Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Interesting. The multi-core gain isn't exactly insignificant, I expected it to be smaller due to all the thermal constraints, and what you say about the temperature is an interesting tidbit. Maybe they did improve the cooling system after all, after listening to all the complaints about the 2018 model?

Looking forward to seeing some teardowns and reviews about it, even if I'm afraid that whatever they did to it will make my 2018 6-core MBP look old in comparison :(:D

Id believe they definitely improved the cooling - they don need to have announced it but we will see in a tear down as you said. if I had to guess they made 2 separate heat pipes for the CPU and the GPU + bigger fans maybe ?
 
Interesting. The multi-core gain isn't exactly insignificant, I expected it to be smaller due to all the thermal constraints, and what you say about the temperature is an interesting tidbit. Maybe they did improve the cooling system after all, after listening to all the complaints about the 2018 model?

Looking forward to seeing some teardowns and reviews about it, even if I'm afraid that whatever they did to it will make my 2018 6-core MBP look old in comparison :(:D

It’s close. 29.5/23.9 = 1.23, 8/6 = 1.33. In this form factor I think we had to expect less then perfect linear scaling, and given the heat issues, almost no gain at all seemed a possibility. 23% is pretty good. I’ll be interested to see how the top end processor does.
 
  • Like
Reactions: afir93
Good stuff. As we know it only takes a couple minutes for the heat to reach a steady state. No need for hours of benchmarking. Thanks for posting this. Very exciting.
Not at all. It takes minutes for 2018 MBP to overheat and drop freq, yes.
But it also takes hours of heavy work to understand difference between Cheesegrate MacPro and 2013 MacPro. First half an hour they both seem running well, so minutes are often not enough.
 
Only just published by the way, so may be a bit fuzzy till it's finished processing.
[doublepost=1558627612][/doublepost]
Not at all. It takes minutes for 2018 MBP to overheat and drop freq, yes.
But it also takes hours of heavy work to understand difference between Cheesegrate MacPro and 2013 MacPro. First half an hour they both seem running well, so minutes are often not enough.

Agreed. My first i9 6 core would throttle back to 2-2.2Ghz within 10 mins or so. It was replaced and the new one doesn't drop between 2.9Ghz from what I've seen.
 
Video up here. If anyone wants anything specific happy to look at - will do a more in depth review shortly.

There is something wrong with your 2018, it looked kind of low, so I just did the yes command x12 on 2.2 MBP and it holds 3.5 Ghz at around 55W. Or the global warming finally got UK and your ambient temp is tropical.
 
Not at all. It takes minutes for 2018 MBP to overheat and drop freq, yes.
But it also takes hours of heavy work to understand difference between Cheesegrate MacPro and 2013 MacPro. First half an hour they both seem running well, so minutes are often not enough.

See below. Double posted after trying to edit.
 
Last edited:
There is something wrong with your 2018, it looked kind of low, so I just did the yes command x12 on 2.2 MBP and it holds 3.5 Ghz at around 55W. Or the global warming finally got UK and your ambient temp is tropical.

Well, the first one I had was far worse. Also, 2.2Ghz MBP is a bit different to a 6 core 2.9Ghz i9.
 
Not at all. It takes minutes for 2018 MBP to overheat and drop freq, yes.
But it also takes hours of heavy work to understand difference between Cheesegrate MacPro and 2013 MacPro. First half an hour they both seem running well, so minutes are often not enough.

Sorry, hog wash, and you don’t understand the physics of heat. After the processor heats the initial cool materials in the laptop, which only takes a small handful of minutes, you are running at steady state. There is no slow linear creep. Big servers obvious have advantages over laptops for large sustain tasks, and CPU heat-related issues are one of them, but more regarding the nature of using a laptop, not the reliability of CPU performance. Maybe if you're talking weeks, months, even years of sustained high workloads, but not hours.
 
Last edited:
Yo
Just a tad smaller cache, shouldn't matter in this test, maybe in primes when they bang cache.

You'd think so, but that's not what everyone was seeing with all the initial problems with these units. Anyhow, happy at least it doesn't throttle below clock now. First version was terrible for it.
 
You'd think so, but that's not what everyone was seeing with all the initial problems with these units. Anyhow, happy at least it doesn't throttle below clock now. First version was terrible for it.
I'm convinced your cooling on the 2018 is still broken, just less than the first one. You won't see this in Geekbench, the tests last mere seconds, but your scores in Corona/Cinebench (especially R20) are probably way down. There was a lot of i9 reported to have issues with thermals initially, I mean even after the throttling patch, it kind of dried out when Vega came out.

