Apple Logic Pro /LogicExpress Discussion group photo

Yahoo Groups archive

Apple Logic Pro /LogicExpress Discussion

Index last updated: 2026-04-28 23:06 UTC

Thread

intel

intel

2005-06-06 by GAmoore@aol.com

Someone chastised me for mentioning the rumor last week, but now its no longer a rumour. I guess the the powermac G3..4..5.. line as we know it is over.

http://edition.cnn.com/2005/BUSINESS/06/06/apple.intel.ap/


They say it will happen by 2007 and that they will need to rewrite code for the new chip, and that when future software comes out (e.g. 10.6 Lion, 10.7 housecat, or whatever, Logic 9, etc) they won't work on our existing macs. Great. I have gone through this a couple of times with apple already.

Re: [Logic_Cafe] intel

2005-06-06 by Chris Coccia

GAmoore@... wrote:
>   Someone chastised me for mentioning the rumor last week, but now its 
> no longer a rumour. I guess the the powermac G3..4..5.. line as we know 
> it is over.
> 
> http://edition.cnn.com/2005/BUSINESS/06/06/apple.intel.ap/
> 
> 
> They say it will happen by 2007 and that they will need to rewrite code 
> for the new chip, and that when future software comes out (e.g. 10.6 
> Lion, 10.7 housecat, or whatever, Logic 9, etc) they won't work on our 
> existing macs. Great. I have gone through this a couple of times with 
> apple already.

Not chastizing, just reminding everyone that weve heard this rumor since 
about 2001 or so hehe.. At least now we can put it to rest..

As for new versions of things not working on 'old' Powermacs, I believe 
this was mentioned in the keynote: You can compile your binaries 
cross-platform and they should work no problem between the PPC and x86 
architecture. The only reason something shouldnt work is if theres some 
new x86 optimizations (ala Altivec) or if the developer doesnt hit that 
check box in XCode when doing the compiling.
I think itll be alot less painless than we think, but then again, well 
have to wait and see til the hardware comes out for sure..
After all OS X itself has been cross-platform and compiled and running 
secretly on an intel box at Apple HQ for 5 years now. :)

All I know is, just two days ago I decided instead of upgrading my 
G4/800 I was going to save and get a new dual 2Ghz G5.. Think Ill go 
back and get the Gigadesigns for the 800 and see what happens in the 
next year before buying a new Mac..

-- 
Chris

http://www.descentrecords.com

Re: [Logic_Cafe] intel

2005-06-06 by Maurits van de Kamp

>  Someone chastised me for mentioning the rumor last week,

As long as you don't mention rumours as facts (especially not their 
consequences) no one needs to be chastised :o)

> but now its no 
> longer a rumour. I guess the the powermac G3..4..5.. line as we know it is
> over.

Why? Will Jobs come to my home, rip the G5 out my PowerMac and put an Intel 
chip in? If they want to make a new line based on Itanium chips or whatever, 
it's fine with me. Not that I would buy one, but I'm happy with my system for 
now anyway. Or you mean future generations? Well, the G3 was over when G4 
came, G4 was almost over when G5 came (if it weren't for the fact that there 
are still no notebook G5 processors, which is one of the reasons for this 
move).

> They say it will happen by 2007 and that they will need to rewrite code for
> the new chip, and that when future software comes out (e.g. 10.6 Lion, 10.7
> housecat, or whatever, Logic 9, etc) they won't work on our existing macs.

Well the next version of OSX is already quite far in development so it would 
surprise me if there wouldn't be a PPC version. After that well.. new 
generations of software require new generations of hardware. I'm glad Apple 
has never been afraid to jump legacy boundaries. Otherwise we'd be working 
with souped-up 2GHz 65xx processors (and I can assure you a G5 is 
faster). ;o) It's a pity they make a similar move to an inferior processor, 
that is if it really will be an x86 chip, but so far only the rumours 
"confirmed" this; it might just as well be Itanium family (makes more sense 
too, fully 64 bit). Anyway, the big advantage of this jump will be price 
rather than quality/performance.

