From Newsgroup: comp.sys.mac.system


Just found out that this iMac (home, late 2012 v.) won't be able to load
Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big
Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting
to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays).

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load
Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big
Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting
to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays).

And the good news is, if your old Mac has a mechanical hard disk, take
note...

Scholle McFarland in “Take Control of Catalina” says: “Get a solid
state drive (or a new Mac): macOS Mojave and Catalina both use Apple’s
APFS system, which doesn’t play well with mechanical hard drives.”

Excerpt From: Scholle Sawyer McFarland. “Take Control of Catalina
(1.1).” Apple Books.

And here's what Joe Kissell in “Take Control of Big Sur” has to add: “APFS does have one significant downside: because it’s designed
primarily for SSDs, its performance on mechanical hard drives is
terrible.”

Excerpt From: Joe Kissell. “Take Control of Big Sur (1.1).” Apple Books.

--
chrispam1@me.com is a filtered spam magnet. Email replies may be lost.
You're better off replying to this newsgroup.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load
Big Sur...

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using
daily for more than a decade. I really got my money's worth on this
thing.

--
K.

Connection closed by foreign host.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <rovk74$drh$1@dont-email.me>, me@home.spamsucks.ca (Kir�ly)
wrote:

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using daily for more than a decade. I really got my money's worth on this thing.

Same here, on this mid-2010 iMac. Changing to an SSD certainly made it peppier. And by installing 10.13 first onto an external rotating hard
drive (can you say S...L...O...W?) and then cloning to the SSD I was
able to retain the use of HFS+ and thus continue to use DiskWarrior. I
don't think that trick will work with 10.14 and later.

But rumour has it that the most recent security update for 10.13 will be
the last one, so I guess this iMac's days are unfortunately numbered.

David

--
David Ryeburn
david_ryeburn@telus.netz
To send e-mail, change "netz" to "net"
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 11/16/20 11:40 PM, Király wrote:

Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load
Big Sur...

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using
daily for more than a decade. I really got my money's worth on this
thing.

My family used a late 2009 for 11 years- right up until I sold it on craigslist earlier this year.

Some guy who was in love with that particular model paid me way too much
money for it too!

--
Karl Marx is an historically prominent figu...but no one ever mentions
his wife Onya who invented the Starter’s Pistol…
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load
Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big
Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting
to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays). >>

And the good news is, if your old Mac has a mechanical hard disk, take note...

Scholle McFarland in “Take Control of Catalina” says: “Get a solid state drive (or a new Mac): macOS Mojave and Catalina both use Apple’s
APFS system, which doesn’t play well with mechanical hard drives.”

Excellent point since I have to do a clean install it will fall to APFS
and, yes, I have a spinner in there. I'll hold off some more and look
into changing the drive. I'll have to order a repair kit from iFixIt to replace the glass gasket. <sigh>.

Upside is an SSD would be a lot quicker of course.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 08:27, Wade Garrett wrote:

On 11/16/20 11:40 PM, Király wrote:

Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load >>> Big Sur...

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using
daily for more than a decade. I really got my money's worth on this
thing.

My family used a late 2009 for 11 years- right up until I sold it on craigslist earlier this year.

Some guy who was in love with that particular model paid me way too much money for it too!

Sold my 2007 iMac for near CAD$1000 in 2013...

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-16 23:40, Király wrote:

Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load
Big Sur...

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using
daily for more than a decade. I really got my money's worth on this
thing.

The 2012 iMac with i7 (4 HT cores) still lifts fine - except for video
editing (final render) which is an increasing workload even.

I was thinking of an 8 core i7 (last intel iMac) but after the M1
presentation I'm determined to not buy anymore intel Macs (nothing
against intel, the 86 architecture has fared very well and done well for me...) but no sense in getting the "last" intel iMac when the "next" Mx
iMac will likely be in a very, very hot version that will last a long,
long time.

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32
GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm
certainly not expecting it.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <JfTsH.4$7D7.1@fx03.iad>, Alan Browne- <Blackhole@entropy.ultimateorg> wrote:

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32
GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm certainly not expecting it.

there likely will be in higher end models.

this is the *beginning* of the transition, not the end.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <rp0j4n$b16$1@dont-email.me> Wade Garrett <wade@cooler.net> wrote:

On 11/16/20 11:40 PM, Király wrote:

Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load >>> Big Sur...

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using
daily for more than a decade. I really got my money's worth on this
thing.

My family used a late 2009 for 11 years- right up until I sold it on craigslist earlier this year.

Some guy who was in love with that particular model paid me way too much money for it too!

Yeah, I sold a 2009 3 years ago for stupid amounts of money.

--
"There's nothin' wrong with bein' a son of a bitch." -- Gaspode the
Wonder Dog
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <171120201240455000%nospam@nospam.invalid> nospam <nospam@nospam.invalid> wrote:

In article <JfTsH.4$7D7.1@fx03.iad>, Alan Browne- <Blackhole@entropy.ultimateorg> wrote:

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32
GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm
certainly not expecting it.

there likely will be in higher end models.

I do not expect to see Mac with upgradable RAM. Maybe the Mac Pro/iMac
Pro (if the iMac Pro continues to exist). The advantages to having the
RAM on the SoC are too great.

this is the *beginning* of the transition, not the end.

People seem to be forgetting this.

I will be very interested in the next round of machines.

--
Oh! I thought they smelled bad on the *outside*!
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On Nov 17, 2020, Alan Browne wrote
(in article <D6TsH.686073$%p.29768@fx33.iad>):

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne<bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays).

And the good news is, if your old Mac has a mechanical hard disk, take note...

Scholle McFarland in “Take Control of Catalina” says: “Get a solid state drive (or a new Mac): macOS Mojave and Catalina both use Apple’s APFS system, which doesn’t play well with mechanical hard drives.”

Excellent point since I have to do a clean install it will fall to APFS
and, yes, I have a spinner in there. I'll hold off some more and look
into changing the drive. I'll have to order a repair kit from iFixIt to replace the glass gasket. <sigh>.

Upside is an SSD would be a lot quicker of course.

Does Big Sur play OK with fusion drives?

Ron, the humblest guy in town.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <D6TsH.686073$%p.29768@fx33.iad> Alan Browne <Blackhole@entropy.ultimateorg> wrote:

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load >>> Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big
Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting
to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays). >>>

And the good news is, if your old Mac has a mechanical hard disk, take
note...

Scholle McFarland in “Take Control of Catalina” says: “Get a solid
state drive (or a new Mac): macOS Mojave and Catalina both use Apple’s
APFS system, which doesn’t play well with mechanical hard drives.”

Excellent point since I have to do a clean install it will fall to APFS
and, yes, I have a spinner in there. I'll hold off some more and look
into changing the drive. I'll have to order a repair kit from iFixIt to replace the glass gasket. <sigh>.

Upside is an SSD would be a lot quicker of course.

I had to boot a machine off a spinning disk recently. It was horrible.
Even a normal USB SSD is painful now.


--
When men talk to their friends, they insult each other. They don't
really mean it. When women talk to their friends, they compliment
each other. They don't really mean it.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 08:04:47 +0000, David Ryeburn said:

In article <rovk74$drh$1@dont-email.me>, me@home.spamsucks.ca (Kir�ly)
wrote:

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been using
daily for more than a decade. I really got my money's worth on this
thing.

Same here, on this mid-2010 iMac. Changing to an SSD certainly made it peppier. And by installing 10.13 first onto an external rotating hard
drive (can you say S...L...O...W?) and then cloning to the SSD I was
able to retain the use of HFS+ and thus continue to use DiskWarrior. I
don't think that trick will work with 10.14 and later.

But rumour has it that the most recent security update for 10.13 will be
the last one, so I guess this iMac's days are unfortunately numbered.

David

I used a PowerMac G3 for 20 years, daily for both my own business and
leisure time. As long as the computer continues to function and does
what you need it to, there's no problem and certainly no need to keep upgrading to the latest toys simply for the sake of it.


