Switched from a maxed out MacBook Pro 13'' late 2013 to the newest 2018 13'' maxed out model. Don't like the case design at all, while synchronizing dropbox the laptop got so hot I had to take it away from my legs. Also insane how fast my battery is going down right now
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.
Tests Confirm Apple's Throttling Fix Improves Performance for 2018 MacBook Pro Models [Updated]
- Thread starter MacRumors
- Start date
- Sort by reaction score
I found that this resonates with my intuition as well:
“PCWorld’s Mah Ung speculates that even more tuning could be around the corner, and I suspect that’s because the performance uplift here, while welcome, isn’t quite where it should be yet.”
https://www.forbes.com/sites/jasone...cbook-pros-performance-problems/#2627b9359051
So my question is if Apple implements it's own processor for laptop use, will this mitigate the heat issue for high end applications that people need an i9 for?
i guess that depends on whether or not the application will even run on the processor natively in the first place
and if not, how well/efficient the emulation mode may be.
idk, i imagine many high end programs for mac will continue to be used with intel chips for the foreseeable future..
if/when the programs run natively on Apple chips, i don’t think anybody knows the answer to your question.. but i’m an optimist so i’ll answer “probably “ : )
Would make sense as heat is the main enemy (leading indicator of power consumption which is the second) and if they can get more performance and better power management out of something that uses less power than the alternative then that’s a case for production.
In view of the usefulness to i9 of more thermal headroom, would this not be an marvelous time to reintroduce a 17 inch MBP ?
I would assume that the chief reasons for using their own professors is that Apple can then tweak them for power and heat efficiency, and further optimising their own software to make maximum use of these processors.
The end result would be that Final Cut Pro runs like a champ, while third party apps like adobe premiere continue to suck, if they even run at all.
One can only hope so.
i think it's clear you have a misunderstanding about rendering and CPU requirements and turbo speeds.
if Apple is touting MBP as a capable rendering machine then it's because of the core count.. not turbo speeds..
these two things fight each other..
the more cores a process is using then the less per core turbo frequency is expected.. the less cores a process is using then the faster per core frequency is to be expected..
rendering is one of the few process that can be divided up effectively to saturate many cores.. it's not about super fast clock rate on a single thread (which is what max turbo speed is about)
also, a lot of rendering software is now utilizing GPGPU.. processing tasks are being offloaded to the GPU since a GPU has thousands of cores albeit much slower per core than a CPU.. the GPU in the MBP with 4GB VRAM is decent to good for rendering on if the software is written that way.. a lot better than many desktops even.. the GPU would be another reason for Apple using 3D rendering in their ad literature with this computer.
----
thing is, at least in a 3D rendering workflow, all of the stuff leading up to the rendering.. the CAD/3D Modeling software.. consists of virtually all single threaded linear processes.. these types of software love super fast single core performance.. they will certainly make use of max turbo speeds.. the speeds advertised by Apple with the i9 (4.8GHz) will be achieved and utilized when using CAD software.. and the user will notice and appreciate these speeds during most of the work.
or, most of the actual work that goes into creating a render is under the conditions where rated turbo speeds will be utilized.
and just by the way-- rendering is easy.. or, compared to the entirety of the workflow, rendering is the easiest part of it to learn and do.. it's not the most 'pro' part of it by any means.. the designing and engineering is the most 'pro' and most creative part and that happens mostly in the user's brain.. once you get to computer, the most pro part of it, the part that's most difficult to master and become efficient at, is the modeling/drafting..
the rendering, that's mostly used to show the client a pretty picture and to sell your idea to them prior to fabrication as they're not trained to interpret a CAD model into what it looks like in real life.. so you render the model to make it easier for them to understand.. so is it important? sure, it helps secure the project and to get money from the client.. but it's definitely not the hardest part of a project nor is it the most time consuming.. i'd guess on an average project, i spend about 1% of the whole thing making renders (or, 1 hour setting up renders for a 2-3week project (the finals are processed via cloud based supercomputers))
point being-- there seems to be a common misguided idea around here that 'pro' workloads correlate with these stress test benchmarks... reality is, the benchmark tests have very little to do with the working conditions of a pro.. i mean, you can easily see this for yourself..
take the Cinebench test for instance.. the test includes a scene.. the modeling and texturing and lighting have already been provided.. and the user/tester just pushes a button and waits for a score..
all of the stuff a pro is responsible for has already been provided... all of the work is already done.
the test waits a minute for the score to return but the score says nothing about the hours of work that have gone into creating the scene.
the test gives you an indication of how the computer works when the designer isn't working.. it says nothing about how the computer works when the user is actually interacting with it and doing work. (ok, it says something.. but i think you need experience in the entirety of the process to understand how to extrapolate the results)
or-- do you actually think a 'pro' goes into work then pushes a button then waits a while for the computer to finish? idk, to me, that doesn't sound very hard at all.. everyone can (and around here, does) do that so it's a bit of an insult to someone who has spent years honing their skills only to have it devolve into this type of comparison or test..
i'm sorry but if you focus on these benchmark scores as to what a pro needs.. and you're upset about 'Apple doesn't listen to me'.. well, the reason they don't listen is because it's clear you don't know what you're talking about.
same thing with the video editing tests.. they're timing how long it take to export or recode a project.. but it says nothing about the actual editing.. you know, the majority of the work and/or the work the user is being paid to do..
