[Very Possibly Solved] Why the Macbook Pro 2016 is slower than 2015?

[Kris.K]

Answer:
The memory and cpu on the new Macbook Pro is holding the machine back. Apple wanted to preserve battery life so they made some compromises at the expense of performance.

1- Memory is slower.
2- CPU has a lower max clock speed.
3- Skylake runs faster on DDR4 not DDR3.

**important point**
Yes they are different chipsets and there are many variables which is why this is just a starting point and possibly solved not actually solved. Over to you world!

Why?
Apple wanted to keep the battery life to 10hours as we have come to expect. Skylake doesn't have support for low powered DDR4 so they had to go with low powered DDR3. However Skylake also doesn't officially support 2133Mhz DDR3 so they had to lower the timings and do what they could to get the higher frequency running and keep the power usage meeting requirements they set. This is why you also can't get 32GB of ram, it will use double the power.

Proof: 1. Memory is slower
The memory on the 2016 model is CL16, the memory on the 2015 is CL11.

In time this means:
2015 DDR3 1600Mhz CL11 latency: 13.75 nanoseconds
2016 DDR3 2133Mhz CL16 latency: 15 nanoseconds
Source: DDR3 RAM ineffectiveness sheet.
(lower is better.)

Latency timing screenshots in CPU-z:

2016
(Source Macrumors member deadworlds Thank you!!)

2015
(Notebookcheck.info Not the most credible source so if someone can do another test that would be great)

Memory Performance:
2015 memory latency Geekbench 4: 12.5 Moperations/sec
2016 memory latency Geekbench 4: 7.64 Moperations/sec

2015 memory bandwidth Geekbench 4: 25.1 GB/sec
2016 memory bandwidth Geekbench 4: 24.9 GB/sec

2015 memory copy Geekbench 4: 14.4 GB/sec
2016 memory copy Geekbench 4: 13.8 GB/sec

Remember this is DDR3 1600Mhz vs 2133Mhz so the numbers shouldn't be this tight.

2015 source: https://browser.primatelabs.com/v4/cpu/942360
2016 source: https://browser.primatelabs.com/v4/cpu/1047696

Proof: 2- CPU has a lower max clock speed
2016 max speed: 3.8Ghz
2015 max speed: 4Ghz
Source: http://ark.intel.com/compare/88972,83503

*Note* I am not saying that the 2015 CPU is a faster CPU. I am saying that in combination with the ram compromises it makes the 2016 i7 perform slower.*

Proof: 3- Skylake runs faster on DDR4 not DDR3.

Source: http://www.corsair.com/en/blog/2015/november/ddr3-vs-ddr4


Suggestion to Apple:
Give your customers the option to choose. Bring out a fix.
When plugged in, we are happy for lower timing if we get faster speeds.
If you want 32GB you will have lower battery performance.

Thoughts moving forward:
Apple is going to release a new Macbook Pro in 2017 on a new Intel chipset which will support LPDDR4. This will be the must faster Macbook and what we have come to expect.

*****Requests to the community*****:
Someone who has a completely upgraded 2015 Macbook Pro 15" Retina, can you please do a CPU-z and get us the memory details?

Also if we can do some memory/cpu benchmarks. Open to suggestion as to what would confirm the above theory.

Special thanks:
Thank you to members of the forum, specifically deadworlds for doing the test to get us the data, without him we wouldn't have been able to confirm anything. And Peter the CTO of ServersAustralia for helping me ask the right questions and identifying where the issues were as it didn't make sense why it was slower.

 

[Samuelsan2001]

Dude give it up the new machines are faster in benchmarks overall, quite a lot faster in real life usage, thinner smaller and lighter and still have a class equalling battery life.

Real life testing has shown them to be faster for just about anything you throw at them what is the problem??

 

[raymond7]

Dude give it up the new machines are faster in benchmarks overall, quite a lot faster in real life usage, thinner smaller and lighter and still have a class equalling battery life.

Real life testing has shown them to be faster for just about anything you throw at them what is the problem??

The new machines may feel faster because of the faster SSD's. If you would give both the same SSD's the 2016 could be actually slower based on this "research".