Please, run the R20 or Corona on the 8 core.
 
I've got two of them by the way. They both now do exactly the same thing, as does the one across the desk from me at work. I've looked. Honestly. I'm not making it up :) The one that was replaced was throttling back ridiculously so - back to 2-2.2Ghz. Took taking in another one for them to replace it. Was quite irritating.

The 8 core seems to behave more inline with the non-i9 2018 units from what I can tell.

I'll run those benchmarks shortly and post up.
 
Sorry, hog wash, and you don’t understand the physics of heat. After the processor heats the initial cool materials in the laptop, which only takes a small handful of minutes, you are running at steady state. There is no slow linear creep. Big servers obvious have advantages over laptops for large sustain tasks, and CPU heat-related issues are one of them, but more regarding the nature of using a laptop, not the reliability of CPU performance. Maybe if you're talking weeks, months, even years of sustained high workloads, but not hours.
Ah yes! I don't understand :) Just try to render something on the MacPro 6.1 :)
First. Heating the entire laptop body isnt a matter of seconds and even minutes.
Second. Laptop isnt a fry pan. Its an electronic device containing conductors and semiconductors. Both of them arent happy when heated, so the stability decreases. Concerning PSU this means unstable power, thus higher consumption and more heat. Concerning battery - the hotter battery is, the more energy leaks, meaning more heat, etc etc.
Anyways, I am happy if MBP 2019 doesnt have overheating issues. 2017 and 2018 were awful :(
 
  • Like
Reactions: Ener Ji
Ah yes! I don't understand :) Just try to render something on the MacPro 6.1 :)
First. Heating the entire laptop body isnt a matter of seconds and even minutes.
Second. Laptop isnt a fry pan. Its an electronic device containing conductors and semiconductors. Both of them arent happy when heated, so the stability decreases. Concerning PSU this means unstable power, thus higher consumption and more heat. Concerning battery - the hotter battery is, the more energy leaks, meaning more heat, etc etc.
Anyways, I am happy if MBP 2019 doesnt have overheating issues. 2017 and 2018 were awful :(

Dude, relax. Why do you think you're the only one to do work on multiple platforms? I've uses cheese graters, the trashcan, multiple MacBook pros, linux boxes, clusters or clouds. And yes, heating your laptop up is minutes. If you're having stability issues, something is wrong with your hardware. I regularly run things overnight on a MBP, its not a huge deal. Now, if you're using a laptop to do week long runs, you're just doing life wrong in the first place.
 
  • Like
Reactions: Perezx
Has anyone actually ordered the new 2019 MBP? If so why? Seems to get a LOT of negative press on MR, but looks like both will be nice (not ideal) but pretty decent upgrades....

I'm ordering one if nothing major happens within 2 weeks of release.

They look pretty good on paper but nothing is perfect - let's see if there will be any deal breakers.
 
  • Like
Reactions: Ener Ji
Has anyone actually ordered the new 2019 MBP? If so why? Seems to get a LOT of negative press on MR, but looks like both will be nice (not ideal) but pretty decent upgrades....

I’m getting ready to pick one up today (upgrading from a 2017 model). It’s definitely not the full redesign I was hoping for (though if one magically shows up at WWDC, I’m doing an exchange for sure). That being said, they did (supposedly) fix the keyboard and they threw in the newest Intel chips which are a nice performance bump. I would have rather they updated the GPU as well but I guess that will come next year.

As it stands, it’s a decent enough upgrade and it’s nice to see Apple upgrade their macs on a regular schedule. However, if I wasn’t so tied to Mac OS (mainly for app development), I would buy the new Razer without thinking twice. What they give you for a couple hundred dollars less than the MBP price is far better.
 
  • Like
Reactions: Ener Ji
I’m getting ready to pick one up today (upgrading from a 2017 model). It’s definitely not the full redesign I was hoping for (though if one magically shows up at WWDC, I’m doing an exchange for sure). That being said, they did (supposedly) fix the keyboard and they threw in the newest Intel chips which are a nice performance bump. I would have rather they updated the GPU as well but I guess that will come next year.

As it stands, it’s a decent enough upgrade and it’s nice to see Apple upgrade their macs on a regular schedule. However, if I wasn’t so tied to Mac OS (mainly for app development), I would buy the new Razer without thinking twice. What they give you for a couple hundred dollars less than the MBP price is far better.
haha - good points, did you go for the 15?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.