be a big pain in the ass.
It would have been so much simpler if the Lord God Jobs had ditched PPC when he ditched OS 9.iDave said:
OSX==Intel, Classic==Transitive+OS9
But, we all know that hindsight is 20-20....
ps: "Transitive" is the company that innovated the technology named "QuickTransit" that Apple is marketing as "Rosetta" -- just in case anyone thinks that any part of "Rosetta" other than the name came from Cupertino.... (Transitive is in Manchester/Los Gatos http://www.transitive.com/)
Only if a change to Airport is needed for some new iPod feature - the iPod Computer Company isn't investing in Macs these days....themacman said:
[/sarcasm]
digitalbiker said:
Well that was one of my arguements with the Rosetta concept orginally because almost everything now has some altivec utilisation. Apps like Photoshop take advantage of it but will also run on G3's with out the need for extra code which isn't included on the disks.
How much of a software applications need for a specific processor is tied to speed as opposed to a specific feature we don't really know. I'm pretty sure Apple will add some kind of altivec support to Rosetta even if it isn't optimized. The problem with the current situation is each developer is exxpected to creat x86 mac apps but the time scale is all over the place. A maker of Pro apps may hurry the process along but may only get Powermacs to use it on late in 2007. It must be a real headache to companies like Adobe who release PC and mac together knowing how to invest their time best.
psycho bob said:
apple is probably going to use stock parts, agreed, but what if... what if apple still thinks about developing altivec by themselves, and asking intel to build such a cpu that has apple-designed simd unit? a lot of multimedia content creation apps depend on altivec and obviously apple needs to address that issue.
joshuawaire said:
let's hope rosetta some day converts altivec instructions into whatever simd unit the apple-chosen x86 cpu is using. currently rosetta only converts ppc instructions to x86 and nothing else.
apps that use altivec WILL NOT RUN VIA ROSETTA.
generik said:
if it has a L3 cache, then it'd be a huge improvement over the previous model. the reason why i could see this rumor happening is that us pro folks are not switching our production systems to a rev-A intel macs and we would really benefit from great last-generation ppc powerbooks and powermacs. and steve knows this, too. it's a possibility.
JFreak said:
The thing is all the apps will be rewritten for x86 so there is no need to use an Altivec based system. We'll be getting used to SSE2/3 and the various Intel technologies. If Intel started incorporating Apple designed element into their chips they are setting up special production lines to do so this does not ensure the consistant stream of mass produced chips SJ wants.
The Altivec patent is part owned by Motorola and IBM as well, they will not license such technology to Intel and it isn't really worth Apple designing custom elements to give to Intel. That's like the professor being taught by the first year student.
psycho bob said:
x86 sucks in computing floats and sse is not good enough either. ever read bencmarks between intel's and amd's offerings? amd totally kills intel with floating point performance -- and float performance (altivec) is the most important benchmark for us multimedia content creators.
apple is in competition with avid in both audio (logic vs protools hd) and video (fcp vs nitris). in other words, apple has the features almost on par but has to compete with host hardware performance compared to dedicated dsp.
apple cannot afford crappy float performance, period. therefore intel must have something better than x86+sse3 to offer.
psycho bob said:
yes it does. it's just as easy than producing a flagship pentium and a crippled one. intel knows the trick, it's easy for them.
psycho bob said:
apple owns the patent partially, so i believe licensing should not cost too much. but nobody knows.
JFreak said:
Well whether future chips improve on floating point through basic design or with the addition of a special unit it does not prevent an Intel product being released now with current x86 technology. Any altivec type unit will be a performance addition basic coding will be the same. A Pentium M Powerbook released now could be thought of as a G3 equivalent albeit considerably faster. If G4 Powerbooks are not going to get substatially faster then why not let pros buy the remaining stock and start to offer an Intel based machine to get tongues wagging and stock prices going up. This would also demonstrate to the rest of us that something is actually happening in Cupertino.
you contradict yourself...
Benchmarks? Look at http://www.spec.org/cpu2000/results/cfp2000.html for the HP Intel and Opteron numbers - Xeons are getting around 1700 to 1800, Opterons are 1800 to 1900. Faster, but "totally kills" is hyperbole worthy of a Stevenote.
Intel could do killer floating point if it needed to - and it could give SSE a separate set of vector math units, it it needed to. Intel's been concentrating on integer performance, but that could change. Maybe the Steve's seen plans for a multi-core chip with SSE3 subsystems that absolutely roast AltiVec (and does it in double precision - something that AltiVec cannot do, it's 32-bit float only).
x86/x64 doesn't require poor floating point - within the x86 and x64 architectures different implementations could do different things. You admit this with your "Intel vs. AMD" statement earlier.
This doesn't make sense - Pentium/Xeon and Athlon/Opteron are all "x86" processors. You can't say "x86 is bad" and "x86 is good" in the same sentence....JFreak said:
Benchmarks? Look at http://www.spec.org/cpu2000/results/cfp2000.html for the HP Intel and Opteron numbers - Xeons are getting around 1700 to 1800, Opterons are 1800 to 1900. Faster, but "totally kills" is hyperbole worthy of a Stevenote.
