How Windows 10 Performs On a 12-inch MacBook 241
An anonymous reader writes: As Microsoft prepares for the launch of Windows 10, review sites have been performing all sorts of benchmarks on the tech preview to evaluate how well the operating system will run. But now a computer science student named Alex King has made the most logical performance evaluation of all: testing Windows 10's performance on a 2015 MacBook. He says, "Here's the real kicker: it's fast. It's smooth. It renders at 60FPS unless you have a lot going on. It's unequivocally better than performance on OS X, further leading me to believe that Apple really needs to overhaul how animations are done. Even when I turn Transparency off in OS X, Mission Control isn't completely smooth. Here, even after some Aero Glass transparency has been added in, everything is smooth. It's remarkable, and it makes me believe in the 12-inch MacBook more than ever before. So maybe it's ironic that in some regards, the new MacBook runs Windows 10 (a prerelease version, at that) better than it runs OS X."
Linux Mint 17 XFCE, SSD (Score:2, Interesting)
and my beat to shit 2007 DEL Inspirion 1720 Core 2 duo also flies smooth. Stange eh?
Re: (Score:2)
That's what LMDE 1 ended being (semi-rolling snapshotted debian) but LMDE 2 and Ubuntu Mint get plenty security updates (from the upstream distros)
Re: (Score:2)
No that's the one that just works. Just like I buy my cars instead of building a kit car when I need a new car.
Re: (Score:2)
Sorry I don't speak acronym.
Comment removed (Score:4, Insightful)
Re:Anecdotal evidence (Score:5, Funny)
Hey everybody! I found the guy who keeps marking every sentence [citation needed] on Wikipedia!
Comment removed (Score:5, Funny)
Re: (Score:3, Funny)
http://apple.slashdot.org/comments.pl?sid=7426477&cid=49709415 [slashdot.org]
Re:Anecdotal evidence (Score:5, Insightful)
True, though there is some precedence. OS-X does not seem to be particularly zippy in the few cross platform app benchmarks that are to be found. A good example is DAW bench's test on Cubase, Protools, and Kontakt: http://dawbench.com/win7-v-osx... [dawbench.com]. What you see is that Cubase has a much more efficient engine than ProTools (no surprise) and that on Windows either one gets a lot more polyphony than the Mac. At any given buffer size (lower buffers are harder to deal with) Windows did better.
Pretty good test too since you are dealing with tools that have long been cross platform. Kontakt has been cross platform for its entire life, Pro Tools was Mac only until version 5 (1998ish), since when it has been cross platform, and Cubase has been cross platform since back in the DOS and Atari ST days. All the software has long development histories on both platforms, yet Windows gives superior results.
None of this means OS-X is unusable or anything, but it doesn't appear to have the performance Windows does, when pushed.
Re:Anecdotal evidence (Score:5, Interesting)
True, though there is some precedence. OS-X does not seem to be particularly zippy in the few cross platform app benchmarks that are to be found. A good example is DAW bench's test on Cubase, Protools, and Kontakt: http://dawbench.com/win7-v-osx... [dawbench.com]. What you see is that Cubase has a much more efficient engine than ProTools (no surprise) and that on Windows either one gets a lot more polyphony than the Mac. At any given buffer size (lower buffers are harder to deal with) Windows did better.
Pretty good test too since you are dealing with tools that have long been cross platform. Kontakt has been cross platform for its entire life, Pro Tools was Mac only until version 5 (1998ish), since when it has been cross platform, and Cubase has been cross platform since back in the DOS and Atari ST days. All the software has long development histories on both platforms, yet Windows gives superior results.
None of this means OS-X is unusable or anything, but it doesn't appear to have the performance Windows does, when pushed.
That benchmark is a bit dated, there must be a newer benchmark than OS X 10.6? Hower, now that I have let fly the obligtory nitpick of your post, the fact that Windows 10 performs well does not surprise me. I have long been aware of the fact that Microsoft quietly sank a lot of resources into rewriting Windows and improving performance so I'm not exactly thunderstruck by news that they succeeded. What I'd like to see is a benchmark shootout between Windows 10 and the most recent OS X AND Linux distros. Something tells me that if Windows 10 also turns out to be faster than Linux the atmosphere around here would shift from shadenfreude to righteous fury and calls for a Jihad.
Re: (Score:3, Informative)
Nothing rigorous that I've found. I've seen some things like a Mac user posting on a forum asking why Cubase was hitting harder on OS-X than Windows along with screenshots of the overall load meters that it has, but little in the way of details on methodology.
While I haven't done extensive looking, I haven't come across anything and it is something I'm interested in.
