Surface Book vs. MacBook Pro: It isn't twice as fast. It's three times as fast (Ruh-roh!)

PC World ^ | Oct 22, 2015 | Gordon Mah Ung

[dayglored]

Of course we had to pit the Surface Book vs. the MacBook Pro. It’s like Ford vs. Chevy, or Coke vs. Pepsi. Each side has its diehard fans, plus others who just want to know which is better.

Microsoft claims its new Surface Book is “twice” as fast as its equivalent MacBook Pro. Well, we ran some benchmarks, and hate to say it, but Microsoft lied. The Surface Book isn’t twice as fast.

It’s three times as fast. Read on for the details.

What Microsoft meant

First, let’s clarify what Microsoft meant when it said the Surface Book would smoke the MacBook Pro. The company specifically means the MacBook Pro 13 inch model. That’s a very important distinction, because the MacBook Pro 15 is a different class of laptop. It’s larger, heavier, and packs a quad-core CPU and fairly beefy AMD discrete graphics. For Microsoft to say the Surface Book out performed the MacBook Pro 15 would be absurd. It would be like Apple saying the MacBook Pro 15 outperforms, oh, an MSI GT 80 Titan SLI laptop in gaming. So the target for the Surface Book is the MacBook Pro 13. Microsoft even compares the two directly in its reviewers guide.

How I tested

...

(Lots and lots of detail at the source...)

[minnesota_bound]

The advantage of the Surface Pro is you have Windows 10 and android and can draw and paint as well. It is expensive!

There are several laptops out there for several hundred less that. They have touch. They do not have the android ability or draw & paint.

[minnesota_bound]

#18 Hillary Clinton really like using the Surface Pro to write her emails on : )

[dayglored]

> in my experience after 2-3 years of use the Windows boxes will slow down substantially due to “creeping crud”.

While that's true to a degree in nearly any consumer desktop OS, it is especially true of Windows because of its accumulation of crap that it can't get rid of, particularly in the Windows Registry and other hashed databases.

And while it's bad after 2-3 years (e.g. drops to half-speed generally and quarter-speed for some operations), I've seen some Windows systems slow down by that much in less than a year (depending on what they're used for), which is inexcusable.

But it's only fair to say that -any- fresh consumer system is going to operate at rated speed, and after that it's all downhill. :-)

The only operating systems I'm familiar with that retain their original speed indefinitely are Unix and Linux servers.

[itsahoot]

Once you switch to Mac, and you like it, there is no way you’ll switch operating systems.

Yep.

[itsahoot]

Malwarebytes is a common install on macs

Prove that statement. If there were malware commonly installed on Macs, the Mac forums I read would be screaming about it.

[Squawk 8888]

Visual Studio runs just fine on my $300 Acer. No need to spend a penny more.

[Squawk 8888]

Calling him a liberal is a seriously low blow.

[itsahoot]

Can’t do that with Apple so I have no dog in this fight.

No but I bet you can run Apple on that box if you are a real tech head.

[Squawk 8888]

I accidentally figured out how to BSoD any version of Windows:

1. Don’t change the VM settings from the default.
2. Set up a VHD to auto-mount at startup.
3. Dismount said VHD during your session.

[dayglored]

Yep, that'll do it.

Sorta like saying your truck stalls out.... when you drive it off the cliff. ;-)

[Squawk 8888]

Macs rarely do updates. Certainly not weekly or daily Neither do Windows machines. Windows Update runs once a month.

[dayglored]

> I bet you can run Apple on that box if you are a real tech head

He's likely gonna need one o'them fancy Apple ROM EFI boot emulator thingies (that's a technical term):

https://en.wikipedia.org/wiki/OSx86

[dayglored]

> Windows Update runs once a month.

Except when it needs to run out of cycle because of a Zero-Day critical vulnerability being announced and it's too far to wait until Patch Tuesday.

They've happened a few times this year.

[Darteaus94025]

you’re very blissful

[itsahoot]

Building an X99 Hackintosh LIVE!

[Swordmaker]

The proof is in the iPhone. It is not the fastest, nor arguably the “best” phone on the market ... but people really don’t switch.

Actually, RightwardHo. . . right now, and for the past almost two years, with Metal, and the graphic chips and true 64bit processors and OS, the iPhone is, and has been, the fastest.