Intel could do killer floating point if it needed to - and it could give SSE a separate set of vector math units, it it needed to. Intel's been concentrating on integer performance, but that could change. Maybe the Steve's seen plans for a multi-core chip with SSE3 subsystems that absolutely roast AltiVec (and does it in double precision - something that AltiVec cannot do, it's 32-bit float only).
x86/x64 doesn't require poor floating point - within the x86 and x64 architectures different implementations could do different things. You admit this with your "Intel vs. AMD" statement earlier.
Now that Steve Jobs has backed out of giving the keynote address, what are the chances that a dual-core 970MP PowerMac will be announced at the Paris Expo? I'm very interested in the answer to this question since I just started a new job and they are going to buy me a new computer. I don't know whether to buy now or wait for the 970MP to make its appearance. Any suggestions or comments?
Chundles said:
JFreak said:
Get your head out of the hyperbole! Apple cant unilaterally license AltiVec without IBM or Motorolas consent. All three own the floating point and SIMD instruction set technology(AltiVec, Velocity Engine, and VMX); therefore, all three have to agree to ANY licensing or selling of the floating point and SIMD instruction set, as developed by the AIM Alliance. Even if they could split up their "portions", what use would it be to a potential buyer?
"Wouldn't cost too much"...?
By your logic, Apple should violate a Patent Agreement by selling/licensing tech it doesnt 100% own and subject itself to a multi-Billion $$ lawsuit, by the partners that CO-own "AltiVec", it will loose. Good thing Apple doesnt have lawyers who use your logic! 
So there won't be a keynote at the Paris thing because he already did his thing for the month today. Maybe they just couldn't wait to bring that stuff out. Or maybe they are coming out with so much stuff in Paris they couldn't fit the phone and the iPod Nano into one event, and that's why they had this little thing. But then why isn't there going to be a keynote in paris? I think I might be wrong with some stuff, but I'm sure everyone else will straighten me out.
AidenShaw said:
exactly. my point was all about that it MUST do better, as apple needs great floating point performance. there is no chance for apple software to compete with dedicated hardware if the future intel systems cannot excel in float performance.
AidenShaw said:
steve did see plans of 3GHz G5's too. promised delivery within a year, now it's already two and a half years and still nothing. steve has great tech with altivec NOW and intel really has to have proven their simd unit superior in real world for steve to believe.
the pro audio world wants that intel system to roast everything apple sells right now. we can never have too much processing power. it's just that altivec does fantastic job in real world audio software and it's going to be one tough job for SSE3 to surpass. in its current implementation it sucks. let's hope for something better.
You should hope for improvements in the SSE3 implementation, since that will benefit all Intel systems (and help Intel outperform AMD). It will also be completely compatible with existing applications on Windows and Linux, and could exceed the performance of AltiVec, especially in 64-bit mode and on 64-bit floating point (where AltiVec fails to help).JFreak said:
Any notion that Apple can get Intel to glue AltiVec onto a special x86 chip is "crazy talk". Intel isn't in the low-volume custom processor business. Apple couldn't afford a special run of AltiVec-enhanced x86 chips, and companies like Adobe would resist special "Apple_x64" optimizations and stick with the SSE3 stuff that works on all the x64 platforms.
BTW, one thing that I find puzzling in the transition to Intel is that Apple will be doing 32-bit x86. If Apple had settled on x64-only, then it wouldn't be facing another 64-bit transition in short order (and fat binaries wouldn't need a third code branch for x64 (in addition to x86 and PPC)).
I can only guess that Apple is so desperate for Yonah that they can't wait for 64-bit dual-core in the laptops. (Note that Intel's desktop and server processor line supports 64-bit on all newer CPUs, even the dual-core. Even the current Celerons are 64-bit.)
AidenShaw said:
probably. but that's always the case when we're talking about rumors, isn't it
what makes it little less crazy is the fact that it is also crazy talk to expect apple building such systems that cannot exceed previous architecture's performance in multimedia content creation, where apple competes with host performance against dedicated dsp.whatever gives me more processing power in protools without sacrificing stability is fine with me.
I hate to read between the lines....but....
Sounds like something is going to be announced if by no other reason that Jobs will be there to take questions. I can't imagine them NOT announcing something, because what the hell is the press going to ask about? "Hey Steve, how was your summer?" Or "Nice new iPods, when are you going to revamp the 9 month old PowerBooks?" Jobs: "Sorry MacCentral, we don't comment on unannounced products."
Sounds like something is going to be announced if by no other reason that Jobs will be there to take questions. I can't imagine them NOT announcing something, because what the hell is the press going to ask about? "Hey Steve, how was your summer?" Or "Nice new iPods, when are you going to revamp the 9 month old PowerBooks?" Jobs: "Sorry MacCentral, we don't comment on unannounced products."
I still believe there will be a PB update. I believe Steve is not going to give a keynote because he's not going to extol a technology he's about to ditch. Most of the other products were recenlty updated. It's going to be a quite release for the PBs.