Sadly, there seems to be little interest in testing. People who own PCs can't really test it, outside of building a hackintosh, and Mac users a
Re:Anecdotal evidence (Score:5, Insightful)
There is also the "fast enough" metric. Mac users generally have mid to high end equipment. Chances are "slower than windows" for them is probably still plenty snappy. It is like the aero, vs OS X versus lightweight and snappy on a 1990's laptop debate: if you have the money to buy the hardware and feel more comfortable with the UI design, available software, heck just want transparency effects or whatever: who cares? It is a matter of preference. We are like whinny car nuts debating the merits of Ford vs Chevy trucks. It's a box you put shit into, if you are happy with it that is good enough I don't have to prove mine is better.
Re: (Score:2)
There is also the "fast enough" metric. Mac users generally have mid to high end equipment.
Careful, he has a real need to believe his money was well spend and does not wish to do something which might challenge that idea.
We'll be on to those silly one button mouses any moment here........
Re: (Score:2)
So salty. I wish I knew how it was to care about a few hundred bucks on the price of hardware. i feel so out of touch with you neck beards
I tend to buy what a I like - and a lot of us do. That's why there are more autos out there than the Toyota Coralla
Wow - getting to be a lot of auto analogies in this thread...
Re: (Score:2)
Well they get away with it I think because they can stop supporting older hardware should they hit a limit vs the Windows ecosystem where you have people buying everything from $200-10k workstations/laptops.
They haven't stopped supporting systems for a long time though, my home desktop is a 2009 iMac and I was able to install Yosemite on it no problem. Didn't find that much difference performance wise from Lion, I guess slow is slow :). New system (Dell XPS 15 maxed out) comes on Tuesday so back to a reason
Re: (Score:3)
People who own PCs can't really test it, outside of building a hackintosh, and Mac users are not very interested in testing particularly since many of them have a real need to believe their money was well spend and do not wish to do something which might challenge that idea.
Yeah, that must be it - I'm afraid to find out something that doesn't agree with my worldview.
Or it could be that there is a lot of us that don't give a flying fig what the benchmarks are. If the computer starts running slow, and the usual culprits are ruled out, it's time to get a new one.
Benchmarking is the computer equivalent of audiophiles listening to test tones on their stereo systems, People who install loudest sound system ever in their cars that won't play music, and rednecks at the corner gas
Re: (Score:2)
Congratulations! And I mean that sincerely. The next step in your transcendence is: Abandon Slashdot. It is the corner gas station.
Re:Anecdotal evidence (Score:4, Informative)
They learned a lot from the Xbox, and developed a lot of tools for optimising game performance that could also be used to optimise Windows. It was most evident with the transition from Vista to 7, where a lot of the major bottlenecks were eliminated and performance vastly improved. With 8 they did a lot of improve application performance too.
Re: (Score:3)
The Mythical Man Month only really discusses using more resources to shorten delivery time (doesn't work). It does not say more resources cause a worse outcome.
Re: (Score:2)
The Mythical Man Month only really discusses using more resources to shorten delivery time (doesn't work). It does not say more resources cause a worse outcome.
yes it does, did you read it? more resources need to be brought up to speed, thus requiring meetings with existing resources, taking time away from them, slowing down the project. new resources mean new priorities, changes in direction and a different (probably worse) outcome. it's all there, read it.
Re: (Score:2)
I have read it. What it says is that if you are time constrained, more developers does not necessarily speed up the process, and may lead to quality problems if you keep the same time constraints. It does NOT say that more resources is always a problem. Time is a resource, does adding time to a project mean it is going to be worse? Money is a resource, does hiring some top developers, and paying them accordingly, lead to a worse outcome? Good leadership is a resource, same question. Does having 10 pe
Re: (Score:2)
Also, learn the difference between may and will. Yes, there are certain situations where adding resources makes it worse. However, there are also situations (probably more situations) where adding resources makes it better. Therefore, the statement 'adding resources makes it worse' is false.
Re: (Score:2)
OSX also tends to perform worse than linux when running various open source cross platform applications... It just seems performance has never been apple's primary concern.
Re: (Score:3, Interesting)
Apple has never been very good at Systems software. The old MacOS was a huge piled-on mess. Their User Interface design took precedence over robust design 'beneath the hood.'
The Macintosh didn't have per-emptive multitasking until they finally gave up on an in-house redesigned new MacOS. There was Taligent, and Butt-Headed-Astronomer (previously known as 'Sagan' until Sagan sued them!)
Apple spent many millions in attempts at a new Operating System before finally giving up and buying NeXT Step, whi
Re: (Score:3)
Snort, Selective memory? At the same points in time Windows is/was also a "huge piled-on mess".
To your Classical MacOS I reply Windows Me.
To your preemptive Windows NT I reply A/UX.
Need I compare OS X & Vista? Windows 8?
There are/were piled on messes on both sides.
Re: (Score:3)
Hardly. Windows Me (and Win 9x), as bad as it was, was lightyears ahead of classic MacOS from an architecture pov. It was fully preemptive, multitasking. And it was to a large degree really 32-bit, though it was bootstrapped from a 16-bit environment, and some of the drivers appear to have been 16-bit. But it did run in protected mode. Wasn't nearly as good as Windows NT of course, which was also pretty darn good, based on the venerable VMS operating system's architecture.
