Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Qualcomm CEO says he’s expecting Apple to use its own modems in iPhones in 2024
8 views
Skip to first unread message
badgolferman
Mar 1, 2023, 11:48:21 AM3/1/23
to
Apple is moving to in-house 5G modem chips for its 2024 iPhones, as far
as the chief executive of Qualcomm — which currently produces them for
the tech giant — is aware.
“We’re making no plans for 2024, my planning assumption is we’re not
providing [Apple] a modem in ’24, but it’s their decision to make,”
Cristiano Amon told CNBC at the Mobile World Congress in Barcelona.
Apple’s most recent iPhone 14 models use Qualcomm modems, but the
company has been looking to go solo in the wireless connectivity market
for some years.
It bought Intel’s modem business in 2019 and there had been speculation
it would begin using in-house parts this year.
In an interview with CNBC’s Karen Tso and Arjun Kharpal, Amon said
Qualcomm had told investors back in 2021 that it did not expect to
provide modems for the iPhone in 2023, but Apple then decided to
continue for another year.
Amon did not confirm whether Apple would pay Qualcomm QTL licenses if
it moves to its own modems, but said royalty was “independent from
providing a chip.”
https://www.cnbc.com/2023/03/01/qualcomm-ceo-says-planning-for-apple-to-make-own-iphone-modems-from-2024-.html
as the chief executive of Qualcomm — which currently produces them for
the tech giant — is aware.
“We’re making no plans for 2024, my planning assumption is we’re not
providing [Apple] a modem in ’24, but it’s their decision to make,”
Cristiano Amon told CNBC at the Mobile World Congress in Barcelona.
Apple’s most recent iPhone 14 models use Qualcomm modems, but the
company has been looking to go solo in the wireless connectivity market
for some years.
It bought Intel’s modem business in 2019 and there had been speculation
it would begin using in-house parts this year.
In an interview with CNBC’s Karen Tso and Arjun Kharpal, Amon said
Qualcomm had told investors back in 2021 that it did not expect to
provide modems for the iPhone in 2023, but Apple then decided to
continue for another year.
Amon did not confirm whether Apple would pay Qualcomm QTL licenses if
it moves to its own modems, but said royalty was “independent from
providing a chip.”
https://www.cnbc.com/2023/03/01/qualcomm-ceo-says-planning-for-apple-to-make-own-iphone-modems-from-2024-.html
badgolferman
Mar 1, 2023, 11:58:32 AM3/1/23
to
of problems? It's not like they can actually simulate real-world
conditions and do the amount of comprehensive testing necessary to work
out the bugs.
--
"Being happy doesn't mean everything's perfect; it just means you've
decided to see beyond the imperfections." ~ Unknown
nospam
Mar 1, 2023, 12:13:54 PM3/1/23
to
In article <xn0nytut...@reader443.eternal-september.org>,
low.
> It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
what makes you say that? they have enormous resources to do all sorts
of testing.
part of that includes employees using pre-release phones all over the
san francisco bay area and well beyond it.
put it in an iphone 14 body and nobody outside of the people carrying
it will know, possibly even them too.
badgolferman <REMOVETHISb...@gmail.com> wrote:
>
> What are the chances Apple's first generation 5G modems will have lots
> of problems?
nothing is perfect, but given their success with apple silicon, it's
>
> What are the chances Apple's first generation 5G modems will have lots
> of problems?
low.
> It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
of testing.
part of that includes employees using pre-release phones all over the
san francisco bay area and well beyond it.
put it in an iphone 14 body and nobody outside of the people carrying
it will know, possibly even them too.
badgolferman
Mar 1, 2023, 12:20:48 PM3/1/23
to
design has many bugs the first couple years. Something as sensitive to
environmental conditions as a modem can't be tested in San Francisco
only. It needs a very wide and long test bed.
nospam
Mar 1, 2023, 12:46:36 PM3/1/23
to
In article <xn0nytvh...@reader443.eternal-september.org>,
motors or honda.
also, designing a new car is more complicated than a single component
in said car.
> Something as sensitive to
> environmental conditions as a modem can't be tested in San Francisco
> only. It needs a very wide and long test bed.
yep, which is why it's tested in more than just san francisco.
it's pretty much guaranteed that the modems, which they've been working
on for many years, are currently in employee iphones and ipads that are
being used over the world right now, possibly in other devices too.
and i think the team who's working on it is based in san diego, not
that it changes much.
badgolferman <REMOVETHISb...@gmail.com> wrote:
> >> It's not like they can actually simulate real-world
> >> conditions and do the amount of comprehensive testing necessary to
> >>work out the bugs.
> >
> >what makes you say that? they have enormous resources to do all sorts
> >of testing.
> >
> >part of that includes employees using pre-release phones all over the
> >san francisco bay area and well beyond it.
> >
> >put it in an iphone 14 body and nobody outside of the people carrying
> >it will know, possibly even them too.
>
>
> Car manufacturers also have enormous resources, but most often a new
> design has many bugs the first couple years.
not as much as apple, who has enough net cash to buy ford, general
> >> It's not like they can actually simulate real-world
> >> conditions and do the amount of comprehensive testing necessary to
> >>work out the bugs.
> >
> >what makes you say that? they have enormous resources to do all sorts
> >of testing.
> >
> >part of that includes employees using pre-release phones all over the
> >san francisco bay area and well beyond it.
> >
> >put it in an iphone 14 body and nobody outside of the people carrying
> >it will know, possibly even them too.
>
>
> Car manufacturers also have enormous resources, but most often a new
> design has many bugs the first couple years.
motors or honda.
also, designing a new car is more complicated than a single component
in said car.
> Something as sensitive to
> environmental conditions as a modem can't be tested in San Francisco
> only. It needs a very wide and long test bed.
it's pretty much guaranteed that the modems, which they've been working
on for many years, are currently in employee iphones and ipads that are
being used over the world right now, possibly in other devices too.
and i think the team who's working on it is based in san diego, not
that it changes much.
Alan
Mar 1, 2023, 12:52:57 PM3/1/23
to
Apple has a very talented silicon design department and has already
shown that it prefers to use its own widgets.
Alan
Mar 1, 2023, 12:54:00 PM3/1/23
to
Alan
Mar 1, 2023, 12:55:38 PM3/1/23
to
On 2023-03-01 09:20, badgolferman wrote:
> nospam wrote:
>
>> In article <xn0nytut...@reader443.eternal-september.org>,
>> badgolferman <REMOVETHISb...@gmail.com> wrote:
>>
>>>
>>> What are the chances Apple's first generation 5G modems will have
>>> lots of problems?
>>
>> nothing is perfect, but given their success with apple silicon, it's
>> low.
>>
>>> It's not like they can actually simulate real-world
>>> conditions and do the amount of comprehensive testing necessary to
>>> work out the bugs.
>>
>> what makes you say that? they have enormous resources to do all sorts
>> of testing.
>>
>> part of that includes employees using pre-release phones all over the
>> san francisco bay area and well beyond it.
>>
>> put it in an iphone 14 body and nobody outside of the people carrying
>> it will know, possibly even them too.
>
>
> Car manufacturers also have enormous resources, but most often a new
> design has many bugs the first couple years.
Give an example of this.
> nospam wrote:
>
>> In article <xn0nytut...@reader443.eternal-september.org>,
>> badgolferman <REMOVETHISb...@gmail.com> wrote:
>>
>>>
>>> What are the chances Apple's first generation 5G modems will have
>>> lots of problems?
>>
>> nothing is perfect, but given their success with apple silicon, it's
>> low.
>>
>>> It's not like they can actually simulate real-world
>>> conditions and do the amount of comprehensive testing necessary to
>>> work out the bugs.
>>
>> what makes you say that? they have enormous resources to do all sorts
>> of testing.