----------
idk, when you're reading these forums, try to recognize the red flags being said by certain posters and take the words with a grain of salt.. "these aren't pro machines" -- " Apple doesn't care about pros" etc.. it's bs..
i mean, in this thread, you have people complaining how every last bit of the i9 isn't being utilized by the stress tests.. and how pros need every last drop of performance to be sucked out of these chips.. and their suggested solutions go something like:
make the laptops 50% thicker.. make them 2x heavier.. then a pro will be able to get 15% more performance during 1% of their workload..
hmmm..
I agree with you but it's like you're forgetting about "all" of us who ARE actually talking about real time computing performance, heat and faster running fans (booo) it's funny you think magically because youtubers and some folks here focus on rendering time, because that's what saves THEM alot of time, compared to previuous gens. When I do agree there are limited real world usage tests, those are also much harder to compare by the end user (the viewer) but don't think people are just rendering, or not redering and not needing the fast performance. THAT'S where I unfortunately see a broken chain, so many of you disregard those of us who DO have a need to push the system REAL TIME, REAL WORK, WHILE WORKING. That's not a joke. That's why we pay 30000 Danish Kroner for a Pro laptop, and THAT'S why people are complaining about heat and fan noise and suggesting slightly thicker chassis etc.
Feel free to push the machine, but the base clock speed, with all-cores at 100%, is 2.9GHz. That’s how Intel spec’ed it and that’s how it appears to run, after the update Apple issued is applied.
I admit I tl;dr the OP’s post so maybe I missed something.
I agree.
Another way to see it is that 6 cores at 2.9Ghz are equivalent to 4 cores running at 4.4/4.5Ghz. How you like them apples now ???
Yup. Before this year, the fastest quad-core on the previous MBP (i7-7920HQ) had a base clock of 3.1GHz. The new CPUs give you 50% more cores and you only have to give up 200MHz.I agree.
Another way to see it is that 6 cores at 2.9Ghz are equivalent to 4 cores running at 4.4/4.5Ghz. How you like them apples now ???
But depending on your workload, the 32GB max RAM can contribute more toward increased performance than the extra cores.
At $3,200 for either, it’s a tough choice between the 2.6/32/512 and the 2.9/32/256. For most, I think the i7 with the 512GB SSD is probably a better value but it really depends on your particular use case. The 2.2/32/256 at $2,800 is also a viable option. But for me 32GB is a must so I wouldn’t consider the $2,400 base machine. Of course, YMMV.
Last edited:
By the way I just tripped over a USB C power adapter cable - the side that plugs into the Mac - and it detached at the AC adapter terminated side. So even though MagSafe is no more Apple has given us a cable that can disconnect at either end and in this moment that seemed to have saved this MacBook Pro from my clumsy rush to check the oven.nothing?
not charging:
View attachment 772755
not charging:
View attachment 772756
sketchy AF looking:
View attachment 772757
-----
i promise you these aren't setup photos in attempt to make it look worse than it really is.. they're exactly real life conditions and i'm always having to make sure my laptop /cord is in a certain position to get a charge.. also, the magsafe connector is spring loaded and will jam up.. i have to buy a new one about once per year.. judging by my third picture, that time is coming up (although the laptop itself will be replaced soon instead of the power cord).. it will start smoking eventually due to the shorting/flakey connection..
i get it that if these things are always used in calm/clean/controlled environments then you may have better luck with them but bring them into a fast paced environment (or in my case, a fast-paced manufacturing environment) and your idea about how good magSafe is will likely change relatively quick..
if you're worried about tripping on the cord and crashing the machine to the floor, coil up the slack near the connection.. you'll then have to trip on the cord and drag it 8' before it snags the laptop.. bonus tip-- this works with all sorts of other tools/machines/plug-in-stuff too..
Basevs makes a USB-C breakaway cable that can support the 87W charging of the MacBook Pro 15. There is also the Griffin BreakSafe for the 12" and 13" laptops (I do wish they offered an 87W version).
This issue has somewhat come back for me on my MBP i9 2.9Ghz. I see it throttling down to 0.8GHz when doing something simple like an iMovie export. Everything becomes unresponsive. I've only noticed for the last two days. The throttling patch initially seemed to work for me but now unfortunately, it is throttling even worse than before, but it does not seem to have a relationship to the temperature like previously, according to the Intel Power Gadget. Anyone experiencing the same recently?
Doesn’t sound like the original problem, which was throttling at about 2.2GHz. Run some cinebench, 4 or 5 in a row and compare them to what everyone was getting after the fix.
Does iMovie use AVX or AVX2 instructions? Intel CPUs will down-clock themselves significantly (as low as 0.8GHz) when performing extended operations with these instructions.
I have no idea unfortunately. Apple seem to think that it might be a machine specific issue. So it might be a swap out tomorrow.