--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 18:05:27 +0000, Lewis said:

I had to boot a machine off a spinning disk recently. It was horrible.
Even a normal USB SSD is painful now.

Ahh ... the impatience of "youngsters" and how soon the "oldies" forget.

I remember having a FileMaker Pro database on a Mac Classic. One
requirement was that the database had to re-calculate the person's age
each day, which meant a 20-30min wait after booting the computer each
morning ... and that was only a few hundred database records.


--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 16:45:23 +0000, Alan Browne said:

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load >>> Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big
Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting
to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays). >>

And the good news is, if your old Mac has a mechanical hard disk, take
note...

Scholle McFarland in "Take Control of Catalina" says: "Get a solid
state drive (or a new Mac): macOS Mojave and Catalina both use Apple's
APFS system, which doesn't play well with mechanical hard drives."

Excellent point since I have to do a clean install it will fall to APFS
and, yes, I have a spinner in there. I'll hold off some more and look
into changing the drive. I'll have to order a repair kit from iFixIt
to replace the glass gasket. <sigh>.

Upside is an SSD would be a lot quicker of course.

The SSD is even quicker on a new Apple Silicon Mac.



--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 12:40, nospam wrote:

In article <JfTsH.4$7D7.1@fx03.iad>, Alan Browne- <Blackhole@entropy.ultimateorg> wrote:

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32
GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm
certainly not expecting it.

there likely will be in higher end models.

this is the *beginning* of the transition, not the end.

The M1 has the memory modules (2) on the same chip carrier as the M1 as
far as I can see. Possibly very tight coupling to the M1 chip. Putting
the memory on a separate carrier might make impedance matching / speed
an issue. To be seen.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 13:05, Lewis wrote:

In message <D6TsH.686073$%p.29768@fx33.iad> Alan Browne <Blackhole@entropy.ultimateorg> wrote:

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to load >>>> Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big >>>> Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting >>>> to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2 displays). >>>>

And the good news is, if your old Mac has a mechanical hard disk, take
note...

Scholle McFarland in “Take Control of Catalina” says: “Get a solid >>> state drive (or a new Mac): macOS Mojave and Catalina both use Apple’s >>> APFS system, which doesn’t play well with mechanical hard drives.”

Excellent point since I have to do a clean install it will fall to APFS
and, yes, I have a spinner in there. I'll hold off some more and look
into changing the drive. I'll have to order a repair kit from iFixIt to
replace the glass gasket. <sigh>.

Upside is an SSD would be a lot quicker of course.

I had to boot a machine off a spinning disk recently. It was horrible.
Even a normal USB SSD is painful now.

No real issue. I boot every 30 - 60 days typically.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 14:40, Your Name wrote:

On 2020-11-17 08:04:47 +0000, David Ryeburn said:

In article <rovk74$drh$1@dont-email.me>, me@home.spamsucks.ca (Király)
wrote:

Hey, I'm stuck using 10.13, on the same late-2009 iMac I've been
using daily for more than a decade. I really got my money's worth on
this thing.

Same here, on this mid-2010 iMac. Changing to an SSD certainly made it
peppier. And by installing 10.13 first onto an external rotating hard
drive (can you say S...L...O...W?) and then cloning to the SSD I was
able to retain the use of HFS+ and thus continue to use DiskWarrior. I
don't think that trick will work with 10.14 and later.

But rumour has it that the most recent security update for 10.13 will be
the last one, so I guess this iMac's days are unfortunately numbered.

David

I used a PowerMac G3 for 20 years, daily for both my own business and leisure time. As long as the computer continues to function and does
what you need it to, there's no problem and certainly no need to keep upgrading to the latest toys simply for the sake of it.

You also miss out on h/w and s/w that simply will never work on a G3 at
all. I'm willing to go 8 years, but 20 would leave to many new and interesting things in the dust - and that includes the near seamless integration of various iOS/Watch/Tv things with Mac OS.

Rendering even 720p on a G3 with minimal effects and transitions would
be deadly long... if even possible.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <Nv_sH.1482$xe4.309@fx41.iad>, Alan Browne
<bitbucket@blackhole.com> wrote:

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32
GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm
certainly not expecting it.

there likely will be in higher end models.

this is the *beginning* of the transition, not the end.

The M1 has the memory modules (2) on the same chip carrier as the M1 as
far as I can see. Possibly very tight coupling to the M1 chip. Putting
the memory on a separate carrier might make impedance matching / speed
an issue. To be seen.

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <nB_sH.1484$xe4.470@fx41.iad>, Alan Browne
<bitbucket@blackhole.com> wrote:

On 2020-11-17 14:40, Your Name wrote:

I used a PowerMac G3 for 20 years, daily for both my own business and leisure time. As long as the computer continues to function and does
what you need it to, there's no problem and certainly no need to keep upgrading to the latest toys simply for the sake of it.

You also miss out on h/w and s/w that simply will never work on a G3 at
all. I'm willing to go 8 years, but 20 would leave to many new and interesting things in the dust - and that includes the near seamless integration of various iOS/Watch/Tv things with Mac OS.

Rendering even 720p on a G3 with minimal effects and transitions would
be deadly long... if even possible.

never mind rendering. simply watching 720p on a g3 dropped frames.
don't even think about 1080p.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 20:54, nospam wrote:

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.

Big picture: competitors will also move ahead. Intel may be stuck ion
molasses, but AMD is forging ahead at high speed. NVIDIA is moving at
light speed for GPUs.

How long it takes for Apple to roll out the full family from the watch
to the Mac Pro will dictate how well they can keep up woth AMD/NVIDIA
and stay ahead of them.

Apple did very well in chip development for IOS devices. Rapid pace and
stayd ahead of the pack. But once you add the full range of computers to
their workload, How often can Apple release new chip for each market
segment? Remember rhat this now includes the GPU as well. So Apple is competing against Intel for laptops, AMD for higher end computers and
NVIDIA for GPUs.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

RonTheGuy <ron@null.invalid> wrote:

Does Big Sur play OK with fusion drives?

I don't know for sure but it must because many of the older machines that
can run Big Sur had those.

I know I read somewhere Fusion is a dead issue on new machines which only
makes sense since the whole idea was from a time that SSD's were only in the 64gb/128gb sizes.

Today with 1TB SSD's for $100 +/- $50, it just doesn't make sense anymore.

-bruce
bje@ripco.com
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <HX3tH.10482$2j.1075@fx38.iad>, JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

Big picture: competitors will also move ahead. Intel may be stuck ion molasses, but AMD is forging ahead at high speed. NVIDIA is moving at
light speed for GPUs.

don't underestimate apple.

How long it takes for Apple to roll out the full family from the watch
to the Mac Pro will dictate how well they can keep up woth AMD/NVIDIA
and stay ahead of them.

apple said two years. they said the same about the intel transition and
it was done much sooner than that.

why do you ignore public statements and fabricate your own reality?

Apple did very well in chip development for IOS devices. Rapid pace and
stayd ahead of the pack. But once you add the full range of computers to their workload, How often can Apple release new chip for each market
segment? Remember rhat this now includes the GPU as well. So Apple is competing against Intel for laptops, AMD for higher end computers and
NVIDIA for GPUs.

apple can release new chips as often as they want. they are not tied to
anyone else's schedule.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <171120202054065457%nospam@nospam.invalid> nospam <nospam@nospam.invalid> wrote:

In article <Nv_sH.1482$xe4.309@fx41.iad>, Alan Browne <bitbucket@blackhole.com> wrote:

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32
GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm
certainly not expecting it.

there likely will be in higher end models.

this is the *beginning* of the transition, not the end.

The M1 has the memory modules (2) on the same chip carrier as the M1 as
far as I can see. Possibly very tight coupling to the M1 chip. Putting
the memory on a separate carrier might make impedance matching / speed
an issue. To be seen.

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.

