Thanks. I just bought a new one. I'm 87 years old, so it might not be wise to wait.
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.
Waiting for the 2018 MBP
- Thread starter benface
- Start date
- Sort by reaction score
Ahaha, that's the best comment so far. Enjoy your new Mac. I am writing from a 2017 MBP as well and I love it.
Well, there are new chips expected in the next couple of months that would be appropriate to use in some models of the MacBook Pro. It is predicted that the MacBook Pros would be updated with these in Q2 or Q3, good timing for a yearly update.
Dark humour! I chuckled.
I feel apple will step up their game in the next mbp release, cpu wise at least. Realistically speaking, what chip could they ve giving us? 6 core really plausible? What about a faster 4 core, maybe 3ghz per core?
6 core in the top end. Lower base clock, but that doesn’t matter.
4 core at the lower and mid end. Lower base clock, but that doesn’t matter.
- Danish law is identical to EU law on that point, with the same 6 month cut-off. In practice and due to previous case law, it is handled in a manner so that it is never an issue for the consumer.
Why wouldnt that matter? base clock is good for apps which are not multithread optimized.
Because with current chips, for single-threaded applications and poorly multi-threaded applications, the chips will run at much faster than base clocks, and for extended periods of time.
For example, Cinebench R15:
8th gen 1.9 GHz Core i7-8650U (4-core 8-thread 15 W chip with UHD 620)
178 - single-thread
675 - multi-thread
7th gen 2.8 GHz Core i7-7600U (2-core 4-thread 15 W chip with HD 620)
164 - single-thread
379 - multi-thread
As you can see, not only does the 8th gen 4-core chip absolutely destroy the 7th gen 2-core chip in multi-thread speeds, it also slightly beats it in single-core as well, even though the 4-core chip has a much, much lower base clock speed. The reason is that in 2018, the base clocks are meaningless for predicting single threaded application speeds, since the chips aren't actually running at those base clocks.
Last edited:
So how can you objectively compare cpu speed? Take the turboboost value into account only?
That’s why I posted benchmarks. And no the turbo boost value may not always be accurate either. They are theoretical max.
Basically I would just forget about it and realize that these 2017 and 2018 multicore CPUs do single thread tasks as well as chips with fewer cores, but do multi threaded tasks much better.
Just as a heads up - NVidia now offers a 10W variant of the MX150. Although the MX150 is still based on a 14nm Pascal architecture, performance of the 10W variant is somewhere between 10 to 30 percent faster than the Iris Plus 650. So using a 15W quad-core 8th gen CPU with a 10W MX150 in the 13“ Touch Bar model would be a realistic alternative to waiting for Intel‘s 28W GT3e chips.
There's no need to wait. The i5-8269U already exists.
[doublepost=1521894814][/doublepost]
It's the same here (Czechia).
[doublepost=1521894921][/doublepost]
If the cooling can keep up, the chip can run on maximal Turbo for whole task, making base clock just a "worst case performance number" these days.
[doublepost=1521895043][/doublepost]
Well, right now Apple has the worst offering in comparison with other competitors. This should suffice Apple as a reason to fix that.
- It's been leaked, but not released. It's also 15W according to the leaks.
Well, Apple several time had exclusive access to some chips (for example the first Air).
According to SOME leaks, I've read (original source) that it should have 28 W TDP. It would make sense - 3GTe with 128 MB eDRAM + 4 Cores needs some thermal room. The main difference between 15 W and 28 W chips even nowadays are reachable clocks and "thermal room" - the chips are the same, they "just" run higher on clocks and keep turbo clocks longer time.
Not saying you're wrong (I hope you're right) but this article seems to think that chip will have regular (U)HD 620
The most important issue for me is whether those will greatly improve performance compared to my 2.8Ghz i7 in my Mid 2015 MBP. Upgrades from 2015 to 2017 have reported to not be significant...
The 2015 to 2017 is not that significant in terms of raw performance. The main benefit is hardware 10-bit HEVC support.
The 2017 to 2018 may be more significant because you'd be going from 4-core to 6-core.
However, it would depend on what you do with it. For poorly threaded applications, the performance going to 6-core would not be lower but it might not be much higher either. OTOH, for well threaded applications, the performance going to 6-core would be noticeably higher.
Mind you, I'm the type of buyer that if I had a 2015, I probably wouldn't upgrade until 2019 or later. I tend to keep my machines a very long time.
How long do we wait to actually buy a new MacBook Pro, once they're announced? The AirPods took quite a while...
Usually, it doesn't take too long with spec bumps.
AirPods aren't such a fitting comparison because they were an entirely new product line with lots of new technology in it that Apple didn't want to mess up, but they were essentially forced to present them alongside the iPhone 7 because they were pushing for wireless headphones and wanted to minimize the backlash over the headphone jack removal.
With new MBPs and generally other Mac spec bumps, Apple isn't under that kind of pressure, they can pretty much choose to announce them whenever they want and they usually don't do that unless it's close to being released. Once Intel's new chips are ready then they are ready, there's no reason to announce the new models months in advance.
IIRC, the 2017 models were announced during WWDC and were available to order on Apple's website the same day. They were available to buy in-person 2 days later at an Apple store. I wouldn't be surprised for them to announce a refresh during WWDC again.
Whether they'll use Coffee Lake (14nm) or Cannon Lake (10nm) CPUs is unknown at this point, Intel claims they've been shipping Cannon Lake since the end of 2017 though. It seems some Coffee Lake-U cpus that fit Apple's requirements have been leaked out, but nothing about Cannon has surfaced yet. I'd like to say Apple would just skip Coffee and go straight to Cannon for better power savings, but I'm obviously not Apple. 4 cores is going to be offered regardless for the base chip as I doubt Apple will use the i3 chips (2 cores/4 threads still) in their bread and butter MB Pro. The i5's are being offered as 4 cores/8 threads from here on as of Coffee Lake.
Personally, I'll hold out. My haswell based Windows laptop is starting to show its age with its i5 4200u, and I really don't want another 2 core/4 thread CPU. If I needed a new laptop right now, I'd probably grab one of the HP Envys that are offered with the 4 core/8 thread Coffee Lake-U cpus. Saw one at Costco with the MX150 and it was really nice. Would be nice to get back into the Mac world though, my last MB Pro was a mid 2010. Not too impressed with the keyboard on the MB Pro's now, but I really liked the build quality. I found the touchbar was pretty useful, but again, I think a touchscreen would be a nicer choice.
Last edited:
Yea, it's looking like WWDC at this point. Pretty psyched to see what apple has planned. I am looking at the 6 core. Memory is not that big of a deal for me (pro-audio). It's all about processor preetty much.
Thanks for the explanation. I wish I could keep mine forever as I love the form factor. However it’s been a while I feel the i7 2.8 is lagging behind in my daily production activities. I wanted at least for the next gen to rival a desktop 4790k performance wise. That was so smooth
New keyboard please. More travel = more reliability?
If you buy the argument that the keyboard failure rate on the rMB is lower, then no, travel and reliability have little correlation. Reliability is correlated to reliability, that's all.