As for Vista, under the hood i
Re:Anecdotal evidence (Score:4, Interesting)
I will preface this by saying that I was a Mac user and developer for over a decade until very recently...
To your Classical MacOS I reply Windows Me.
Sorry, will have to call this one out. At the time that classical MacOS was around, most users were either running Windows 98 or Windows 2000. Few people ran ME... I worked in the industry and I think at the time for every 10 Win2K boxes I saw, I saw maybe one ME and 15 Windows 98. Noone saw a compelling need for ME so they didn't upgrade. And early versions of OSX up to Jaguar (10.2) were horrendous messes under the hood. The interface was gorgeous but some of the APIs were kludged together wrecks.
Now, I'm not saying that Windows was much better, at least on the 98/ME track... but the Windows 2000 API cleanup actually produced some really nice results. This was at the same time (roughly) as Mac OS9 which was one of the most awful kludges of an OS I think I've ever had the displeasure to work on. It was obvious that OS9 was created in a place where OSX was getting all the attention, but even OSX wasn't really properly clean until 10.2... and even then it was slow. Have you ever used 10.0 or 10.1 on hardware of the time? It was horrifyingly slow to do anything and the only thing it really had going for it was the interface.
Speaking of the interface... seriously... Finder is significantly worse than Explorer in terms of threading, resource utilization and stability. You want to see a kludged mess, check out Finder circa... well... any version of OSX actually. Of course, I've recently abandoned the Mac platform for Windows 8 for various reasons so I can't speak to Yosemite... but every time I work on a machine running Yosemite I just feel like the entire OS is going in a direction I don't appreciate.
Need I compare OS X & Vista? Windows 8?
Vista and Windows 8 both had huge improvements under the hood. Windows 8 in particular has gotten a bad rep simply because it has a UI that people find really polarizing... but it's seriously a fast and efficient OS that really takes advantage of the underlying hardware. It actually is a better operating system than the much-vaunted Windows 7 (which was itself an improvement over Vista) but most people never get to see it because they get hung up on the UI.
I have played with Windows 10, and I like it. I run 8.1 on my computers today but will switch to 10 when it comes out. That's not to say it's a fundamentally "better" operating system than OSX... but for my needs the priorities are all screwed up in OSX. They're both modern, stable and secure operating systems... and if that's all you need then great. However, running the same applications on both platforms does show the weaknesses of OSX; memory management is of questionable value in OSX and the storage management is kludgy at best.
Windows has been faster than OSX on the same hardware since Windows 7. I know; I've always had a Windows Boot Camp installation on my Macs. My last Mac is still a good one (2012 15" MBP) but has now been surpassed significantly. It used to be that Macs had a 5 year lifespan whereas Windows had a 3... that's one reason I liked Macs. Nowadays, not so much... and it's not changing workloads that are killing it but rather the overly heavy APIs and core problems with OSX that just don't scale quite so well... so new versions get heavier and slower on the same hardware.
Re:Anecdotal evidence (Score:5, Interesting)
Apple has been optimising OS X for years, but not for Graphics Performance. They've been optimising it for longer battery duration.
It's been years since I've needed or even noticed needing better graphics performance yet battery performance of OS X pure versus Windows? Oh yeah, THAT I notice. On the occasions I forget a charger I have to minimise running Windows or I'll be running out of battery at least twice as fast as when using OS X. I can get work 7 Hours using just the battery on my rMBP with occasional excursions to Windows to check mail but or use corporate windows only tools but running windows will only give me 3 hours. Windows isn't even doing much, it just does it all the time & never lets the CPU sleep for any significant time - see here [arstechnica.com] for reasons why.
Yeah, the pure graphics performance may be poorer on OS X. Most people don't care as long as it's good enough.
Re:Anecdotal evidence (Score:5, Informative)
The 15" rMBP has a 95 Wh battery and lasts 7-8 hours. The Dell XPS 15 with similar hardware and a higher res screen has a 91 Wh battery and lasts 6-7 hours under Windows. If you're only getting 3 hours in Windows on your rMBP, that's more an indication that Apple has put very little effort into optimizing their Windows drivers. Not an indication that Windows sucks.
If you want to compare the 13" rMBP, it has a 75 Wh battery and lasts about 10-12 hours. The Dell XPS 13 manages 9-10 hours with a battery only 2/3rds the size (52 Wh) and a higher res screen (3200x1800 vs 2560x1600). If you get the lower res screen (1920x1080) it'll go 15 hours.
Re: (Score:2)
Nope, the problem is that Windows does not have task aggregation like Grand Central Dispatch so the CPU is constantly getting interrupted & never actually spends much time in low power states. Read the article I linked. Driver non-optimisation is not the problem, less OS optimisation in Windows is.