On a guess, I look at what Apple did with the Mac Pro and I think
something along those lines with two M1s basically meshed together into
one super-high speed SoC might be what we see in the next round of MBPs,
giving 16 cores of CPU, 16 cores of GPU and 4 USB4 ports.

Memory seems to be much more efficient, as things that were not really
doable without at least 16GB of RAM and really you wanted more are
doable on the M1 with 8 without breaking a sweat.

Another report from the low-end M1 MBP, "more than 100 tabs open in
Chrome and editing 4K video while Xcode [is also open] is like editing
on my iMac Pro."

--
The only good thing ever to come out of religion was the music.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <HX3tH.10482$2j.1075@fx38.iad> JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

On 2020-11-17 20:54, nospam wrote:

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.

Big picture: competitors will also move ahead.

Unlikely that anyone can replicate what Apple has done, and they won't
be able to even start to try for years.

Intel may be stuck ion molasses, but AMD is forging ahead at high
speed. NVIDIA is moving at light speed for GPUs.

It doedsn't seem to matter, the one advantage that seems clear from the
M1 si that the tight integration yeilds massive benefits, not just the
tight integration of the M1, but the fact that the chips are optimized
for the operating system and the operating system is optimized for the
chips. Qualcomm is still years behind Apple, and every year the gap
widens.

Intel can't do that, they don't write the OS. NVidia can't do that. AMD
can't do that.

How long it takes for Apple to roll out the full family from the watch
to the Mac Pro will dictate how well they can keep up woth AMD/NVIDIA
and stay ahead of them.

The Apple Silicon transition, Apple has said repeatedly, is 2 years.

Watches, phones, and iPads have been Apple Silicon for years. The watch
has always been Apple Silicon.

Apple did very well in chip development for IOS devices. Rapid pace and
stayd ahead of the pack. But once you add the full range of computers to their workload, How often can Apple release new chip for each market
segment? Remember rhat this now includes the GPU as well. So Apple is competing against Intel for laptops, AMD for higher end computers and
NVIDIA for GPUs.

And kicking their asses.

You literally cannot buy a non-Apple computer with the capabilities of
the $699 Mac mini for under $2000, and you would have to build that
yourself, and I am not actually sure it is possible because I can't find
any support out there for USB4 on the white-box market. Granted, didn't
look that hard, but fuck it's hard to find support for non-crippled TB3.

--
I WILL NOT INSTIGATE REVOLUTION Bart chalkboard Ep. 7G06
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <slrnrradl7.4f3.g.kreme@ProMini.lan> Lewis <g.kreme@kreme.dont-email.me> wrote:

In message <171120202054065457%nospam@nospam.invalid> nospam <nospam@nospam.invalid> wrote:

In article <Nv_sH.1482$xe4.309@fx41.iad>, Alan Browne
<bitbucket@blackhole.com> wrote:

I don't like the fixed memory option though - I liked that I could
purchase the iMac with min mem and then order upgrades at much lower
cost from crucial or the other guys. My next iMac will be at least 32 >>> >> GB if not 64 (assuming that's offered - I wouldn't be shocked to see
128). I'd be delighted if there is user upgradeable memory but I'm
certainly not expecting it.

there likely will be in higher end models.

this is the *beginning* of the transition, not the end.

The M1 has the memory modules (2) on the same chip carrier as the M1 as >>> far as I can see. Possibly very tight coupling to the M1 chip. Putting >>> the memory on a separate carrier might make impedance matching / speed
an issue. To be seen.

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.

On a guess, I look at what Apple did with the Mac Pro and I think
something along those lines with two M1s basically meshed together into
one super-high speed SoC might be what we see in the next round of MBPs, giving 16 cores of CPU, 16 cores of GPU and 4 USB4 ports.

Memory seems to be much more efficient, as things that were not really
doable without at least 16GB of RAM and really you wanted more are
doable on the M1 with 8 without breaking a sweat.

Another report from the low-end M1 MBP, "more than 100 tabs open in
Chrome and editing 4K video while Xcode [is also open] is like editing
on my iMac Pro."

Then there is this:
<https://sixcolors.com/post/2020/11/m1-macs-review/>
"Based on my testing, it’s also safe to say that all three M1-based
Macs, these low-end systems at the bottom of Apple’s price lists, are
among the fastest Macs ever made."

and

"To be clear, all three of these M1-based Macs are faster than every
currently shipping Mac except for the Mac Pro, the iMac Pro, and the
very highest-end configurations of the 5K iMac—the 8- and 10-core i9 configurations released earlier this year and the 8-core i9 model
released last year."


--
And east is east and west is west and if you take cranberries and
stew them like applesauce they taste much more like prunes than
rhubarb does.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18 02:21, JF Mezei wrote:

Big picture: competitors will also move ahead. Intel may be stuck ion molasses, but AMD is forging ahead at high speed. NVIDIA is moving at
light speed for GPUs.

Better picture:

In Mx Macs the core functionality is in a tiny area of the system
whereas x86 integration is by disparate parts making extreme tight
integration much less likely or possible.

That's before bringing up the OS...



--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18 11:23, Lewis wrote:

In message <slrnrradl7.4f3.g.kreme@ProMini.lan> Lewis <g.kreme@kreme.dont-email.me> wrote:

In message <171120202054065457%nospam@nospam.invalid> nospam <nospam@nospam.invalid> wrote:

In article <Nv_sH.1482$xe4.309@fx41.iad>, Alan Browne
<bitbucket@blackhole.com> wrote:

I don't like the fixed memory option though - I liked that I could >>>>>> purchase the iMac with min mem and then order upgrades at much lower >>>>>> cost from crucial or the other guys. My next iMac will be at least 32 >>>>>> GB if not 64 (assuming that's offered - I wouldn't be shocked to see >>>>>> 128). I'd be delighted if there is user upgradeable memory but I'm >>>>>> certainly not expecting it.

there likely will be in higher end models.

this is the *beginning* of the transition, not the end.

The M1 has the memory modules (2) on the same chip carrier as the M1 as >>>> far as I can see. Possibly very tight coupling to the M1 chip. Putting >>>> the memory on a separate carrier might make impedance matching / speed >>>> an issue. To be seen.

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.

On a guess, I look at what Apple did with the Mac Pro and I think
something along those lines with two M1s basically meshed together into
one super-high speed SoC might be what we see in the next round of MBPs,
giving 16 cores of CPU, 16 cores of GPU and 4 USB4 ports.

Memory seems to be much more efficient, as things that were not really
doable without at least 16GB of RAM and really you wanted more are
doable on the M1 with 8 without breaking a sweat.

Another report from the low-end M1 MBP, "more than 100 tabs open in
Chrome and editing 4K video while Xcode [is also open] is like editing
on my iMac Pro."

Then there is this:
<https://sixcolors.com/post/2020/11/m1-macs-review/>
"Based on my testing, it’s also safe to say that all three M1-based
Macs, these low-end systems at the bottom of Apple’s price lists, are
among the fastest Macs ever made."

and

"To be clear, all three of these M1-based Macs are faster than every currently shipping Mac except for the Mac Pro, the iMac Pro, and the
very highest-end configurations of the 5K iMac—the 8- and 10-core i9 configurations released earlier this year and the 8-core i9 model
released last year."

Which puts a stopper in me even thinking about the i7 8 core HT.

I'll wait for the Mx iMac or higher end mini.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-17 14:46, Your Name wrote:

On 2020-11-17 16:45:23 +0000, Alan Browne said:

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to
load
Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big >>>> Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting >>>> to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2
displays).

And the good news is, if your old Mac has a mechanical hard disk, take
note...

Scholle  McFarland in "Take Control of Catalina" says: "Get a solid
state drive (or a new Mac): macOS Mojave and Catalina both use Apple's
APFS system, which doesn't play well with mechanical hard drives."

Excellent point since I have to do a clean install it will fall to
APFS and, yes, I have a spinner in there.  I'll hold off some more and
look into changing the drive.  I'll have to order a repair kit from
iFixIt to replace the glass gasket.  <sigh>.

