Sign in with
Sign up | Sign in
Your question

Mac to Switch to Intel?

Last response: in Home Audio
Share
Anonymous
June 6, 2005 11:54:49 AM

Archived from groups: rec.audio.pro (More info?)

Announcement today. Anyone holding your breath? Anyone own Intel
stock? Anyone own Apple stock?


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

More about : mac switch intel

Anonymous
June 6, 2005 11:54:50 AM

Archived from groups: rec.audio.pro (More info?)

In article <znr1118055742k@trad>, mrivers@d-and-d.com wrote:

> Announcement today. Anyone holding your breath? Anyone own Intel
> stock? Anyone own Apple stock?

I've seen a couple suggestions on this topic that add some more
information. The top theories are that 1) Apple will use a "just in time"
code translator called Quick Transit (which would allegedly allow x86 to
run PPC code with no modification or performance penalty) and 2) that
Intel will just *manufacture* PPC chips for Apple, then create new, faster
PPC designs into the future. Another variant on this latter theory is that
Intel will add Altivec to x86.

--
Jedd Haas - Artist
http://www.gallerytungsten.com
http://www.epsno.com
Anonymous
June 6, 2005 12:54:15 PM

Archived from groups: rec.audio.pro (More info?)

In article <znr1118055742k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
>
>Announcement today. Anyone holding your breath? Anyone own Intel
>stock? Anyone own Apple stock?

It's kind of sad and depressing, in that it means the whole world is
basically now dependant on a single hardware architecture for everything.

Having grown up in a world where hundreds of different instruction sets
existed, and where VLIW machines and stack machines were mainstream
architectures, I find it sad to see all of the things that made computers
interesting and exciting have been reduced to this.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Related resources
Anonymous
June 6, 2005 12:55:22 PM

Archived from groups: rec.audio.pro (More info?)

Jedd Haas <jnh@epsno.com> wrote:
>In article <znr1118055742k@trad>, mrivers@d-and-d.com wrote:
>> Announcement today. Anyone holding your breath? Anyone own Intel
>> stock? Anyone own Apple stock?
>
>I've seen a couple suggestions on this topic that add some more
>information. The top theories are that 1) Apple will use a "just in time"
>code translator called Quick Transit (which would allegedly allow x86 to
>run PPC code with no modification or performance penalty) and 2) that
>Intel will just *manufacture* PPC chips for Apple, then create new, faster
>PPC designs into the future. Another variant on this latter theory is that
>Intel will add Altivec to x86.

What about the huge number of people already running 68k applications on
their PPC machine with translation? Are those applications going to be
supported in the new era?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
June 6, 2005 3:04:02 PM

Archived from groups: rec.audio.pro (More info?)

I was hoping the announcement would be about a proprietary two-button
mouse....I dream big!

Scott Dorsey wrote:
> In article <znr1118055742k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
> >
> >Announcement today. Anyone holding your breath? Anyone own Intel
> >stock? Anyone own Apple stock?
>
> It's kind of sad and depressing, in that it means the whole world is
> basically now dependant on a single hardware architecture for everything.
>
> Having grown up in a world where hundreds of different instruction sets
> existed, and where VLIW machines and stack machines were mainstream
> architectures, I find it sad to see all of the things that made computers
> interesting and exciting have been reduced to this.
> --scott
>
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
June 6, 2005 4:21:21 PM

Archived from groups: rec.audio.pro (More info?)

"Mike Rivers" <mrivers@d-and-d.com> wrote in message
news:znr1118055742k@trad...
>
> Announcement today. Anyone holding your breath? Anyone own Intel
> stock? Anyone own Apple stock?



Industry analysts predict this could cost IBM sales of as many as dozens
of units per year.

When asked how the loss of Apple's business would affect IBM, CEO B. Lou
Biggs replied, "Who? Gee, I don't know... you'd have to check with
Sales. Of course, we hate to lose *any* customer, even... I'm sorry,
who did you say it was again? Apple, was it?"

Shares of Apple, IBM and Intel all dropped sharply, since the reports
included no grossly exaggerated claims whatsoever. All news is bad news
on Wall Street, and shareholders were quick to pee their pants over
suggestions that something could change, either for the better or worse.

No one at Motorola has been reached for comment. No one has tried to
contact them.

Sources at Apple have denied claims that future generations of the
McIntosh operating system will be named OSXp.

--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good

(Remove spamblock to reply)
Anonymous
June 6, 2005 4:21:22 PM

Archived from groups: rec.audio.pro (More info?)

"Lorin David Schultz" wrote ...

> No one at Motorola has been reached for comment. No one has tried to
> contact them.

Who? You mean the cell-phone people?
Anonymous
June 6, 2005 9:24:54 PM

Archived from groups: rec.audio.pro (More info?)

On 6 Jun 2005 08:55:22 -0400, kludge@panix.com (Scott Dorsey) wrote:

>Jedd Haas <jnh@epsno.com> wrote:
>>In article <znr1118055742k@trad>, mrivers@d-and-d.com wrote:
>>> Announcement today. Anyone holding your breath? Anyone own Intel
>>> stock? Anyone own Apple stock?
>>
>>I've seen a couple suggestions on this topic that add some more
>>information. The top theories are that 1) Apple will use a "just in time"
>>code translator called Quick Transit (which would allegedly allow x86 to
>>run PPC code with no modification or performance penalty) and 2) that
>>Intel will just *manufacture* PPC chips for Apple, then create new, faster
>>PPC designs into the future. Another variant on this latter theory is that
>>Intel will add Altivec to x86.
>
>What about the huge number of people already running 68k applications on
>their PPC machine with translation? Are those applications going to be
>supported in the new era?

