That’s a bit of an extreme position.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Upcoming 13-Inch MacBook Pro Models to Use Intel's 10th-Generation Ice Lake Chips
- Thread starter MacRumors
- Start date
- Sort by reaction score
I returned my 16" and decided to wait for the refresh to include 802.11ax, I figured it was worth waiting for and I think it will be even more important to have for resale a few years down the road. I also figured that would allow me to choose between the 14" if Apple decided to make one.
I think the new processor (the improved performance) will be much more important than 802.11ax.
As a former AMD CPU designer, I disagree. AMD never stays in the lead for long, and it’s more of a commitment to support both chip providers than you’d think. If they are going to make the leap, it’s time for ARM.
Normally I'd agree with you, but when nearly every current laptop has (for me) one big dealbreaker, the butterfly keyboard, then I'm back in the waiting/speculation game unfortunately.
Elimination of MagSafe was a mistake. Hopefully with Ive gone the war on ports and functionality will once again be favored over looks and/or thinness.
I just traded in my 15" with the odious scissor keyboard to get a new 16", and wow, I can type again. Trackpad palm rejection is better too. So, whatever they do with the processor or screen pales in comparison to a reliable, working keyboard. And a ginormous trackpad less likely to misfire.
I think people are putting way too much emphasis on .11ax. Most network connections are bottlenecked by the broadband pipe and unless both ends can transmit and receive, you will never actually be connecting at that limit. I have 200mb connection into my house... .11ac doesn’t max out.. so until we all have the higher end 5G or better access to fiber... it’s not actually addressing where you speed bottleneck exists..
The headlines I hate most are like: You won’t believe what this man found under his car! Or: The tech giants don’t want you to see this device! Or: celebrities are making money with this secret app!
What kind of shady scum uses those headlines.
[automerge]1582321443[/automerge]
With all those devices in my house the internal speed is way more important than my internet speed going out. So yeah Wifi6 is important to me.
AMD had a good lead back in the day with the Athlon and Opteron chips. They got lazy and had a poor CEO. AMD's chips are ahead of Intel and will be for the forseeable future. Ryzen rocks on the desktop, will be great on the laptop when the new chips come out in Q2, EPYC is a datacenter dream with high performance and low power draw. They finally are getting the well deserved traction. It's no wonder they are winning supercomputer contracts left and right.
Intel had name recognition and that is it. It's chips are mediocre and have been for a long time. Hell the X86-64 instruction set we all use today is AMD's baby, not Intel.
Why are you telling a guy who helped design x86-64 at AMD what x86-64 is?
Sigh. The internet. Everyone is an expert.
AMD is about to face some serious competition in the server space
Intel 10nm Whitley Lake CPU Benchmark Leaked - Decimates 14nm With Up To 100% Performance Increase Core For Core
Looks like it's Intel leak season because a benchmark of Intel's upcoming Whitley platform just leaked out, and boy does it have some impressive results. Manufactured on the 10nm process, the Whitley platform will succeed Purley and will (finally) transition Intel's server ecosystem to 10nm...
wccftech.com
Intel 10nm 'Ice Lake SP' 14 Core Server CPU Gets First Selfie On SiSoft Sandra, Over 54% IPC Improvement
Intel's 10nm Ice Lake server CPU has been benchmarked for the first time on Sisfot Sandra and features an insane amount of IPC improvement!
wccftech.com
Losing compared to what? The Mac's market share has been stable for years.
Sure, but that was a different world where desktops ruled.
For the Mac, the laptop is the most interesting, and AMD has yet to deliver. There have been no independent benchmarks of Ryzen Mobile 4000.
AMD is about to face some serious competition in the server space
Intel 10nm Whitley Lake CPU Benchmark Leaked - Decimates 14nm With Up To 100% Performance Increase Core For Core
Looks like it's Intel leak season because a benchmark of Intel's upcoming Whitley platform just leaked out, and boy does it have some impressive results. Manufactured on the 10nm process, the Whitley platform will succeed Purley and will (finally) transition Intel's server ecosystem to 10nm...
Intel 10nm 'Ice Lake SP' 14 Core Server CPU Gets First Selfie On SiSoft Sandra, Over 54% IPC Improvement
Intel's 10nm Ice Lake server CPU has been benchmarked for the first time on Sisfot Sandra and features an insane amount of IPC improvement!
Those articles are pretty weird. For starters, I'm not sure why the benchmarks are run on Geekbench 4 (if anyone is wondering why those single-core numbers are so high, this is why — you can't compare these numbers to the Geekbench charts, which are Geekbench 5 now!).
Second, a more reasonable comparison really isn't that great of a bump. Clock for clock, the alleged Ice Lake chip is 5.5% faster in single-core and 8.2% faster in multi-core, compared to a CPU that's almost three years old. And that's assuming they could ship a 3.2 GHz Ice Lake server at all.
We'll see.
(But also, I don't find the server space that interesting.)
[automerge]1582328947[/automerge]
Probably won't see 10nm 45W until at least 2021.
Last edited:
No doubt early 2021 is ambitious from intel. It does sound like from internal comms though that the golden cove chip is super high visibility for intel execs so maybe it's possible. By that point they'll already be on the 10nm process so I think an architecture change is more straightforward.
[automerge]1582331322[/automerge]
Looks like you're right. I guess you'd mostly just be missing out on the xe graphics/av1 hardware decoding then compared to what's likely to come in fall (assuming Apple even does another cpu refresh on the 13" models at that point)
every cache design improvement equals 2 percent performance improvement.
every cache design improvement equals 2 percent performance improvement.
Out of interest... but without enough time to read... is that 2% an average based on historical improvements, some arbitrary limit or the design goal? Why not make twice as many improvements at once for 4%
It’s a weird thing i noticed when modeling this stuff years ago. Every bright idea improved cache performance by about 10 percent, and processor performance by 2 percent. It turned into a running joke.
combining things didn’t work by addition. You multiplied. So do two things and you end up with 10 percent of 10 percent on top of 10 percent (which is 11 percent).
I’m sure that one can violate Maier’s Law quite easily, but it is funny how often it works.