First 2018 MacBook Air Benchmark Shows Moderate Gains Over 2017 MacBook Air and MacBook

[Val-kyrie]

From a performance perspective, this MBA lacks the value proposition of the previous MBA:

(1) The new 7W CPU may have similar raw performance to a previous generation 15W, but the iGPU will hold back the performance of this machine. Real world benchmarks will make this evident and are much more reliable than synthetic benchmarks like GeekBench.

(2) Apple chose to neuter the performance of the MBA for this generation by changing from 7th generation 15W CPUs to 8th generation 7W CPUs. The new MBA should have received new 8th generation 15W CPUs. In so doing, this is no longer a MBA, but rather a glorified MB with an "Air" moniker to assuage the masses who don't understand specs.

(3) Comparable high end Windows ultrabooks like the Dell XPS 13 and Lenovo X1 Carbon are running 8th generation quad-core 15W CPUs with better iGPUs (Intel UHD 620 0r 630), which offer much better performance for prices similar to the new MBA and nTB MBP.

 

[FrostyF]

From a performance perspective, this MBA lacks the value proposition of the previous MBA:

(1) The new 7W CPU may have similar raw performance to a previous generation 15W, but the iGPU will hold back the performance of this machine. Real world benchmarks will make this evident and are much more reliable than synthetic benchmarks like GeekBench.

(2) Apple chose to neuter the performance of the MBA for this generation by changing from 7th generation 15W CPUs to 8th generation 7W CPUs. The new MBA should have received new 8th generation 15W CPUs. In so doing, this is no longer a MBA, but rather a glorified MB with an "Air" moniker to assuage the masses who don't understand specs.

(3) Comparable high end Windows ultrabooks like the Dell XPS 13 and Lenovo X1 Carbon are running 8th generation quad-core 15W CPUs with better iGPUs (Intel UHD 620 0r 630), which offer much better performance for prices similar to the new MBA and nTB MBP.

Funny, real world videos of it in use does not show any performance hit. Not everyone can afford a BMW. Go buy a Hyundai if that’s what you want. Comparing hardware spec for spec alone is not the full story. The customer service, design and software are thrown in, but they aren’t free to produce.

 

[- rob -]

. Comparing hardware spec for spec alone is not the full story. The customer service, design and software are thrown in, but they aren’t free to produce.

This is not what is being discussed here. No one is questioning the value of the Apple experience. This thread is about raw performance.

If I am running two pycharm projects, 18 chrome windows / tabs, 8 terminal sessions, docker, am streaming iTunes to a homepod, little snitch, a variety of other apps in the tray, excel, and then need to fire up Xcode in my 4th Space how will this computer handle this?

I’m asking because I do all of this on a 2015 11” mba 8gb and the computer handles it. Then I snap it closed throw into in my bag and go to a meeting across town.

I bought this new machine to reduce lag in these instances. Because after a few days of this or if my browser sessions grow to 25, I start seeing beach balls.

And it truly needs to handle this or the thing has to be returned.

Of course it has to have a great bug free OS Apple’s privacy values etc etc. That’s why the option for me is not some windows machine. The option for me is to stick w my 2015 11” and get the Mac mini for my primary docked office instead.

I am starting to realize what I really needed was the 2015 11” MBA with 16 GB of RAM. I could hold on with a non retina screen until a the legit arm MacBook arrives.

It just sucks Apple didn’t put a better gpu in and a stronger cpu. I’m still hoping it performs better than the spec sheet reads.
[doublepost=1541260369][/doublepost]

but the iGPU will hold back the performance of this machine. Real world benchmarks will make this evident and are much more reliable than synthetic benchmarks like GeekBench.

Which quantitative real world benchmarks do you think will best demonstrate performance?

 

[FrostyF]

This is not what is being discussed here. No one is questioning the value of the Apple experience. This thread is about raw performance.

If I am running two pycharm projects, 18 chrome windows / tabs, 8 terminal sessions, docker, am streaming iTunes to a homepod, little snitch, a variety of other apps in the tray, excel, and then need to fire up Xcode in my 4th Space how will this computer handle this?

I’m asking because I do all of this on a 2015 11” mba 8gb and the computer handles it. Then I snap it closed throw into in my bag and go to a meeting across town.

I bought this new machine to reduce lag in these instances. Because after a few days of this or if my browser sessions grow to 25, I start seeing beach balls.

And it truly needs to handle this or the thing has to be returned.