How do PPC machines run 68k code? I recall when those came out, I
thought it was just with a 68k emulator program running on the PPC. My
last Mac was a 68k machine, then I converted to the Dork Side. I mean,
uh...
And just what is Quick Transit, if it's not an emulator?

>--scott

-----
http://mindspring.com/~benbradley
Anonymous
June 6, 2005 9:24:55 PM

Archived from groups: rec.audio.pro (More info?)

In article <kg19a1p4r11m0t6hscsrjtok6rlqck7dpk@4ax.com>,
Ben Bradley <ben_nospam_bradley@frontiernet.net> wrote:
>On 6 Jun 2005 08:55:22 -0400, kludge@panix.com (Scott Dorsey) wrote:
>
>>Jedd Haas <jnh@epsno.com> wrote:
>>>In article <znr1118055742k@trad>, mrivers@d-and-d.com wrote:
>>>> Announcement today. Anyone holding your breath? Anyone own Intel
>>>> stock? Anyone own Apple stock?
>>>
>>>I've seen a couple suggestions on this topic that add some more
>>>information. The top theories are that 1) Apple will use a "just in time"
>>>code translator called Quick Transit (which would allegedly allow x86 to
>>>run PPC code with no modification or performance penalty) and 2) that
>>>Intel will just *manufacture* PPC chips for Apple, then create new, faster
>>>PPC designs into the future. Another variant on this latter theory is that
>>>Intel will add Altivec to x86.
>>
>>What about the huge number of people already running 68k applications on
>>their PPC machine with translation? Are those applications going to be
>>supported in the new era?
>
> How do PPC machines run 68k code? I recall when those came out, I
>thought it was just with a 68k emulator program running on the PPC. My
>last Mac was a 68k machine, then I converted to the Dork Side. I mean,
>uh...

It's an emulator, but it's a very good and extremely fast emulator. But
it's not so good that you'd be wanting to run it in PPC emulation mode.

> And just what is Quick Transit, if it's not an emulator?

There are emulators, and there are translators, and today there are a lot
of things that are combinations of the two. It is almost certain a combination.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
June 6, 2005 9:40:09 PM

Archived from groups: rec.audio.pro (More info?)

On Mon, 06 Jun 2005 12:21:21 GMT, "Lorin David Schultz"
<Lorin@DAMNSPAM!v5v.ca> wrote:

>"Mike Rivers" <mrivers@d-and-d.com> wrote in message
>news:znr1118055742k@trad...
>>
>> Announcement today. Anyone holding your breath? Anyone own Intel
>> stock? Anyone own Apple stock?

No, no, and no. I haven't kept up with the computer industry press,
but I understand that an announcement like this has been rumored for
years.

>Industry analysts predict this could cost IBM sales of as many as dozens
>of units per year.

The Power PC architecture is popular in embedded products, so they
won't notice - total sales could even increase.

>When asked how the loss of Apple's business would affect IBM, CEO B. Lou
>Biggs replied, "Who? Gee, I don't know... you'd have to check with
>Sales. Of course, we hate to lose *any* customer, even... I'm sorry,
>who did you say it was again? Apple, was it?"
>
>Shares of Apple, IBM and Intel all dropped sharply, since the reports
>included no grossly exaggerated claims whatsoever. All news is bad news
>on Wall Street, and shareholders were quick to pee their pants over
>suggestions that something could change, either for the better or worse.
>
>No one at Motorola has been reached for comment. No one has tried to
>contact them.

Did anyone think to call Freescale? Did anyone even know that
Motorola doesn't make PPC's anymore? They spun off their processor
division to a separate company and named it Freescale, just as they
did with their jellybean parts with On Semi.
Does Motorola still make anything anymore?

>Sources at Apple have denied claims that future generations of the
>McIntosh operating system will be named OSXp.

Knowing how long these things take, I would have thought it would
be OSXhorn, but it could even be Yhorn or Zhorn.

-----
http://mindspring.com/~benbradley
Anonymous
June 6, 2005 9:40:10 PM

Archived from groups: rec.audio.pro (More info?)

"Ben Bradley" <ben_nospam_bradley@frontiernet.net> wrote in message
news:4o19a1df2f07sq5o8tdveqjoag6p180lr6@4ax.com...
>
> Knowing how long these things take, I would have thought it would
> be OSXhorn, but it could even be Yhorn or Zhorn.
>

Actually, the work is essentially already done.
Anonymous
June 6, 2005 11:03:21 PM

Archived from groups: rec.audio.pro (More info?)

"Mike Rivers" wrote ...
> Announcement today. Anyone holding your breath? Anyone own
> Intel stock? Anyone own Apple stock?

It is only a psychological shift. Won't affect the stock of IBM or
Intel as Apple was only 2% of IBM's production. Likely down
in the noise as a percentage of capacity of Intel.

It might have a longer-term positive effect on Apple stock as they
will finally have a consistent source of leading-edge CPUs to design
products around.
Anonymous
June 6, 2005 11:12:00 PM

Archived from groups: rec.audio.pro (More info?)

In article <d81h1n$pke$1@panix2.panix.com>, kludge@panix.com (Scott
Dorsey) wrote:

> It's kind of sad and depressing, in that it means the whole world is
> basically now dependant on a single hardware architecture for
> everything.
>
> Having grown up in a world where hundreds of different instruction sets
> existed, and where VLIW machines and stack machines were mainstream
> architectures, I find it sad to see all of the things that made
> computers
> interesting and exciting have been reduced to this.

