Slashdot Mirror
Windows 7 Benchmarks Show Little Improvement On Vista
snydeq writes "InfoWorld's Randall Kennedy examines Windows 7 from the kernel up, subjecting the 'pre-beta' to a battery of benchmarks to find any signs that the OS will be faster, more responsive, and less resource-intensive than the bloated Vista, as Microsoft suggests. Identical thread counts at the kernel level suggest to Kennedy that Windows 7 is a 'minor point-type of release, as opposed to a major update or rewrite.' Memory footprint for the kernel proved eerily similar to that of Vista as well. 'In fact, as I worked my way through the process lists of the two operating systems, I was struck by the extent of the similarities,' Kennedy writes, before discussing the results of a nine-way workload test scenario he performed on Windows 7 — the same scenario that showed Vista was 40 percent slower than Windows XP. 'In a nutshell, Windows 7 M3 is a virtual twin of Vista when it comes to performance,' Kennedy concludes. 'In other words, Microsoft's follow-up to its most unpopular OS release since Windows Me threatens to deliver zero measurable performance benefits while introducing new and potentially crippling compatibility issues.'"
Some facts:
- Vista is barely slower than XP on hardware bought within the last 2 years. It was fairly slower on RTM for many reasons, but vastly improved drivers & some colossal patches have put that to bed now.
- Vista in fact speeds up some operations over XP by pre-caching commonly used stuff. This uses more memory, and is often confused for being "bloated" by actually using the memory that you blessed your computer with being able to use, for what in fact it was designed for - speed increase.
- Windows 7 is taking Vista and putting it on a diet while not fundamentally changing the architecture. If it works on Vista it'll work on W7. That's a stated design goal.
Thus, for performance: Expect Windows 7 to be more responsive to user-input, work on lower-ended machines, start up quicker, etc. Don't expect: CPU intensive apps (games for example) to suddenly speed up 50%; memory intensive apps to use any less memory. They won't - Windows 7 is an operating system, not an overclockers kit.
throw new NoSignatureException();
"Don't expect: CPU intensive apps (games for example) to suddenly speed up 50%;"
Indeed , 50% is absurd. But they might speed up 5% or so depending on whether the process schedular and memory management have had a rewrite. For a machine with a lot of processes running and an app using a lot of memory those page and cache miss percentage can make a noticable difference as well as how intellgently the OS swaps in and out processes of varying priorities.
The very interesting thing about OS X 10.5 (Leopard) boot process is: It does nothing in order. It is parallel booting, firing all OS startup stuff at once and expects to do their jobs. That happens thanks to launchd architecture which I have no clue why not adopted by Linux or *BSD.
Here is its presentation by the inventor of launchd
http://video.google.com/videoplay?docid=1781045834610400422
(in 8:00")
That is one of underrated features/changes of Leopard. Now the term "photocopy" comes from this: They do something like launchd without using the underlying Unix logic and architecture. So, there is a huge chance that it won't be scaled. I have really lost count of how many kernel extensions, startup items, daemons running on my Leopard but it boots exactly same speed as it was cleanly installed for first time. Just like I really don't care about 1000+ .plist (pref) files on my user directory.
They named it "parallel booting" or something, some story about it on http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9119230&intsrc=hm_list
"Come on guys, its a pre-beta! ... did you really expect them to actually do any thing significant so far?"
Yes. I pulled some facts off the Wiki but I think they are pretty accurate.
Windows Vista RTM: November 8, 2006.
Microsoft stated in 2007 that it is "scoping Windows 7 development to a three-year timeframe"
Release dates are supposed to be in the region of 2009 or 2010.
So, to me, that says that it's *at least* eighteen-months, two-years into development (or thereabouts). It's got another year to eighteen months to go. So, halfway through it's development process, we have *zip* that is actually useful to the average user (which is who it is supposedly aimed at) and nothing to entice business users. There are *no* performance improvements. None. Programmers don't magically add 50% performance after-the-fact, it's *design* that gives you performance.
Halfway through and we don't have a single groundbreaking feature. Nothing. Not even something to show off temporarily. Seriously, read through the Wiki page on "new features in Windows 7" and have a look at the features that are actually *HERE*, not the ones "promised"... remember, Windows Vista was going to have WinFS etc. It's completely embarassing. Instead of a "new operating system", we just have:
Vista, with no better performance, some unnecessary UI changes (purely to make gullible people pay to "retrain" on the new OS in my opinion), removal of lots of built-in applications, a "Health Centre", some claims about fantastic new features that this article proves aren't even in there yet (better performance, threading, etc.) or that only a handful of people in the world could get excited about.
What that tells me is that all these marvellous new features DO NOT EXIST in a reliable form. But I'd be showing them everywhere if they did just work, even only on one machine - I'd be booting it up in conferences, showing it in trade shows, making people WANT that feature that I haven't finished yet and which only works on 25% of machines while the programmers hack on it. But there's *nothing*.
Fortunately, I saw the Vista thing coming.