Of course it has to have a great bug free OS Apple’s privacy values etc etc. That’s why the option for me is not some windows machine. The option for me is to stick w my 2015 11” and get the Mac mini for my primary docked office instead.

I am starting to realize what I really needed was the 2015 11” MBA with 16 GB of RAM. I could hold on with a non retina screen until a the legit arm MacBook arrives.

It just sucks Apple didn’t put a better gpu in and a stronger cpu. I’m still hoping it performs better than the spec sheet reads.
[doublepost=1541260369][/doublepost]

Which quantitative real world benchmarks do you think will best demonstrate performance?

Nobody can assure you if it’s going to be fast enough based on your expectations. You have the option to try it and return it if it isn’t.

 

[- rob -]

Nobody can assure you if it’s going to be fast enough based on your expectations. You have the option to try it and return it if it isn’t.

If the GPU is at least equal to the 2015 11" air, and the CPU is equal to the air, I think I could live with that just to get the extra ram. But I'm not sure--I'm wondering if I should have just gone with a beefed up mini at the same price point ~1900.

 

[manu chao]

Exactly. The Air is not really a power user machine. Many will never run a game on it, or anything more taxing than a browser. The RAM and SSD make more of a difference when it comes to “feel.” If I were to complain, it would be the price hike, but even that was a long time coming.

Starting price:
2008 MBA 13": $1799
2009 MBA 13": $1499
2010 MBA 13:" $1299
2011 MBA 13:" $1299
2012 MBA 13:" $1199
2013 MBA 13:" $1099
2014 MBA 13:" $999
2015 MBA 13:" $999

2010 MB 13": $999
2011 MBP 13": $1199
2012 MBP 13": $1199

If you consider 13" the minimum useful screen size, then starting with the model year 2011, the cheapest Apple laptop was the 13" MBP at $1199. In 2012, it got joined by a $1199 13" MBA. It took another year for that to drop to $1099 and yet another year for it to drop to $999. For three model years (2011, 2012, 2013), there was thus no sub-$1000 Apple laptop. Or viewed differently, there were only two model years (2014, 2015), when there was annually updated 13" Apple laptop below $1000.

 

[FrostyF]

If the GPU is at least equal to the 2015 11" air, and the CPU is equal to the air, I think I could live with that just to get the extra ram. But I'm not sure--I'm wondering if I should have just gone with a beefed up mini at the same price point ~1900.

???? Now you’re comparing a desktop to a laptop???? I don’t understand. The choice here would either be an Air or a Pro.

 

[Falhófnir]

???? Now you’re comparing a desktop to a laptop???? I don’t understand. The choice here would either be an Air or a Pro.

Because god forbid anyone actually considers different options before buying a computer

 

[FrostyF]

Because god forbid anyone actually considers different options before buying a computer

No, it makes no sense. If you're ok with a desktop and don't need a laptop, that would be the clear choice every time. The Air sacrifices performance for weight and portability.

 

[Falhófnir]

No, it makes no sense. If you're ok with a desktop and don't need a laptop, that would be the clear choice every time.

It’s all about trade offs - in the case of the mini you lose the portability but gain more power, in the case of the air you lose power and some screen quality for battery life and Touch ID and the pro provides the best screen and is in the middle on power but you don’t get Touch ID and the battery life will/ might be noticeably shorter. I don’t see the problem with considering whether a desktop might be a better option if it’s in the same price category as the laptop you’re looking at getting.

 

[- rob -]

???? Now you’re comparing a desktop to a laptop???? I don’t understand. The choice here would either be an Air or a Pro.

I use the machine docked much of the time.

So the comparison for me isn't simply old MBA vs new MBA. it is keep the old MBA and get a new Mac Mini--OR stick with the new MBA I purchased as a single solution.

If the 2015 MBA 11" is roughly equal to the new MBA 13", I can keep using the 2015 and get a Mac Mini in the primary office--rather than dock this portable. I could deal with having the 2015 for for on the go for another year or two--until this new ARM machine drops.