Nicely put, I feel the same way (I guess we must be similar vintages).
Anonymous
June 7, 2005 12:07:43 AM

Archived from groups: rec.audio.pro (More info?)

Richard Crowley wrote:
> "Mike Rivers" wrote ...
> > Announcement today. Anyone holding your breath? Anyone own
> > Intel stock? Anyone own Apple stock?
>
> It is only a psychological shift. Won't affect the stock of IBM or
> Intel as Apple was only 2% of IBM's production. Likely down
> in the noise as a percentage of capacity of Intel.
>
> It might have a longer-term positive effect on Apple stock as they
> will finally have a consistent source of leading-edge CPUs to design
> products around.

When I looked at about 3:30PM Chicago time today, Apple's stock was
down .92. What that says to me is that the market isn't exactly behind
this move at the moment.

It also says that there might be some fear that some of the people who
already are heavily invested in Mac systems are worried about having to
re-invest in about 2 years when previously, the re-investment time
frame was more like 4 to 5 years. ROI concerns are serious ones for
Mac people, just like they are for any other businesses.

Lee Darrow, C.H.
http://www.leedarrow.com
Anonymous
June 7, 2005 12:10:40 AM

Archived from groups: rec.audio.pro (More info?)

Scott Dorsey <kludge@panix.com> wrote:


> What about the huge number of people already running 68k applications on
> their PPC machine with translation? Are those applications going to be
> supported in the new era?

I suspect Classic and its 68K emulator may be discontinued, but this is
only my guess.

When I worked for Apple in the OS X I/O Group, we were required to
always keep our code little-endian-compatible, with either the correct
code or compile-time switches. A different group used to create Intel
builds of the OS for testing. Jobs' mention of a five-year planning
process is true.

I have mixed feelings about the switch. On the one hand, I'm sure Jobs
& Company didn't make the decision lightly. But, from a technical
perspective, there's gonna be *lots* of work to do. Quite a bit has
been done, because they've got Tiger ready for developer testing, but
it'll be an...interesting couple of years.

--
- rick http://www.cfcl.com/~rick/
Rick Auricchio Macs Only: Macintosh support rick@cfcl.com
I acknowledge the existence of a higher power, and have therefore installed
surge suppressors.
Anonymous
June 7, 2005 12:17:09 AM

Archived from groups: rec.audio.pro (More info?)

Well, that's a great opportunity for IBM to get off its duff and finally
come out with a line of Power-based machines running Linux... (No, I
haven't heard any rumors; I'd just like to see it happen.)
Anonymous
June 7, 2005 12:34:06 AM

Archived from groups: rec.audio.pro (More info?)

Roger Christie <rochrist@ wrote:
> "Ben Bradley" <ben_nospam_bradley@frontiernet.net> wrote in message
> news:4o19a1df2f07sq5o8tdveqjoag6p180lr6@4ax.com...
>
>> Knowing how long these things take, I would have thought it would
>>be OSXhorn, but it could even be Yhorn or Zhorn.
>>
>
>
> Actually, the work is essentially already done.

....except for the thousands of applications that need to be ported --
and re-purchased. :-(
Anonymous
June 7, 2005 12:44:48 AM

Archived from groups: rec.audio.pro (More info?)

<mstrhypno@earthlink.net> wrote in message
news:1118113662.992442.165880@g49g2000cwa.googlegroups.com...
>
>
> Richard Crowley wrote:
>> "Mike Rivers" wrote ...
>> > Announcement today. Anyone holding your breath? Anyone own
>> > Intel stock? Anyone own Apple stock?
>>
>> It is only a psychological shift. Won't affect the stock of IBM or
>> Intel as Apple was only 2% of IBM's production. Likely down
>> in the noise as a percentage of capacity of Intel.
>>
>> It might have a longer-term positive effect on Apple stock as they
>> will finally have a consistent source of leading-edge CPUs to design
>> products around.
>
> When I looked at about 3:30PM Chicago time today, Apple's stock was
> down .92. What that says to me is that the market isn't exactly
> behind
> this move at the moment.

Short term is noise. I said "longer-term".
Anonymous
June 7, 2005 1:18:43 AM

Archived from groups: rec.audio.pro (More info?)

Scott Dorsey wrote:

> In article <znr1118055742k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
>>
>>Announcement today. Anyone holding your breath? Anyone own Intel
>>stock? Anyone own Apple stock?
>
> It's kind of sad and depressing, in that it means the whole world is
> basically now dependant on a single hardware architecture for everything.

Indeed. Having to support multiple architectures (little/big endian
in particular) encourages better programming practices. It seems a shame
that it will be going away (on the desktop, at least).

I don't think anyone else has mentioned this, which is a feed of minutes
from the developer conference, and currently the main source of info:

http://www4.macnn.com/macnn/wwdc/05/


> --scott
>

--
JP Morris - aka DOUG the Eagle (Dragon) -=UDIC=- jpm@it-he.org
Fun things to do with the Ultima games http://www.it-he.org
Reign of the Just - An Ultima clone http://rotj.it-he.org
d+++ e+ N+ T++ Om U1234!56!7'!S'!8!9!KAW u++ uC+++ uF+++ uG---- uLB----
uA--- nC+ nR---- nH+++ nP++ nI nPT nS nT wM- wC- y a(YEAR - 1976)
Anonymous
June 7, 2005 3:13:06 AM

Archived from groups: rec.audio.pro (More info?)