 

[sinoka56]

Dude give it up the new machines are faster in benchmarks overall, quite a lot faster in real life usage, thinner smaller and lighter and still have a class equalling battery life.

Real life testing has shown them to be faster for just about anything you throw at them what is the problem??

equal battery life?

 

[Jaekae]

Why does all tests show that the new macbook is faster then

 

[Kris.K]

Dude give it up the new machines are faster in benchmarks overall, quite a lot faster in real life usage, thinner smaller and lighter and still have a class equalling battery life.

Real life testing has shown them to be faster for just about anything you throw at them what is the problem??

This isn't aimed to complain about it, it's aimed to explain it.

They are faster because they are compensating from other parts of the machine. Also I'm actually very happy with my purchase (all upgraded 2016 Macbook Pro 15").

With this said, I think it's important to see what went on as you can start to see Apple's constraints and what they'll want to do in the next Macbook pro. (More efficient and faster CPU/Ram combination which will allow more ram which if you're about to drop 3-4k on a laptop that should last you a while (more than a year), this kind of information is relevant don't you think?)
[doublepost=1479727036][/doublepost]

Why does all tests show that the new macbook is faster then

Faster GPU, Faster SSD? I'm sure they have done some nice improvements on the chipset too and made things faster there.

This is just pointing out the 2 parts that held this back from being blown away performance.

 

[dk808]

The new machines may feel faster because of the faster SSD's. If you would give both the same SSD's the 2016 could be actually slower based on this "research".

But they dont have the same ssd... therefore the 2016 is faster

 

[Kris.K]

But they dont have the same ssd... therefore the 2016 is faster

The new 2016 has PCI-E which allows it to run faster. 2015 doesn't have that.

The entire PC works in combination with each other, one bottle neck and the entire thing suffers.

 

[leman]

Once our machines will arrive I will test their CPU/RAM performance on real-world tasks. Unfortunately, this is unlikely to happen before January

 

[PBG4 Dude]

The new 2016 has PCI-E which allows it to run faster. 2015 doesn't have that.

The entire PC works in combination with each other, one bottle neck and the entire thing suffers.

Apple computers have been using PCI-E since 2013.

 

[maflynn]

The new machines may feel faster

From the reviews, I've read, the machines are faster, regardless of the reason, the machines are doing tasks faster then the prior models and isn't that what people want?

I thought the GPU was faster as well.

 

[leman]

Proof: 3- Skylake runs faster on DDR4 not DDR3.
Source: http://www.corsair.com/en/blog/2015/november/ddr3-vs-ddr4

Actually, your link shows exactly the opposite. According to it, at 2133Mhz, the DDR3 faster than DDR4. Sure, faster DDR4 is faster, but laptops usually come with 2133Mhz RAM.

At the same time, Apple doesn't use DDR3. It uses LPDDR3. And it is entirely possible that that RAM has lower bandwidth. I'd like to see other benchmarks on it than Geekbench, but so far nobody has delivered them.

 

[Kris.K]

Actually, your link shows exactly the opposite. According to it, at 2133Mhz, the DDR3 faster than DDR4. Sure, faster DDR4 is faster, but laptops usually come with 2133Mhz RAM.

At the same time, Apple doesn't use DDR3. It uses LPDDR3. And it is entirely possible that that RAM has lower bandwidth. I'd like to see other benchmarks on it than Geekbench, but so far nobody has delivered them.

Yes at the lowest point DDR4 can go, DDR3 is faster than it lol...

However DDR4 has alot of room and Skylake was built to run on it, so nearly 50% increase in bandwidth at the top end which is what I meant.

 

[aevan]

Suggestion to Apple:
Give your customers the option to choose. Bring out a fix.
When plugged in, we are happy for lower timing if we get faster speeds.
If you want 32GB you will have lower battery performance.