But frankly I'd rather just have one great portable that docks and goes with me to other places. That's the machine I've been looking for since 2016.
[doublepost=1541267012][/doublepost]

Starting price:
2008 MBA 13": $1799
2009 MBA 13": $1499
2010 MBA 13:" $1299
2011 MBA 13:" $1299
2012 MBA 13:" $1199
2013 MBA 13:" $1099
2014 MBA 13:" $999
2015 MBA 13:" $999

2010 MB 13": $999
2011 MBP 13": $1199
2012 MBP 13": $1199

If you consider 13" the minimum useful screen size, then starting with the model year 2011, the cheapest Apple laptop was the 13" MBP at $1199. In 2012, it got joined by a $1199 13" MBA. It took another year for that to drop to $1099 and yet another year for it to drop to $999. For three model years (2011, 2012, 2013), there was thus no sub-$1000 Apple laptop. Or viewed differently, there were only two model years (2014, 2015), when there was annually updated 13" Apple laptop below $1000.

Looking at this, I still contend the 2015 MBA line was the best value / performance in laptops from Apple in the past several years.

 

[Val-kyrie]

This is not what is being discussed here. No one is questioning the value of the Apple experience. This thread is about raw performance.

If I am running two pycharm projects, 18 chrome windows / tabs, 8 terminal sessions, docker, am streaming iTunes to a homepod, little snitch, a variety of other apps in the tray, excel, and then need to fire up Xcode in my 4th Space how will this computer handle this?

I’m asking because I do all of this on a 2015 11” mba 8gb and the computer handles it. Then I snap it closed throw into in my bag and go to a meeting across town.

I bought this new machine to reduce lag in these instances. Because after a few days of this or if my browser sessions grow to 25, I start seeing beach balls.

And it truly needs to handle this or the thing has to be returned.

Of course it has to have a great bug free OS Apple’s privacy values etc etc. That’s why the option for me is not some windows machine. The option for me is to stick w my 2015 11” and get the Mac mini for my primary docked office instead.

I am starting to realize what I really needed was the 2015 11” MBA with 16 GB of RAM. I could hold on with a non retina screen until a the legit arm MacBook arrives.

It just sucks Apple didn’t put a better gpu in and a stronger cpu. I’m still hoping it performs better than the spec sheet reads.
[doublepost=1541260369][/doublepost]

Which quantitative real world benchmarks do you think will best demonstrate performance?

It simply depends on what you plan to do with it. There are few issues with business apps unless you get into complicated and extensive Excel spreadsheets. Bigger differences arise with photo editing, video editing, and when multi-tasking. Usually, the difference in performance between computers comes down to thermal constraints (which should not be a problem for the new MBA), IO (which can be limited by the CPU's frequency), and the GPU (which is the weakest point for the new MBA). There are several very good websites that post a variety of real world tests.

Oh--and RAM. Many apps will perform differently based on the amount of RAM available. Unlike the CPU and iGPU, the consumer has some choice in the amount of RAM when purchasing the MBA. More is always better.

 

[- rob -]

It simply depends on what you plan to do with it. There are few issues with business apps unless you get into complicated and extensive Excel spreadsheets. Bigger differences arise with photo editing, video editing, and when multi-tasking. Usually, the difference in performance between computers comes down to thermal constraints (which should not be a problem for the new MBA), IO (which can be limited by the CPU's frequency), and the GPU (which is the weakest point for the new MBA). There are several very good websites that post a variety of real world tests.

Oh--and RAM. Many apps will perform differently based on the amount of RAM available. Unlike the CPU and iGPU, the consumer has some choice in the amount of RAM when purchasing the MBA. More is always better.

I also think that RAM has been the biggest limiter on the 2015 MBA 11". If they had offered that in 16GB at the time, it would have been very, very powerful. There wasn't even a question in my mind that I'd put 16GB on the MBA I just ordered. I'm a bit surprised Apple is still selling 8GB. It feels like when they were selling their 16GB iPhones--stingy and not in the best interests of the customer even if it offers a lower pricepoint.

If my 2016 11" air had 16GB of ram, I think it would have been harder to justify this upgrade because much of the time I can't see the retina screen when I'm docked, nor use the touch ID.

So I am now very curious about how thermal / IO / GPU performance on this machine. Can you recommend any sites that do some of the tests you're describing? I'll run the tests when the machine arrives Wednesday and compare them directly w 2015 MBA 11".

I'm now starting to understand why Apple cut the 11" MBA in 2015. The price and performance was actually too good. Price wise it competed too directly with the retina macbook and was on track to perform better than the MBP line. If they had kept selling it, I don't think they would have sold many nTB MBP 13's as people opted for the old keyboard, old screen but won on processor / ram / graphics performance.

It isn't talked about, but I really do think the removal of the 11" MBA in 2015 left a gaping performance / price hole in Apple's portable line. I don't think it will be filled again until we see this ARM machine.