Re: (Score:3)
It's been years since I've needed or even noticed needing better graphics performance yet battery performance of OS X pure versus Windows? Oh yeah, THAT I notice. On the occasions I forget a charger I have to minimise running Windows or I'll be running out of battery at least twice as fast as when using OS X. I can get work 7 Hours using just the battery on my rMBP with occasional excursions to Windows to check mail but or use corporate windows only tools but running windows will only give me 3 hours. Windows isn't even doing much, it just does it all the time & never lets the CPU sleep for any significant time - see here [arstechnica.com] for reasons why.
What you're seeing here really is driver issues. Apple doesn't put the resources into developing the Boot Camp drivers properly for Windows... they're good enough so they're shipped. Apple doesn't want you to run Windows, or Linux or anything else. They want their beautiful machines running their beautiful OS, to hell with what the end user actually needs. A good set of power management drivers would go a long way to fix Windows on Mac hardware but they're not going to put the time and energy into it.
This p
Re: (Score:2)
Read the article that I linked. Grand Central Dispatch & other optimisations that Apple has brought to OS X are what makes the difference in Battery life. It goes far beyond mere driver optimisation. The entire GUI stack has changed & you cannot bring equivalent optimisations to other OSs by "optimising the drivers".
The level of ignorance ("it's a driver optimisation problem) & bias ("thats why I've abandoned Apple") you display are unfortunately all too common.
Re: (Score:2)
Yet other laptops running Windows get just fine battery life with the same or smaller battery. Simply put; Apple develops their own EFI to manage fans and power states and when the driver doesn't exist or is non-optimal, the hardware runs "hot". Grand Central Dispatch is a nice technology but does NOT magically increase your battery life despite what Apple wants to claim. It merely provides a common framework for addressing the EFI and power management customizations in a way that is pretty nice, but not a
Re: (Score:2)
Re: Anecdotal evidence (Score:2)
Huh... and here I've been hearing that CoreAudio makes everything pro audio so much better on Macs. I can't really complain on Win7, but I suppose the ASIO drivers for a lot of interfaces could be more stable...
Re: (Score:2)
None of this means OS-X is unusable or anything, but it doesn't appear to have the performance Windows does, when pushed.
Where do we get jobs running benchmarks?
Re: (Score:2)
Where do we get jobs running benchmarks?
if you have to ask, you're not worthy
Re: (Score:2)
Where do we get jobs running benchmarks?
if you have to ask, you're not worthy
(walks away sulking) heheh.
Re: (Score:2)
Pretty good test too since you are dealing with tools that have long been cross platform. Kontakt has been cross platform for its entire life, Pro Tools was Mac only until version 5 (1998ish), since when it has been cross platform, and Cubase has been cross platform since back in the DOS and Atari ST days. All the software has long development histories on both platforms, yet Windows gives superior results.
You mean more likely a pretty good test of the particular Development Teams. Who knows whether each of these Products has dedicated Dev. teams for each platform, what their relative skill-levels are, whether (as is often the case) the Mac versions are contracted-out to who-knows-who, etc. etc.
I'm not saying that these anecdotal results aren't valid; just that they need more in the way of validation.
Re: (Score:3)
Both right. I have a late 2009 iMac bootcamped to Win 8.1 pro. 1-2min before I can start opening up my own applications (have autohotkeys, and iTunes autostarting). But at work, SSD box. The system restarts faster than the monitor can detect that the computer has dropped the connection (~10-15 restart from desktop-bounce-login screen). I log in and as fast as I can get a mouse over a shortcut the computer is responsive. It all comes down to hardware at that level particularly ssd vs hdd.
Re: (Score:2)
What about getting to the real desktop instead of that Metro splash screen?
Win7 is where the time lag is especially painful even with an SSD and recent i7, but Win8 still has the underlying problem and launching anything that needs more than trivial resources has to fight with the antivirus, skype and a pile of notification shit starting up.
Re:Anecdotal evidence (Score:4, Insightful)
There doesn't seem to be much factual evidence to make the claim that "It's unequivocally better than performance on OS X,"... The claim, by it's very language deletes a lot of information making the claim worthless.
My guess is that he means "user perception," and I don't find that claim hard to believe at all. Notice the comments in TFS about 'animations?' One of the (to me) most annoying features of every windows from XP is the extensive use of fade in the user interface. Click on the start button, and it spends 300ms fading into existence. Click on an item, another 300 ms fading a sub-menu into existence. This makes the UI feel horribly sluggish and is the first thing I turn off on a new system. OSX has its own bits of animation, like bouncing a task bar item while it starts up. Maybe these things look great at product demos, but they get in the way of me working: always waiting, just a little bit, for the computer to get around to drawing the menu I asked for.
Point is: if MS turns down or off task bar and menu fade in Win10, it will "feel" much faster than other Windows, and very possibly OSX.