Besides, you're misinterpreting what it says. Of course when you port binaries 
to another processor it will not work on the first one, but no one is 
preventing any developper from keeping two versions. Logic ran on OS9 and OSX 
for 2 major versions for example. (And in fact, OSX has been Intel-ready for 
a while now, according to another article).

> Great. I have gone through this a couple of times with apple already.

Yes and I still don't see the problem some people have with this. Why is it 
some people either want the latest hardware with old software, or the latest 
software with old hardware.. I'd say either stick with what you have if 
you're happy, or get something new if you're not. It's pretty naive to expect 
software written in 3-4 years not to run out of power on today's machines 
anyway. PC users simply buy a new PC every two years because their system is 
too slow for current software, but they don't complain because there's no 
real compatibility issue (so they don't feel "forced"). But the result is the 
same, except that on a PC you have all the complications that come with 
compatibility to old systems.

Besides, in all the articles people are waving with now, all I see is "..for 
the low-end machines.." (mac mini, ibook) because PPC chips are too 
expensive. I still have to see whether this means there will never be a 
PowerMac G6 for example.

Don't get me wrong, I think Intel architecture sucks. But in general I'm not 
worried about processor jumps and in fact what I like about Apple is that 
they're not afraid of them and actually has managed to make those jumps in a 
reasonably painless way.

Maurits.

[Logic_Cafe] intel

2005-06-06 by Kamm Schreiner

> You can compile 
> your binaries cross-platform and they should work no problem 
> between the PPC and x86 architecture. The only reason 
> something shouldnt work is if theres some new x86 
> optimizations (ala Altivec) or if the developer doesnt hit 

If any program uses assembly language routines, I can't see any way to
simply recompile that code. It will have to be completely rewritten in x86
assembly language.

I don't know this, but I wouldn't at all be surprised if Logic has many
assembly language routines to optimize its audio engine. Anybody on this
list know the answer to that Q for sure?

Only time will tell...

Kamm

Re: [Logic_Cafe] intel

2005-06-06 by Peter Duemmler

Kamm Schreiner wrote:
> I don't know this, but I wouldn't at all be surprised if Logic has
> many assembly language routines to optimize its audio engine.

Hey, that would be a chance to rewrite a lot of old buggy code!
That\ufffds actually a good thing.
But what do I know, I\ufffdm no programmer.

Peter
---
http://www.merlinsound.de

Re: [Logic_Cafe] intel

2005-06-06 by Chris Coccia

Kamm Schreiner wrote:

> If any program uses assembly language routines, I can't see any way to
> simply recompile that code. It will have to be completely rewritten in x86
> assembly language.
> 
> I don't know this, but I wouldn't at all be surprised if Logic has many
> assembly language routines to optimize its audio engine. Anybody on this
> list know the answer to that Q for sure?
> 

According to Jobs, if devs have been using XCode its only a 'matter of 
minutes to recompile'.. Microsoft Office apparently needs no recompiling 
already.. Wolfram took a matter of minutes and only needed about 20 
lines of code changed to bring it up to speed.. I would HOPE Logic is 
being developed in a similar matter, you would think Apple would have 
been planning on this and made appropriate measures awhile ago..


-- 
Chris

http://www.descentrecords.com

[Logic_Cafe] intel

2005-06-06 by Kamm Schreiner

> According to Jobs, if devs have been using XCode its only a 
> 'matter of minutes to recompile'.. Microsoft Office 
> apparently needs no recompiling already.. Wolfram took a 
> matter of minutes and only needed about 20 lines of code 
> changed to bring it up to speed.. I would HOPE Logic is being 
> developed in a similar matter, you would think Apple would 
> have been planning on this and made appropriate measures awhile ago..