Data from Anandtech, reported by BGR:
"Who needs octa-core CPUs and 4GB of RAM? iPhone 6s is the fastest phone on the planet"
By Yoni Heisler—BGR (Boy Genius Report)—Sep 28, 2015

These results bring into question the entire premise of the PCWorld article. Gordon Mah Ung's very amateurish testing shows Apple's MacBook Pro could only produce 18.5 frames per second using theoretical tests and 23.6 FPS in a "real world game," yet a new state of the art iPhone 5S from two years ago was doing far better with far higher screen resolution. And a year old phone is even better and the latest iPhone 6S is more than double the real world game. Yes, I know the screen of the iPhone 6S is smaller, but pixels are pixels!

This brings up my analysis of this PC World comparison between the off-the-shelf Macbook Pro from March 2015 using the Broadwell intel core i5 which the author claims is an Intel Core i5 5752U. It isn't, because Apple has never used such a processor in their 13" or even 15" Macbook pros. . . According to EveryMac.com's Specs for this model, the processor Apple uses is the Intel Core i5 5257U. Since even Intel does not list a processor with the article author's listed designator, it has to be a sloppy type on his part, but it is indicative of his approach to the subject matter.

Ung compounds his error by doing a comparison of that off-the-shelf Macbook Pro to a kludged together hybrid Microsoft Surface Book. He claimed that to make an equivalent laptop to the i5 MacBook Pro for his test (he really did not), he merely ". . .just plugged the Clipboard section with the Core i5 into the base (from an i7) with the graphics chip in it. Neat."

He made an unwarranted assumption that I do not believe could possibly be true: that the graphics section, the GPU of the Microsoft Surface Book system, is located in the keyboard section of the hybrid laptop/tablet system. This may explain the dismal battery life of the tablet section when separated from the keyboard. The tear-down videos show almost all the electronics of the device are in the tablet section which has a minuscule battery. . . and only keyboard and large batteries in the keyboard section. To have a large powerful graphic chip in the keyboard section would require many more connections between the two for bandwidth. I think his assumption is probably basically wrong. Perhaps MS has perfected the low power graphic chip sufficiently with a PCIe transfer to do this, but the overhead concerns say it it would be simpler to keep it closer to the processor.

Finally, Ung's next trick, which explains why the MacBook Pro turned in such dismal frames per second when much less power full iPhone 5S, 6, and 6S can blow the doors off of the MacBook with far less power, is that he deliberately uses obscure speed tests that are NOT cross-platform optimized, choosing to use instead tests that have been optimized for Windows' DirectX. He also tosses in a little zinger.

"Next I ran Unigine’s Heaven 4.0 graphics test. The test was run at 1366 x 768 resolution with 2x AA, no tessellation and medium quality. I did this because the MacBook Pro 13 defaulted to many of those settings when started. On the Mac, the only graphics API is OpenGL, while the PC has DirectX and OpenGL.

Uh, no, not by a long shot. The MacBook Pro's has a native Retina resolution of 2560 X 1600 and can easily support It also supports scaled resolutions of 1680 by 1050, 1440 by 900 (the historical resolution of non-Retina MacBooks), and 1024 by 640. But to support the odd resolution of 1366 X 768 the MacBook Pro has to calculate every frame on the fly, appreciably putting on a huge performance overhead. Can you say Gordon Mah Ung cheated??? I certainly can!

Now, let's examine his bald lie that the ". . . only graphics API is OpenGL. . ." Again, how disingenuous." He wants the readers to accept his self-serving lie because he wants to ignore Apple's METAL GRAPHICS API which is why Apple's iPhones and MacBooks running OS X.11.1 El Capitan are so fast in graphics. On the other hand, he wants them to ignore that the tests he's selected are optimized for Windows' OpenCL and DirectX functions. He just slipped that little detail past the readers. . .

"I opted for DirectX, as I don’t think it would have been fair to use OpenGL on the PC—Windows is all about DirectX, and it’s a big advantage for the platform."

WOW! If it isn't fair, Norman, why are you choosing to use it???? Amazing!

Apple introduces Metal for Mac, promises huge leap in graphics performance
By AppleInsider Staff— Monday, June 08, 2015, 10:41 am PT (01:41 pm ET)

At WWDC on Monday, Apple SVP of Software Engineering Craig Federighi announced Metal for Mac, which combines the power of OpenCL and graphics crunching capability of OpenGL into a unified API that reduces draw rendering times by 50 percent.

First introduced as a feature in iOS 8, Metal is a core-level graphics technology that allows developers nearly untethered access to system GPU hardware for highly efficient processing. Metal for Mac works almost identically to its iOS counterpart, as developers can stack on apps on core animation and core graphics, as well as built-in OpenGL and OpenCL API support.

