On Fri, 3 Jan 2020 20:02:20 -0500, JF Mezei wrote:
> You accused Apple of failing in modems. It didn't try to make modems.
> When relations with Qualcomm soured, Apple turned to Intel who decided
> to try to make modems.
Hi J.F. Mezei,
It's rare to find an adult on this newsgroup, which I appreciate, since a
completely different conversation can ensue where we learn from each other.
I agreed with you, and I already conceded that Apple doesn't (yet) make
smartphone modems (although nospam's JPEGs were hilarious):
Where we all used the full-duplex & half-duplex put-the-handset-in-the-grip
modems way back in the mid 1970's and early 1980's.
I explained that modems aren't the main topic; they're just an example of
the main point, which is that Apple MARKETS far more than Engineering can
deliver.
a. Apple MARKETING claims best-in-class solutions, and yet,
b. Apple Engineering can't deliver them (i.e., poor "design choices").
Yet, if we discuss modems, we must state the fact that Apple spent
_billions_ last year alone to buy out Intel's modem business & to start
building a modem-design facility next door to Qualcomm; and, you must be
aware that the Apple-selected Intel modems were inferior to Qualcomm
modems, such that Apple had to _throttle_ the Qualcomm modems in the
following iPhones, as Steve Scharf explained:
My point on modems is simply the fact that Apple has had little success in
almost any best-in-class engineering endeavor; so what makes adults think
that Apple can be successful with GPUs?
The fact is, as I see those facts, if you remove the hugely influential
MARKETING claims Apple is so great at, we find almost zero best-in-class
solutions coming out of Apple.
*That statement shocks people who own purely imaginary belief systems.*
o But it does not shock people who comprehend facts.
BTW, I don't actually blame engineering for Apple's lack of best-in-class
solutions; Federighi just can't deliver what MARKETING promises since what
MARKETING promises is a tight schedule of purely imaginary functionality.
It's a lot easier to advertise modem performance than it is to deliver it.
o That's why Apple surrendered to Qualcomm on 5G modems, for example.
> Apple Maps was a combination of problems (political, management and
> legacy of Steve Jobs) as well as grossly underestimating the challenge
> for Apple to build a mapping service from scratch.
I don't disagree with any logical assessment of the facts, as you're well
aware, J.F. Mezei; hence I don't disagree with your assessment of what
happened with Apple Maps.
Admittedly, Google is as good at Maps as Apple is at marketing; where it's
amazing to me that Google kicked Garmin's ass, and Garmin had the nav
market sewn up prior to the advent of Google.
I don't blame Apple for losing to Google on Maps, since _everyone_ loses to
Google on maps. For example, as you're likely aware, I remove everything I
can on my phone that is related to Google (even the Google Account), and
yet, I still am forced (at times), to use Google Maps on my new $100 Moto
G7, which luckily has plenty of storage for programs (64GB + 512GB sd):
<
https://i.postimg.cc/tgn4z5L4/googlemaps.jpg>
> For Montréal, the satelite imagery is from 2010, giving great
> historical/museum visuals of many sites that have radically changed
> since. So Apple bought imagery that is older than Apple Maps.
Thank you for that factual data and logical assessment of those facts.
o I don't disagree with tenable logical assessments based on facts.
The good news is that Apple MARKETING claims they'll have best-in-class
quality Maps soon, and I actually am friends with coders on their map team
in Cupertino, who "say" Apple _will_ catch up; where I'm a fan of
competition, particularly when a company spends millions to give me a free
product.
Hence, I _hope_ Apple creates a best-in-class Map product, and, in fact, I
hope Apple creates lots of best-in-class products - because that means the
competition will give me better free products overall.
But my main point of bringing up Maps was simply as yet another example
where it's easier to loudly claim best in class than it is to deliver it.
> Your criticism does not apply to chips because Apple has done extremely
> well with chips. And that is a very difficult business. Apple's
> bragging about its Ax chips is very humble compared to the amount of
> heroic work that is needed to produce new chips every year, with new
> logic, not just speed boosts.
I'm not sure if you made your case on chips, where I don't claim to be an
expert in chip design, but, I do know that Apple has had to secretly
throttle millions of CPUs due to nothing but poor "design choices".
The case of the modems was a "design choice" by Apple (to throttle Qualcomm
modems in Apple iPhones).
Likewise, Apple has a well known penchant for the "design choice" of paltry
RAM, and few CPUs, where, for example, my $100 Android 8-core Moto G7 has
4GB of RAM as a matter of habit (and 64GB of storage, plus expansion of
512GB), all of which are decent "design choices" for a $100 phone.
Unfortunately, the thread on the topic of the closest 'design choice' from
Apple didn't come to any useful conclusion on which iPhone is closest:
o *What is the closest Apple iPhone comparison to the $100 64GB 4GB RAM Motorola G7?*
<
https://groups.google.com/d/msg/misc.phone.mobile.iphone/0hqJt3SOcAI/NlqW_3WyCAAJ>
> Samsung bragged about its phones until they started to catch fire and
> their phones were banned from aircraft. Apple is very conservative when
> it comes to batteries, anf on the Xs, the lightning port shuts down when
> it is cold now (very annoying when listening to music) to disable
> charging when battery is cold which can damage it.
You have facts about cold weather, which I appreciate, since Apple seems to
have only tested the iPhones in Cupertino, which, let's face it, has one of
the nicest climates on this planet.
I understand why you "think" Apple is "very conservative" when it comes to
batteries, where I'd say it's more obvious that Apple is just as stingy
with battery capacity as Apple is stingy with RAM & CPU cores.
You assess the facts as "conservative"; while I assess the facts as
'stingy'.
o Stingy batteries
o Stingy RAM
o Stingy CPU cores
etc.
In fact, the "design choice" of stingy batteries were one of the main
reasons Apple felt the intense need to secretly, drastically, and
permanently throttle CPU speeds in iPhones after only about a year of use.
I wouldn't call secretly, permanently, and drastically throttling CPUs to
about half speed in about a year of use as a "conservative" approach to the
solution, would you?
I'd assess CPU throttling facts as an extremely drastic solution by Apple.
But I do agree with you that Apple doesn't seem to have tested their
iPhones in a refrigerator up in Cupertino (based on the evidence so far).
> We'll never know how much engineers had warned Apple about iPhone 6s
> batteries being underpowered for the CPU after a year.
Again, I have to agree with you J.F. Mezei, since I know plenty of
engineers at Apple (many of my neighbors work for either Apple or Google
where I discuss this with them all the time).
Your assessment of the facts is reasonable.
I suspect _lots_ of engineers worried about the "design choices" Apple
MARKETING foisted upon Engineering. As I've said all along, Apple's claim
to fame is their MARKETING.
The problem is that it's a lot easier to market superior performance than
it is to deliver it, even at the astronomical prices of iPhones today.
> Apple reacted by
> ensuring components stop drawing too much power on older batteries. The
> throttling can now be disabled if you want, and as long as you have an
> Apple installed battery, you get battery health information. (not
> providing it for other batteries is bad tough).
Yes. But. The problem remains on those phones. Forever.
It was a poor "design choice", which only the apologists can try to deny.
>> I'm not in the least surprised that Apple doesn't make the best GPU,
> Where is this claim of yours coming from?
My main claim is that Apple does two things, consistently:
a. Apple MARKETING claims best-in-class solutions, and yet,
b. Apple Engineering can't deliver them (i.e., poor "design choices").
The proof is what we've been discussing (e.g., modems, Maps, GPUs, Power
Delivery, etc.).