Upside is an SSD would be a lot quicker of course.

The SSD is even quicker on a new Apple Silicon Mac.

The point is to move my iMac to Catalina with a fresh install and keep
it running for a few more years. In the meantime I'll wait to see what
the new Mx iMac or Mini will be like next year.

I can see all sorts of uses for an updated iMac.

IAC- kit ordered a few minutes ago.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18 12:20:57 +0000, bje@ripco.com said:

RonTheGuy <ron@null.invalid> wrote:

Does Big Sur play OK with fusion drives?

I don't know for sure but it must because many of the older machines that
can run Big Sur had those.

I know I read somewhere Fusion is a dead issue on new machines which only makes sense since the whole idea was from a time that SSD's were only in the 64gb/128gb sizes.

Today with 1TB SSD's for $100 +/- $50, it just doesn't make sense anymore.

Except both RAM and SSD are soldered to the circuit board, and Apple
charges rather ridiculous amounts of money to increase RAM and storage
at time of ordering. The price of the two M1 Mac Mini models is US$699
for 256GB and US$899 for 512GB ... that's a $200 price jump. Changing
the 512GB SSD to a 1TB drive costs another US$200 while a 2TB drive
sets you back US$600. :-(

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On Nov 18, 2020 at 11:20:45 AM PST, "Alan Browne" <Blackhole@entropy.ultimateorg> wrote:

On 2020-11-17 14:46, Your Name wrote:

On 2020-11-17 16:45:23 +0000, Alan Browne said:

On 2020-11-16 19:53, Chris Schram wrote:

On 2020-11-16, Alan Browne <bitbucket@blackhole.com> wrote:

Just found out that this iMac (home, late 2012 v.) won't be able to >>>>> load
Big Sur...

Still haven't updated it to Catalina, was going to jump straight to Big >>>>> Sur from Mojave...

I have to do a clean install in any case - I had some hangup attempting >>>>> to update to Catalina some months ago. (already discussed).

But that should get me through to the next iMac (or Mini with 2
displays).

And the good news is, if your old Mac has a mechanical hard disk, take >>>> note...

Scholle  McFarland in "Take Control of Catalina" says: "Get a solid
state drive (or a new Mac): macOS Mojave and Catalina both use Apple's >>>> APFS system, which doesn't play well with mechanical hard drives."

Excellent point since I have to do a clean install it will fall to
APFS and, yes, I have a spinner in there.  I'll hold off some more and >>> look into changing the drive.  I'll have to order a repair kit from
iFixIt to replace the glass gasket.  <sigh>.

Upside is an SSD would be a lot quicker of course.

The SSD is even quicker on a new Apple Silicon Mac.

The point is to move my iMac to Catalina with a fresh install and keep
it running for a few more years. In the meantime I'll wait to see what
the new Mx iMac or Mini will be like next year.

I can see all sorts of uses for an updated iMac.

IAC- kit ordered a few minutes ago.

I made the leap from "High Sierra" skipping "Mojave" to "Catalina" a few weeks ago, mainly to continue Adobe CC, as the latest Adobe PS & LRc are no longer supported in macOS 10.13 "High Sierra".

I found the transition to Catalina problem free on this 2017 Quad-Core i7 iMac with a Fusion drive, and I am happy to live with that for a while. I do not plan to be an early adopter of "Big Sur" and I will probably wait into the new year before testing the new laptop waters.

--
Regards,
Savageduck


--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18, Alan Browne <Blackhole@entropy.ultimateorg> wrote:

On 2020-11-18 02:21, JF Mezei wrote:

Big picture: competitors will also move ahead. Intel may be stuck ion
molasses, but AMD is forging ahead at high speed. NVIDIA is moving at
light speed for GPUs.

Better picture:

In Mx Macs the core functionality is in a tiny area of the system
whereas x86 integration is by disparate parts making extreme tight integration much less likely or possible.

That's before bringing up the OS...

FUDsters gotta FUD...

From: <https://daringfireball.net/2020/11/the_m1_macs>

Rosetta is a marvel — the exemplification of “it just works”. x86-compiled apps run just about as fast as they do on the most
expensive Intel-based MacBook Pros, and faster, by a long shot, than on
the last and best Intel-based MacBook Air.

Let that sink in: apps compiled for Intel run *faster* in *translation*
on the M1 than they do on *actual* Intel CPUs in MacBook Airs and most
MacBook Pros.

Here, fine, I’ll show a few benchmarks from Geekbench (single-core / multi-core):

16-inch MacBook Pro (2019, 8-core Core i9): 1,160 / 7,160
MacBook Air (2020, 4-core Core i5): 1,050 / 2,410
M1 MacBook Pro (x86 benchmark running in Rosetta): 1,260 / 5,600
M1 MacBook Pro (native M1 benchmark): 1,730 / 7,530

You just don’t have to worry about Rosetta, period. The first time you
launch an Intel app, MacOS will ask you if you want to install Rosetta.
Agree, and installation flies by in a jiff — mere seconds — and after
that, you’ll just never notice whether something is compiled for Intel
or native Apple Silicon. You don’t have to worry about extensions (like
for Safari or Finder) being matched — Intel extensions run just fine in Apple Silicon apps. Dropbox, for example, is still an Intel app but it
runs just fine, including Finder integration.

Rosetta performs great and compatibility is seemingly seamless. This is
a technical tour de force, a home run for Apple that will mostly go
unheralded by typical users because the entire point is that they
shouldn't notice or care.

--
E-mail sent to this address may be devoured by my ravenous SPAM filter.
I often ignore posts from Google. Use a real news client instead.

JR
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18 07:24, nospam wrote:

apple can release new chips as often as they want. they are not tied to anyone else's schedule.

Meaningful improvements take time. Just because Apple has been able to
have one meaningful improvement every year since the A4 does not mean
that they have infinite capacity to release a new chip whenever they
want, especially now that its archifecture has diverged into multiple
branches from the watch, iPhone, the iPad/low end Macbooks, high end
MacBoosk with iMacs, and hi end iMacs with Mac Pro. And then add the
Apple TV, Mac Mini and what not. Don't forget their chip team also busy
with earphones, home speakers and whateer else Apple may be working on.



--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <ZTwtH.28215$kM7.16792@fx43.iad>, JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

apple can release new chips as often as they want. they are not tied to anyone else's schedule.

Meaningful improvements take time.

that's why the m1 macs were released this year and not earlier.

people have been expecting them for years.

Just because Apple has been able to
have one meaningful improvement every year since the A4 does not mean
that they have infinite capacity to release a new chip whenever they
want, especially now that its archifecture has diverged into multiple branches from the watch, iPhone, the iPad/low end Macbooks, high end
MacBoosk with iMacs, and hi end iMacs with Mac Pro. And then add the
Apple TV, Mac Mini and what not.

straw man. nobody said infinite capacity. the reality is that apple has
plenty of talent and capacity to release all sorts of things.

Don't forget their chip team also busy
with earphones, home speakers and whateer else Apple may be working on.

you do realize that their chip team is big enough to do multiple chips,
right? apparently not.

here's a clue: their chip design team is several thousand people.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18 10:06, Lewis wrote:

Unlikely that anyone can replicate what Apple has done, and they won't
be able to even start to try for years.

Consider what NVIDIA has just come out with.
Consider what AMD has just released for the Ryzen 8086s.
Don't underestimate their ability at the high end even if Intel is comatose.

It doedsn't seem to matter, the one advantage that seems clear from the
M1 si that the tight integration yeilds massive benefits, not just the
tight integration of the M1, but the fact that the chips are optimized
for the operating system and the operating system is optimized for the
chips.

Also remember that Apple proprietary apps are written for the hardware
and make use of the proprietary stuff like Metal, neural engine and
image processing sub processors. Normal multiplatform apps are far less
likely to be optiomized to use those features and will rely much more on
the CPU, and some may go as far as supporting Apple's Metal.