Apple notes performance advantages can be found in any graphics-intensive app. For example, Adobe found huge improvements by stacking After Effects and Illustrator on Metal.

Metal is bundled into OS X El Capitan, which was made available to developers today.

OS X.11 El Capitan was released on September 30th, and it's been out in a public Beta since July. . . and Ung is using it. But he doesn't mention Metal, and claims that the only API in Apple's graphics arsenal is OpenGL, ignoring the Metal, and OpenCL API. . . because he does not want the reader to notice he is selecting tests that don't take advantage of those APIs and are optimized only for Windows DirectX. He's not using the full panoply of Geek Bench tests, only a select few that have been optimized for Windows such as the Heaven 4.0 test which, as I pointed out above, he further "fixed" by setting the resolution to a MacBook's non-native format. . . in fact to one the MacBook as far as I can tell has never been used natively. He used tests that do not use the APIs that actually utilize the Mac to it's fullest abilities, while DID use the APIs that utilize the Surface to its fullest capabilities being optimized for its DirectX strengths.

I reiterate his slipped in decision, which I am certain he hoped no one would notice:

"I opted for DirectX, as I don’t think it would have been fair to use OpenGL on the PC—Windows is all about DirectX, and it’s a big advantage for the platform."

I assure you, I noticed, and know what he's doing! It's called cheating.

Another dis-ingenuity: "Both manufacturers actually claim 12 hours of battery life for movie play back." Apple claims up to 12 hours of battery usage on the MacBook Pro when playing back iTunes movies. . . but Ung decided to NOT use iTunes for his test and use QuickTime player instead because of what he perceives as a "bug" in iTunes in not allowing replaying of iTunes movies. When it doesn't meet Apple's claim, because he says the MacBook Pro was at 19% after only eight hours, he gives up and goes home, giving the 29% battery life of the Microsoft Surface the win. Not many people watch the same movie over-and-over unless they suffer from Autism. In actuality, the MS Surface claims up to NINE hours of Video Playback using a Surface with only 8GB of RAM, and 10 hours on a 4GB unit. 12 hours was achieved only with auto-brightness brightness and WIFI disabled per Microsoft Surface Book Specs , none of which is required on a MacBook Pro! OOPS!

Finally, he is comparing an Intel Core i5 Apple MacBook Pro with Intel Iris 6100 Graphics priced at only $1299 MacBook Pro to an Intel Core i5 Microsoft Surface Book priced at $1699, both have 8GB of RAM and a 128GB SSD. and the MS hybrid is $400 more expensive.

[Harmless Teddy Bear]

Awww dangit... Can’t we just talk about something less controversial, like religion or politics?

All of MacBook users are progressive heretics who love KC Barbecue.

All Surface Pro users are hidebound legalists who drip mustard sauce on their keyboards.

Discuss.

The perfect flame thread. :)

[Squawk 8888]

I’ve never seen any measurable performance hits due to registry bloat. The real problem is the amount of software out there that loads something at startup, at minimum the update checker.

Another more serious issue I’ve been encountering with my customers over the past year has been the bundling of crapware with downloads from sites such as CNET; the installers are designed so that the average user won’t notice he’s adding junk. A pox on CNET for failing to enforce their own TOS in this regard.

[Hodar]

I just downloaded El Capitan to my '09 MacBook Pro. Runs flawlessly. Thing was built like a tank. Let's revisit this thread six years from now to see how these Surface Books are holding up.

Perhaps the most sincere, bold, succinct and clear comparison of all the posts made. Nice job.

I have spent 25 years working R&D as an engineer - great deal of it in the PC industry. For work, I have found that the world revolves around MSFT. Be it manufacturing, medicine, instrumentation, factory automation - not even considering the office suites. MSFT utterly decimates the competition. But, for me, at home, on MY dime - I just prefer my Mac.

[Swordmaker]

List of known updates, true less than Windows but not what I would call rare.

• The first Ten updates on your list of security updates, released on October 21, 2015, are for ELEVEN different products. . .
• The next 3 updates released on September 30, 2015 were in just three packages and covers four products. Two updates for Safari.
• The next update release on September 21, 2015, was a major update to the Apple Watch.
• The next release, on September 16, 2015, were major update to the iOS Operating system, iOS 8 to iOS 9, plus Xcode 7, Windows iTunes, and OS X SERVER.

We are now back to August, then June, and is essentially a once a month schedule of minor upgrades. . . spread around six different products. Some list.

So, Bidimus1, where are all these major updates to your devices???? Most of the updates are minor. Your point you were hoping to make leaks like a sieve.