Re: (Score:2)
There doesn't seem to be much factual evidence to make the claim that "It's unequivocally better than performance on OS X,"... The claim, by it's very language deletes a lot of information making the claim worthless.
Unequivocally ... (says who?)
Better... (by what standard?)
Performance (by what metric?)
I know that this is probably just a personal blog with an opinion.... and he does want to quantify the claim with stats... but it's a bit too early to make the claim.
Not to mention that benchmarking a pre release is just about irrelevant to the final software performance.
Re: (Score:2, Interesting)
yes, but... (Score:5, Informative)
... there's still a long-ish way to go until Windows 10 is out. And I'm afraid it'll come with surprises that we don't want (more bloatware? Advertising?).
I'm impressed by the performance boosts Windows got through 8, 8.1 and now 10, but unfortunately that is not enough for an OS. I'm uncomfortable with navigating the OS, something which should be seamless, logical and extremely easy to do; imagine if you had to think about every step you take whilst you go shopping.
I've also installed Windows 7, 8, 8.1 on a Macbook Pro and it's terrible. Oh, it is fast, trouble is the energy management is so poor the processor is overheating so the fans go turbo-mode. Not a pleasant experience.
I'll stick to 7 for the moment and OS X, they do the job properly without the hassle of a sad smilie BSOD.
Embed controllers (Score:5, Insightful)
Oh, it is fast, trouble is the energy management is so poor the processor is overheating so the fans go turbo-mode. Not a pleasant experience.
Welcome to the fantastic world of "embed controllers" (EC). The small custom chip sitting in the middle of a laptop, and in charge with all the peculiar functionality that are peculiar to this laptop, but don't exist in standard desktop/workstations. (like battery management, etc.)
The problem is that there is absolutely no standardisation of ECs. Every model is its own special snow flake (and when I say "model" I mean model of motherboard. In some case, specially consumer oriented laptop, some product range might have the same model name and the same plastic case and looks absolutely the same from the outside, but is actually different revisions which looks completely different under the hood, depending on which parts were the cheapest during the month this one was produced) (that's why for the same "Model" you have a few different BIOS downloads depending on part number, revision, etc.)
To get it working the manufacturer could write a specific driver. Usually this is done by the hardware manufacturer who write drivers for the target OS they have. Most laptop manufacturer write drivers for Windows, because they produce windows laptops. Here it's an *Apple* don't expect much.
To make things worse: usually these aren't your garden variety of drivers. Very often, platform functionality like ECs are handled by ACPI (now part of UEFI). i.e.: by firmware that is byte-code interpreted by the running system. In theory it should make things more OS-agnostic and portable. In practice it's a nightmare as every ACPI implementation is buggy in its own way, and every OS has a different variety of quirks. So writers of firmware (BIOS/UEFI/ACPI) for laptops have to release new versions of firmware (again, one per model of EC on mother board).
You can count on big brands to release a new BIOS pack download to cover the major flavours of Windows that they ship with this model. Maybe cover an update.
But don't count on engineers working for Apple to scram to release a new firmware update, just because some random schmuck decided to install a newer version of windows whose ACPI implementation is broken in a subtly different way than the preceding.
Their official OS that they support is OS X, they might have decided to add as a bonus a version or two supported as part of their Boot Camp offering.
And that's about it, don't count much more from them.
Funny that *windows* is now at the receiving end of this firmware/EC problem, that usually haunts Linux users on laptop that mostly run Windows.
(The problem that you report trying to run Windows on Apple hardware ? That's the daily plight of Linux users on most Windows laptops).
--------------
Also that might be the reason of the performance gain and "overall smoothness" reported by TFA:
- when running under OS X, the OS balances performance with battery life, thermal limits, etc. hardware runs at an equilibrium. Thus isn't as smooth as it theoretically could, because OS decides to save a bit power.
- Windows 10 has very probably an ACPI implementation that is subtly broken in a different way that its predecessor. Power management doesn't work. GPU is run at max power profile, CPU is run at max frequency. Results are probably smoother, but if the guy had actually carred to measure it in details, he would probably have observed shittier battery life.
Relevant quote (emphasis by me):
Yup definitely a possibility that shitty battery life and heat that you describe and the high performance that the author got are the
Re: (Score:2)
The days of custom chips are in recession, not advancing. Apple's laptops don't contain secret custom chips, it's all off the shelf and anyone's laptops could be made out of the same parts. The only difference between a macbook motherboard and anyone else's Foxconn-built laptop is the BIOS. The BIOS and the use of the TPM (which is now built into every modern CPU, rather than having to be integrated separately) are what make the system 'special'. Even supposedly custom chips (like console GPUs) are just cob
Re: (Score:2)
That just is not true, especially in the SoC and embedded world.
No, it is completely true. Those SoCs are slapped together from shit they had lying around already, they're not totally custom. The issue of closed-source drivers is real, but not the issue that was raised.
