2018 i9 MacBook Pro throttling discussion [merged]

[dan9700]

I have just wasted £3300 on a machine thats slower than my 2017 i sold for £2150 apple you have done it i wanna re think everything with apple now im coming to a end

 

[Eason85]

I have just wasted £3300 on a machine thats slower than my 2017 i sold for £2150 apple you have done it i wanna re think everything with apple now im coming to a end

It pays to do research on any purchase. At least you got a decent price on your old MacBook.

 

[dan9700]

It pays to do research on any purchase. At least you got a decent price on your old MacBook.

Yea now i think i made a mistake i never thought apple make a top flagship thata slower what exactly i paying to wnjoy fron new one

 

[Keebler]

Yea now i think i made a mistake i never thought apple make a top flagship thata slower what exactly i paying to wnjoy fron new one

Don't you have 15 or 30 days to return your purchase? I believe we have that here in Canada. I'd look into it if I were you.

I ordered the 2.6 6 core. Hoping it doesn't have the same issue.

Good luck!

Cheers,
Brian

 

[sennomulo]

Lisa from MobileTechReview has just released her review of the 2018 MBP, discussing the thermal issues at length:

In the video, she recommends a free app called "Macs Fan Control" (https://www.crystalidea.com/macs-fan-control) which gives you manual control over your fan speeds. I would be interested to hear how well this app works for owners of the 2018 MBP.

 

[mr.anthonyramos]

Yea now i think i made a mistake i never thought apple make a top flagship thata slower what exactly i paying to wnjoy fron new one

Hi Dan. I have the 2018 i9. I think you need to read a little more and wait for more data before getting into an absolute resolution here regarding your i9. For my workflow, that would usually take around 15 minutes to process on my maxed out 2017 i7, took me just over 5 minutes with my i9. Compared with my colleagues, 2018 i7, It took his around 7 minutes to do the exact same process. I won't go into more detail as you can search my other posts.

There are other tests that show that the i9 is still the most capable MacBook Pro in the market BUT what the issue is, and is my issue as well with Apple is that they would advertise the i9 to being able to keep a 2.9 clock base and boost up to 4.8, whereas underload, we are getting a 2.6-2.7 load with boasts that don't go higher than 4.3. Which to me, is a deceiving number to bring out without a disclaimer about the cooling system limitations on the MacBook Pro Chassis.

With that said, nevertheless for my use, I'm happy because it boils down to me actually getting the best performance for my need beating both the 2017 and 2018 i7. Now, those using other applications and what not, I have no idea why, but it seems like a whole lot of other stuff happening. Now whether the small bump is worth the additional cost of the i9, that is totally up to you.

So here is a tip for you, and others who are considering returning or cancelling your i9. Take it, use it to hell, if it doesn't satisfy you, return it. I was considering returning it when I saw it was throttling but when I did real life tests to see how it compared to my older mac, I found out it performed better. So, I am keeping it.

 

[zshane1125]

Hi Dan. I have the 2018 i9. I think you need to read a little more and wait for more data before getting into an absolute resolution here regarding your i9. For my workflow, that would usually take around 15 minutes to process on my maxed out 2017 i7, took me just over 5 minutes with my i9. Compared with my colleagues, 2018 i7, It took his around 7 minutes to do the exact same process. I won't go into more detail as you can search my other posts.

There are other tests that show that the i9 is still the most capable MacBook Pro in the market BUT what the issue is, and is my issue as well with Apple is that they would advertise the i9 to being able to keep a 2.9 clock base and boost up to 4.8, whereas underload, we are getting a 2.6-2.7 load with boasts that don't go higher than 4.3. Which to me, is a deceiving number to bring out without a disclaimer about the cooling system limitations on the MacBook Pro Chassis.

With that said, nevertheless for my use, I'm happy because it boils down to me actually getting the best performance for my need beating both the 2017 and 2018 i7. Now, those using other applications and what not, I have no idea why, but it seems like a whole lot of other stuff happening. Now whether the small bump is worth the additional cost of the i9, that is totally up to you.

So here is a tip for you, and others who are considering returning or cancelling your i9. Take it, use it to hell, if it doesn't satisfy you, return it. I was considering returning it when I saw it was throttling but when I did real life tests to see how it compared to my older mac, I found out it performed better. So, I am keeping it.

Or you could return it and get the base i7 2.2 with the same performance, and save $400.

 

[Aea]

Or you could return it and get the base i7 2.2 with the same performance, and save $400.

I was typing the exact same thing. That's my beef with this. The site (https://www.apple.com/macbook-pro/) heavily advertises performance and specs. It stands to reason that you should get better performance when you add $400 to the price.

The caveat emptor attitude that some posters have on here is a bit disappointing. Certainly the flagship model from the best laptop manufacturer in the world shouldn't come with a long list of caveats regarding the basic performance Apple is advertising so heavily.

 

[ashcairo]

Guys, he literally just said that his 2018 i9 takes 5 minutes, his mate's 2018 i7 takes 7 minutes, and his 2017 took 15 minutes.

So he IS getting better performance for his use case - not sure what it is - but hopefully it's not frying an egg.