 

[FrostyF]

https://browser.geekbench.com/v4/cpu/10604887

 

[Penlocky]

https://browser.geekbench.com/v4/cpu/10604887

Comparing your 11" Air GPU HD6000 to a similar GPU to the new Air HD615 (similar to the new Air UHD617).

 

[FrostyF]

View attachment 801064

Comparing your 11" Air GPU HD6000 to a similar GPU to the new Air HD615 (similar to the new Air UHD617).

Wrong person

 

[darrens]

That's hardly surprising, when you consider Intel's CPUs haven't actually improved at all the last few years, except for when they increase the number of CPU cores, requiring a larger battery that doesn't make sense on a MacBook Air.

Yes - exactly the reason I don't see a need to "upgrade" my Early 2014 11" MacBook Air - my eyes aren't good enough to notice a retina screen!!

 

[Kimcha]

If I am running two pycharm projects, 18 chrome windows / tabs, 8 terminal sessions, docker, am streaming iTunes to a homepod, little snitch, a variety of other apps in the tray, excel, and then need to fire up Xcode in my 4th Space how will this computer handle this?

I’m asking because I do all of this on a 2015 11” mba 8gb and the computer handles it. Then I snap it closed throw into in my bag and go to a meeting across town.

I bought this new machine to reduce lag in these instances. Because after a few days of this or if my browser sessions grow to 25, I start seeing beach balls.

Your issue is RAM, not CPU or GPU.

The new MBA will be totally fine for that. And the increased SSD speed will also help in day to day operation even when RAM is running low.

The only concern I have is how well the new MBA will be able to handle 4K or even 5k displays. I’m very doubtful it can run both the internal and an external 5k displays at the same time without lag. But hopefully it will be able to do it in clamshell mode.

 

[FrostyF]

4393 7902
notbadobama.jpg
https://browser.geekbench.com/v4/cpu/10617631

 

[- rob -]

Your issue is RAM, not CPU or GPU.

The new MBA will be totally fine for that. And the increased SSD speed will also help in day to day operation even when RAM is running low.

The only concern I have is how well the new MBA will be able to handle 4K or even 5k displays. I’m very doubtful it can run both the internal and an external 5k displays at the same time without lag. But hopefully it will be able to do it in clamshell mode.

#iwantotobelieve.

Given how the iGpu has been disparaged so far, i would be amazed if it runs 5k smoothly in clamshell at all.
[doublepost=1541314038][/doublepost]

4393 7902
notbadobama.jpg
https://browser.geekbench.com/v4/cpu/10617631

It is possible that some of the scores were brought down because of background processes during setup of what must be embargoed press review machines or something else. If you look at all of the possible Macbook Air 2018 scores so far, there are outliers at the bottom too, specifically attributed to one user:

If you kick out these lower numbers, (the ones scoring below 4200 on single core) we're left with six readings so far with an average single core of 4300, multicore of 7865. Here's the spreadsheet, and a link to the most up to date list of benchmarks generally attributed to the new machine.

My primary comparison is the 2.2ghz 11 inch early 2015. I still believe this new machine is the natural upgrade path for people who used the 11" MBA.

It may be odd I'm focused on this, but given the direction that particular computer was headed price / performance wise, I really do feel like Apple did not give this machine a reasonable send off with only paths through the Macbook and 13" MBP. I think it somewhat shows in that the company never really explained why the 11" was going away or got behind the idea that 11" customers should now be 13" MBP customers.

Geekbench has the average for the 2015 11" at 3543 single and 6747 for multicore.

If the above averages for the new machine hold then the new MBA is roughly 20% faster single core, and 17% faster multicore over that 3.5 year old machine. Not anything to write home about but I think with increased RAM the overall feel could be a lot further ahead.

Frankly, I think the 11" should have just gotten regular yearly upgrades like the 13" ever since, and then the line should have been collapsed into this model this year. But then Apple would have made less progress pushing usb-c in macbook and reduced the number of people who tried out that silly touchbar.

 

[Dave245]

#iwantotobelieve.

