prismatics

I believe apple might consider switching to AMD processors (e.g. ZEN 2); or maybe an unannounced APU with integrated VEGA graphics, that would sound interesting from an Apple perspective. Getting the operating system scheduler in place takes its time since the desktop ryzen chips (or ryzen pro if you want, they are the same) are effectively 2 NUMA nodes on a single die; Other than that I believe there is no reason to keep us waiting.

Well the temperature ramp shows exceedingly clearly that there is a problem of thermal transfer between the heatsink and the die, considering that the Mac Mini heatsink has considerably higher thermal capacity than the insult of a cooling solution in the MacBook Pro

sergioz said:

I don’t think 5G will be super popular with cellphones in the beginning. As we develop new experiences and tech, demand will rise, but right now I don’t see why you would need access to 1 gigabit connection in you pocket? Plus as tricky 5G as a technology, even when it becomes widely available, it’ll be like the icing on the cake to have it.

The thing is, and what will make you understand why this matters is, that the gigabit is not for every individual device, but multiples of it (e.g. 4 or 8) are shared between many, many more subscribers. As data plans become increasingly more aware of increased media consumption (e.g. basic stuff as browsing, watching videos or streaming music), so do providers noticing that the higher data allowances result in increasing traffic in the existing mobile infrastructure.

The headroom that 4G technologies gave to 3G is exhausted, mobile broadband networks now face increasing cell congestion where there was plenty of capacity 3 or 4 years ago.

Continuing in existing cell topologies is no longer feasable, as in bigger cells, the ratio of available frequency spectrum (conclusively the available data rate) against subscriber count is rapidly going against unmaintainability of services in a useful manner and we are approaching a barrier which defines the maximum information density one can achieve with given parameters of a cell in traditional 4G.

5G is taking a different approach. Rather than creating big cells that have many clients share the same frequency spectrum like 4G and older technologies did, the concept of 5G is to create much, much smaller cells which have a much smaller range.

The obvious disadvantage is, that it's necessary to create more cells and it is much harder to coordinate this big amount of cells. Clients need to switch cells much more often and always maintain multiple connections to multiple cells nearby, which is its own challenge when it comes to energy consumption and the real-time requirements for the 5G backbone as all cells must cooperate.

But, as you might have understood from the explanation, the advantage of this approach is, that every cell can use the frequency spectrum it uses to transmit data in its much reduced range much more effectively, since less subscribers are active in each cell, providing much lower latency, much, much higher data rates as well as better quality of service to each customer.

The thing is, and many will be upset about this is, that the core benefits of 5G technology only apply to use-cases (or environments) where high cell congestion is an issue (e.g. airports, universities, cities, indoor areas). In areas outside of high population people will not see much change.

I hope I could address your concerns accurately.

polymnia said:

rcfa said:

The one thing that is missing: ECC RAM, otherwise, particularly in combination with an eGPU, this is a killer machine.

MacPro feature. 

Even the cheapest of the cheapest AMD processors (which are both faster and use less power than intel processors in their respective use-cases they have been designed for, while being offered at a lower price) support ECC.

MacPro feature my ass.

danielchow said:

This was probably why Apple is discontinuing their current hardware because it’d be a waste of resources to keep manufacturing the Airports when a new hardware standard is going to come out (?)

WPA3 can be easily implemented in todays WPA2 compatible systems since it is based on the same hardware acceleration for the cryptography happening under the hood.