Xcode is a 'C' based programming language and if a program is written
totally with XCode, then yea, I suspect it would be dead easy to recompile.
I can't see Office needing the kind of optimization that Logic might.
However, I *don't* know that Logic uses any assembly language routines. It
might not. I'm only suggesting that if it does, it may not be trivial to
convert it to run on an x86 CPU.

We're just going to have to wait to see I suspect.

Kamm

Re: [Logic_Cafe] intel

2005-06-07 by GAmoore@aol.com

Logic is an old program. And all of my experience with old programs is that they have old parts ('if it ain't broke, don't fix it'). In industry they are using 20 year old programs - because they work and there is no point to spending hundreds of thousands for a little more efficiency. Logic does seem to suffer from a lack of efficiency compared to the other major music programs - which could mean that they did not use assembly language (which can be much faster) - or it means Logic has a lot of old code. Also, I recall a few years ago when I was asking for multiple undo, and f-erenc responded that it would take a major re-write to get that...which they evidently did...but its unclear how much they did.

If I were Apple, I would re-do Logic from scratch - they have a big investment in it and its part of their constellation of applications, and a showpiece to sell macs. they need it topnotch. There are a lot of buggy things about the new version (sometimes I long for version 4 or 5 which were stable).

Re: [Logic_Cafe] intel

2005-06-07 by Maurits van de Kamp

> If I were Apple, I would re-do Logic from scratch

I think they more or less did that with 7 (most of the changes were about 
replacing Emagic software components with Apple ones).

> showpiece to sell macs. they need it topnotch. There are a lot of buggy
> things about the new version (sometimes I long for version 4 or 5 which
> were stable).

Because it's rewritten; that also means new bugs. :o)

Maurits.

Re: [Logic_Cafe] intel

2005-06-07 by Eddie Sullivan

On Monday, June 6, 2005, at 07:01 PM, Kamm Schreiner wrote:

> If any program uses assembly language routines, I can't see any way to
> simply recompile that code. It will have to be completely rewritten in 
> x86
> assembly language.
>
> I don't know this, but I wouldn't at all be surprised if Logic has many
> assembly language routines to optimize its audio engine. Anybody on 
> this
> list know the answer to that Q for sure?


Isn't this recompilation what this Rosetta thingy is all about?

Eddie
IMS

RE: [Logic_Cafe] intel

2005-06-07 by Kamm Schreiner

> Isn't this recompilation what this Rosetta thingy is all about?

I'm getting the impression that there are actually two things being talked
about. One is actually recompiling the software using xcode and the other is
a hardware emulator. My comments were with respect to the prior. The prior
method provides a native application for the CPU being used and would not
have any speed impact. The emulator allows a non-native application to run
on the CPU, but has an impact on speed.

Kamm

Re: [Logic_Cafe] intel

2005-06-07 by Chris Coccia

Kamm Schreiner wrote:

> I'm getting the impression that there are actually two things being talked
> about. One is actually recompiling the software using xcode and the other is
> a hardware emulator. My comments were with respect to the prior. The prior
> method provides a native application for the CPU being used and would not
> have any speed impact. The emulator allows a non-native application to run
> on the CPU, but has an impact on speed.
> 

Im thinking this dual binary thing is gonna be like the old Carbon 
applications we used in the beginning of the OS X days that ran under 
both OS 9 and X.. Guess that makes Rosetta kinda like the new Classic 
environment eh?


-- 
Chris

http://www.descentrecords.com

Re: [Logic_Cafe] intel

2005-06-07 by amgmamgma

GAmoore@... <GAmoore@...> wrote:
: 
: They say it will happen by 2007 and that they will need to rewrite code for 
: the new chip, and that when future software comes out (e.g. 10.6 Lion, 10.7 
: housecat, or whatever, Logic 9, etc) they won't work on our existing macs. Great. 
: I have gone through this a couple of times with apple already.

I wouldn't worry about future software not running on your existing Macs
any more rapidly than it has been happening anyway. (i.e., you can't run
10.3 on a beige PowerMac G3; you can't run 10.4 on a pre-firewire Mac,
etc.)