Given how the iGpu has been disparaged so far, i would be amazed if it runs 5k smoothly in clamshell at all.
[doublepost=1541314038][/doublepost]
It is possible that some of the scores were brought down because of background processes during setup of what must be embargoed press review machines or something else. If you look at all of the possible Macbook Air 2018 scores so far, there are outliers at the bottom too, specifically attributed to one user:

If you kick out these lower numbers, (the ones scoring below 4200 on single core) we're left with six readings so far with an average single core of 4300, multicore of 7865. Here's the spreadsheet, and a link to the most up to date list of benchmarks generally attributed to the new machine.

My primary comparison is the 2.2ghz 11 inch early 2015. I still believe this new machine is the natural upgrade path for people who used the 11" MBA.

Geekbench has the average for the 2015 11" at 3543 single and 6747 for multicore.

If the above averages for the new machine hold then the new MBA is roughly 20% faster single core, and 17% faster multicore over that 3.5 year old machine. Not anything to write home about but I think with increased RAM the overall feel could be a lot further ahead.

The thing is that the average user ie the person who will buy this new Air, won't care or even know what these scores are, hell i don't really know what they mean, all i know is that i want a machine that will sit nicely with my 2012 iMac (that i will replace when Apple get around to updating) work well for long writing sessions, web browsing, email, Netflix, Youtube, some Pixelmator photo editing (nothing too heavy) and thats about it really.

Yes i get that for an extra couple hundred £ i could get the MacBook Pro with Touch Bar and have a lot more power, but for that i'd be trading battery life and weight, also the power probably wouldn't be used all that much. I'm not even sure if i use all the power of my 2012 iMac which still seems fine for the things i do on it, even when playing the games i like, such as Two Point Hospital and Tropico.

The new Macbook Air will probably sell well to students and the general consumer who just want a laptop, all they will probably see is that it has 12-13 hours battery life, Touch ID, is less volume, thinner lighter and now comes in new colours.

 

[M-1]

About the thermal fan in the MacBook Air..

What I'm most curious about is the 7W CPU and the inclusion of a thermal fan in the MacBook Air. Since the 12" MacBook has a 5W CPU and is able to run fanless, does this mean the MacBook Air is able to run mostly cool and silent without the fan? There's nothing I dislike more than hearing fans going off while working. I actually would prefer lower performance and no audible fan noise. The 12" MacBook has a anisotropic graphite sheet under the logic board for heat dispersion and I wonder if the MacBook Air includes it as well.

I really was waiting and hoping for a 12" MacBook spec upgrade and an additional usb-c port. In my opinion, I think it's the perfect light weight portable mac. It's crazy to me that Apple didn't even consider lowering the price of the 2017 MacBook considering it was released 17 months ago but I guess they want to keep the same pricing for the next MacBook update. However paying the same price for specs that are 17 months old today just doesn't feel right.

I am now considering the 2018 MacBook Air but can't stand fan noise. However if the MacBook Air is able to run mostly silent like the 12" MacBook but only rarely kicks in the fan for highly intensive tasks and sustained performance to stay cool, that's a design I can accept. I really hope the MacBook Air is a beefed up MacBook with 95% silent operation. That will be my deciding factor.

In theory a 7W CPU doesn't need a fan like a 15W CPU in a MacBook Pro does. I wonder if Thunderbolt 3 is the reason for the fan. Would thunderbolt 3 be possible on the next 12" MacBook with fanless design? I wonder if that's the main reason for the single port and exclusion of Thunderbolt 3 so far.

 

[Kimcha]

Since the 12" MacBook has a 5W CPU and is able to run fanless, does this mean the MacBook Air is able to run mostly cool and silent without the fan?

I am using the 2017 tb-MBP and I rarely hear the fan. Just now I was doing some light photoshop work and have had safari and my usual apps running in the background. I couldn't hear the fan in a quiet room. It's currently running at 1,245 RPM, but to me that's inaudible.

Even holding the MBP to my ear I can't hear it.

 

[- rob -]

About the thermal fan in the MacBook Air..

The fan on the 2015 11" doesn't go on that often. I don't like the fan but it doesn't bother me when it does come on. When it comes on it it is usually pretty easy to identify why.

About the thermal fan in the MacBook Air..I guess they want to keep the same pricing for the next MacBook update.

If the next macbook update is ARM based then I am guessing the price could actually go up.

 

[agentsmithone]

Excited to upgrade my MacBook Pro (Late 2013) [Single Core: 4193] to the MacBook Air (Late 2018) [Single Core: 4248]. I'm happy to get a lighter Mac with Thunderbolt 3, similar single core performance, longer battery life, I have to check on the 'improved' keyboard. Buying up my USB-C device cables now lol