>>
>> part of that includes employees using pre-release phones all over the
>> san francisco bay area and well beyond it.
>>
>> put it in an iphone 14 body and nobody outside of the people carrying
>> it will know, possibly even them too.
>
>
> Car manufacturers also have enormous resources, but most often a new
> design has many bugs the first couple years.
> Something as sensitive to
> environmental conditions as a modem can't be tested in San Francisco
> only. It needs a very wide and long test bed.
Alan
Mar 1, 2023, 12:56:43 PM3/1/23
to
On 2023-03-01 09:20, badgolferman wrote:
What would make Apple less able to do the kind of testing required than
Qualcomm is?
badgolferman
Mar 1, 2023, 1:05:18 PM3/1/23
to
Joerg Lorenz
Mar 1, 2023, 1:12:50 PM3/1/23
to
Am 01.03.23 um 17:58 schrieb badgolferman:
development work.
--
Gutta cavat lapidem (Ovid)
> What are the chances Apple's first generation 5G modems will have lots
> of problems? It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
It is obvious that you do not understand industrial processes and
> of problems? It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
development work.
--
Gutta cavat lapidem (Ovid)
nospam
Mar 1, 2023, 1:21:17 PM3/1/23
to
In article <xn0nytwo...@reader443.eternal-september.org>,
cellphone were circulating, as well as shortly after its introduction.
what does apple know about phones? nokia, motorola, etc. have been
doing it for years. blackberry has the market locked up. steve ballmer
famously laughed, claiming that people would not want to buy a $600
phone without a keyboard.
badgolferman <REMOVETHISb...@gmail.com> wrote:
> >Put another way:
> >
> >What would make Apple less able to do the kind of testing required
> >than Qualcomm is?
>
> Experience.
that's exactly what people said when rumours of apple making a
> >Put another way:
> >
> >What would make Apple less able to do the kind of testing required
> >than Qualcomm is?
>
> Experience.
cellphone were circulating, as well as shortly after its introduction.
what does apple know about phones? nokia, motorola, etc. have been
doing it for years. blackberry has the market locked up. steve ballmer
famously laughed, claiming that people would not want to buy a $600
phone without a keyboard.
Andy Burnelli
Mar 1, 2023, 2:04:29 PM3/1/23
to
nospam wrote:
>> Car manufacturers also have enormous resources, but most often a new
>> design has many bugs the first couple years.
>
> not as much as apple, who has enough net cash to buy ford, general
> motors or honda.
Thank God Apple has so much money. That must be why Apple's first
>> Car manufacturers also have enormous resources, but most often a new
>> design has many bugs the first couple years.
>
> not as much as apple, who has enough net cash to buy ford, general
> motors or honda.
generation Maps app was so much better than the existing Google app.
>> Something as sensitive to
>> environmental conditions as a modem can't be tested in San Francisco
>> only. It needs a very wide and long test bed.
>
> yep, which is why it's tested in more than just san francisco.
wide, because Apple tested out the iPhone power delivery system so well.
> and i think the team who's working on it is based in san diego, not
> that it changes much.
The main thing it "changes" is they can steal from Qualcomm's ranks (IMHO).
Andy Burnelli
Mar 1, 2023, 2:04:47 PM3/1/23
to
badgolferman wrote:
>>What would make Apple less able to do the kind of testing required
>>than Qualcomm is?
>
> Experience.
Hi badgolferman,
>>What would make Apple less able to do the kind of testing required
>>than Qualcomm is?
>
> Experience.
This is an adult assessment of fact...
How many times has Apple tried to source (via Intel perhaps) their own
modem and patents and failed to create a competitive chip (& therefore
throttled the competitive chip so that it wouldn't be so obvious after
all)?
Who came out with the first competitive 5G modem and patents and what
generation are they on down in socal?
Who is their closest competitor and what generation are they on in Korea?
Now, what generation is Apple's 5G modem chip currently on?
--
This has been an adult assessment of fact...
Andy Burnelli
Mar 1, 2023, 2:05:01 PM3/1/23
to
badgolferman wrote:
> What are the chances Apple's first generation 5G modems will have lots
> of problems? It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
> What are the chances Apple's first generation 5G modems will have lots
> of problems? It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
Hi badgolferman,
This is an adult assessment of fact...
Do you know Apple's record on GPUs, badgolferman?
This is an adult assessment of fact...
I do. We discussed this on this newsgroup. HINT: Not good.
Do you know Apple's record on modem chips (albeit via Intel)?
I do. We discussed this on this newsgroup. HINT: Not good.
Do you know Apple's record on integrating modem ICs?
I do. We discussed this on this newsgroup. HINT: Not good.
Do you know Apple's record on unpatchable secure enclave holes?
I do. We discussed this on this newsgroup. HINT: Not good.
Do you know Apple's record on unpatchable M1/M2 holes?
I do. We discussed this on this newsgroup. HINT: Not good.
Do you know Apple's record on unpatchable Ax Bionic holes?
I do. We discussed this on this newsgroup. HINT: Not good.
The question to ask is has Apple _ever_ made a best in class SOC?
HINT: The answer is not good when you strip away the marketing.
Alan
Mar 1, 2023, 2:29:09 PM3/1/23
to
I do. We discussed this on this newsgroup. HINT: Not good...
...for Arlen's credibility.
Alan
Mar 1, 2023, 2:30:00 PM3/1/23
to
On 2023-03-01 11:04, Andy Burnelli wrote:
> nospam wrote:
>
>>> Car manufacturers also have enormous resources, but most often a new
>>> design has many bugs the first couple years.
>>
>> not as much as apple, who has enough net cash to buy ford, general
>> motors or honda.
>
> Thank God Apple has so much money. That must be why Apple's first
> generation Maps app was so much better than the existing Google app.
"first generation" vs "existing".
> nospam wrote:
>
>>> Car manufacturers also have enormous resources, but most often a new
>>> design has many bugs the first couple years.
>>
>> not as much as apple, who has enough net cash to buy ford, general
>> motors or honda.
>
> Thank God Apple has so much money. That must be why Apple's first
> generation Maps app was so much better than the existing Google app.
Do you see the issue?
nospam
Mar 1, 2023, 2:55:39 PM3/1/23
to
In article <tto7jq$1jfgg$1...@paganini.bofh.team>, Andy Burnelli
google refused to continue licensing their maps to apple, thereby
forcing apple to release their own maps when they did. apple did not
have the luxury of waiting until all of the issues were resolved.
*something* had to ship.
<nos...@nospam.net> wrote:
> Thank God Apple has so much money. That must be why Apple's first
> generation Maps app was so much better than the existing Google app.
that was a very different situation.
> Thank God Apple has so much money. That must be why Apple's first
> generation Maps app was so much better than the existing Google app.
google refused to continue licensing their maps to apple, thereby
forcing apple to release their own maps when they did. apple did not
have the luxury of waiting until all of the issues were resolved.
*something* had to ship.
Andy Burnelli
Mar 1, 2023, 3:26:03 PM3/1/23
to
nospam wrote:
> In article <tto7jq$1jfgg$1...@paganini.bofh.team>, Andy Burnelli
> <nos...@nospam.net> wrote:
>
>> Thank God Apple has so much money. That must be why Apple's first
>> generation Maps app was so much better than the existing Google app.
>
> that was a very different situation.
What? Apple didn't have so much money back then, nospam?
> In article <tto7jq$1jfgg$1...@paganini.bofh.team>, Andy Burnelli
> <nos...@nospam.net> wrote:
>
>> Thank God Apple has so much money. That must be why Apple's first
>> generation Maps app was so much better than the existing Google app.
>
> that was a very different situation.
> google refused to continue licensing their maps to apple, thereby
> forcing apple to release their own maps when they did.
short time _after_ the highly touted well marketed typical Apple rollout?
