Apple is addressing the 2018 Macbook Pro Throttling problem

[Hunter259]

https://www.macrumors.com/2018/07/24/apple-addresses-macbook-pro-throttling/?utm_source=osx&utm_medium=push&utm_campaign=front

 

Apple's Explanation:

Quote

Following extensive performance testing under numerous workloads, we've identified that there is a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads on the new MacBook Pro. A bug fix is included in today's macOS High Sierra 10.13.6 Supplemental Update and is recommended. We apologize to any customer who has experienced less than optimal performance on their new systems. Customers can expect the new 15-inch MacBook Pro to be up to 70% faster, and the 13-inch MacBook Pro with Touch Bar to be up to 2X faster, as shown in the performance results on our website.

 

TLDR: Apple has released an update to High Sierra that should help or completely alleviate the atrocious throttling issues that the 15" Pro's have been facing but also says it should help the 13" Pro's as well. From what Apple is saying, it seems that this was either a mistake that was not caught in time for release. Apple themselves started to look into this issue after David Lee's video about his 15" Pro. They contacted Lee so they could replicate his workflow and this is the result of their findings.

 

Still incredibly frustrating for anyone who did buy these machines. This is something that should have been caught before a wide release like this. Hopefully reviewers update their reviews to better show the product and it's potential performance.

[Crunchy Dragon]

Please edit your post to match Tech News guidelines.

 

https://linustechtips.com/main/topic/11724-posting-guidelines-read-before-posting/

[D13H4RD]

Quote

Following extensive performance testing under numerous workloads, we've identified that there is a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads on the new MacBook Pro.

What

[captain_to_fire]

I’m surprised they didn’t pull something like this one 

Edit: Oh no wait they did 

Quote

Apple for its part said it never experienced any issues with excessive performance degradation in its pre-production testing of the 2018 MacBook Pro, completed in June. The company did identify what it calls an isolated bug after digging deeper, and came up with what it calls a simple fix. 

To be expected, Apple also talked up the new MacBook Pro, noting how most of what it has been hearing about the notebook from customers—including many professionals—has been very positive, including about performance.

 

Edited July 24, 2018 by captain_to_fire

[laminutederire]

Okay so well have a mass recall in 2019 for thermal issues grilling vrms. Apple is just counting on people being idiots enough never to use their machinesfor anything heavy in the mean time.

[TheOriginalHero]

1 minute ago, D13H4RD2L1V3 said:

What

"Write something techie as an excuse for the article" *Actual fix: Turns up fan speed*

[BrinkGG]

2 minutes ago, D13H4RD2L1V3 said:

What

This is apple we're talking about. I'm honestly not surprised by BS like this anymore. 

[jiyeon]

Quote

Following extensive performance testing under numerous workloads, we've identified that there is a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads on the new MacBook Pro.

Have I not been paying attention in my 5 years of computer science & hardware classes in high school or is this entire claim garbage?

[DrMacintosh]

Seems good. Looking forward to more benchmarks after everyone installs the macOS update.

[Sniperfox47]

INB4 they just tweak it with an undervolt and drop the core clocks causing it to still perform terribly, just not as terribly.

[Max_Settings]

BS, software update isn't going to do anything unless they are just limiting the turbo clock or doing a serious undervolt. It's the cooler's fault, not the software.

[Snipergod87]

I love software fixes for bad design from apple, just like the whole iphone 5 antenna issue, their software fix was to change how the signal strength bars are measured to make people think the issue was solved.  Apple QC #1

Fan speeds can help but if it thermal throttles in less than a minute at reasonable fan speeds a bigger heatsink is going to be needed (its just physics) unless they undervolt which could make the whole system unstable.

Let alone a sustained CPU+GPU load for heavy encoding.

[TheGlenlivet]

Something has to give right?

Higher fan speeds and more noise or lower voltages and CPU speed.

Is there something we are missing here?

[PeterBocan]

9 minutes ago, DrMacintosh said:

Seems good. Looking forward to more benchmarks after everyone installs the macOS update.

Yeah, we've seen the benchmarks already. Pretty shit.

[D13H4RD]

Well, I expect at least some changes in voltage.

 

Those damn VRMs aren't going to last very long at all at stock settings

[AskTJ]

Digital key? smh Apple.

[mynameisjuan]

14 minutes ago, TheGlenlivet said:

Something has to give right?

Higher fan speeds and more noise or lower voltages and CPU speed.

Is there something we are missing here?

Something hardware related does have to change to make any impact. 

 

Its either going to be twice as loud or going to crash because voltage is too low. 

[maartendc]

22 minutes ago, TheGlenlivet said:

Something has to give right?

Higher fan speeds and more noise or lower voltages and CPU speed.

