2018 i9 MacBook Pro throttling discussion [merged]

[Nozuka]

You guys are seeing very high temps (80c) at quite low clocks (2.2 ghz). I just ran a quick test on my XPS 15 9570 (also one of the thinnest & lightest and thus most thermally-constrained laptops with the i7-8750H), running a 12-thread stress test at 2.2 GHz. After 1 minute, temperatures on all cores held at 52C and the fans did not come on. Again, the MBP 15's behavior is telling me that the cooling solution is just woefully underdesigned, and there is nothing that can be done to get these chips to hit their potential. And these are brand new macs, with no dust in the fans, no dry thermal grease, etc.

Now i googled the XPS 15 9570 and it's quite funny, how people are talking about thermal problems and undervolting it and how Dell did not change anything to deal with the increased heat.

And the temperature you seem to be getting, is the idle temperature for other guys with this device. So you must have quite a special one.

Some quotes:

So after pushing the computer a bit at stock:

I get an ambient sensor telling me 121 C and cpu maxing out at 93 C. This results in CPU throttling all the way down to 800 mhz.

It is my belief that on heavy long workload, the i9 will throttle to the same level of the i7 (even if you undervolt etc etc). Save the money and get an i7.

Oh and they realized that they are not alone:

Looks like the Macbook i9's also throttle

Sound familiar?
http://forum.notebookreview.com/threads/xps-15-9570-owners-thread.817008/page-71


And also a youtube video, talking about it

You can find plenty more on google.

 

[limo79]

Can anyone saw that Weta studio use a MacBook Pros for rendering? As I remember it was used just for simple tasks like video scene preview. If you really focus on work efficiency and labour cost per hour you do not work on laptop but use a workstation with large monitor with matte screen at work and you can use a laptop when travelling but 13 inch laptop is more handy than holding 1.8kg 15" or 17" machine especially if you travel over Europe with Wizz

 

[Ploki]

ROFL - This is not an iPhone. Even if it were, you know those upgrades you get all the time on your iPhone? Those are firmware updates that add features, fix bugs, and improve performance all the time. This happens on laptop computers, desktops, etc.

WTF - Has no one upgraded the bios on a PC to accept a new CPU? It happens all the time. I don't think there is a thing to see here.

Regardless, I just ran a compile that took ~25 minutes on my mid-2015 i7 2.8. It was ~17 minutes on my new i9. I do this several times a day. I think I'll keep it.

improve performance.

keyword, IMPROVE. not bring to baseline.
this is more an iphone than an old PC tower - everything is soldered, there's absolutely nothing they couldn't have accounted for during testing.

 

[ofarlig]

I am actually very interested in if and how fast Apple would fix the power limit thing. Since it seems to be a new behaviour from before it is possible it is a bug introduced with the T2 chip that they can just fix.

 

[maflynn]

[MOD NOTE]
Closed for review, this thread is off the rails.

 

[maflynn]

[MOD NOTE]
The thread has been re-opened, please stay on topic, avoid attacking other members and be respectful.


Rules for Appropriate Debate

Respect
Guidelines: Show respect for your fellow posters. Expect and accept that other users may have strongly held opinions that differ from yours.

Rules:

1. Name-calling. Name-calling falls into the category of insults and will be treated as such according to the forum rules, your own opinion about another member notwithstanding. You can't call a bigot a bigot, a troll a troll, or a fanboy a fanboy, any more than you can call an idiot an idiot. You can disagree with the content of another member's statement or give your evidence or opinion to dispute their claims, but you may not make a negative personal characterization about that member.
2. Insults. Slurs and insults against groups of people based on negative-stereotyping and obvious generalizations fall into the category of trolling and will be treated as such.
3. Taunting. Mocking or taunting another forum member is not acceptable. Posts that ridicule another member or obviously exaggerate or misstate their views may be removed.

 

[The Mercurian]

Any bets on when Apple will deliver this performance update everyone is talking about? Let alone issue a statement about all of this?

No guarantee they will!

Is a mobile solution even ready to handle your needs?

Until Ryanair allow me to bring my desktop, monitor, keyboard as carryon, a mobile solution is a requirement for my needs.

 

[shavou]

Guys, very good findings here ... see what the fellow member ‘Aea’ discovered!!

 

[mr.anthonyramos]

So we all know that the 2018 15 inch MacBooks Throttle most evident on the 2.6 i7 and the i9 due to the fact that these do run hotter.

Now let me be clear, I’m not talking about throttling, I’m not talking about doing Geekbench Benchmarks, Cinebench benchmarks that we’ve seen over and over.

I enjoyed Jonathan Morrison’s video because he tried to be as diverse as possible comparing the different MacBook Pros and is currently doing more tests.

Here is his video comparing the base 2018 15 inch with the i9:

And here are some initial findings on his twitter comparing the top spec 2017 model with the i9. We can expect the video soon:

https://twitter.com/tldtoday/status/1020887972780969985?s=21

And of course here is a new video by Tailosive Tech that gives a little technical knowhow regarding real life performance and why results happen why they do:

If any of you have more info to share please do!

Let the games begin!

 

[Schranke]

so you want to discuss performance without discussing throttling? and real life at that?
How about we have seen throttling in real life performance such as rendering?

 

[Elektrofone]

Has anyone seen any comparisons between the 2.6GHz 2018 and the 2.9GHz i9 models?

The Jonathan Morrison video was interesting but he only compared the 2.2GHz i7. Also, a technical spec that’s left out of a lot of these videos is RAM. Some application’s render times, i.e. After Effects, depend heavily on the amount of RAM.