> apple did not
> have the luxury of waiting until all of the issues were resolved.
"predicted" schedules numerous times (according to Kuo anyway)?
> *something* had to ship.
You forget all that highly touted marketing bullshit, exactly like the
"Apple Silicon" bullshit for the unpatchably permanently flawed M1, nospam.
My point, nospam, if you are an adult who can comprehend that point, is
that oodles of money alone does not make a MARKETING company into a DESIGN
house.
Just look at the huge failures in almost every SOC design Apple ever did
(where you can defined "design" any way you want no matter who fabs it).
Case in point:
Why is Apple's iOS the _worst_ of all smartphone OS's nospam, in terms
of zero-day holes that Apple doesn't even bother to look for?
If money was all that was needed, the iPhone wouldn't be so insecure.
--
This has been an attempt of an adult conversation with an iKook who
believes that only money is needed to have design a 5G modem and nothing
else.
nospam
Mar 1, 2023, 3:38:44 PM3/1/23
to
In article <ttoccp$1jvur$1...@paganini.bofh.team>, Andy Burnelli
apple's market cap is currently ~$2.3 trillion, peaking at $3 trillion
last year. in 2011 it was $377b.
that's ~16% of what it is now, or about one sixth the value, and about
one eighth from its peak.
> > google refused to continue licensing their maps to apple, thereby
> > forcing apple to release their own maps when they did.
>
> Didn't Apple fire the leader of the Maps group for incompetence just a very
> short time _after_ the highly touted well marketed typical Apple rollout?
they did.
google's actions were several years prior.
do try to keep up rather than dig yourself an even deeper hole.
> > apple did not
> > have the luxury of waiting until all of the issues were resolved.
>
> What's different with the modem where Apple already slipped their
> "predicted" schedules numerous times (according to Kuo anyway)?
quite a lot is different, and 'predicted' is meaningless because the
schedules are not public. he's guessing.
>
> If psychological counseling was all that was needed, I wouldn't be so insecure.
ftfy
<nos...@nospam.net> wrote:
> >> Thank God Apple has so much money. That must be why Apple's first
> >> generation Maps app was so much better than the existing Google app.
> >
> > that was a very different situation.
>
> What? Apple didn't have so much money back then, nospam?
correct, they did not.
> >> Thank God Apple has so much money. That must be why Apple's first
> >> generation Maps app was so much better than the existing Google app.
> >
> > that was a very different situation.
>
> What? Apple didn't have so much money back then, nospam?
apple's market cap is currently ~$2.3 trillion, peaking at $3 trillion
last year. in 2011 it was $377b.
that's ~16% of what it is now, or about one sixth the value, and about
one eighth from its peak.
> > google refused to continue licensing their maps to apple, thereby
> > forcing apple to release their own maps when they did.
>
> Didn't Apple fire the leader of the Maps group for incompetence just a very
> short time _after_ the highly touted well marketed typical Apple rollout?
google's actions were several years prior.
do try to keep up rather than dig yourself an even deeper hole.
> > apple did not
> > have the luxury of waiting until all of the issues were resolved.
>
> What's different with the modem where Apple already slipped their
> "predicted" schedules numerous times (according to Kuo anyway)?
schedules are not public. he's guessing.
>
> If psychological counseling was all that was needed, I wouldn't be so insecure.
ftfy
Alan
Mar 1, 2023, 3:55:32 PM3/1/23
to
On 2023-03-01 12:26, Andy Burnelli wrote:
> nospam wrote:
>
>> In article <tto7jq$1jfgg$1...@paganini.bofh.team>, Andy Burnelli
>> <nos...@nospam.net> wrote:
>>
>>> Thank God Apple has so much money. That must be why Apple's first
>>> generation Maps app was so much better than the existing Google app.
>>
>> that was a very different situation.
>
> What? Apple didn't have so much money back then, nospam?
How is that even relevant?
> nospam wrote:
>
>> In article <tto7jq$1jfgg$1...@paganini.bofh.team>, Andy Burnelli
>> <nos...@nospam.net> wrote:
>>
>>> Thank God Apple has so much money. That must be why Apple's first
>>> generation Maps app was so much better than the existing Google app.
>>
>> that was a very different situation.
>
> What? Apple didn't have so much money back then, nospam?
Good software takes time.
>
>> google refused to continue licensing their maps to apple, thereby
>> forcing apple to release their own maps when they did.
>
> Didn't Apple fire the leader of the Maps group for incompetence just a very
> short time _after_ the highly touted well marketed typical Apple rollout?
>
>> apple did not
>> have the luxury of waiting until all of the issues were resolved.
>
> What's different with the modem where Apple already slipped their
> "predicted" schedules numerous times (according to Kuo anyway)?
>
>> *something* had to ship.
>
> You forget all that highly touted marketing bullshit, exactly like the
> "Apple Silicon" bullshit for the unpatchably permanently flawed M1, nospam.
Andy Burnelli
Mar 1, 2023, 5:13:50 PM3/1/23
to
nospam wrote:
> do try to keep up rather than dig yourself an even deeper hole.
I will respond with the astute _adult_ point that I am making about your
> do try to keep up rather than dig yourself an even deeper hole.
claim that money alone buys quality for a MARKETING outfit like Apple.
Apple has _never_ made a best-in-class app, nospam, let alone an IC.
Apple Maps is a classic case of where Apple took years to catch up.
Just like CocaCola/Pepsi are MARKETING powerhouses, Apple's ooodles of
money doesn't mean that Apple can design a competitive 5G modem, nospam.
In fact, you can't find a _single_ best-in-class SOC that Apple has ever
had fabricated under their name, nospam, and certainly no modem ICs.
The only thing Apple does well is MARKETING (like CocaCola/Pepsi does).
Apple is the best there is... in marketing what amounts to sugar water.
Alan
Mar 1, 2023, 5:38:37 PM3/1/23
to
On 2023-03-01 14:13, Andy Burnelli wrote:
> nospam wrote:
>
>> do try to keep up rather than dig yourself an even deeper hole.
>
> I will respond with the astute _adult_ point that I am making about your
> claim that money alone buys quality for a MARKETING outfit like Apple.
No one claimed that, so...
> nospam wrote:
>
>> do try to keep up rather than dig yourself an even deeper hole.
>
> I will respond with the astute _adult_ point that I am making about your
> claim that money alone buys quality for a MARKETING outfit like Apple.
>
> Apple has _never_ made a best-in-class app, nospam, let alone an IC.
> Apple Maps is a classic case of where Apple took years to catch up.
>
> Just like CocaCola/Pepsi are MARKETING powerhouses, Apple's ooodles of
> money doesn't mean that Apple can design a competitive 5G modem, nospam.
>
> In fact, you can't find a _single_ best-in-class SOC that Apple has ever
> had fabricated under their name, nospam, and certainly no modem ICs.
sms
Mar 1, 2023, 6:08:03 PM3/1/23
to
that the 15 would not. In that way they could work out any issues with
their modem using an entry level phone where any performance issues
would be less newsworthy. If all goes well then the 16 could use the
Apple modem, but it would be surprising if one year were sufficient to
do enough revisions of the silicon and the microcode to work out all the
issues.
The other issue is that all the other phone manufacturers have already
integrated the modem in with the processor. That will be Apple's next
goal once the standalone modem is working sufficiently well. Perhaps
doing an MCM (Multi-Chip-Module) prior to putting everything on one die.
sms
Mar 1, 2023, 6:13:02 PM3/1/23
to
On 3/1/2023 8:58 AM, badgolferman wrote:
<snip>
> What are the chances Apple's first generation 5G modems will have lots
> of problems? It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
That's why they are predicted to release it first on the SE4. There is
no way that they can do enough worldwide testing, on all of the
different 4G and 5G bands, without the devices being in general
circulation. I also wonder if they are going to drop 2G and 3G
capability from their modem. There are only a few countries that have 3G
but not LTE.