When an app on Windows makes use of NVIDIA's proprietary CUDA GPU, but
on the MAC is 100% CPU, then the Windows version has an advantage
irrespective of how well the Apple ARM cores perform.


Consider NVIDIA,s new RTX 3080. https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3080/

8704 cores at between 1.44 and 1.71 gHz. 10BG of RAM. The thing
consumes 400 watts. And that is just the GPU.

Apple's M1 has 8 GPU cores and might be so power efficient it can run on
a CR 2032 pill battery for 10 weeks, but somehow, I doubt it can compare against a behemoth of a GPU with a gazillion cores that consume 400
watts (with all the cooling challenges accompanying this).

Qualcomm is still years behind Apple, and every year the gap
widens.

At the core performance, I think they are less than a year apart. What
happens to the Wintel market will be very interesting. Will AMD and
NVIDIA give tghe 8086 new life, or is the market moving to ARM, at which
pointl does Qualcomm become a major force and what happens of Intel.

Intel can't do that, they don't write the OS. NVidia can't do that. AMD
can't do that.

I am pretty sure there are some serious conversations between AMD and
Microsoft and between Intel and Microsoft on what features are needed.
Intel can also do a lot of performan metric tests when rining windows on
its machines. (though being comatose, they may no lonegr be doing it,
but AMD sure are).

The Apple Silicon transition, Apple has said repeatedly, is 2 years.

I was talking about sustainable rate of developmeht for a wide range of
chips not just the first iteration which likely has been in the works
for many years for the peripherals, waiting only to get the latest cores
and neural engine/GPU pasted into the design.

Going forward, Apple will have to maintain/improve a lot of different
chips. So it is a lot of work.

This is why careful planning of development becomes important. I
suspect cores will be common design, but other chips will have different prripheral processors for IO, and likely different memory controller.
When it gets a big question is what IO options larger macs will have.


--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <GMxtH.11057$2j.3742@fx38.iad>, JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

Also remember that Apple proprietary apps are written for the hardware
and make use of the proprietary stuff like Metal, neural engine and
image processing sub processors.

those are public apis which anyone can use.

Normal multiplatform apps are far less
likely to be optiomized to use those features and will rely much more on
the CPU, and some may go as far as supporting Apple's Metal.

wrong.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <GMxtH.11057$2j.3742@fx38.iad> JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

On 2020-11-18 10:06, Lewis wrote:

Unlikely that anyone can replicate what Apple has done, and they won't
be able to even start to try for years.

Consider what NVIDIA has just come out with.

Cards that cost as much as the Mac mini.

Consider what AMD has just released for the Ryzen 8086s.

There is no such thing as Ryzen 8086.

Don't underestimate their ability at the high end even if Intel is comatose.

So you want to compare high end to the LOWEST END of the Apple Silicone
Macs? Yep, that does sound like you.

:"Why, this $700 Mac is barely as fast as this $3500 system I can build,
if I ignore the lack of USB-4 and the 600 Watt PSU I'll need to run it
off of."

Now, compare your "high end" bullshit ot a sub $1000 laptop with a 17
hour battery life. You can throw all the money in the world at your
NVidia and AMD and you still will get nowhere near that.

Your FUD game is pathetic.

Also remember that Apple proprietary apps are written for the hardware
and make use of the proprietary stuff like Metal, neural engine and

You mean unlike Adobe Proprietary apps and Nvidia proprietary CUDA?

What a pathetic troll.

Consider NVIDIA,s new RTX 3080. https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3080/

Consider the $800 price tag, dipshit. Consider that you can't put that
card in a laptop, dumbass.

8704 cores at between 1.44 and 1.71 gHz. 10BG of RAM. The thing
consumes 400 watts. And that is just the GPU.

Apple's M1 has 8 GPU cores and might be so power efficient it can run on
a CR 2032 pill battery for 10 weeks, but somehow, I doubt it can compare against a behemoth of a GPU with a gazillion cores that consume 400
watts (with all the cooling challenges accompanying this).

It compares very favorably, in fact. You, of course will ignore the
power draw, the cost, and the lack of portability.

Qualcomm is still years behind Apple, and every year the gap
widens.

At the core performance, I think they are less than a year apart.

Yes, but you are a fool.

I am pretty sure there are some serious conversations between AMD and Microsoft and between Intel and Microsoft on what features are needed.
Intel can also do a lot of performan metric tests when rining windows on
its machines. (though being comatose, they may no lonegr be doing it,
but AMD sure are).

the M1 is several generations ahead of the newest AMD chips, and it is the lowest end Mac chip Apple will ever make.

The Apple Silicon transition, Apple has said repeatedly, is 2 years.

I was talking about sustainable rate of developmeht for a wide range of
chips

No, you were making up vague FUD like you always do. It's all you are
capable of.

--
"Are you pondering what I'm pondering?"
"I think so, Brain, but shouldn't the bat boy be wearing a cape?"
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-19 11:16, JF Mezei wrote:

On 2020-11-18 07:24, nospam wrote:

apple can release new chips as often as they want. they are not tied to
anyone else's schedule.

Meaningful improvements take time. Just because Apple has been able to
have one meaningful improvement every year since the A4 does not mean
that they have infinite capacity to release a new chip whenever they
want, especially now that its archifecture has diverged into multiple branches from the watch, iPhone, the iPad/low end Macbooks, high end
MacBoosk with iMacs, and hi end iMacs with Mac Pro.

Who claimed infinitely? Advances in CPU's have been mainly through core number growth and for that there is plenty of room to grow over the next decade+.

Those are actually advantages since CPU architecture is much like code:
re-use is high. So an advance in one product CPU design can be quickly adapted to another. This was not possible with intel CPU's, obviously.
I won't belabour the OS/HW integration that now has a very fine seam.

As to team sizes, that is not only scalable but it is desirable that
they enlarge - again akin to software teams.

The real challenge is managing it all. But of course (something you've
never seemed to grasp) Apple don't "suddenly" come out with things.
They are years in the planning and execution. Work on whatever comes
after M1 began long before M1 was even close to final design. Whatever
"comes next" is certainly running on half a dozen candidate products as
they wend there way to final configuration.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-19 12:17, JF Mezei wrote:

On 2020-11-18 10:06, Lewis wrote:

Unlikely that anyone can replicate what Apple has done, and they won't
be able to even start to try for years.

Consider what NVIDIA has just come out with.

Consider the irrelevancy of your statements and move on. You're pedaling
in reverse. As nospam correctly points out the M1 is _the beginning_.

To whit, I had the 2007 core duo dual core. Nice. Then I had the i7
quad core HT. Blows the former away to the point where I'm still in no
rush to replace it near 8 years later.

Now given Apple's history, do you really, really believe that things are
not going to get better than what is already a stellar start?

The competition is now in a pretty sad place where Apple is concerned
because Apple have pulled most of the computing elements of the computer
in close. In close to every important i/o element and to the OS as well.

Further providing open API's to things like Metal and a generously large system dev kit for developers to use - not only in Mac OS but across the device family from Apple.

For developers this makes things 'cheap' to copy across devices where applicable. Indeed, as Apple say, all your iPhone/iPad apps will
already run on M1 w/o change. For a developer, applying that to a "appropriate" to type version, it becomes much cheaper in every way (to
be clear, via XCode it was already quite easy. Just got easier).

The rest of the personal computing market is stuck gluing together x86
to GPU's and memory and I/O in a loose, expensive to tie together
environment. Wasted cycles, wasted power, wasted time. Even if intel
offer a comparable x86 SoC (it's not impossible) it will consume much
more power than the ARM based system and will be a more expensive fab to
boot.

The MBA at US$699 is looking more appealing than ever to Windows laptop
users now...

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-18 02:21, JF Mezei wrote:

On 2020-11-17 20:54, nospam wrote:

the m1 is the first of 'a family'.

future members of the clan will either have more onboard memory or a
way to add it separately, perhaps even both.

Big picture: competitors will also move ahead. Intel may be stuck ion molasses, but AMD is forging ahead at high speed. NVIDIA is moving at
light speed for GPUs.