Re: (Score:2)
These days there is a more standard way of doing thermal management in laptops. Intel built the capability into their CPUs with configuration data loaded by the BIOS. Fan control is via the Intel chip set in the same manner.
You used to need a driver back in the Core 2 days, but most (all?) Core i devices don't.
Re: (Score:2)
... there's still a long-ish way to go until Windows 10 is out. And I'm afraid it'll come with surprises that we don't want .
Problem is, mow that a pre release version has presumably beat a OSX install, now will enter the PC fanbois lexicon, and will be brought up forever and anon, just like we still see them bitching about one button mice, which is something like 18 years irrelevant.
Perhaps we might wait until the final version, and then keep comparing them after each new update. If it is anything like any previous Windows OS I've used, it will slow down plenty between pre release, and 2 years from now.
let me know in 6 months (Score:2)
I just switched. Not going back. (Score:4, Informative)
My beloved 7 crashed, bad, again, this past week and I finally decided to try out 10, I am running 10074 and it made my ....5? year old pavilion DV6 6135 feel like a brand new computer. It really is gorgeous and has a lower overhead. Not sure that this is relevant other than additional praise for 10, but I thought you would like to know.
Lets cut through to the chase (Score:4, Funny)
Windows fanboys: We told you so...
Mac fanboys: Pfft buncha Windows fanboys
Linux fanboys: Whatever, Linux is free and runs on anything lightning fast
Re:Lets cut through to the chase (Score:4, Informative)
Re:Lets cut through to the chase (Score:5, Insightful)
The difference is that Linux and Mac GUIs get choppy under the slightest load, while Windows says smooth like butter in every situation. Some people have even installed Windows on their Macs and they are getting more graphics performance from the same hardware.
You're conflating a couple of different things. Windows general-purpose multitasking is terrible compared to Linux/MacOS (at least based on Win7 experiences).
On the other hand, Microsoft has had a laser focus on Windows gaming, with the obvious tie-in to Xbox gaming. This has resulted in very fast graphics drivers for Windows.
Linux seems to be doing well lately, with some Steam games getting higher frame rates on Linux than on Windows. Linux may end up being the best of all possible worlds (well out of three worlds anyhow) given its lean design, performance and stability. Perhaps eventually it'll see more proprietary software ports like Solidworks and ProEngineer.
Re: (Score:2)
Re: (Score:2)
MBP never had a problem. This is about the seriously underpowered Macbook (one-port wonder, mobile grade RAM, motherboard same size as that of a phone, no fan). Since Windows 10 runs on everything from $50 phones to $10k workstations, I am not surprised that it is much better optimized than OSX.
one burning question (Score:2)
How is testing windows 10 on a macbook "the most logical performance evaluation of all"?
Re: (Score:3)
It is the most logical evaluation of performance of OS X. How else are you going to test it?
Used.... (Score:2)
This submission (as some of the others recently) are there to troll slashdotters. We're being used as THEY know that there'll be factions and fanbois sprouting stuff about their favourite OS. It's a matter of respect and there is little of that around any more.
This judgment is entirely subjective (Score:5, Informative)
Not that snappy in virtual environment on linux (Score:4, Interesting)
I've recently installed the latest 64bit Windows 10 technical preview on a new computer running a gen 5 Intel processer with 6 cores and hyperthreading, SSD raid array and nVidia graphics card and the performance appears to be suboptimal. It requires DEP on by default, minimum RAM and other weirdness like specifying the type of cpu name (which I set to "core2duo") just to install but does allow all the usual hardware acceleration. I assigned several CPUs and a good chunk of DDR4 RAM to it with 16GB of hard drive space.
Running next to my trusty virtual XP windows 10 is a dog, even with all the settings for performance switched on. The integrated "search my computer and Internet" is painfully slow and the start button brings up this wacky osd box with embedded "metro style" elements. If you click a couple of times as it loads (can take any random amount of time) it doesn't ever appear again. I suppose this is a bug that will be fixed, but the whole thing feels slow to respond (including the mouse pointer and drag+drop snappiness). Some of this seems to be due to clunky design rather than the optimization of the codebase.
The only thing I can say is quicker is the startup and shutdown. I suppose the new "Spartan" thing loads alright, but I'm not a fan.
Re: (Score:2)
"virtual environment" means fuck-all. Are you using QEMU-KVM? Virtualbox? Or are you actually using VMware, which is what you do when you care about performance?
Running next to my trusty virtual XP windows 10 is a dog,
Again, which environment are you using? What's your config look like?
Re:Not that snappy in virtual environment on linux (Score:4, Interesting)
I am running QEMU-KVM. My Windows XP virtual machine and other machines run really smooth and quick with the same type of settings (actually less much RAM and CPU allocated), adjusting for 32bit XP and 64bit Windows 10.
I don't think most of my performance issues with Windows 10 can be solved by moving to VMWare, or even running it native. The problem for me is that it is just a clunky interface that slows down the process of doing things (e.g. waiting for another mode to open for the 2 different control panels, WTF?). Some of the things actually appear to be time based in waiting, because I have a really fast setup hardware-wise and it should "just work" as the other virtual machines I've tried do.