 

[mr.anthonyramos]

Or you could return it and get the base i7 2.2 with the same performance, and save $400.

I'm getting better results with my i9 compared with my colleagues 2018 i7 when processing 200 ARW files. And if you also missed what I said, it takes up nearly double the time with my 2017 MacBooks Pro for a similar workflow situation which is why I said I am happy. IF my work flow results are equal to the 2018 base model then I would definitely return the machine and get the base and give my wife and I a nice dinner date somewhere and a little more.

 

[zshane1125]

make senses now...apple is throwing some expensive smoke grenade at us man, just gotta spray and pray now.

 

[Aea]

make senses now, this is now a bunch of smoke.

Guys what temperature would these things physically start to melt at?

 

[anshuvorty]

Guys what temperature would these things physically start to melt at?

I think anything close to 100 degrees C.

 

[zshane1125]

Really debating now between the returning the i9 512GB and getting the i7 2.2 1TB, or keep the i9 for 2 weeks to see of the smoke grenades dissipates a bit.

 

[ashcairo]

Really debating now between the returning the i9 512GB and getting the i7 2.2 1TB, or keep the i9 for 2 weeks to see of the smoke grenades dissipates a bit.

1TB will be way more useful, opens your computer up to doing more things like installing Windows Bootcamp and video editing. Everyone always needs more space, so the person in the future who buy's the machine off you will appreciate it too.

 

[mr.anthonyramos]

make senses now, this is now a bunch of smoke.

Now what I am curious about is how more real world tests pan out for the new macs.

Really debating now between the returning the i9 512GB and getting the i7 2.2 1TB, or keep the i9 for 2 weeks to see of the smoke grenades dissipates a bit.

I'd say keep it for another two weeks. What will you lose? Nothing much. Again, use it to hell, and if it doesn't do what you want to do then return it, or maybe with what you need to be doing with your mac can be done on the base model, get a refund and go for that.

 

[joebclash]

Look at the comment from macworld. This is starting to snowball. Apple needs to get infront of this.

 

[zshane1125]

Now what I am curious about is how more real world tests pan out for the new macs.


I'd say keep it for another two weeks. What will you lose? Nothing much. Again, use it to hell, and if it doesn't do what you want to do then return it, or maybe with what you need to be doing with your mac can be done on the base model, get a refund and go for that.

Exactly, my thinking was to get the max specced machine which I did without the 4 TB storage. Since you cannot upgrade the CPU, the RAM, and the GPU later, but you CAN get external storage, it's the best option for my intentions. Now you might argue that the i9 throttle bad, but it's still the fastest overall CPU.

Now, there is a possibility that Apple will address that with either a thermal paste service or some kinda firmware / software update, and the i9 will reach its max potential, it would be fantastic. But, if I went ahead and return it and got the 2.2 i7, I might regret it later if the improvements roll out, but I would never regret not having larger storage, because I can always get more.

 

[anshuvorty]

Look at the comment from macworld. This is starting to snowball. Apple needs to get infront of this.

But you know they won't. They will silently release a cumulative update that among "security and bug fixes also includes firmware changes that ups the fan curve or something. History has shown us how Apple usually responds to these kinds of debacles.

 

[mr.anthonyramos]

Exactly, my thinking was to get the max specced machine which I did without the 4 TB storage. Since you cannot upgrade the CPU, the RAM, and the GPU later, but you CAN get external storage, it's the best option for my intentions. Now you might argue that the i9 throttle bad, but it's still the fastest overall CPU.

Now, there is a possibility that Apple will address that with either a thermal paste service or some kinda firmware / software update, and the i9 will reach its max potential, it would be fantastic. But, if I went ahead and return it and got the 2.2 i7, I might regret it later if the improvements roll out, but I would never regret not having larger storage, because I can always get more.

A little off topic but did you get the 2TB? I struggled with the decision and ended up getting the 1TB. My personal 2014 MacBook Pro has 1TB and after nearly 4 years I have a total of 300GB of files. So I’m thinking, 1TB will do and after I clear up some clutter I might get back another 60-80GB. I’m thinking maybe just maybe I’ll send this back and then go 2TB.

 

[zshane1125]

A little off topic but did you get the 2TB? I struggled with the decision and ended up getting the 1TB. My personal 2014 MacBook Pro has 1TB and after nearly 4 years I have a total of 300GB of files. So I’m thinking, 1TB will do and after I clear up some clutter I might get back another 60-80GB. I’m thinking maybe just maybe I’ll send this back and then go 2TB.

I have 512GB, even I think that's excessive, I do everything in cloud and external storage.

 

[semajm85]

Lisa from MobileTechReview has just released her review of the 2018 MBP, discussing the thermal issues at length:

In the video, she recommends a free app called "Macs Fan Control" (https://www.crystalidea.com/macs-fan-control) which gives you manual control over your fan speeds. I would be interested to hear how well this app works for owners of the 2018 MBP.

so following this, I personally set the fans to the read the sensor off the "CPU PECI" sensor and the fans to kick in @ 50 deg C.