<snip>
> What are the chances Apple's first generation 5G modems will have lots
> of problems? It's not like they can actually simulate real-world
> conditions and do the amount of comprehensive testing necessary to work
> out the bugs.
no way that they can do enough worldwide testing, on all of the
different 4G and 5G bands, without the devices being in general
circulation. I also wonder if they are going to drop 2G and 3G
capability from their modem. There are only a few countries that have 3G
but not LTE.
nospam
Mar 1, 2023, 6:25:23 PM3/1/23
to
In article <ttoimr$1kl4b$1...@paganini.bofh.team>, Andy Burnelli
so much for trying to be an adult. recess is over.
<nos...@nospam.net> wrote:
>
> I will respond with the astute _adult_ point that I am making about your
> claim that money alone buys quality for a MARKETING outfit like Apple.
nobody ever made such a claim.
>
> I will respond with the astute _adult_ point that I am making about your
> claim that money alone buys quality for a MARKETING outfit like Apple.
so much for trying to be an adult. recess is over.
nospam
Mar 1, 2023, 6:25:25 PM3/1/23
to
In article <ttolsi$2pot$5...@dont-email.me>, sms
version will be limited. for the modem to be used across all iphones,
they need ~250 million units per year. that's a lot of modems and that
won't be something that's suddenly available day one.
> The other issue is that all the other phone manufacturers have already
> integrated the modem in with the processor. That will be Apple's next
> goal once the standalone modem is working sufficiently well. Perhaps
> doing an MCM (Multi-Chip-Module) prior to putting everything on one die.
that is not an issue nor is it even a goal. keeping them separate
offers far more flexibility for various products. for example, ipads
*without* cellular. if the modem was integrated, then people would be
forced to pay for something they don't want.
<scharf...@geemail.com> wrote:
> The prediction that I saw was that the SE4 would use an Apple modem but
> that the 15 would not. In that way they could work out any issues with
> their modem using an entry level phone where any performance issues
> would be less newsworthy. If all goes well then the 16 could use the
> Apple modem, but it would be surprising if one year were sufficient to
> do enough revisions of the silicon and the microcode to work out all the
> issues.
a far more likely reason are supply constraints, which for the first
> The prediction that I saw was that the SE4 would use an Apple modem but
> that the 15 would not. In that way they could work out any issues with
> their modem using an entry level phone where any performance issues
> would be less newsworthy. If all goes well then the 16 could use the
> Apple modem, but it would be surprising if one year were sufficient to
> do enough revisions of the silicon and the microcode to work out all the
> issues.
version will be limited. for the modem to be used across all iphones,
they need ~250 million units per year. that's a lot of modems and that
won't be something that's suddenly available day one.
> The other issue is that all the other phone manufacturers have already
> integrated the modem in with the processor. That will be Apple's next
> goal once the standalone modem is working sufficiently well. Perhaps
> doing an MCM (Multi-Chip-Module) prior to putting everything on one die.
offers far more flexibility for various products. for example, ipads
*without* cellular. if the modem was integrated, then people would be
forced to pay for something they don't want.
nospam
Mar 1, 2023, 6:25:26 PM3/1/23
to
In article <ttom5s$2pot$6...@dont-email.me>, sms
> I also wonder if they are going to drop 2G and 3G
> capability from their modem. There are only a few countries that have 3G
> but not LTE.
unlikely.
<scharf...@geemail.com> wrote:
> That's why they are predicted to release it first on the SE4. There is
> no way that they can do enough worldwide testing, on all of the
> different 4G and 5G bands, without the devices being in general
> circulation.
yes they can, just as other companies can.
> That's why they are predicted to release it first on the SE4. There is
> no way that they can do enough worldwide testing, on all of the
> different 4G and 5G bands, without the devices being in general
> circulation.
> I also wonder if they are going to drop 2G and 3G
> capability from their modem. There are only a few countries that have 3G
> but not LTE.
Alan Browne
Mar 2, 2023, 8:29:20 AM3/2/23
to
On 2023-03-01 18:25, nospam wrote:
> lexibility for various products. for example, ipads
> *without* cellular. if the modem was integrated, then people would be
> forced to pay for something they don't want.
If integrated within the Apple processor, it's cheaper to make them all
> lexibility for various products. for example, ipads
> *without* cellular. if the modem was integrated, then people would be
> forced to pay for something they don't want.
the same and simply deactivate the modem portion for models "without"
cellular.
Indeed, could be an "upgrade later" option - just pay for it to be
activated. That could have additional sales appeal as well as increase
the resale value.
--
“Donald Trump and his allies and supporters are a clear and present
danger to American democracy.”
- J Michael Luttig - 2022-06-16
- Former US appellate court judge (R) testifying to the January 6
committee
nospam
Mar 2, 2023, 8:55:19 AM3/2/23
to
In article <Nw1ML.900125$gGD7....@fx11.iad>, Alan Browne
every a* and m* processor. that's just silly. it's possible that they
could make an iphone-specific processor that includes a modem due to
space constraints (not power as has incorrectly been claimed) but will
still want a separate modem chip for other products. whether it's worth
it to do that is something nobody outside of apple can determine.
> Indeed, could be an "upgrade later" option - just pay for it to be
> activated. That could have additional sales appeal as well as increase
> the resale value.
that would cause a huge negative reaction and not how apple works
either.
<bitb...@blackhole.com> wrote:
> > lexibility for various products. for example, ipads
> > *without* cellular. if the modem was integrated, then people would be
> > forced to pay for something they don't want.
>
> If integrated within the Apple processor, it's cheaper to make them all
> the same and simply deactivate the modem portion for models "without"
> cellular.
it limits their flexibility. they're not going to include a modem into
> > lexibility for various products. for example, ipads
> > *without* cellular. if the modem was integrated, then people would be
> > forced to pay for something they don't want.
>
> If integrated within the Apple processor, it's cheaper to make them all
> the same and simply deactivate the modem portion for models "without"
> cellular.
every a* and m* processor. that's just silly. it's possible that they
could make an iphone-specific processor that includes a modem due to
space constraints (not power as has incorrectly been claimed) but will
still want a separate modem chip for other products. whether it's worth
it to do that is something nobody outside of apple can determine.
> Indeed, could be an "upgrade later" option - just pay for it to be
> activated. That could have additional sales appeal as well as increase
> the resale value.
either.
Alan Browne
Mar 2, 2023, 9:05:09 AM3/2/23
to
On 2023-03-02 08:55, nospam wrote:
> In article <Nw1ML.900125$gGD7....@fx11.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>> lexibility for various products. for example, ipads
>>> *without* cellular. if the modem was integrated, then people would be
>>> forced to pay for something they don't want.
>>
>> If integrated within the Apple processor, it's cheaper to make them all
>> the same and simply deactivate the modem portion for models "without"
>> cellular.
>
> it limits their flexibility. they're not going to include a modem into
> every a* and m* processor. that's just silly. it's possible that they
Not at all. The modem as a chip "package" is mainly packaging. As chip
> In article <Nw1ML.900125$gGD7....@fx11.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>> lexibility for various products. for example, ipads
>>> *without* cellular. if the modem was integrated, then people would be
>>> forced to pay for something they don't want.
>>
>> If integrated within the Apple processor, it's cheaper to make them all
>> the same and simply deactivate the modem portion for models "without"
>> cellular.
>
> it limits their flexibility. they're not going to include a modem into
> every a* and m* processor. that's just silly. it's possible that they
area it is pretty small. Thus cheaper to make all Ax (and possibly Mx)
chips with the modem in place and configuration controlled when the
device is made either fused in the chip or by device config.