 

[mr.anthonyramos]

Has anyone seen any comparisons between the 2.6GHz 2018 and the 2.9GHz i9 models?

The Jonathan Morrison video was interesting but he only compared the 2.2GHz i7. Also, a technical spec that’s left out of a lot of these videos is RAM. Some application’s render times, i.e. After Effects, depend heavily on the amount of RAM.

He mentions that comparison is on the way. Currently he is focusing on comparing the top spec 2017 15 inch and the i9.

 

[Elektrofone]

Is there any evidence that the i9s battery life is worse than the i7s when being used to just browse the web, watch videos, etc...?

 

[Queen6]

It's your perogative, equally I don't see any need for this thread, especially when you are explicitly putting limitations on the the dialogue.

The current throttling issues will affect many users with heavy workflow and therfore must be included. Frankly in some respects your only inviting more responses that you don't care for.

The fewer threads on this subject the better as helps to stop fragmentation across the MBP forum, more importantly puts all the solutions in one place, inclusive of real world workflow and results which is important.

Last point in no way am I suggesting your point of view is not relivent or should be suppressed in any way. I just feel we're reinventing the wheel here.

Q-6

 

[Ploki]

Is there any evidence that the i9s battery life is worse than the i7s when being used to just browse the web, watch videos, etc...?

why do you even opt for i9 for such use?

 

[Jorg1312]

I been flicking through almost all the 43 pages here but haven’t found an answer.

Does anyone know if the base entry i7 2.2 is able to hold it’s speed or does it throttle badly too?

 

[agaskew]

I been flicking through almost all the 43 pages here but haven’t found an answer.

Does anyone know if the base entry i7 2.2 is able to hold it’s speed or does it throttle badly too?

The smart money seems to be on the base model 2.2.

 

[mr.anthonyramos]

I been flicking through almost all the 43 pages here but haven’t found an answer.

Does anyone know if the base entry i7 2.2 is able to hold it’s speed or does it throttle badly too?

It will not throttle as much or as early as the 2.6 or the i9 models.

 

[Eason85]

Now i googled the XPS 15 9570 and it's quite funny, how people are talking about thermal problems and undervolting it and how Dell did not change anything to deal with the increased heat.

And the temperature you seem to be getting, is the idle temperature for other guys with this device. So you must have quite a special one.

Some quotes:




Oh and they realized that they are not alone:


Sound familiar?
http://forum.notebookreview.com/threads/xps-15-9570-owners-thread.817008/page-71


And also a youtube video, talking about it

You can find plenty more on google.

I am a frequent participant of the thread you just linked. the 9570 does not throttle below base clocks in my experience. You see, we tend to have high expectations for our machines and we like to see them stay at or close to their turbo speeds under load. Myself and many others consider non-turbo boost clocks to be throttling, so the word gets thrown around a lot for PCs.

I can say that after undervolting and repasting, I get no throttling on my 8750H XPS 15. A 30 minute loop of cinebench is a nice stable 1200+ for me. Sure, that's after tweaking you may say, which is fair. When I conducted my review of the XPS 15 9570, I ran all tests completely stock; that same 30-minute loop of CB R15 averaged 1128 points. What are the stock MBP 15s scoring for that, again? 800-900?

The 9550 and 9560 did have thermal and VRM throttling issues, to be sure -- which is why I wrote several guides on how to undervolt, pad VRMs, and set up frequency profiles to fix the problems. I've been dealing with manufacturers pulling this ******** for years; I didn't accept it 6 years ago and I don't accept it now.

 

[mr.anthonyramos]

I am a frequent participant of the thread you just linked. the 9570 does not throttle below base clocks in my experience. You see, we tend to have high expectations for our machines and we like to see them stay at or close to their turbo speeds under load. Myself and many others consider non-turbo boost clocks to be throttling, so the word gets thrown around a lot for PCs.

I can say that after undervolting and repasting, I get no throttling on my 8750H XPS 15. A 30 minute loop of cinebench is a nice stable 1200+ for me. Sure, that's after tweaking you may say, which is fair. When I conducted my review of the XPS 15 9570, I ran all tests completely stock; that same 30-minute loop of CB R15 averaged 1128 points. What are the stock MBP 15s scoring for that, again? 800-900?

The 9550 and 9560 did have thermal and VRM throttling issues, to be sure -- which is why I wrote several guides on how to undervolt, pad VRMs, and set up frequency profiles to fix the problems. I've been dealing with manufacturers pulling this ******** for years; I didn't accept it 6 years ago and I don't accept it now.

Has anyone tried running cinebench under bootcamp on their Macs and see if the results come out diffeeently?

 

[Eason85]

I would be willing to try installing OSX in a VM on my xps 15 9570 and run some benchmarks, provided that it isn't super difficult to do so.

edit: seems latest macOS isn't easy to run with vmware?

 

[mr.anthonyramos]

I would be willing to try installing OSX in a VM on my xps 15 9570 and run some benchmarks, provided that it isn't super difficult to do so.

edit: seems latest macOS isn't easy to run with vmware?

Edit: misread this. So nvm. Lol.

 

[br0adband]

I would be willing to try installing OSX in a VM on my xps 15 9570 and run some benchmarks, provided that it isn't super difficult to do so.

edit: seems latest macOS isn't easy to run with vmware?

Works fine for me with Unlocker 2.1, no issues with High Sierra 10.13.6 so far. Of course I'm just doing this on an old ThinkPad T420s and not some brand spankin' new i9 powered machine, however.

 

[MakeAppleAwesomeAgain]

Intel updated its tool to 3.5.3.

 

[mr.anthonyramos]

Intel updated its tool to 3.5.3.

No difference apparently.