I'm not a big fan of the Mac OSX interface either, but at least it appears smooth when you stick an SSD in an old Mac. I use Mac at work with Windows 7 in a virtual machine (VMWare) (I also use linux on servers) and am evaluating Windows 10 with enterprise deployment in mind. I wonder if the author of the article actually clicked the start button or tried to type much in the new integrated search bar of Windows 10. I mean things like this -- Firefox works well, but if you try to set it as a default application Windows appears to freeze for half a second as it loads some weird "default application" selecting application that is like it is part of a completely different OS that has to load with it.
Re: (Score:2)
I'm not surprised that windows 10 doesn't run well in KVM yet. That's never surprising. Give it some time. Try it in vmware before you decide the performance is bad.
I don't think most of my performance issues with Windows 10 can be solved by moving to VMWare, or even running it native. The problem for me is that it is just a clunky interface that slows down the process of doing things
That's not a performance issue, then. It's a gewgaw bullshit issue. That's a real issue, but not the smae thing.
Firefox works well, but if you try to set it as a default application Windows appears to freeze for half a second as it loads some weird "default application" selecting application that is like it is part of a completely different OS that has to load with it.
Every time it loads, or just then?
Re: (Score:2)
I am running QEMU-KVM. My Windows XP virtual machine and other machines run really smooth and quick with the same type of settings (actually less much RAM and CPU allocated), adjusting for 32bit XP and 64bit Windows 10.
So what you're saying is that a 14 year old operating system performs better in a suboptimal virtual environment where the programmers have specifically targeted the performance profiles of said OS for about 10 of those years? Stop the freaking presses, mate.
Seriously? Windows 10 isn't final, and QEMU-KVM optimization for Win10 will probably lag at least 6 months to a year after release. Let me know then how it performs given your adjustments. I'm not a Microsoft fanboy but even I can tell you that your tes
Not sure how I can believe this. (Score:3)
The writer is saying Apple came out with a MacBook that cannot smoothly run the OS that was preinstalled on it?
Frankly, I can understand if it was a 4 year old laptop with the latest OS, but Apple's usually pretty good about integration. Can anyone else verify that the latest MacBook that OS animations aren't smooth?
parts of me (Score:4, Insightful)
The realist in me says "You paid too much for the Windows laptop."
Isn't the Apple motto "It just works"? Not it works well, or quickly.
The optimist in me is still sleeping.
speed isn't everything (Score:2, Flamebait)
Granted, for many users, speed matters. But, here are some other concerns.
1) The Registry File. Enough said.
2) Under OS X, open any kind of file in any kind of editor. Go back to the Finder window, rename the file, move it to a different folder no problem. Can't be done under Windows. Half the time, even after you close the file (not the editor app), the app fails to 'release' the lock and you STILL can't rename the file.
3) None of Microsoft's pseudo-shell implementations come close to bash/csh/ksh
Re:speed isn't everything (Score:5, Interesting)
Well for registry file, we'll see, I think it might only be for universal apps but win 10 is supposed to have a virtual registry similar to how WoW works. An app sees a registry but in actuality it is a private clone of the registry. When the app is uninstalled the OS just blasts away that clone and no cruft is left. At least that is the theory. For the desktop user, I think Mac has the cleanest installation process. Drag and drop. Things are installed in a single folder. Drag and drop the app to the trash and it is gone. For the power user: debian like is my favorite: apt-get is great. But still always having to remember what's in bin, whats in sbin, etc etc. Things do get scattered all over the place in Linux. Oddly it is the command line friendly OS that requires a lot of jumping between directories (which is a pain in the ass from command line) to get anything done. VS windows where everything is pretty much dumped in the windows folder and or available in the ui from a single (albeit huge) widget (control panel).
Re: (Score:3)
2) Under OS X, open any kind of file in any kind of editor. Go back to the Finder window, rename the file, move it to a different folder no problem. Can't be done under Windows. Half the time, even after you close the file (not the editor app), the app fails to 'release' the lock and you STILL can't rename the file.
So what you're saying is that Windows implements logical and useful locking mechanisms whereas OSX implements something bound to break eventually? Wow... what a crime. I've used Mac OS, Windows, Linux, BSD and more flavours of UNIX than you can shake a stick at... I think Windows' file locking semantics are just fine the way they are. Anything else can lead to confusion, corruption and lost files.
Having said that, I do wish Windows would implement a "Can't do this because file is open in (blah).exe" error m
Re: (Score:2)
It happens to me occasionally on my Win7 box at work, and sometimes requires firing up Process Explorer to find out who still has a lock on the file if the file had been closed a while back. Getting out of the editor is the only thing that releases the lock in those cases. It's happened in Word, Notepad++, and UltraEdit.