JOhn wrote:
> I was hoping the announcement would be about a proprietary two-button
> mouse....I dream big!
>
> Scott Dorsey wrote:
>
>>In article <znr1118055742k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
>>
>>>Announcement today. Anyone holding your breath? Anyone own Intel
>>>stock? Anyone own Apple stock?
>>
>>It's kind of sad and depressing, in that it means the whole world is
>>basically now dependant on a single hardware architecture for everything.
>>
>>Having grown up in a world where hundreds of different instruction sets
>>existed, and where VLIW machines and stack machines were mainstream
>>architectures, I find it sad to see all of the things that made computers
>>interesting and exciting have been reduced to this.
>>--scott
>>
>>--
>>"C'est un Nagra. C'est suisse, et tres, tres precis."
>
>
Just plug any USB mouse into the back, Macs havr known how to use two
buttons for years just never told anyone important
Anonymous
June 7, 2005 3:36:43 AM

Archived from groups: rec.audio.pro (More info?)

http://stream.apple.akadns.net/

Jobs talks about it about 27 minutes into the keynote speech.

Looks impressive. They've been compiling OS X and running it on Intel
processors for 5 years now, apparently..

Just in case. ;-)

I wonder if they're trying AMD, too? I also wonder if they'll allow
for wider hardware support, or if it has to be a Mac motherboard?
Anonymous
June 7, 2005 4:42:19 AM

Archived from groups: rec.audio.pro (More info?)

JOhn <jpayne@berrypr.com> wrote:

> I was hoping the announcement would be about a proprietary two-button
> mouse....I dream big!

No, thanks. I'm totally happy *without* carpal tunnel syndrome.

--
Jazzman DX7II - 01/W - S3000XL - Ion

" Without geometry, life is pointless "
Anonymous
June 7, 2005 4:42:52 AM

Archived from groups: rec.audio.pro (More info?)

On 2005-06-06, J. P. Morris <jpm@it-he.org> wrote:
> Scott Dorsey wrote:
>
>> In article <znr1118055742k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
>>>
>>>Announcement today. Anyone holding your breath? Anyone own Intel
>>>stock? Anyone own Apple stock?
>>
>> It's kind of sad and depressing, in that it means the whole world is
>> basically now dependant on a single hardware architecture for everything.
>
> Indeed. Having to support multiple architectures (little/big
> endian in particular) encourages better programming practices.
> It seems a shame that it will be going away (on the desktop,
> at least).

Amen.

At least 99.9% of all running code is written in portable
high-level languages. Architecture-neutral, source-oriented
platforms like Linux have more market share than they have had
in decades. We even have architecture-neutral buses (PCI).

It's ironical to see such a push to standardise on a single
architecture when supporting multiple architectures is easier
than it has ever been.

--
André Majorel <URL:http://www.teaser.fr/~amajorel/&gt;
(Counterfeit: yzoc@increasable.org ecufuhuw@kivu.com)
What worries me is not the violence of the few, but the
indifference of the many. -- M. L. King
Anonymous
June 7, 2005 4:42:53 AM

Archived from groups: rec.audio.pro (More info?)

"Andre Majorel" wrote ...
> It's ironical to see such a push to standardise on a single
> architecture when supporting multiple architectures is easier
> than it has ever been.

Baloney. Its no "push to standardize on a single architecture".

Its simply that Motorola/Freescale/IBM proved too inept at
keeping up with CPU technology. Even AMD can do it (kinda).

If Apple wants to keep what's left of its market (which has
shrunk to 2%) they have to compete with viable CPUs.
Anonymous
June 7, 2005 4:42:54 AM

Archived from groups: rec.audio.pro (More info?)

"Richard Crowley" <rcrowley7@xprt.net> wrote in message
news:11a9vl2mbgnaqcd@corp.supernews.com...
> "Andre Majorel" wrote ...
> > It's ironical to see such a push to standardise on a single
> > architecture when supporting multiple architectures is easier
> > than it has ever been.
>
> Baloney. Its no "push to standardize on a single architecture".
>
> Its simply that Motorola/Freescale/IBM proved too inept at
> keeping up with CPU technology.

IBM? Hardly. They're just not particularly concerned with CPU's for
desktops, and Intel is primarily concerned with desktop processors. Intel
will pay more attention to Apple than IBM has.

>Even AMD can do it (kinda).

AMD makes the 'best' desktop processors in conventional terms, if you don't
mind a lot of heat. Personally, I do, but Intel isn't much better.

jb
Anonymous
June 7, 2005 5:33:42 AM

Archived from groups: rec.audio.pro (More info?)

"Lorin David Schultz" <Lorin@DAMNSPAM!v5v.ca> wrote in message...

> Industry analysts predict this could cost IBM sales of as many as dozens
> of units per year.
>
> When asked how the loss of Apple's business would affect IBM, CEO B. Lou
> Biggs replied, "Who? Gee, I don't know... you'd have to check with
> Sales. Of course, we hate to lose *any* customer, even... I'm sorry,
> who did you say it was again? Apple, was it?"
>
> Shares of Apple, IBM and Intel all dropped sharply, since the reports
> included no grossly exaggerated claims whatsoever. All news is bad news
> on Wall Street, and shareholders were quick to pee their pants over
> suggestions that something could change, either for the better or worse.
>
> No one at Motorola has been reached for comment. No one has tried to
> contact them.
>
> Sources at Apple have denied claims that future generations of the
> McIntosh operating system will be named OSXp.



Most Excellent reporting !

;-)
Anonymous
June 7, 2005 12:40:40 PM

Archived from groups: rec.audio.pro (More info?)