How long it takes for Apple to roll out the full family from the watch
to the Mac Pro will dictate how well they can keep up woth AMD/NVIDIA
and stay ahead of them.

Apple did very well in chip development for IOS devices. Rapid pace and
stayd ahead of the pack. But once you add the full range of computers to their workload, How often can Apple release new chip for each market
segment? Remember rhat this now includes the GPU as well. So Apple is competing against Intel for laptops, AMD for higher end computers and
NVIDIA for GPUs.

Apple broke the graphics mold years ago on iPads (iPad pro has been
performing better than most laptops (of the same year) for at least 2
years. Once Apple stop supporting intel based Macs (5 - 7 years), then
they will be free of everything except peripheral i/o concerns (external storage, printers, other devices). Those will become more and more
wireless as well.

Apple will have absolutely no issue outfitting the entire line of Macs
with new processors that outperform most if not all intel/AMD based x86 designs. When it comes to desktop (Mini, iMac, Pro) they will be pretty
much unconstrained power wise and can simply throw cores at it until the
cows come home...

And by the time 18 months rolls around, they'll be announcing the fresh
round to replace M1 in the MBA and lower end MBP and mini...

I don't understand your obsession with trying (badly) to pick holes in
what Apple are doing. Is it because you just wasted $400 on updating
your old Mac Pro?

(Full disclosure: I will be updating this iMac very soon with an SSD as
a "life extension" project. Also repairing the tilt stand while I'm in there... too bad the SSD won't be here this weekend).

The only losers in all this is the small "hackintosh" community. But
they knew this was coming some day.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <_3StH.657$se1.64@fx27.iad> Alan Browne <bitbucket@blackhole.com> wrote:

I don't understand your obsession with trying (badly) to pick holes in
what Apple are doing. Is it because you just wasted $400 on updating
your old Mac Pro?

I have had a list of parts for an intended AMD low end machine that I
was going to stick my old GTX 950 in and use for h.265 encoding via
NVENC and for maybe using as a spare machine for the occasional wintendo
game.

That machine parts out at $653 without a video card (sinc I already own
it). Or, I can spend $46 more and get a mac mini that does everything
that machine will do and far more.

Hmm.. decisions. I mean, it's a tough one, isn't it? (No, it is not a
tough one at all). Duh.

The only losers in all this is the small "hackintosh" community. But
they knew this was coming some day.

Even they aren't losers as they were building machines to try to get
better performance.



--
Forgive your enemies, but remember their names.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-20 11:39, Lewis wrote:

In message <_3StH.657$se1.64@fx27.iad> Alan Browne <bitbucket@blackhole.com> wrote:

I don't understand your obsession with trying (badly) to pick holes in
what Apple are doing. Is it because you just wasted $400 on updating
your old Mac Pro?

I have had a list of parts for an intended AMD low end machine that I
was going to stick my old GTX 950 in and use for h.265 encoding via
NVENC and for maybe using as a spare machine for the occasional wintendo game.

That machine parts out at $653 without a video card (sinc I already own
it). Or, I can spend $46 more and get a mac mini that does everything
that machine will do and far more.

Hmm.. decisions. I mean, it's a tough one, isn't it? (No, it is not a
tough one at all). Duh.

The only losers in all this is the small "hackintosh" community. But
they knew this was coming some day.

Even they aren't losers as they were building machines to try to get
better performance.

Certainly. I meant in the sense that their never ending quest to build machines that run the latest Mac OS is now in it's last phase. There
will be a "last" x86 Mac OS just as there was a last PowerPC OS X.

3 years? 5 maybe?

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <iFStH.12059$sW6.3343@fx47.iad> Alan Browne <bitbucket@blackhole.com> wrote:

On 2020-11-20 11:39, Lewis wrote:

In message <_3StH.657$se1.64@fx27.iad> Alan Browne <bitbucket@blackhole.com> wrote:

I don't understand your obsession with trying (badly) to pick holes in
what Apple are doing. Is it because you just wasted $400 on updating
your old Mac Pro?

I have had a list of parts for an intended AMD low end machine that I
was going to stick my old GTX 950 in and use for h.265 encoding via
NVENC and for maybe using as a spare machine for the occasional wintendo
game.

That machine parts out at $653 without a video card (sinc I already own
it). Or, I can spend $46 more and get a mac mini that does everything
that machine will do and far more.

Hmm.. decisions. I mean, it's a tough one, isn't it? (No, it is not a
tough one at all). Duh.

The only losers in all this is the small "hackintosh" community. But
they knew this was coming some day.

Even they aren't losers as they were building machines to try to get
better performance.

Certainly. I meant in the sense that their never ending quest to build machines that run the latest Mac OS is now in it's last phase. There
will be a "last" x86 Mac OS just as there was a last PowerPC OS X.

3 years? 5 maybe?

Apple was pretty fast to stop support on PowerPC Macs, but even then I
think it was 7?

Let's see, Tiger came out in Jan 2006 with support for Intel Mac and
10.7 dropped Intel macs in July 2011, so 5 years?

I suspect something like that, though I would not be surprised to see a
much more pronounced bifurcation in the OS between AS and Intel with the
AS machines getting a lot more features and capabilities just because
they have processing power to burn.

--
Do not meddle in the affairs of wizards for they are subtle and quick
to anger.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <slrnrrg0ab.dag.g.kreme@ProMini.lan>, Lewis <g.kreme@kreme.dont-email.me> wrote:

Apple was pretty fast to stop support on PowerPC Macs, but even then I
think it was 7?

Let's see, Tiger came out in Jan 2006 with support for Intel Mac and
10.7 dropped Intel macs in July 2011, so 5 years?

part of that was due to ibm buying rosetta and refusing to license it
to apple (or anyone else for that matter). apple had no choice but to
cease offering it. that said, demand had certainly dropped off.

rosetta 2 is apple's own technology and they can keep it going as long
as they see a reason to do so.

keep in mind that there are a *lot* more mac users with a *lot* more
intel apps than in the previous transition.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 20 Nov 2020 at 19:05:33 GMT, nospam <nospam@nospam.invalid> wrote:

rosetta 2 is apple's own technology and they can keep it going as long
as they see a reason to do so.

Original Rosetta translated on the fly, and required Apple to supply PowerPC versions of all framworks and APIs, too. Since Rosetta 2 translates once only, and that before the app actually runs, does this mean that Big Sur running on AS doesn't need to include Intel versions of all the frameworks and APIs?

--
Tim


--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-20 14:18, Tim wrote:

On 20 Nov 2020 at 19:05:33 GMT, nospam <nospam@nospam.invalid> wrote:

rosetta 2 is apple's own technology and they can keep it going as long
as they see a reason to do so.

Original Rosetta translated on the fly, and required Apple to supply PowerPC versions of all framworks and APIs, too. Since Rosetta 2 translates once only,
and that before the app actually runs, does this mean that Big Sur running on AS doesn't need to include Intel versions of all the frameworks and APIs?

Rosetta 2 will still do JIT conversions in some cases.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <201120201405339798%nospam@nospam.invalid> nospam <nospam@nospam.invalid> wrote:

In article <slrnrrg0ab.dag.g.kreme@ProMini.lan>, Lewis <g.kreme@kreme.dont-email.me> wrote:

Apple was pretty fast to stop support on PowerPC Macs, but even then I
think it was 7?

Let's see, Tiger came out in Jan 2006 with support for Intel Mac and
10.7 dropped Intel macs in July 2011, so 5 years?

part of that was due to ibm buying rosetta and refusing to license it
to apple (or anyone else for that matter). apple had no choice but to
cease offering it. that said, demand had certainly dropped off.

Yes to both of those.

rosetta 2 is apple's own technology and they can keep it going as long
as they see a reason to do so.

Apple does not call it Rosetta 2 AFAIK.

<https://developer.apple.com/documentation/apple_silicon/about_the_rosetta_translation_environment>

keep in mind that there are a *lot* more mac users with a *lot* more
intel apps than in the previous transition.