Some things never change (Score:4, Insightful)
but how about Xubuntu versus OSX? (Score:3)
I am pretty sure it will blow Windows 10 out of the water and you won't have to do work to install all the extras (assuming you like free software). If someone gives me a MacBook, I'm happy to run this experiment ;-)
OSX performance is dismal nowadays (Score:5, Interesting)
It has long appeared to me that all the best talent at Cupertino is being assigned to work on iOS, and OSX is suffering as a result.
One upon a time, a new MacOSX release meant you might need to put more RAM in the box, but even some seven-year-old piece of hardware would go faster, and there would be fewer bugs or crashes. [Consider a late G3 iMac, upgraded all the way from Public Beta to Tiger.] These days, you put more RAM in (if you can if you have a machine old enough that Apple didn't solder all the memory in during manufacture) just to make it less sluggish.
Drawing performance seems especially poor, which makes me wonder: if I bought a top-end top-of-the-range Mac Pro, would it be any better at being a desktop computer than what I have now? I'm not sure it would.
Laugh (Score:2)
Reminds of another study some years ago where WIndows 7 was found to be more secure out of the box than OS X was at the time, or for that matter likely still is.
Maybe there's a place but I never seem to see any info on who is doing the in house OS X coding, the age line, would be interesting to see.
"Renders at 60FPS" (Score:2)
That's not how Windows works, it's not running a continuous render-loop like a game.
Re: (Score:2)
That's not how Windows works, it's not running a continuous render-loop like a game.
Are you sure... I mean it would explain a lot :)
(Sorry, I just long for the good old days where we could get away with windows bashing)
Hey Mac Users... (Score:2)
If you really needed smooth animation and graphics, Apple would tell you that was important and then you would have it. In the meantime, this is clearly not important. Remember it's much more important to be stylish and cool than effective and efficient.
I believe it... (Score:2)
Up until Windows 8 I would say that OSX seemed a bit faster and smoother. I've got an old iMac (circa 2007). It has 4GB of RAM and a 256GB SSD. When I run OSX it feels slow and sluggish. I have it dual booted with Windows 8 and Windows, much to my surprise, runs much faster. It consistently uses 1.5 - 2 GB of memory. It never crashes.
It runs well enough that it is my everyday home office computer. I can do everything on it that I need. Granted, most of what I do involves connecting via Remote Desktop to som
Lowering expectations (Score:2)
20 years ago I was impressed when I was first able to move entire windows around the screen in real-time on a Pentium 90mhz machine running NT 3.5.1.
Today I half heartedly expect articles to talk about the responsiveness of calculator applications.
Good News for Mac users (Score:2)
I have never understood why Windows users wanted to "win" by having all competition sink without trace, nor Mac users for that matter. If there's only one game in town, why would anyone ever want to make it any better? Healthier for everyone if there ar
Re: (Score:2, Insightful)
Re: News for shills, stuff that costs money (Score:2)
That is the Trouble with Gribbles.
Re: (Score:3)
cygwin? you're a horrible person (Score:2)
I wouldn't wish cygwin on my worst enemy.
Re: (Score:2)
Why? Seriously?
First tool I install on my Windows boxes is MobaXTerm which contains a fully configured and set up cygwin including X Server. Works like a champ for me managing my Linux boxes, and all I had to do was install it. SSH works, remoted X works, and I can run bash scripts even referencing /proc entries from my Linux box and most of them work with no modifications.
Re: (Score:2)
Tried that but wasn't able to get something useful from "cat /proc/cpuinfo".
I had exactly that experience! Though mine was on Linux and was one of the things that pushed me to *BSD. An unstable text-based format that varies between architectures and between kernel versions turns out to be a piss-poor way of getting information from the kernel.
Re: (Score:2)
Re: (Score:2)
Probably because it supports a lot of .Net functionality.
Re: (Score:3)
So he did the needful?
Re: (Score:2)
So he did the needful?
Kindly?
Re: (Score:3)
Seriously though, I like the things Nadella has been doing. Reverting the Windows 8 start screen, releasing Windows 10 for free
Let's not forget that a lot of those decisions have actually been made by the Operating Systems Group, which is lead by Joe Belfiore and Terry Myerson.
Re: (Score:2)
It starts at the top. There's a reason why those things didn't happen when Ballmer was in charge.
Re: (Score:2)
I thought this "dirty regions" system went out of style with Windows XP in favor of Windows Vista's composited Aero environment.
Re: (Score:2)
Test of interactive responsiveness (Score:2)
we used to measure compute performance in flops or #tx/second, alas performance to get real work done.
Correct me if I'm wrong, but it appears you're assuming that "real work" means batch processing. Dragging a window around is a test of the responsiveness of an operating system's event handling and graphics components, which are correlated with the responsiveness of gaming and multimedia production applications. It's in fact a form of the "#tx/second" you mention, as each frame is the result of a "move window object to these coordinates" transaction.
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)