In article <1118126202.987910.185180@f14g2000cwb.googlegroups.com>,
"jatougas@yahoo.com" <jatougas@yahoo.com> wrote:

> http://stream.apple.akadns.net/
>
> Jobs talks about it about 27 minutes into the keynote speech.
>
> Looks impressive. They've been compiling OS X and running it on Intel
> processors for 5 years now, apparently..
>
> Just in case. ;-)
>
> I wonder if they're trying AMD, too? I also wonder if they'll allow
> for wider hardware support, or if it has to be a Mac motherboard?

It's likely that only Apple x86 boxes will run OSX "out of the box."
However, I believe an Apple rep has already admitted that other x86 boxes
could run OSX with some hacking. Other commenters suggest that Intel will
provide DRM hooks on the chip to only allow OSX if it's authorized. Notice
how the "Developer Kit" being sold for $999 is apparently an off-the-shelf
x86 box.

Additionally, some people suggest that the DRM hooks are part of Apple's
grand strategy for the movie version of iTunes--iFlicks, if you will.
Notice how the first Mac to be 86'd (haha) is going to be the Mini. The
theory is that the Mini will become a combination Tivo/iTunes/home "media
hub"/game station (etc) that will be the be-all end-all in those
categories. With the DRM, the movie studios will allow an iTunes-like
movie store. You order your movie, it downloads in the background, you
play it, it gets deleted after a preset time, etc. No more need to return
the DVD or VHS to Blockbuster.

As noted previously, Apple has had this in the works for years. It seems
they noticed how the ship date for Longhorn keeps slipping further into
the future. They probably figure they can finish the transition before
Longhorn appears (or before it gets established to any degree) and then go
head-to-head with MS. Then perhaps we'll see an insanely clever
Intel/Apple deal whereby Apple licenses their OSX DRM to Intel; Intel
charges a fee for every board; OSX-enabled boards are everywhere and they
both laugh all the way to the bank.

(As for Scott's question about 68k, I suspect that both Classic and 68k
will be supported via emulation for the forseeable future.)

--
Jedd Haas - Artist
http://www.gallerytungsten.com
http://www.epsno.com
Anonymous
June 7, 2005 2:57:59 PM

Archived from groups: rec.audio.pro (More info?)

"Ed Anson" <EdAnson@comcast.net> wrote in message
news:HuydnSM8osPgdjnfRVn-uw@comcast.com...
> Roger Christie <rochrist@ wrote:
> > "Ben Bradley" <ben_nospam_bradley@frontiernet.net> wrote in message
> > news:4o19a1df2f07sq5o8tdveqjoag6p180lr6@4ax.com...
> >
> >> Knowing how long these things take, I would have thought it would
> >>be OSXhorn, but it could even be Yhorn or Zhorn.
> >>
> >
> >
> > Actually, the work is essentially already done.
>
> ...except for the thousands of applications that need to be ported --
> and re-purchased. :-(

Reportedly, it took two hours to port Mathematica, a /very/ complex app.
Its supposedly requires not much more than straight recompiling.
Anonymous
June 7, 2005 7:57:59 PM

Archived from groups: rec.audio.pro (More info?)

"Michael" <ra3035@NOTfreescale.com> wrote:
>
> If I had my druthers, I'd be running our studio on a Mac. But
> the stuff was put together by our keyboardist (an ex-Dell worker)
> so it's PC-based, and it has proven much more problematic than Mac
> setups I've used.



I wonder if that's because your system is set up by someone who is at
least as interested in the computer as the job it's there to do.
Propellerheads can't resist "customizing."

I set up my systems to be as stupid simple as possible. If there's an
issue with a certain piece of hardware I just replace it with something
else rather than struggle with workarounds that invariably turn into
mid-session "gotchas." Same with software. My studio machines do NOT
run "entertainment" product or office apps. I stay with tried-and-true,
mainstream products and setups rather than always trying out the latest
utility or tweak. The result has been stable, productive systems. That
hasn't always been the case for some of my colleagues who like to screw
around with the computer for fun.

There's also the issue of budget. It's harder to "cheap out" a Mac.
I've found that the same rules apply to computer audio as other pro
audio gear -- sometimes cheap gear doesn't work very well and it's worth
paying more for better peripherals. I've had lots of problems with
cheap USB and PCMCIA products on my laptop. In every case but one,
replacing the offender with a more expensive name-brand product has
solved the problem.

The net result is that in my day-to-day work there has been no
significant advantage evident between our OSX and XP rigs. There are
subtle differences, but no compelling reason to recommend one over the
other. Either one can be as useful as the other.

--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good

(Remove spamblock to reply)
Anonymous
June 7, 2005 8:58:57 PM

Archived from groups: rec.audio.pro (More info?)

jatougas@yahoo.com <jatougas@yahoo.com> wrote:

> Looks impressive. They've been compiling OS X and running it on Intel
> processors for 5 years now, apparently..

At least... The very first versions of OSX (called Rhapsody DR1 and DR2
(or was it PR...)) delivered to all developers were X86, not PPC. My
first contact with OSX from Apple was on a PC. These pre-pre-releases
also implemented the universal binaries. It's all tested and tried in
production in the name of NeXT. I take it for granted they never took it
out, but maintained these features all the time.

Lars


--
lars farm // http://www.farm.se
lars is also a mail-account on the server farm.se
aim: larsfarm@mac.com
Anonymous
June 8, 2005 4:13:32 PM

Archived from groups: rec.audio.pro (More info?)