Is there something we are missing here?

 

24 minutes ago, Max_Settings said:

BS, software update isn't going to do anything unless they are just limiting the turbo clock or doing a serious undervolt. It's the cooler's fault, not the software.

 

6 minutes ago, mynameisjuan said:

Something hardware related does have to change to make any impact. 

 

Its either going to be twice as loud or going to crash because voltage is too low. 

 

I think you guys missed the story about how the website Notebookcheck had managed to get the throttling under control by just putting a Power limit on the CPU instead of just a temperature limit:

 

I would imagine Apples fix to this is:

 

- Set power limit on CPU

- Perhaps undervolt the CPU slightly (although that can cause instability)

- Perhaps undervolt the VRMs slightly (instability?)

- Adjust the fan curve to ramp up faster under sustained load. (although they don't even need to this, seeing as Notebookcheck has proven that setting the power limit already has a big impact). 

 

This would all result in higher sustained clocks, potentially with a bit more noise.

And done.

 

The funny thing is: This just shows that Apple didn't do any testing of the i9 model, because it is blatantly obvious that it throttles under sustained workloads. The fact that they say it is a "minor bug" that only happened under certain rare conditions is just a lie, it happened under very common conditions and for all users. It also shows that, had they spent perhaps an hour or 2 thinking about it, they could have come up with this solution in the first place. Pretty terrible engineering if you ask me. Someone is getting fired over this as we speak.

 

[mynameisjuan]

[D13H4RD]

5 minutes ago, maartendc said:

I think you guys missed the story about how the website Notebookcheck had managed to get the throttling under control by just putting a Power limit on the CPU instead of just a temperature limit:

There's also the Reddit post regarding the VRMs

 

https://www.reddit.com/r/macbookpro/comments/91256u/optimal_cpu_tuning_settings_for_i9_mbp_to_stop/

 

[Fasterthannothing]

Asus remembers the turbo key 

Quote

Advertised feature for M4N72-E: 
Turbo Key
ASUS Turbo Key allows the user to turn the PC power button into an overclocking button. After the easy setup, Turbo Key can boost performances without interrupting ongoing work or games - with just one touch!!

.... Breaking news from Apple they now have courageously removed the power button and instead opted to have a turbo key!

 

 

[Master Disaster]

Props where props are due, they've fixed it in a very timely manner.

 

Can't wait for the community to pull the fix apart and tell us what they actually did though. I can't wait to hear they copied the community fix and included it as official.

[dizmo]

5 minutes ago, Shorty88jr said:

Asus remembers the turbo key 

Advertised feature for M4N72-E: 

Turbo Key

ASUS Turbo Key allows the user to turn the PC power button into an overclocking button. After the easy setup, Turbo Key can boost performances without interrupting ongoing work or games - with just one touch!!

Pretty sure I used that with their later boards Or something similar. Just Windows software that'd apply an overclock.

It's also the only time I've messed around with overclocking.

[Tech Enthusiast]

29 minutes ago, Master Disaster said:

Props where props are due, they've fixed it in a very timely manner.

 

Can't wait for the community to pull the fix apart and tell us what they actually did though. I can't wait to hear they copied the community fix and included it as official.

I fully expect to have grilled MacBooks by early 2019 as a result.

Will be fun to watch the iSheep suffer tbh, but meanwhile let me just say:

 

"Oh nice Apple, fixing an exposed hardware issue with a [key] in firmware. I totally buy that!"

[Blade of Grass]

1 hour ago, maartendc said:

 

 

 

I think you guys missed the story about how the website Notebookcheck had managed to get the throttling under control by just putting a Power limit on the CPU instead of just a temperature limit:

 

I would imagine Apples fix to this is:

 

- Set power limit on CPU

- Perhaps undervolt the CPU slightly (although that can cause instability)

- Perhaps undervolt the VRMs slightly (instability?)

- Adjust the fan curve to ramp up faster under sustained load. (although they don't even need to this, seeing as Notebookcheck has proven that setting the power limit already has a big impact). 

 

This would all result in higher sustained clocks, potentially with a bit more noise.

And done.

 

The funny thing is: This just shows that Apple didn't do any testing of the i9 model, because it is blatantly obvious that it throttles under sustained workloads. The fact that they say it is a "minor bug" that only happened under certain rare conditions is just a lie, it happened under very common conditions and for all users. It also shows that, had they spent perhaps an hour or 2 thinking about it, they could have come up with this solution in the first place. Pretty terrible engineering if you ask me. Someone is getting fired over this as we speak.

 

It sounds like they never did any real world testing and instead only used some synthetic in-house benchmark which didn’t pick it ip... why? Who even knows. 

 

Lets see see if this actually fixed it ??‍♂️