Yes, but this Rosetta is 1) all apple and 2) runs intel apps faster
than intel chips do.


--
Twentieth century? Why, I could pick a century out of a hat,
blindfolded, and come up with a better one.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-20 21:38:54 +0000, Lewis said:

In message <201120201405339798%nospam@nospam.invalid> nospam <nospam@nospam.invalid> wrote:

In article <slrnrrg0ab.dag.g.kreme@ProMini.lan>, Lewis
<g.kreme@kreme.dont-email.me> wrote:

Apple was pretty fast to stop support on PowerPC Macs, but even then I
think it was 7?

Let's see, Tiger came out in Jan 2006 with support for Intel Mac and
10.7 dropped Intel macs in July 2011, so 5 years?

part of that was due to ibm buying rosetta and refusing to license it
to apple (or anyone else for that matter). apple had no choice but to
cease offering it. that said, demand had certainly dropped off.

Yes to both of those.

rosetta 2 is apple's own technology and they can keep it going as long
as they see a reason to do so.

Apple does not call it Rosetta 2 AFAIK.

<https://developer.apple.com/documentation/apple_silicon/about_the_rosetta_translation_environment>

keep in mind that there are a *lot* more mac users with a *lot* more
intel apps than in the previous transition.

Yes, but this Rosetta is 1) all apple and 2) runs intel apps faster
than intel chips do.

Rosetta 2 doesn't "run" Intel apps at all. It translates apps into a
Universal Binary on the first opening. After the translation process
during that first opening, thea apps are running *natively* on Apple
Silicon.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <slrnrrgdre.14g6.g.kreme@ProMini.lan>, Lewis <g.kreme@kreme.dont-email.me> wrote:

Apple was pretty fast to stop support on PowerPC Macs, but even then I
think it was 7?

Let's see, Tiger came out in Jan 2006 with support for Intel Mac and
10.7 dropped Intel macs in July 2011, so 5 years?

part of that was due to ibm buying rosetta and refusing to license it
to apple (or anyone else for that matter). apple had no choice but to
cease offering it. that said, demand had certainly dropped off.

Yes to both of those.

rosetta 2 is apple's own technology and they can keep it going as long
as they see a reason to do so.

Apple does not call it Rosetta 2 AFAIK.

<https://developer.apple.com/documentation/apple_silicon/about_the_rosetta_tra
nslation_environment>

looks like they're calling it both rosetta and rosetta 2:

<https://support.apple.com/en-us/HT211861>
If you need to install Rosetta on your Mac

Rosetta 2 enables a Mac with Apple silicon to use apps built for
a Mac with an Intel processor.
...
About Rosetta
Rosetta 2 works in the background whenever you use an app built
only for Mac computers with an Intel processor. It automatically
translates the app for use with Apple silicon.

keep in mind that there are a *lot* more mac users with a *lot* more
intel apps than in the previous transition.

Yes, but this Rosetta is 1) all apple and 2) runs intel apps faster
than intel chips do.

true, but eventually, nearly all apps will have been updated for apple
silicon, so at some point in the future, apple may decide that it's not
worth maintaining for the tiny, tiny minority that aren't.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-20 07:49, Lewis wrote:

So you want to compare high end to the LOWEST END of the Apple Silicone
Macs? Yep, that does sound like you.

If you exlude the Mac pro, the 3 new Macs are actually at the high end
of the Mac lineup in terms of performance (based on what has been
published so far).

Until Apple releases iMacs and MacPro, we can't really know how much
faster those machines will go vs the 3 released so far. We don't know
what memory model they will choose (all within the Soc, or will higher
end machines get external RAM slots?

Same with the GPU. Will higher end machines have external GPU (either
Apple or 3rd party) ? or will all of them be limited to in-chip apple
GPUs ?

We don't know that.

But what we do know is that on the wintel side, AMD/Ryzen and NVIDIA
have made a huge leap forward these past couple of months.

Now, compare your "high end" bullshit ot a sub $1000 laptop with a 17
hour battery life.

It is important to put the M1 performance in perspective. It may be very
very impressive compared to other Macs, but one must not forget there is computing power available oustide of the Mac marketplace.

M1 doesn't fill the full gamut of CPU needs. This is why I brought in
the reality check with the new RYZEN and NVIDIA products. As Apple moves
in its transition, it will eventually reach the need to release Macs
that compete against Ryzen and NVIDIA's performance.

You mean unlike Adobe Proprietary apps and Nvidia proprietary CUDA?

Yes, NVIDIA's CUDA is proprietary. And it leaves AMD with OpenCL (which
Apple is dropping). But NVIDIA is king right now. First to see
Appleication support.

Apple has to nudge developpers to support Metal. And now, Apple is also
the new kid on the block in terms of GPU hardware.

There is no question that M1 has performance that exceeds what one
expects in laptops, and by a big margin.

Compared to the Intel built-in GPU, Apple's built-in GPU is far better
and consumes less power.


But until Apple releases the iMac and Mac Pro, we can't know how well
they will do against Ryzen/NVidia Wintel boxes.

There are a lot of possibilities, but until Apple releases the rpducts,
we can't know for sure.

Could be that the new products will blast performance for those levels
of cimputers the same way M1 blaster theoirugh olaptop performance. And
Apple may have topped what it can do with its ARM cores and we won't see
much change for Imacs. We don't know.

It would be very interesting to do an electron microscope of the 8 cores
in M1. I wouldn't be susprised if the same cores will be used to do an
iMac with 8 high performance cores (aka: bost power to the 4 low power
cores to make them high performance), and the 8 with hiogher close
rates. Once you are in a pplugged in environment, battery life no longer
an issue.

And with desktops being lower volume, Apple can take the higher quality
chips (lower yield) for desktops, and use the lower quality ones for
laptops with 4 cores brrought down to "energy efficient" speed and the 4
high performance core are lower clock rate to same on power.

or Apple could have a truly different chip for them. (which might be
necessary if you have external RAM, more IO controllers etc).

The point is that no matter how great the laptop performance is with M1,
it does not garantee that Apple will be able to scale it up to the Mac
Pro and compete against Ryzen and NVIDIA. Only time will tell.

(and before you should "FUD", I am not saying Apple won't succeed. What
I am saying is that we are not at a point where we can know).

It compares very favorably, in fact. You, of course will ignore the
power draw, the cost, and the lack of portability.

Apple has proven it can do laptops incredibly well with M1. The
question is how well will it do for iMacs and Mac Pro.

At the core performance, I think they are less than a year apart.

Yes, but you are a fool.

So you agree, but call me a fool, right ?

the M1 is several generations ahead of the newest AMD chips, and it is the lowest end Mac chip Apple will ever make.

Ryzen 9 5000 series suchj as 5850X is done at 7nm, so half a generation
behind Apple's 5nm.
https://www.amd.com/en/products/cpu/amd-ryzen-9-5950x

16 cores (32 threads), 3.4ghz boost to 4.9.



Intel is ages behind stills tuck at 14nm and some chips at 10.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

Original Rosetta translated on the fly, and required Apple to supply PowerPC >> versions of all framworks and APIs, too. Since Rosetta 2 translates once only,
and that before the app actually runs, does this mean that Big Sur running on
AS doesn't need to include Intel versions of all the frameworks and APIs?

There are Intel versions of all frameworks supplied.

The translator relinks all system calls to the Intel version of the
system routine.

The translated image still uses the Intel subroutine calling standard
(argument passing, register containing return address etc).

Upon being called, the Intel version of the system routine unpacks the
Intel format argument list and builds a native ARM argument list and
then calls the real system routine (and does the same when it returns so
the return code can be returned to the translated image).

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <wUYtH.2474$7D7.1308@fx03.iad> JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

On 2020-11-20 07:49, Lewis wrote:

So you want to compare high end to the LOWEST END of the Apple Silicone
Macs? Yep, that does sound like you.