Just wanted to comment this - a lot of people think Apple should give them "a choice" between performance/weight/battery life. While some people would indeed benefit from such a choice (not debating that), Apple will never do that and it would be a bad business decision for them. The reason is simple: a lot of people have no idea how much RAM or performance they need. People still perceive "bigger numbers" as better value. They would walk into the store, get the most specs they can, then nag how their battery life is awful or leave their notebook at home because it's too heavy. Customer satisfaction would drop, we would have even more criticism, and this time from the majority of users, not just a small group of vocal people.

Their approach is to make the combination of power, portability and battery life they think is most optimal. If you think they haven't considered your 'suggestions' - you're wrong. They have. This (the new MBPs) is what they decided to do.
[doublepost=1479730504][/doublepost]

(More efficient and faster CPU/Ram combination which will allow more ram which if you're about to drop 3-4k on a laptop that should last you a while (more than a year),

This laptop will last you at least 3 years, when it comes to performance.

 

[Kris.K]

Just wanted to comment this - a lot of people think Apple should give them "a choice" between performance/weight/battery life. While some people would indeed benefit from such a choice (not debating that), Apple will never do that and it would be a bad business decision for them. The reason is simple: a lot of people have no idea how much RAM or performance they need. People still perceive "bigger numbers" as better value. They would walk into the store, get the most specs they can, then nag how their battery life is awful or leave their notebook at home because it's too heavy. Customer satisfaction would drop, we would have even more criticism, and this time from the majority of users, not just a small group of vocal people.

I 100% agree with all that you've said. Doesn't mean I won't try / wishful thinking lol.

Their approach is to make the combination of power, portability and battery life they think is most optimal. If you think they haven't considered your 'suggestions' - you're wrong. They have. This (the new MBPs) is what they decided to do.

This laptop will last you at least 3 years, when it comes to performance.

Yes it will, but why do people upgrade their phones each year? It isn't because its slow or lacking features.

 

[Charlesje]

It is a bit misleading to compare cas latency timings as they are measured in clock cycles, thus depending on the actual clock frequency. Actual time differences are indeed 13,75 vs 15 nanosec, which amounts to a 9% difference. However in the same cpu-z memory tab row refresh timings are faster on the 2016 (7%). We need realworld memory performance tests as other members request.

This doen't necessarily mean this memory would bottleneck cpu performance. Actually the 2016 cpu performs really well in the very limited tests we have yet (people interested in proving extra tests?). I don't know why you state notebookcheck is not a credible reviewer -i'm not an expert, but they seem to provide the most extensive reviews of notebooks- but the mid tier macbook pros for 2015 and 2016 on the ars technica review shows the following cinebench r15 cpu (this is purely cpu based render math on four cores) results:

2015 4870hq (2,5 Ghz) mbp: 577
2016 6820hq (2,7 ghz) mbp: 696
(http://arstechnica.com/apple/2016/11/the-2016-13-and-15-inch-touch-bar-macbook-pros-reviewed/)

This amounts two a signifant 20% cpu realworld performance under load increase for the mid-tiers.
My guess (maybe just hope ) is that the new maxed-out mps will have around 15% advantage in cpu perf. compared to their 2015 counterparts.

 

[leman]

Yes at the lowest point DDR4 can go, DDR3 is faster than it lol...

However DDR4 has alot of room and Skylake was built to run on it, so nearly 50% increase in bandwidth at the top end which is what I meant.

Are we talking about now or about some potential future? Because right now, laptops are commonly shipped with DDR 2133. I think its a bit weird to criticise DDR3 RAM now based on the fact that one can potentially use faster DDR4 RAM in the future. Now, if Apple would still use 2133 RAM when everyone else moves to 2666 DDR4 or so, then yes, we definitely have a problem. But right now? Even MS is using LPDDR3 in their new laptops.

 

[Charlesje]

A question for the experts: anybody knows if intel turboboost affects memory performance on the previous gen notebooks? I'm wondering of this could explain part of the geekbench moperations/sec score.

 

[edgeQV]

*****Requests to the community*****:
Someone who has a completely upgraded 2015 Macbook Pro 15" Retina, can you please do a CPU-z and get us the memory details?

My late 2013 baseline 15'' rMBP has CL11, so the 2015 should have too. Unfortunately as you go up with freqs latency increases too...