Roger Christie <rochrist@<REMOVETOEMAIL> wrote:
>"Ed Anson" <EdAnson@comcast.net> wrote in message
>> Roger Christie <rochrist@ wrote:
>> > "Ben Bradley" <ben_nospam_bradley@frontiernet.net> wrote in message
>> >
>> >> Knowing how long these things take, I would have thought it would
>> >>be OSXhorn, but it could even be Yhorn or Zhorn.
>> >>
>> > Actually, the work is essentially already done.
>>
>> ...except for the thousands of applications that need to be ported --
>> and re-purchased. :-(
>
>Reportedly, it took two hours to port Mathematica, a /very/ complex app.
>Its supposedly requires not much more than straight recompiling.

For the most part, code that is well-written will be trivial to port,
and code that is badly written will be a nightmare to port.

Sadly, most of the code in the world today is pretty badly-written
and very architecture-specific.

The Wolfram crew writes to somewhat higher standards than the typical
applications guys are used to, I suspect. And it probably saves them a
lot of headaches in the long run, too.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
June 9, 2005 9:33:52 AM

Archived from groups: rec.audio.pro (More info?)

"Roger Christie" <rochrist@<REMOVETOEMAIL>charter.net> writes:

> Reportedly, it took two hours to port Mathematica, a /very/ complex
> app. Its supposedly requires not much more than straight
> recompiling.

Yes, but mm ships on all osrts of systems, 32 and 64 bit, big and
little endian, so lots of the rought edges got hammered down long
ago. This is not a common case.

Why Jobs is jumping in front of a road roller with M$ across the front
is a mystery to me.

--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
comp.os.vms,- The Older, Grumpier Slashdot
Raw, Cooked or Well-done, it's all half baked.
EPIC, The Architecture of the future, always has been, always will be.
Anonymous
June 9, 2005 11:58:23 AM

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey" <kludge@panix.com> wrote in message
>
> Sadly, most of the code in the world today is pretty badly-written
> and very architecture-specific.
>
> The Wolfram crew writes to somewhat higher standards than the typical
> applications guys are used to, I suspect. And it probably saves them a
> lot of headaches in the long run, too.

I think it's just so hilarious that the Macoids have been laughuing at the
Wintels for so many years now, crowing about how superior there computer
choice wa because it's RISC architecture was so inherently superior to CISC
..

geoff
Anonymous
June 9, 2005 11:58:24 AM

Archived from groups: rec.audio.pro (More info?)

Geoff Wood <geoff@nospam-paf.co.nz> wrote:
>"Scott Dorsey" <kludge@panix.com> wrote in message
>>
>> Sadly, most of the code in the world today is pretty badly-written
>> and very architecture-specific.
>>
>> The Wolfram crew writes to somewhat higher standards than the typical
>> applications guys are used to, I suspect. And it probably saves them a
>> lot of headaches in the long run, too.
>
>I think it's just so hilarious that the Macoids have been laughuing at the
>Wintels for so many years now, crowing about how superior there computer
>choice wa because it's RISC architecture was so inherently superior to CISC
>.

For the past two decades the line between CISC and RISC have been so horribly
blurred that it's not really an important distinction any longer. These
days we have microcoded machines with register stacks, and machines with
instructions that all execute in one cycle but still with context-switching
overhead.

Hell, the PDP-8 matches most of the "RISC" requirements, right? Seven
instructions, all executing in fixed time with no microcode?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
June 9, 2005 11:58:24 AM

Archived from groups: rec.audio.pro (More info?)

Geoff Wood wrote:
> I think it's just so hilarious that the Macoids have been laughuing at the
> Wintels for so many years now, crowing about how superior there computer
> choice wa because it's RISC architecture was so inherently superior to CISC

It still *is* superior. However, if you can't get a manufacturer
willing to make low-power RISC chips suitable for laptops in volume,
then it's a moot point since laptops are such a large percentage of
the sales of personal computers these days. (Apple is not a chip
maker, so they have very little control over what chips get made.)

- Logan
Anonymous
June 9, 2005 11:58:25 AM

Archived from groups: rec.audio.pro (More info?)

Logan Shaw <lshaw-usenet@austin.rr.com> wrote:
>Geoff Wood wrote:
>> I think it's just so hilarious that the Macoids have been laughuing at the
>> Wintels for so many years now, crowing about how superior there computer
>> choice wa because it's RISC architecture was so inherently superior to CISC
>
>It still *is* superior. However, if you can't get a manufacturer
>willing to make low-power RISC chips suitable for laptops in volume,
>then it's a moot point since laptops are such a large percentage of
>the sales of personal computers these days. (Apple is not a chip
>maker, so they have very little control over what chips get made.)

I dunno, I still like CISC machines. But the x86 is not exactly the most
elegant CISC architecture around.. it is bloated with enormous numbers of
compatibility features.

The sad part is that some of the really neat features that were added to
the 386 never got used. As far as I know, OS/2 is the only operating system
that ever used the ring protection stuff. It's really a shame. The history
of the x86 is basically one of bad design compromises that get perpetuated
for years, and improvements that are ignored.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
June 9, 2005 11:58:25 AM

Archived from groups: rec.audio.pro (More info?)

"Logan Shaw" wrote ...
> Geoff Wood wrote:
>> I think it's just so hilarious that the Macoids have been laughuing
>> at the Wintels for so many years now, crowing about how superior
>> there computer choice wa because it's RISC architecture was so
>> inherently superior to CISC
>
> It still *is* superior.