Apple and third party vendors will continue to provide apps that run on
PPC probably until the last shipping PPC machine (and Macs with PPCs in
them will continue to ship into early 2007) is at least three years old or
so, maybe longer depending on how this really rolls out. By the time your
dual G5 is no longer supported it will be pretty old.

-- 
 agreenbu @ nyx . net                             andrew michael greenburg

RE: [Logic_Cafe] intel

2005-06-07 by Kamm Schreiner

> Guess that makes Rosetta 
> kinda like the new Classic environment eh?

Yes. I agree, it does seem that way.

Re: [Logic_Cafe] intel

2005-06-08 by GAmoore@aol.com

> Guess that makes Rosetta
> kinda like the new Classic environment eh?

Yes. I agree, it does seem that way.



classis is an emulation that runs in x. Rosetta is a translation compiler which converts code for what processor to the other. Rosetta does not run in real time on your machine. it runs on the developer's machine.

Looks like G5's will be going dirt cheap. who would buy a new G5 now knowing its obsolete. just wait 18 months and get the new generation.

RE: [Logic_Cafe] intel

2005-06-08 by Kamm Schreiner

> classis is an emulation that runs in x. Rosetta is a 
> translation compiler which converts code for what processor 
> to the other. Rosetta does not run in real time on your 
> machine. it runs on the developer's machine.

I'd like to read more about Rosetta. It appears that my impressions of it
are wrong. Where'd you get that info?

> Looks like G5's will be going dirt cheap. who would buy a new 
> G5 now knowing its obsolete.

Yea. I hope so! ;-)

Kamm

Re: [Logic_Cafe] intel

2005-06-08 by Garth McDougall

On 6/7/05 6:01 PM, "GAmoore@..." <GAmoore@...> wrote:

>>> Guess that makes Rosetta
>>> kinda like the new Classic environment eh?
>> 
>> Yes. I agree, it does seem that way.
>> 
> 
> 
> classis is an emulation that runs in x. Rosetta is a translation compiler
> which converts code for what processor to the other. Rosetta does not run in
> real time on your machine. it runs on the developer's machine.

Actually, Rosetta is an application that will run on Intel based Macs in
order to execute PPC coded programs that have not yet been ported. There is
a performance hit, but performance will be ³fast enough² according to Steve
Jobs.

He actually used it at the keynote to run Photoshop CS2 (and a couple of
other programs) on the Intel based Mac he was using. The keynote video is
here:  <http://www.apple.com/quicktime/qtv/wwdc05/>. I think the Intel Mac
stuff starts at around the 30 minute mark (the whole video is about an
hour).

Also, you can go to apple.com/developer and do a search for Rosetta and find
out more than you ever wanted to know. It seems that Rosetta actually
translates the PPC code in real time - so it isn't really an emulator - and
it launches automatically when needed and runs invisibly.

Xcode 2.1 is the program used by developers to produce universal binaries
that will install native code for whatever processor the machine is using.

Garth

[Logic_Cafe] intel

2005-06-09 by Kamm Schreiner

> Actually, Rosetta is an application that will run on Intel 
> based Macs in order to execute PPC coded programs that have 
> not yet been ported. There is a performance hit, but 
> performance will be ³fast enough² according to Steve Jobs.

Thanks Garth. That's what I thought, but I was starting to think I had the
wrong impression.

Kamm

Re: [Logic_Cafe] intel

2005-07-16 by kenneth brill

It's so offensive. I waited a year in a half for a
64bit operating system for my G5. So much hype. Two
weeks after Tiger comes out, I find out "the wabe f
the future" has been already been issued an expiration
date. I wish I could sue. Not for the money, just the
principle.