 

[MacBH928]

Their approach is to make the combination of power, portability and battery life they think is most optimal. If you think they haven't considered your 'suggestions' - you're wrong. They have. This (the new MBPs) is what they decided to do.

If battery is the issue, how about when its connected to the electricity it kicks in high power on CPU and GPU and not limit the CPU potential just because someone is on the go. I believe they can do this easily through software and users will not even be aware of it.

 

[leman]

If battery is the issue, how about when its connected to the electricity it kicks in high power on CPU and GPU and not limit the CPU potential just because someone is on the go. I believe they can do this easily through software and users will not even be aware of it.

They are already using fastest available current-gen CPUs :/ Really, I am sometimes wondering about how people think. And then 'they can do this easily through software'. HOW?

 

[Kris.K]

They are already using fastest available current-gen CPUs :/ Really, I am sometimes wondering about how people think. And then 'they can do this easily through software'. HOW?

Most motherboards allow you to do settings via the bios. Then there are in OS software that allows you to also edit and tweak performance settings. It's quite standard. However not as much on mac if at all.
[doublepost=1479733471][/doublepost]

Are we talking about now or about some potential future? Because right now, laptops are commonly shipped with DDR 2133. I think its a bit weird to criticise DDR3 RAM now based on the fact that one can potentially use faster DDR4 RAM in the future. Now, if Apple would still use 2133 RAM when everyone else moves to 2666 DDR4 or so, then yes, we definitely have a problem. But right now? Even MS is using LPDDR3 in their new laptops.

I'm talking about if DDR4 ram was used with a skylake CPU.

My late 2013 baseline 15'' rMBP has CL11, so the 2015 should have too. Unfortunately as you go up with freqs latency increases too...

You're awesome man! I will add this to the original post. Question, what CPU are you running? Can you screencap that too?

 

[leman]

Most motherboards allow you to do settings via the bios. Then there are in OS software that allows you to also edit and tweak performance settings. It's quite standard. However not as much on mac if at all.

I repeat, they are already using the highest-performance CPUs. How do you suggest they tweak it further? By underclocking the CPU when running on battery? Is that really something that would make the users happy?

I'm talking about if DDR4 ram was used with a skylake CPU.

If Apple were using DDR4 whey would, just as every other laptop manufacturer, use DDR4 2133, which according to your link is slower than DDR3 2133. How would that help the performance problem?

I just don't understand your logic. You are linking a test which shows that fast desktop DDR4 memory is faster. Then you critisize Apple for using DDR3 in their laptops, even though it is — again, according to your link — faster than the DDR4 currently used in other laptops. So what exactly is your point? That Apple should use desktop RAM? Or that they should go back to SO-DIMMs and use DDR4 so that users can upgrade to faster DDR4 SO-DIMMs once they become more widespread?

 

[Kris.K]

Leman man, i'm not suggesting anything! This post is to explain why it's slow. Give an answer to the question only.

When I said how it can be done I was explaining to you how it's done. Not that we should now go ahead and get it done. You asked HOW.

Also remember it's not just about the Mhz, its a combination. The linked test is DDR3 CL9 and DDR4 CL15.

Leman, I am only sharing what I've found to be the most plausible reason so far. Welcome to take it or leave it but at this point I am not going to try and persuade you or anyone else. I just share what I found and up to you to make your own decision about it.

 

[Samuelsan2001]

Leman man, i'm not suggesting anything! This post is to explain why it's slow. Give an answer to the question only.

When I said how it can be done I was explaining to you how it's done. Not that we should now go ahead and get it done. You asked HOW.

Also remember it's not just about the Mhz, its a combination. The linked test is DDR3 CL9 and DDR4 CL15.

Leman, I am only sharing what I've found to be the most plausible reason so far. Welcome to take it or leave it but at this point I am not going to try and persuade you or anyone else. I just share what I found and up to you to make your own decision about it.

And leman is explaining why what you found has been interpreted by you wrong and would make no difference to apples decisions. You don't want to accept that Apple made the most sensible decision given the current technology and their design parameters, thats fine but it doesn't mean you have made a single interesting or even relevant point.