I had a job interview the other day where the main technically-literate person on the panel asked my opinion on Vista. Needless to say, I was wary of giving my reply in case it was interpreted as belligerent or dismissive, but the interviewer and I laughed and joked and told Vista anecdotes for about ten minutes *in the interview* once he realised that I shared his very-low opinion of the OS. (I got the job, by the way.) I'm pretty sure, at this point, that Windows 7 will be more of the same or worse. Promises, promises, promises and then sting the customer before they realise that they've bought a turkey and that actually it was only useful for the little sticker with the Product Key on it that lets you use its predecessor instead.
(especially given that OpenOffice is at least as good as MS Word)
Afraid I've got to interject here. I'm in the early stages of writing a dissertation, and OOo3 Writer just does not have the same feature set as even Word 2003 (which I'm using for it, under wine) for serious document composure.
I use Linux and have done for years, as my only OS, and I've used and support OOo and have done for years. I can't comment on the other portions of either office suite, because I've never put them to serious work. But, having spent a few hours really teaching myself Word 2003, then trying to see where the same functionality was in Writer, it became apparent that some of it just wasn't there.
It's a shame, but until OOo Writer gets (for example) something akin to Outline mode, it's just not able to match Word for advanced features. That said, OOo is very solid software, and will get there with regards to said features sooner or later I'm sure. Some may even say I'm using the wrong tool for the job.
I recently tried Vista (for the second time) because so many monkeys like you keep telling us Vista is much, much better now.
What a bunch of hooey. Vista still makes my (pretty nice) laptop run like a dog. From slow video, to audio stuttering, to far too much hard drive thrashing, to disappointing program startup times...hell, sometimes I can't even track my mouse across the screen without it pausing half way while Vista does God knows what.
And yes, my laptop is "Vista compatible", and yes, I had all the correct drivers installed for my hardware.
I went back to XP (again) and the performance is so much improved, it's like getting a new computer.
Sorry, buddy, but Vista still sucks, despite your claims otherwise. And if Windows 7 is more of the same, I'm going to have to tell Microsoft, "Thanks, but no thanks."
Why are there so many of these pseudo-science-voodoo style reviews/benchmarks floating around? They're not touching on any real or user-meaningful metrics for performance, usability, compatibility, or anything else.
Getting near-identical performance on a pre-beta OS is damn near a miracle, as most people who've been this befeore can attest.
SPTD refuses to run on anything that's a beta, it's well known, nothing new, and isn't a compatibility issue. Why is someone expecting a ring-0 SCSI emulation driver to work on Windows 7 as soon as any developer builds are out the door, anyway?
Inherent multi-core scalability, DWM/Aero, WDDM, Resource Monitor, Explorer, and the kernel have all received pretty major upgrades.
Does anyone remember NT4 to Win2K differences? XP to Vista was like that. Win2K to XP differences were fairly minor, but incremental, and very useful, and everyone loves them now...called WinXP 'the worst OS ever', and 'another WinME', on day one (and before), too. Windows 7 more represents Win2K to XP, but isn't shying away from meaningful changes.
Let's take ReadyBoost, for instance. It was introduced in Vista with a great deal of hype...which was mostly disappointing for limitations. In this release, they've enhanced it, enabled dedicating a USB flash drive to ReadyBoost specifically, allowing the use of -multiple- USB drives, allowed the use of ExFAT, allowed the use of slower drives (particularly with FAT16/ExFAT). A lot of the claimed "Windows 7 boots faster"...can already be experienced with a pair of sludge-cheap $5 2GB usb keys used in tandem with ReadyBoost. Everything seriously launches oodles faster, but Windows 7 tends to launch and boot significantly faster than Vista with a single 2GB ReadyBoost key.
Windows 7's kernel received a few meaningful enhancements, like some heap error correction. DWM takes advantage of DirectX 10.1 class hardware, has little overhead or compatibility issues now. Sound drivers have sampling rate enforced more sanely to prevent needless resampling issues. Filesystem operations tend to scale far better with more than one CPU (finally).
Aside from the pre-beta "unfinished UI" issues, I'd be happy to use the PDC build every day to replace Vista completely in a heartbeat for full-time everyday use.
I'm tired of the bloody nit-picking. We're at least 7 months away from Windows 7 RTM, can't the so-called bloggers find something more useful to do than claim imaginary faults with an OS not even close to being out yet and stir up yet more drama and controversy?
I'm just as tired of people doing it with various aspects/versions of Linux/BSD/Solaris/wine.
Slashdot, frankly, should know a bit better. A article like that isn't news, it's a troll.
I think the bottom line is that the majority of the focus on Windows 7 has been usability, with a fair amount on performance/functionality, with a very small subset focusing on 'eye candy'.
SuperBar isn't flashy. It focuses almost exclusively on UI functionality, doesn't look any different really than regular taskbar. There are a few new 'user visible' Aero features (like the 'Shake' thing?), but the real bulk of changes have been under the hood, with a surprising number of applications and utilities getting improved.
The article's kind of fear mongering is simply assinine.
"A Goddess rarely smiles for she is forced by others to be an island unto herself." - Zephiris