Chip makers have been doing this for 3 decades (at least) to control the
market price while keeping chip model count down.
> could make an iphone-specific processor that includes a modem due to
> space constraints (not power as has incorrectly been claimed) but will
> still want a separate modem chip for other products. whether it's worth
> it to do that is something nobody outside of apple can determine.
>
>> Indeed, could be an "upgrade later" option - just pay for it to be
>> activated. That could have additional sales appeal as well as increase
>> the resale value.
>
> that would cause a huge negative reaction and not how apple works
> either.
it. Still cheaper than having two versions of the chips. The modem is
a comparatively small amount of silicon once you strip the packaging.
Apple always creates outrage somewhere, so that's not that much of an issue.
nospam
Mar 2, 2023, 9:47:49 AM3/2/23
to
In article <o22ML.1391189$9sn9.1...@fx17.iad>, Alan Browne
there are reasons to integrate it and reasons to not integrate it.
it's clear that keeping them separate offers flexibility that they
otherwise would not have, especially with a diverse product line.
keep in mind that qualcomm makes chips and apple makes products.
they're two very different goals, and what makes sense for one company
does not always make sense for the other.
> >> Indeed, could be an "upgrade later" option - just pay for it to be
> >> activated. That could have additional sales appeal as well as increase
> >> the resale value.
> >
> > that would cause a huge negative reaction and not how apple works
> > either.
>
> It's a decision to make. They could also ship it thus and never mention
> it.
there are companies who x-ray chips (not just the ones apple produces).
something like that won't go undetected for very long, as in a week or
two at the most.
it would be a huge public relations disaster.
even the teardowns happen within hours of the product's release, with
people flying to new zealand to have something to show *before* they're
released in north america.
<bitb...@blackhole.com> wrote:
> >
> >>
> >> If integrated within the Apple processor, it's cheaper to make them all
> >> the same and simply deactivate the modem portion for models "without"
> >> cellular.
> >
> > it limits their flexibility. they're not going to include a modem into
> > every a* and m* processor. that's just silly. it's possible that they
>
> Not at all. The modem as a chip "package" is mainly packaging. As chip
> area it is pretty small. Thus cheaper to make all Ax (and possibly Mx)
> chips with the modem in place and configuration controlled when the
> device is made either fused in the chip or by device config.
taken to extreme, they'd put the entire logic board on one chip.
> >
> >>
> >> If integrated within the Apple processor, it's cheaper to make them all
> >> the same and simply deactivate the modem portion for models "without"
> >> cellular.
> >
> > it limits their flexibility. they're not going to include a modem into
> > every a* and m* processor. that's just silly. it's possible that they
>
> Not at all. The modem as a chip "package" is mainly packaging. As chip
> area it is pretty small. Thus cheaper to make all Ax (and possibly Mx)
> chips with the modem in place and configuration controlled when the
> device is made either fused in the chip or by device config.
there are reasons to integrate it and reasons to not integrate it.
it's clear that keeping them separate offers flexibility that they
otherwise would not have, especially with a diverse product line.
keep in mind that qualcomm makes chips and apple makes products.
they're two very different goals, and what makes sense for one company
does not always make sense for the other.
> >> Indeed, could be an "upgrade later" option - just pay for it to be
> >> activated. That could have additional sales appeal as well as increase
> >> the resale value.
> >
> > that would cause a huge negative reaction and not how apple works
> > either.
>
> It's a decision to make. They could also ship it thus and never mention
> it.
something like that won't go undetected for very long, as in a week or
two at the most.
it would be a huge public relations disaster.
even the teardowns happen within hours of the product's release, with
people flying to new zealand to have something to show *before* they're
released in north america.
Alan Browne
Mar 2, 2023, 10:12:21 AM3/2/23
to
On 2023-03-02 09:47, nospam wrote:
> In article <o22ML.1391189$9sn9.1...@fx17.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>>
>>>>
>>>> If integrated within the Apple processor, it's cheaper to make them all
>>>> the same and simply deactivate the modem portion for models "without"
>>>> cellular.
>>>
>>> it limits their flexibility. they're not going to include a modem into
>>> every a* and m* processor. that's just silly. it's possible that they
>>
>> Not at all. The modem as a chip "package" is mainly packaging. As chip
>> area it is pretty small. Thus cheaper to make all Ax (and possibly Mx)
>> chips with the modem in place and configuration controlled when the
>> device is made either fused in the chip or by device config.
>
> taken to extreme, they'd put the entire logic board on one chip.
That's a dodge. Your notion is that there would be separate Ax chips
> In article <o22ML.1391189$9sn9.1...@fx17.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>>
>>>>
>>>> If integrated within the Apple processor, it's cheaper to make them all
>>>> the same and simply deactivate the modem portion for models "without"
>>>> cellular.
>>>
>>> it limits their flexibility. they're not going to include a modem into
>>> every a* and m* processor. that's just silly. it's possible that they
>>
>> Not at all. The modem as a chip "package" is mainly packaging. As chip
>> area it is pretty small. Thus cheaper to make all Ax (and possibly Mx)
>> chips with the modem in place and configuration controlled when the
>> device is made either fused in the chip or by device config.
>
> taken to extreme, they'd put the entire logic board on one chip.
for devices with or without the modem. Say Ax and Axm.
Has nothing to do with integrating more functions (which will also occur
over time, but that's not the issue here).
And of course, it's not just the modem, but external components from the
edge of the function to the appropriate antenna(s). So leaving out the
antenna may be part of this approach as well. (I don't know offhand if
there is a filter/amplifier/filter stage off of the modem chip as well
but would not be surprised. Further the antenna is likely compounded to
include WiFi, BT, etc. and is in place regardless).
If so, the disabling of the modem portion on the Ax chip would be not
only desirable, but required.
Indeed if there is a financial advantage to it (hardly an if), then that
weighs into the decision as well.
> there are reasons to integrate it and reasons to not integrate it.
that is a primary driver for cost, reliability, manufacturing, etc.
>
> it's clear that keeping them separate offers flexibility that they
> otherwise would not have, especially with a diverse product line.
>
> keep in mind that qualcomm makes chips and apple makes products.
> they're two very different goals, and what makes sense for one company
> does not always make sense for the other.
>>>> Indeed, could be an "upgrade later" option - just pay for it to be
>>>> activated. That could have additional sales appeal as well as increase
>>>> the resale value.
>>>
>>> that would cause a huge negative reaction and not how apple works
>>> either.
>>
>> It's a decision to make. They could also ship it thus and never mention
>> it.
>
> there are companies who x-ray chips (not just the ones apple produces).
>
> something like that won't go undetected for very long, as in a week or
> two at the most.
>
> it would be a huge public relations disaster.
cost while making a broader market offering. Decades of it.
>
> even the teardowns happen within hours of the product's release, with
> people flying to new zealand to have something to show *before* they're
> released in north america.
nospam
Mar 2, 2023, 11:02:50 AM3/2/23
to
In article <n13ML.1734842$GNG9.1...@fx18.iad>, Alan Browne
business. what apple does is make chips for their own use in a variety
of products. two very different goals.
qualcomm, intel, amd, etc. are in the chip-making business and they're
not about to make custom chips for each device maker that wants to buy
their chips. what they do is add in every feature knowing someone might
end up using it. that's wasteful.
apple makes chips for their own needs and can include only what's
needed. also, the hardware team works with the software team to
implement certain features in hardware (which android devices cannot
possibly do). thus, it's cost-effective to separate the modem from the
main processor and only include the modem in products that use it.
if part of the apple-qualcomm settlement agreement requires apple to
continue paying the qualcomm ransom (because of the patents the apple
modem uses) for any device with a modem, then they *can't* integrate it
for products that don't need it.