--- GAmoore@... wrote:

>  Someone chastised me for mentioning the rumor last
> week, but now its no 
> longer a rumour. I guess the the powermac G3..4..5..
> line as we know it is over.
> 
>
http://edition.cnn.com/2005/BUSINESS/06/06/apple.intel.ap/
> 
> 
> They say it will happen by 2007 and that they will
> need to rewrite code for 
> the new chip, and that when future software comes
> out (e.g. 10.6 Lion, 10.7 
> housecat, or whatever, Logic 9, etc) they won't work
> on our existing macs. Great. 
> I have gone through this a couple of times with
> apple already.
> 



	
		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - You care about security. So do we. 
http://promotions.yahoo.com/new_mail

Re: [Logic_Cafe] intel

2005-07-16 by Maurits van de Kamp

> It's so offensive. I waited a year in a half for a
> 64bit operating system for my G5. So much hype. Two
> weeks after Tiger comes out, I find out "the wabe f
> the future" has been already been issued an expiration
> date. I wish I could sue. Not for the money, just the
> principle.

Sigh. Get real. You thought there is ever going to be a processor or computer 
generation without an expiration date? The Intel Macs will be the generation 
coming after the current one. Does it matter whether your G5 will be 
obsoleted by a G6 or an Intel chip? The only way companies can keep going is 
if they start designing a followup of their product as soon as the current 
version hits the market. So yes, when you buy it, the expiration date is 
already set. Whether it's a coffee machine, a car or a computer.

By the way, the Intel Macs will still run OSX (by then Tiger will probably be 
obsoleted by OS10.5 with whatever predator name they give it) so I don't see 
what you're complaining about.

Ok back when the 386 came out, it was understandable that people thought this 
was THE cpu that would last for decades. But after 486, PI, PII, PIII, PIV 
and on the PPC side G3, G4 and G5, we know better. And if it weren't for the 
Intel story, the expiration on the G5 would have still been close at hand 
because of IBM's announcement of dual core PPC's, be it called G6 or not, and 
since this announcement still two series of G5 machines have been released, 
maybe even the one you bought.

A new generation of computers will do nothing to change the one you currently 
have (even if they have a different processor family). The only change is in 
your head, when you'll be considering your system lacking of features or 
possibilities once you see them in the new one, or in software requiring the 
new machine. You'll still have all the equipment and software you paid for, 
and it will be doing at least what it's doing now.

Sure you could sue, if it weren't for the total lack of anything to sue about.

Maurits.

Re: [Logic_Cafe] intel

2005-07-16 by GAmoore@aol.com

Get real. You thought there is ever going to be a processor or computer
generation without an expiration date?


I just tried running the Reaktor demo on my ibook G3 but got an error message that altivec is needed, so it would not run at all. Altivec came with the G4 and I think its included in the G5 in some way. If Apple had somehow been able to get the G5's up to 3ghz has Jobs promised a year ago, and then move to a G6, then we could pretty well assume that future OS updates and software will continue to work with our G5's for a few years. However, in a year from now, when Mactels come out, our days are numbered, so we will have to unload the G5's at a big loss, and buy new machines. Of course, I do not blame apple, since they were dependent on IBM and these are business decisions. But it will indeed cost us if we want to keep on the cutting edge.

Re: [Logic_Cafe] intel

2005-07-16 by Maurits van de Kamp

> However, in a year from now,
> when Mactels come out, our days are numbered, so we will have to unload the
> G5's at a big loss, and buy new machines.

Why? Seeing that compiling software for OSX/Intel or OSX/PPC is just a matter 
of setting a compiler macro in most cases, I doubt that any new software 
written a year from now will not support PPC anymore. Software manufacturers 
want to sell their software, so it would be stupid of them not to give that 
little bit of effort. Sure the PPC share will decline, but as history teaches 
us that will take a few years more.Besides, nobody's putting a gun to your 
head to force you to run the latest software only. In total that means your 
G5 becomes obsolete in the amount of time you should normally expect it to 
anyway. 

If in a year the G6 PowerMac would have come out, our G5s still would be worth 
a fraction of what they cost in the first place. It's how things work. It's 
called progress. Either you need the progress so you should buy a new 
machine, or you don't need it and stick to what you've got.

Maurits.

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.