Dunno how. A RISC processor will have to execute so many more
instructions to do the same work as a CISC, seems like a wash at best.
Anonymous
June 9, 2005 11:58:26 AM

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey" wrote ...
> I dunno, I still like CISC machines. But the x86 is not exactly the
> most
> elegant CISC architecture around.. it is bloated with enormous numbers
> of
> compatibility features.
>
> The sad part is that some of the really neat features that were added
> to
> the 386 never got used. As far as I know, OS/2 is the only operating
> system
> that ever used the ring protection stuff. It's really a shame. The
> history
> of the x86 is basically one of bad design compromises that get
> perpetuated
> for years, and improvements that are ignored.

But that is the choice of the market. Backwards compatibility is more
important than forward-looking changes in architecture. Can't leave
all those players of "Grand Theft Auto" out in the cold, can we? :-/
Anonymous
June 9, 2005 12:35:25 PM

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey" <kludge@panix.com> wrote in message
news:D 87jf8$q0m$1@panix2.panix.com...
> Geoff Wood <geoff@nospam-paf.co.nz> wrote:
> >"Scott Dorsey" <kludge@panix.com> wrote in message
> >>
> >> Sadly, most of the code in the world today is pretty
badly-written
> >> and very architecture-specific.
> >>
> >> The Wolfram crew writes to somewhat higher standards
than the typical
> >> applications guys are used to, I suspect. And it
probably saves them a
> >> lot of headaches in the long run, too.
> >
> >I think it's just so hilarious that the Macoids have been
laughuing at the
> >Wintels for so many years now, crowing about how superior
there computer
> >choice wa because it's RISC architecture was so
inherently superior to CISC
> >.
>
> For the past two decades the line between CISC and RISC
have been so horribly
> blurred that it's not really an important distinction any
longer. These
> days we have microcoded machines with register stacks, and
machines with
> instructions that all execute in one cycle but still with
context-switching
> overhead.
>
> Hell, the PDP-8 matches most of the "RISC" requirements,
right? Seven
> instructions, all executing in fixed time with no
microcode?

Nicely said. Under the covers of the 360/30 which was about
as CISC as they come, lay something about as complex as a
PDP-8, maybe a little bit less complex because it was an
8-bit core. Odd machine - most implementations had far more
microcode than real memory.
Anonymous
June 9, 2005 12:49:01 PM

Archived from groups: rec.audio.pro (More info?)

Richard Crowley wrote:
> "Logan Shaw" wrote ...
>> Geoff Wood wrote:

>>> I think it's just so hilarious that the Macoids have been laughuing
>>> at the Wintels for so many years now, crowing about how superior
>>> there computer choice wa because it's RISC architecture was so
>>> inherently superior to CISC

>> It still *is* superior.

> Dunno how. A RISC processor will have to execute so many more
> instructions to do the same work as a CISC, seems like a wash at best.

Well, in reality RISC vs. CISC is an oversimplification that is often
used to describe several different changes in instruction set architecture
that all happened about the same time. At best, RISC is a nebulous term
that refers to several lessons learned about CPU design, such as:

* having lots of registers that are all general-purpose (instead of
special-purpose accumulator registers and stuff) aids in register
allocation and makes it possible to avoid stashing temporary values
in memory (such as on the stack) much less often. 32 registers is
a good number. even more might be better.

* having instructions whose operands' and result's source and destination
registers can all be specified explicitly (as in "add.l r0, r1, r2" to
add r0 and r1 and store the result into r2, as opposed to say, an
instruction that takes one argument and adds that register's value to
the accumulator register) makes life better because it avoids the need
to do register-to-register moves to juggle things around.

* adding complex addressing modes (register, register relative, register
indirect autoincrement, register indirect half-caff autoincrement with
a twist of lemon) to a whole bunch of instructions a la x86 and 680x0
is a mistake because it makes things complex and hard to optimize.
making most instructions just operate on registers and then having a
load and store instruction makes things easier.

* keeping things simple enables fixed-length opcodes, which keeps things
simpler and cleaner and guarantees that you can load a steady stream
of instructions

A lot of this has to do with the invention of pipelining. Yes, RISC
instructions do less, but they are also much easier to pipeline efficiently
because they are uniform. Most RISC instructions on a given architecture
have similar phases to each other. They vary from architecture to
architecture, but it might be something like this:

phase 1: fetch instruction from instruction cache
phase 2: decode instruction
phase 3: read operands from registers (or in the case of
load instruction, read data from cache if possible)
phase 4: do computation in ALU (or one of several ALUs).
phase 5: write back result in register (or in the case of
store instruction, put data into write buffer
to be written to cache and/or memory)

With hopefully few exceptions (due to branches and other pipeline hazards),
when your first instruction is finishing up with phase 1 and starting
phase 2, you can start the second instruction on phase 1. And when you
are starting the first instruction on phase 3, the second instruction can
be entering phase 2 and the third can be entering phase 1.

The key is, if you can keep this pipeline running properly, which the
uniformity of RISC instructions is meant to make possible, you can get
better performance than a machine that tries to do a whole instruction
in a clock cycle and thus has to make the clock cycles realllly loooong
(or a machine that totally completes one instruction before even starting
on the next).

The phases are all designed to be simple bite-size pieces that can be
accomplished quickly, keeping the clock cycle short, but because of the
nature of the pipeline, you can still (excepting cache misses, etc.)
come close to averaging one instruction per clock cycle. This is a
huge win because you are able to do about the same number of instructions
per clock cycle but you are also able to dramatically shorten the clock
cycle, because smaller pieces are happening in each clock.

Anyway, the thing is that these days, the folks over at Intel (and
at AMD) have figured out how to do hardware translation of the x86
instruction set into what amounts to RISC instructions that can be
pipelined efficiently. So, there is a little extra overhead, but they
have managed to shoehorn the x86 instruction set into the pipeline
paradigm and get really good performance by using RISC ideas. This
just proves that RISC has won the design war, though: even though
Intel chips still have a CISC instruction set, internally they turn
everything into RISC to get the speed they need.