> > keep in mind that qualcomm makes chips and apple makes products.
> > they're two very different goals, and what makes sense for one company
> > does not always make sense for the other.
>
> Has nothing to do with this particular design choice.
has everything to do with it. see above.
> >>>> Indeed, could be an "upgrade later" option - just pay for it to be
> >>>> activated. That could have additional sales appeal as well as increase
> >>>> the resale value.
> >>>
> >>> that would cause a huge negative reaction and not how apple works
> >>> either.
> >>
> >> It's a decision to make. They could also ship it thus and never mention
> >> it.
> >
> > there are companies who x-ray chips (not just the ones apple produces).
> >
> > something like that won't go undetected for very long, as in a week or
> > two at the most.
> >
> > it would be a huge public relations disaster.
>
> Not really. It's just another thing chip makers do to reduce overall
> cost while making a broader market offering. Decades of it.
yes really. the usual suspects will have a field day with that.
<bitb...@blackhole.com> wrote:
> > it's clear that keeping them separate offers flexibility that they
> > otherwise would not have, especially with a diverse product line.
>
> Not at all - deselecting "parts" internally is a common practice.
for companies that sell chips, sure. apple isn't in the chip-making
> > it's clear that keeping them separate offers flexibility that they
> > otherwise would not have, especially with a diverse product line.
>
> Not at all - deselecting "parts" internally is a common practice.
business. what apple does is make chips for their own use in a variety
of products. two very different goals.
qualcomm, intel, amd, etc. are in the chip-making business and they're
not about to make custom chips for each device maker that wants to buy
their chips. what they do is add in every feature knowing someone might
end up using it. that's wasteful.
apple makes chips for their own needs and can include only what's
needed. also, the hardware team works with the software team to
implement certain features in hardware (which android devices cannot
possibly do). thus, it's cost-effective to separate the modem from the
main processor and only include the modem in products that use it.
if part of the apple-qualcomm settlement agreement requires apple to
continue paying the qualcomm ransom (because of the patents the apple
modem uses) for any device with a modem, then they *can't* integrate it
for products that don't need it.
> > keep in mind that qualcomm makes chips and apple makes products.
> > they're two very different goals, and what makes sense for one company
> > does not always make sense for the other.
>
> Has nothing to do with this particular design choice.
> >>>> Indeed, could be an "upgrade later" option - just pay for it to be
> >>>> activated. That could have additional sales appeal as well as increase
> >>>> the resale value.
> >>>
> >>> that would cause a huge negative reaction and not how apple works
> >>> either.
> >>
> >> It's a decision to make. They could also ship it thus and never mention
> >> it.
> >
> > there are companies who x-ray chips (not just the ones apple produces).
> >
> > something like that won't go undetected for very long, as in a week or
> > two at the most.
> >
> > it would be a huge public relations disaster.
>
> Not really. It's just another thing chip makers do to reduce overall
> cost while making a broader market offering. Decades of it.
Andy Burnelli
Mar 2, 2023, 12:34:48 PM3/2/23
to
nospam wrote:
> that is not an issue nor is it even a goal. keeping them separate
> offers far more flexibility for various products. for example, ipads
> *without* cellular. if the modem was integrated, then people would be
> forced to pay for something they don't want.
I'm going to treat this convo as if you're an actual adult, nospam. OK?
> that is not an issue nor is it even a goal. keeping them separate
> offers far more flexibility for various products. for example, ipads
> *without* cellular. if the modem was integrated, then people would be
> forced to pay for something they don't want.
Apple can't design a best-in-class SOC and the proof is they never have.
The advantage to integration is well known and if Apple wanted to not have
a modem in any particular device they could disable the modem on the board.
The real reason Apple hasn't yet integrated the modem is that they can't.
Think about it, nospam, as an actual _adult_ would think about it.
*Apple can't even get the modem itself out the door*
*How the heck is Apple going to _also_ integrate it into an SOC*
Even when Apple gets the modem out the door on the el cheapo iPhone,
it still will have huge problems (just like _all_ Apple's chips do).
Must I run down the list of unpatchable flaws again for you nospam?
--
This has been an adult observation using an assessment of known facts.
Alan
Mar 2, 2023, 12:41:51 PM3/2/23
to
On 2023-03-02 09:34, Andy Burnelli wrote:
> nospam wrote:
>
>> that is not an issue nor is it even a goal. keeping them separate
>> offers far more flexibility for various products. for example, ipads
>> *without* cellular. if the modem was integrated, then people would be
>> forced to pay for something they don't want.
>
> I'm going to treat this convo as if you're an actual adult, nospam. OK?
> Apple can't design a best-in-class SOC and the proof is they never have.
The Apple Silicon CPUs are considered among the very best CPUs
> nospam wrote:
>
>> that is not an issue nor is it even a goal. keeping them separate
>> offers far more flexibility for various products. for example, ipads
>> *without* cellular. if the modem was integrated, then people would be
>> forced to pay for something they don't want.
>
> I'm going to treat this convo as if you're an actual adult, nospam. OK?
> Apple can't design a best-in-class SOC and the proof is they never have.
>
> The advantage to integration is well known and if Apple wanted to not have
> a modem in any particular device they could disable the modem on the board.
>
> The real reason Apple hasn't yet integrated the modem is that they can't.
>
> Think about it, nospam, as an actual _adult_ would think about it.
> *Apple can't even get the modem itself out the door*
> *How the heck is Apple going to _also_ integrate it into an SOC*
>
> Even when Apple gets the modem out the door on the el cheapo iPhone, it
> still will have huge problems (just like _all_ Apple's chips do).
>
> Must I run down the list of unpatchable flaws again for you nospam?
And show that they only exist in APPLE'S CPUs.
Andy Burnelli
Mar 2, 2023, 12:45:49 PM3/2/23
to
nospam wrote:
> there are reasons to integrate it and reasons to not integrate it.
This will be an adult observation of assessment of well-known facts.
> there are reasons to integrate it and reasons to not integrate it.
The main reason, IMHO, which nospam is skirting around, is that Apple can't
even get the modem out the door yet, let alone integrate it in an SOC.
I predicted Apple will never get the modem out until the Qualcomm patents
expire but I also predicted Kiev would fall within weeks so let's wait.
*Apple's inability to develop 5G smartphone modem is due to two Qualcomm patents*
<https://www.phonearena.com/news/qualcomm-patent-puts-wrench-into-apples-plan-to-build-5g-modem_id141078 >
But if Apple ever does get a modem out, it's not hard to predict, based on
Apple's own record, the first generation will be as bad as the first
generation of Apple Maps was compared to the Google Maps at that time.
What generation is Qualcomm, Samsung, MediaTek & Huawei on already?
Given Apple's highly touted Bionic, Secure Enclave and M1/M2 chips all
contain unpatchable holes, Apple's history isn't in their favor on modems.
--
This has been an adult assessment of well known facts about Apple chips.
Alan Browne
Mar 2, 2023, 2:36:19 PM3/2/23
to
On 2023-03-02 11:02, nospam wrote:
> In article <n13ML.1734842$GNG9.1...@fx18.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>> it's clear that keeping them separate offers flexibility that they
>>> otherwise would not have, especially with a diverse product line.
>>
>> Not at all - deselecting "parts" internally is a common practice.
>
> for companies that sell chips, sure. apple isn't in the chip-making
> business. what apple does is make chips for their own use in a variety
> of products. two very different goals.
BS reply. As you're all so quick to point out, Apple design their
> In article <n13ML.1734842$GNG9.1...@fx18.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>> it's clear that keeping them separate offers flexibility that they
>>> otherwise would not have, especially with a diverse product line.
>>
>> Not at all - deselecting "parts" internally is a common practice.