If you exlude the Mac pro, the 3 new Macs are actually at the high end
of the Mac lineup in terms of performance (based on what has been
published so far).

The are the lowest end of the new Mac line featuring Apple Silicon
chips, but nice try FIDmeister.

We don't know that.

You continue to spread nothing but FUD.

But what we do know is that on the wintel side, AMD/Ryzen and NVIDIA
have made a huge leap forward these past couple of months.

Nothing compared to what Apple has done.

M1 doesn't fill the full gamut of CPU needs. This is why I brought in
the reality check with the new RYZEN and NVIDIA products.

No, you did it to spread absurd informed garbage FUD.

As Apple moves in its transition, it will eventually reach the need to release Macs that compete against Ryzen and NVIDIA's performance.

They already have.

At the core performance, I think they are less than a year apart.

Yes, but you are a fool.

So you agree, but call me a fool, right ?

I agree that you think nonsense and are a fool.

the M1 is several generations ahead of the newest AMD chips, and it is the >> lowest end Mac chip Apple will ever make.

Ryzen 9 5000 series suchj as 5850X is done at 7nm, so half a generation behind Apple's 5nm.

Apple SoC is just about as fast and much smaller and much lower power
and can be put in a sub $1000 laptop and a sub $700 desktop.


--
What's another word for Thesaurus?
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-21 00:46:18 +0000, Lewis said:

In message <wUYtH.2474$7D7.1308@fx03.iad> JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

On 2020-11-20 07:49, Lewis wrote:

So you want to compare high end to the LOWEST END of the Apple Silicone
Macs? Yep, that does sound like you.

If you exlude the Mac pro, the 3 new Macs are actually at the high end
of the Mac lineup in terms of performance (based on what has been
published so far).

The are the lowest end of the new Mac line featuring Apple Silicon
chips, but nice try FIDmeister.

It says "in terms of performance" ... so, yes, the M1 Macs are the
low-end price-wise, but *performance-wise* they're faster than every
existing Intel Mac model except the top-end Mac Pro and iMac Pro models.

--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In message <IjZtH.13313$J92.5230@fx48.iad> JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

Original Rosetta translated on the fly, and required Apple to supply PowerPC
versions of all framworks and APIs, too. Since Rosetta 2 translates once only,
and that before the app actually runs, does this mean that Big Sur running on
AS doesn't need to include Intel versions of all the frameworks and APIs?

There are Intel versions of all frameworks supplied.

The translator relinks all system calls to the Intel version of the
system routine.

The translated image still uses the Intel subroutine calling standard (argument passing, register containing return address etc).

Upon being called, the Intel version of the system routine unpacks the
Intel format argument list and builds a native ARM argument list and
then calls the real system routine (and does the same when it returns so
the return code can be returned to the translated image).

Cite? Or are you making shit up?

--
Magic doesn't come from talent, it comes from pain.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

In article <slrnrrh29k.c7c.g.kreme@ProMini.lan>, Lewis <g.kreme@kreme.dont-email.me> wrote:

There are Intel versions of all frameworks supplied.

The translator relinks all system calls to the Intel version of the
system routine.

The translated image still uses the Intel subroutine calling standard (argument passing, register containing return address etc).

Upon being called, the Intel version of the system routine unpacks the Intel format argument list and builds a native ARM argument list and
then calls the real system routine (and does the same when it returns so the return code can be returned to the translated image).

Cite? Or are you making shit up?

the answer is obvious.
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-20 19:38, JF Mezei wrote:

Original Rosetta translated on the fly, and required Apple to supply PowerPC
versions of all framworks and APIs, too. Since Rosetta 2 translates once only,
and that before the app actually runs, does this mean that Big Sur running on
AS doesn't need to include Intel versions of all the frameworks and APIs?

There are Intel versions of all frameworks supplied.

The translator relinks all system calls to the Intel version of the
system routine.

The translated image still uses the Intel subroutine calling standard (argument passing, register containing return address etc).

Upon being called, the Intel version of the system routine unpacks the
Intel format argument list and builds a native ARM argument list and
then calls the real system routine (and does the same when it returns so
the return code can be returned to the translated image).

You're thinking this as an emulation when it is a translation. Rosetta
is not a generic, 3rd party app. It is an Apple product designed to
make the transition as seamless as possible. Where Rosetta 1 was
excellent, Rosetta 2 will be far better. Apple aren't throwing RD&E at everything else while making Rosetta 2 a 2nd class endeavour.

So, in the instance where Rosetta translates the intel code once and
once only, it will most likely re-interpret the passing/return of data
via registers and stack to the most efficient way possible. It will not emulate the x86 code instruction by instruction. It will replace it
both at the call and the callee to be functionally identical with the
best ARM instructions for the call frame.

IOW: Rosetta does not have to be as primitive as you suggest. And I
would bet that the Rosetta s/w engineers have looked at thousands of
cases (using appropriate tools) to design the translation to be very
efficient where all operations are concerned, esp. the relatively
expensive case of a complex function/procedure call. An investment
there pays off in a huge way. There may be some statistically rare
cases where the investment doesn't pay off and a purely emulated call
frame is used. But I doubt it very much. Such conventions have been
pretty straightforward over the past 30 years with superficial
difference between OS' (where a given app language is concerned) and
tweaks along the way but little more.

And again (since you don't seem to get it), the fact that the M1 offers
apps 13 more 64bit registers to use over the 16 from intel, Rosetta will likely use those in lieu of stack based vars wherever possible further speeding up the converted app. Pointers in register are very fast v.
stack located pointers. Very often, esp. where arrays or structures are passed to a s/r, pointers are used - and in register for the call.
Where this can be done in register for the call, the call is lightning
fast as memory ops are greatly reduced. Functions typically return
results in a register, very often as a pointer to the resulting
structure (or array) as provided by the caller; where the function has multiple parameters returned the registers are usually more than ample
for simple variables as well as pointers to the result.

(There are usually register protection ops that must be done and those
of course are pushes to the stack).

The function/procedure that is called will historically allocate its
operating variables on the stack. These may be simple to complex. But structurally little different from x86 to ARM"86". And again:
registers. Optimizing compilers typically allocate to registers what
used to go on stack. I posted a fragment of assembler code generated by compiler recently.

Simple variables can also be variable located and with 29 available
registers, such will be used by Rosetta w/o direct emulation of x86
code. It will be translated to the most efficient v. possible in almost
all cases.

Translating the x86 v. of that to the M1 version of that will not pose
any huge engineering challenge for the likes of Apple.

And while I'm thinking about it, I would surmise that Rosetta will
detect XCode developed apps and have a precise idea of how code is
generated for nearly 100% of cases. In such cases, the translation can
be (and probably will be) as perfect as if the code was originally
compiled for ARM.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113

From Newsgroup: comp.sys.mac.system

On 2020-11-20 22:27, Lewis wrote:

In message <IjZtH.13313$J92.5230@fx48.iad> JF Mezei <jfmezei.spamnot@vaxination.ca> wrote:

Original Rosetta translated on the fly, and required Apple to supply PowerPC
versions of all framworks and APIs, too. Since Rosetta 2 translates once only,
and that before the app actually runs, does this mean that Big Sur running on
AS doesn't need to include Intel versions of all the frameworks and APIs?

There are Intel versions of all frameworks supplied.

The translator relinks all system calls to the Intel version of the
system routine.

The translated image still uses the Intel subroutine calling standard
(argument passing, register containing return address etc).

Upon being called, the Intel version of the system routine unpacks the
Intel format argument list and builds a native ARM argument list and
then calls the real system routine (and does the same when it returns so
the return code can be returned to the translated image).

Cite? Or are you making shit up?

Making it up. I think the one time he faced assembler code in his life
(on a HAL-9000 I think), informed his entire worldview of computer architecture.

--
"...there are many humorous things in this world; among them the white
man's notion that he is less savage than the other savages."
-Samuel Clemens
--- Synchronet 3.18b-Win32 NewsLink 1.113