Cinebench( fans @ auto) Peak temp @ 95 Deg C
Test 1 : 859
Test 2 : 808 (Fans Kick in)
Test 3 : 824
Test 4 : 767
Test 5 : 758
Test 6 : 839
Test 7 : 744
Test 8 : 683(starts to spike all over the place)
Test 9 : 681
Test 10 :745


Cinebench (with mac fan control) Peak Temp @ 85 Deg C
Test 1 : 892
Test 2 : 888
Test 3 : 898
Test 4 : 875
Test 5 : 826
Test 6 : 823
Test 7 : 873
Test 8 : 830
Test 9 : 793
Test 10 :864

then I jump over to prime 95 and i see a mark improvement.
in test 1 & 3, no change, rock solid 2.9ghz across all cores.

when we hit test 2 & 4, thats where it gets interesting. With the fans on auto, the throttle starts immediately, however using FAN SMC Controller, it hits much later, almost at the end of the test. I also noticed that it goes crazy as the thermal draw exceeds 45w and goes up, then it agressively throttles.

i'm quite happy to see that this is throttling can be mitigated to a certain degree. we will have to test more and perhaps see what happens over the course of the next few days/weeks.

will wait for more people to chime in with their observations.

update: included my prime95 screenshots

 

[anshuvorty]

so following this, I personally set the fans to the read the sensor off the "CPU PECI" sensor and the fans to kick in @ 50 deg C.

Cinebench( fans @ auto) Peak temp @ 95 Deg C
Test 1 : 859
Test 2 : 808 (Fans Kick in)
Test 3 : 824
Test 4 : 767
Test 5 : 758
Test 6 : 839
Test 7 : 744
Test 8 : 683(starts to spike all over the place)
Test 9 : 681
Test 10 :745


Cinebench (with mac fan control) Peak Temp @ 85 Deg C
Test 1 : 892
Test 2 : 888
Test 3 : 898
Test 4 : 875
Test 5 : 826
Test 6 : 823
Test 7 : 873
Test 8 : 830
Test 9 : 793
Test 10 :864

then I jump over to prime 95 and i see a mark improvement.
in test 1 & 3, no change, rock solid 2.9ghz across all cores.

when we hit test 2 & 4, thats where it gets interesting. With the fans on auto, the throttle starts immediately, however using FAN SMC Controller, it hits much later, almost at the end of the test. I also noticed that it goes crazy as the thermal draw exceeds 45w and goes up, then it agressively throttles.

i'm quite happy to see that this is throttling can be mitigated to a certain degree. we will have to test more and perhaps see what happens over the course of the next few days/weeks.

will wait for more people to chime in with their observations.

Great insight and a great roadmap that Apple could implement in the next firmware update. Nicely laid out. Great job.

 

[Mathematig]

I realized that I had another background process running during that benchmark... Did a new one. Thermals and throttling below. Counted this time - throttling started within about 9 seconds.

View attachment 771196

Edit: So it pretty much maintained at least the base 2.2ghz clock speed... I don't know what to think of that... Others?

This seems a little better than the i9 2.9GHz one.
[doublepost=1531980491][/doublepost]

I am in a similar situation. Currently, I have a custom built pc with a GTX 1080 for 3d creation, primarily game art. It's great, no issues. I purchased an MSI 4k gaming laptop to use a mobile workstation. Performance is great, but I could never get over the build quality and how heavy it was. 17" in a 15" chassis. It was my first non Apple laptop, ever. Big shock, and I don't say that as a fan boy.

My concern is this. I prefer a small laptop. The 13" would be my poison. I'm wondering how long 16 GB of RAM will last me for content creation. The biggest challenge I see is with texture maps. If they stay at 4k for a few years, I suspect I'll get great life out of the 13" paired with an eGPU while working in Substance.

Thanks to a member who sent me a benchmark of their new 13", I can say that the new quad core is considerably faster than my 7th gen 7700hq in my MSI.

If I spec out the 13" the way I want it, it's $3000 vs $3500 for a base cpu, 32gb ram 15". On paper the 15" sounds like a dream, but, after seeing these early reports of thermal throttling, I am strongly leaning towards the 13". I want something light weight, powerful, and can just be plugged into an eGPU when at a desk. I will wait to see more reviews and go see these machines in person. My last MBP was a 2015 15". I want to see if these machines are lighter.

Lastly, is it true OS X handles ram usage better than Windows 10?

I also set my desktop with GTX1080+32GB Memory. I have used it for one year, and I think as long as your memory is big enough, it does not matter.
Bascially, I use my desktop to run a lot of code, and did not see that memory will affact the performance. So I think 32GB memory is enough right now.

 

[Feenician]

I realized that I had another background process running during that benchmark... Did a new one. Thermals and throttling below. Counted this time - throttling started within about 9 seconds.

View attachment 771196

Edit: So it pretty much maintained at least the base 2.2ghz clock speed... I don't know what to think of that... Others?

It averaged well above 2.2. The very large spikes are indicative of a cooling system that’s a little too eager to slow down/quiet down. A small amount of extra fan speed (and noise) should smooth that out a bit. At least the 2.2 might be fixed up fairly easily in the constraints of that cooling system.