So, whether PowerPC is superior is a tricky question. It depends on
what you mean by superior. It's a clean design that's closer to what
it really should be in order to get good performance. But Intel has
developed very effective coping mechanisms for the limitations of
CISC, so in practice it is possible to get RISC-like performance out
of a CISC instruction set. Since there exist real chips that do
this, the advantage that PowerPC holds is just one of being a clean
design. This is primarily good for aesthetic reasons, because it
saves silicon (not having to do the translation), and because it
makes life easier for the compiler. But economies of scale with x86
(like that Intel is rich and can afford the best physical processes)
pretty much offset all those advantages.

- Logan
Anonymous
June 9, 2005 12:49:02 PM

Archived from groups: rec.audio.pro (More info?)

"Logan Shaw" <lshaw-usenet@austin.rr.com> wrote in message
news:1oTpe.38070$j51.19395@tornado.texas.rr.com...



> This
> just proves that RISC has won the design war, though:
even though
> Intel chips still have a CISC instruction set, internally
they turn
> everything into RISC to get the speed they need.

Ironically, this is essentially how the original 360/30 and
40 worked. The published architecture was heavily CISC, but
the hardware architecture was very stripped-back basically
an 8 (360/30) or 16 (360/40) bit RISC. The major difference
was that the low-end 360s did the translation and
optimization of CISC instructions in real time via hardware,
while a true RISC does the translation and much optimization
at compile time with software.

As the argument goes, you can do more global optimization at
compile-time with software, and do it more efficiently
because you only do it once. Back in the real world, don't
count the hardware designers out. As a rule, their products
are more reliable.
Anonymous
June 9, 2005 4:50:29 PM

Archived from groups: rec.audio.pro (More info?)

> Dunno how. A RISC processor will have to execute so many more
> instructions to do the same work as a CISC, seems like a wash at best.

In concept, it's the same as serial vs. parallel. The new serial
technologies blow away old ideas that more channels MUST be faster.

-John O
Anonymous
June 9, 2005 8:14:27 PM

Archived from groups: rec.audio.pro (More info?)

Logan Shaw <lshaw-usenet@austin.rr.com> wrote:
>
> Anyway, the thing is that these days, the folks over at Intel (and
> at AMD) have figured out how to do hardware translation of the x86
> instruction set into what amounts to RISC instructions that can be
> pipelined efficiently. So, there is a little extra overhead, but they
> have managed to shoehorn the x86 instruction set into the pipeline
> paradigm and get really good performance by using RISC ideas. This
> just proves that RISC has won the design war, though: even though
> Intel chips still have a CISC instruction set, internally they turn
> everything into RISC to get the speed they need.

One must also remember the effects of things like out of order execution,
predication, replay, etc. on the ability of machines to fill pipelines
efficiently.

--
Aaron
Anonymous
June 10, 2005 12:34:12 PM

Archived from groups: rec.audio.pro (More info?)

"John O" <johno@!noSPAM!heathkit.com> wrote in message
news:p WWpe.2965$jS1.2277@newssvr17.news.prodigy.com...
>> Dunno how. A RISC processor will have to execute so many more
>> instructions to do the same work as a CISC, seems like a wash at best.
>
> In concept, it's the same as serial vs. parallel. The new serial
> technologies blow away old ideas that more channels MUST be faster.


How about putting 16 SATA cables in parallel and moving data sideways
through that ?

geoff
Anonymous
June 10, 2005 12:34:13 PM

Archived from groups: rec.audio.pro (More info?)

> How about putting 16 SATA cables in parallel and moving data sideways
> through that ?

I'll take that computer.

I've read that PCI Express (kinda the same approach as you described) can
theoretically reach the data carrying limits of copper.

John O
Anonymous
June 10, 2005 8:45:02 PM

Archived from groups: rec.audio.pro (More info?)

Geoff Wood <geoff@nospam-paf.co.nz> wrote:

> I think it's just so hilarious that the Macoids have been laughuing at the
> Wintels for so many years now, crowing about how superior there computer
> choice wa because it's RISC architecture was so inherently superior to CISC

As if we know what the actual chip will be. There are 300 ex-DEC chip
engineers working at Intel for a few years now. DEC had their 64-bit
chip a long time ago. 2007 is a ways off yet. Conclusions drawn right
now might not fit well later on.

--
ha
Anonymous
July 27, 2005 10:09:40 AM

Archived from groups: rec.audio.pro (More info?)

Logan Shaw <lshaw-usenet@austin.rr.com> wrote:
> [RISC] still *is* superior [to CISC]. However, if you can't get a
> manufacturer willing to make low-power RISC chips suitable for laptops
> in volume, then it's a moot point since laptops are such a large
> percentage of the sales of personal computers these days.

I wonder if multi-GHz ARMs will ever appear...

--
Aaron J. Grier | "Not your ordinary poofy goof." | agrier@poofygoof.com
Anonymous
July 27, 2005 10:10:58 AM

Archived from groups: rec.audio.pro (More info?)

Joe Kesselman <keshlam-nospam@comcast.net> wrote:
> Well, that's a great opportunity for IBM to get off its duff and
> finally come out with a line of Power-based machines running Linux...
> (No, I haven't heard any rumors; I'd just like to see it happen.)

can't you still buy power machines running AIX?

--
Aaron J. Grier | "Not your ordinary poofy goof." | agrier@poofygoof.com
!