>
> for companies that sell chips, sure. apple isn't in the chip-making
> business. what apple does is make chips for their own use in a variety
> of products. two very different goals.
silicon to their needs. They sub the fabrication.
Other than limitations of manufacturing processes that the fabs impose,
Apple will do everything they can and want in the design of the chips
and the fabs will make it to that design if it is feasible.
> qualcomm, intel, amd, etc. are in the chip-making business and they're
> not about to make custom chips for each device maker that wants to buy
> their chips. what they do is add in every feature knowing someone might
> end up using it. that's wasteful.
Period. What options Apple puts in there is Apple's business as long as
the fab can do it.
> apple makes chips for their own needs and can include only what's
> needed. also, the hardware team works with the software team to
modem integrated, _and_ the option to disable that modem (by fuse,
firmware or s/w, that is Apple's decision to make.
> implement certain features in hardware (which android devices cannot
> possibly do). thus, it's cost-effective to separate the modem from the
> main processor and only include the modem in products that use it.
always has the modem, and disable it when not sold as a feature in a
given model.
>
> if part of the apple-qualcomm settlement agreement requires apple to
> continue paying the qualcomm ransom (because of the patents the apple
> modem uses) for any device with a modem, then they *can't* integrate it
> for products that don't need it.
based modem design, not Qualcomm's design.
>
>>> keep in mind that qualcomm makes chips and apple makes products.
>>> they're two very different goals, and what makes sense for one company
>>> does not always make sense for the other.
>>
>> Has nothing to do with this particular design choice.
>
> has everything to do with it. see above.
choice and that is all that matters.
>>>>>> Indeed, could be an "upgrade later" option - just pay for it to be
>>>>>> activated. That could have additional sales appeal as well as increase
>>>>>> the resale value.
>>>>>
>>>>> that would cause a huge negative reaction and not how apple works
>>>>> either.
>>>>
>>>> It's a decision to make. They could also ship it thus and never mention
>>>> it.
>>>
>>> there are companies who x-ray chips (not just the ones apple produces).
>>>
>>> something like that won't go undetected for very long, as in a week or
>>> two at the most.
>>>
>>> it would be a huge public relations disaster.
>>
>> Not really. It's just another thing chip makers do to reduce overall
>> cost while making a broader market offering. Decades of it.
>
> yes really. the usual suspects will have a field day with that.
nospam
Mar 2, 2023, 4:22:16 PM3/2/23
to
In article <ttqncb$214fs$1...@paganini.bofh.team>, Andy Burnelli
that's because it's too big to fit through ordinary doors.
once they are able to use a smaller process, it should be ok.
that's because it's too big to fit through ordinary doors.
once they are able to use a smaller process, it should be ok.
nospam
Mar 2, 2023, 4:22:20 PM3/2/23
to
In article <RU6ML.296046$5CY7....@fx46.iad>, Alan Browne
> As you're all so quick to point out, Apple design their
> silicon to their needs. They sub the fabrication.
yep, which is why apple can make processors without a modem for the
devices that don't have one, such as macbooks and ipads.
they might choose to integrate a modem at some point, but as i have
said before, the advantages in doing so don't help apple.
> > implement certain features in hardware (which android devices cannot
> > possibly do). thus, it's cost-effective to separate the modem from the
> > main processor and only include the modem in products that use it.
>
> Yes, that is _one_ option. The other is to reduce to one part that
> always has the modem, and disable it when not sold as a feature in a
> given model.
that's another option. apple will decide which path to take.
the evidence does not support integrated modems at this time. maybe one
day, maybe not at all.
> > if part of the apple-qualcomm settlement agreement requires apple to
> > continue paying the qualcomm ransom (because of the patents the apple
> > modem uses) for any device with a modem, then they *can't* integrate it
> > for products that don't need it.
>
> That's a negotiable item.
negotiable only works when both parties want to negotiate.
qualcomm did not, which is part of what led to a major lawsuit that was
ultimately settled just as it went to trial.
> IAC, I take it they will use Apple's intel
> based modem design, not Qualcomm's design.
apple bought that from intel, so that's a given, however, it still uses
qualcomm's patents.
qualcomm wants to be paid twice, once for the patent and again for
using the chip in a product based on the price of said product.
thus the predatory 'fee' for an apple watch is much less than for an
iphone 14 pro max or ipad pro.
<bitb...@blackhole.com> wrote:
> >>> it's clear that keeping them separate offers flexibility that they
> >>> otherwise would not have, especially with a diverse product line.
> >>
> >> Not at all - deselecting "parts" internally is a common practice.
> >
> > for companies that sell chips, sure. apple isn't in the chip-making
> > business. what apple does is make chips for their own use in a variety
> > of products. two very different goals.
>
> BS reply.
nothing bs about it.
> >>> it's clear that keeping them separate offers flexibility that they
> >>> otherwise would not have, especially with a diverse product line.
> >>
> >> Not at all - deselecting "parts" internally is a common practice.
> >
> > for companies that sell chips, sure. apple isn't in the chip-making
> > business. what apple does is make chips for their own use in a variety
> > of products. two very different goals.
>
> BS reply.
> As you're all so quick to point out, Apple design their
> silicon to their needs. They sub the fabrication.
devices that don't have one, such as macbooks and ipads.
they might choose to integrate a modem at some point, but as i have
said before, the advantages in doing so don't help apple.
> > implement certain features in hardware (which android devices cannot
> > possibly do). thus, it's cost-effective to separate the modem from the
> > main processor and only include the modem in products that use it.
>
> Yes, that is _one_ option. The other is to reduce to one part that
> always has the modem, and disable it when not sold as a feature in a
> given model.
the evidence does not support integrated modems at this time. maybe one
day, maybe not at all.
> > if part of the apple-qualcomm settlement agreement requires apple to
> > continue paying the qualcomm ransom (because of the patents the apple
> > modem uses) for any device with a modem, then they *can't* integrate it
> > for products that don't need it.
>
> That's a negotiable item.
qualcomm did not, which is part of what led to a major lawsuit that was
ultimately settled just as it went to trial.
> IAC, I take it they will use Apple's intel
> based modem design, not Qualcomm's design.
qualcomm's patents.
qualcomm wants to be paid twice, once for the patent and again for
using the chip in a product based on the price of said product.
thus the predatory 'fee' for an apple watch is much less than for an
iphone 14 pro max or ipad pro.
Alan Browne
Mar 2, 2023, 5:37:56 PM3/2/23
to
On 2023-03-02 16:22, nospam wrote:
> In article <RU6ML.296046$5CY7....@fx46.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>>>> it's clear that keeping them separate offers flexibility that they
>>>>> otherwise would not have, especially with a diverse product line.
>>>>
>>>> Not at all - deselecting "parts" internally is a common practice.
>>>
>>> for companies that sell chips, sure. apple isn't in the chip-making
>>> business. what apple does is make chips for their own use in a variety
>>> of products. two very different goals.
>>
>> BS reply.
>
> nothing bs about it.
>
>> As you're all so quick to point out, Apple design their
>> silicon to their needs. They sub the fabrication.
>
> yep, which is why apple can make processors without a modem for the
> devices that don't have one, such as macbooks and ipads.
Of course. But we're talking about iPhones. So don't wander off of the
> In article <RU6ML.296046$5CY7....@fx46.iad>, Alan Browne
> <bitb...@blackhole.com> wrote:
>
>>>>> it's clear that keeping them separate offers flexibility that they
>>>>> otherwise would not have, especially with a diverse product line.
>>>>
>>>> Not at all - deselecting "parts" internally is a common practice.
>>>
>>> for companies that sell chips, sure. apple isn't in the chip-making
>>> business. what apple does is make chips for their own use in a variety
>>> of products. two very different goals.
>>
>> BS reply.
>
> nothing bs about it.
>
>> As you're all so quick to point out, Apple design their
>> silicon to their needs. They sub the fabrication.
>
> yep, which is why apple can make processors without a modem for the
> devices that don't have one, such as macbooks and ipads.
ranch as you're prone to do.
>
> they might choose to integrate a modem at some point, but as i have
> said before, the advantages in doing so don't help apple.
packaging overall.
>>> implement certain features in hardware (which android devices cannot
>>> possibly do). thus, it's cost-effective to separate the modem from the
>>> main processor and only include the modem in products that use it.
>>
>> Yes, that is _one_ option. The other is to reduce to one part that
>> always has the modem, and disable it when not sold as a feature in a
>> given model.
>
> that's another option. apple will decide which path to take.
>
> the evidence does not support integrated modems at this time. maybe one
> day, maybe not at all.
least, in that there is greatly reduced memory BW needed. As 5G and
future designs point to ever increasing BW, having the function on chip
just makes more and more sense.
I would expect the Ax chips to go towards the unified memory scheme of
the Mx line at some point complementing the above.
>
>>> if part of the apple-qualcomm settlement agreement requires apple to
>>> continue paying the qualcomm ransom (because of the patents the apple
>>> modem uses) for any device with a modem, then they *can't* integrate it
>>> for products that don't need it.
>>
>> That's a negotiable item.
>
> negotiable only works when both parties want to negotiate.
The must negotiate, do so fairly and in a non-discriminatory manner.
> qualcomm did not, which is part of what led to a major lawsuit that was
> ultimately settled just as it went to trial.
>
>> IAC, I take it they will use Apple's intel
>> based modem design, not Qualcomm's design.
>
> apple bought that from intel, so that's a given, however, it still uses
> qualcomm's patents.
>
> qualcomm wants to be paid twice, once for the patent and again for
> using the chip in a product based on the price of said product.
they "want" and what they are allowed are two different things. Apple
and Qualcomm settled at some "middle ground", but I'd see Apple going
for the final stroke in the courts.
Apple can make the case that deliberately unused h/w is not subject to
the patent as well.
nospam
Mar 2, 2023, 6:54:39 PM3/2/23
to
In article <6z9ML.896078$MVg8....@fx12.iad>, Alan Browne
combining them is not an advantage.
combining them when there's a tangible benefit is a good idea, and if
so, then apple will likely do it (assuming it's not outweighed by the
disadvantages).
so far, nobody has presented compelling evidence that it's a good idea
other than 'but that's what other companies do'.
one claim was to reduce power consumption, except that isn't an issue
because apple's chips are already more power efficient than qualcomm,
(and with a 3nm process in the next revision, even more so).
on the other hand, apple having a diverse product line that shares
chips is reason to keep them separate.
on the other other hand, the apple watch is *highly* space constrained,
far more than an iphone, so combining them for the s* series might make
sense for the apple watch. then again, the s* series is also used in
the homepod, so perhaps not.
it's nowhere near as clear cut as you and others claim. there are a
number of factors involved.
>
> >
> >>> if part of the apple-qualcomm settlement agreement requires apple to
> >>> continue paying the qualcomm ransom (because of the patents the apple
> >>> modem uses) for any device with a modem, then they *can't* integrate it
> >>> for products that don't need it.
> >>
> >> That's a negotiable item.
> >
> > negotiable only works when both parties want to negotiate.
>
> Qualcomm is bound by FRAND because they are members of the TIA and ATIS.
> The must negotiate, do so fairly and in a non-discriminatory manner.
tell that to qualcomm's lawyers. perhaps you'll have better luck than
apple did (although they both settled so it's moot).
> >
> > qualcomm wants to be paid twice, once for the patent and again for
> > using the chip in a product based on the price of said product.
>
> The price of the product it's used in goes against FRAND (above).
i agree, but that's what qualcomm does and one reason why apple sued
them.
this is one key reason why macbooks do not have cellular, which can be
spec'ed out to $5000+ for the top of the line, requiring a fee that's
more than 10 times higher than an entry level iphone that has a modem.
> What
> they "want" and what they are allowed are two different things. Apple
> and Qualcomm settled at some "middle ground", but I'd see Apple going
> for the final stroke in the courts.
they settled, so that final stroke is not going to happen.
> Apple can make the case that deliberately unused h/w is not subject to
> the patent as well.
they can, and qualcomm can argue that apple built it into the chip.
not that it matters anymore because they've both settled and agreed to
terms, which i don't think are public.
anyway, what i said was that *if* there is still a qualcomm ransom
(which as far as i know is not public), apple has a very strong
financial incentive to *not* integrate it, which outweighs any possible
benefit for parts count, power or anything else.
<bitb...@blackhole.com> wrote:
> >
> > they might choose to integrate a modem at some point, but as i have
> > said before, the advantages in doing so don't help apple.
>
> Minimizing part count is always an advantage esp. when it reduces
> packaging overall.
generally true, however, combining two parts just for the sake of
> >
> > they might choose to integrate a modem at some point, but as i have
> > said before, the advantages in doing so don't help apple.
>
> Minimizing part count is always an advantage esp. when it reduces
> packaging overall.
combining them is not an advantage.
combining them when there's a tangible benefit is a good idea, and if
so, then apple will likely do it (assuming it's not outweighed by the
disadvantages).
so far, nobody has presented compelling evidence that it's a good idea
other than 'but that's what other companies do'.
one claim was to reduce power consumption, except that isn't an issue
because apple's chips are already more power efficient than qualcomm,
(and with a 3nm process in the next revision, even more so).
on the other hand, apple having a diverse product line that shares
chips is reason to keep them separate.
on the other other hand, the apple watch is *highly* space constrained,
far more than an iphone, so combining them for the s* series might make
sense for the apple watch. then again, the s* series is also used in
the homepod, so perhaps not.
it's nowhere near as clear cut as you and others claim. there are a
number of factors involved.
>
> >
> >>> if part of the apple-qualcomm settlement agreement requires apple to
> >>> continue paying the qualcomm ransom (because of the patents the apple
> >>> modem uses) for any device with a modem, then they *can't* integrate it
> >>> for products that don't need it.
> >>
> >> That's a negotiable item.
> >
> > negotiable only works when both parties want to negotiate.
>
> Qualcomm is bound by FRAND because they are members of the TIA and ATIS.
> The must negotiate, do so fairly and in a non-discriminatory manner.
apple did (although they both settled so it's moot).
> >
> > qualcomm wants to be paid twice, once for the patent and again for
> > using the chip in a product based on the price of said product.
>
> The price of the product it's used in goes against FRAND (above).
them.
this is one key reason why macbooks do not have cellular, which can be
spec'ed out to $5000+ for the top of the line, requiring a fee that's
more than 10 times higher than an entry level iphone that has a modem.
> What
> they "want" and what they are allowed are two different things. Apple
> and Qualcomm settled at some "middle ground", but I'd see Apple going
> for the final stroke in the courts.
> Apple can make the case that deliberately unused h/w is not subject to
> the patent as well.
not that it matters anymore because they've both settled and agreed to
terms, which i don't think are public.
anyway, what i said was that *if* there is still a qualcomm ransom
(which as far as i know is not public), apple has a very strong
financial incentive to *not* integrate it, which outweighs any possible
benefit for parts count, power or anything else.
Andy Burnelli
Mar 2, 2023, 7:19:46 PM3/2/23
to
nospam wrote:
> one claim was to reduce power consumption, except that isn't an issue
> because apple's chips are already more power efficient than qualcomm,
Every one of your claims doesn't hold water to the facts, nospam.
> one claim was to reduce power consumption, except that isn't an issue
> because apple's chips are already more power efficient than qualcomm,
If Apple designs are so "power efficient", why did they have to be
(secretly) throttled due to their atrocious power delivery design?
0 new messages