32bit Win7 Vs. Vista Vs. XP 641
An anonymous reader writes "ZDNet's Adrian Kingsley-Hughes tested the latest Win7 build against XP and Vista and came to a surprising conclusion: Win7 performs better than the other 2 OSs in the vast majority of the 23 tasks tested. Even installation. 'Rather than publish a series of benchmark results for the three operating systems (something which Microsoft frowns upon for beta builds, not to mention the fact that the final numbers only really matter for the release candidate and RTM builds), I've decided to put Windows 7, Vista and XP head-to-head in a series of real-world tests...'" This review shows only a 1-2-3 ranking for each test, so there's no sense of the quantitative level of improvement.
Still making 32 bit? (Score:5, Interesting)
Re:I question the results. (Score:4, Interesting)
that just sounds like a fisherman tale....
Rating is bull (Score:1, Interesting)
How can vista boot up faster than XP.. i have never seen this is a real setup... windows 7 faster vista (I can believe that) but vista & windows 7 faster than XP.. Like above said.. take with a grain of salt..
win7 rocks (Score:2, Interesting)
No 64bit test and a 4gb system? (Score:3, Interesting)
Do a 64bit test as well as most system today with with 3-4gb ram + video ram and other system stuff go over the 4gb limit 32bit.
Poor (Score:1, Interesting)
Okay, so ignoring the fact that actually XP is "better" than Vista on machines with 4Gb and somehow "Windows 7" is top on virtually all tests on both types of machines (does the word "bollocks" mean anything to you?), this is ENTIRELY subjective.
For instance, one category is "Burning a DVD" with CDBurnerXP Pro. Somehow, XP gets a 3rd place or a 2 while Vista gets a 3 or a 1st and Windows 7 gets a 1 or a 2 (and, in fact, is one of the very few "non-1st" marks awarded to Windows 7). WTF were you measuring? How can you "rank" the burning of an ISO to DVD in the same software on two seperate machines differently between ANY vaguely similar OS's?
The *only* factor that differs is speed, so you're telling me that Windows 7 can burn disks faster than XP or Vista? Fine... show me the statistics, because I don't believe that XP or Windows 7 are that different when it comes to throwing some data down an IDE/SATA cable, yet somehow this idiot has "ranked" the OS's by some criteria and declared Windows 7 a winner.
Subjective, zero evidence for the reasons of the rankings, stupid scale (1st, 2nd, 3rd, then add up the place rankings and see who got lowest - not one single entry where there's a tied-first or other place, so it takes two "first places" to recover from one "third" place on another category), stupid benchmarks in the first place (i.e. burning a DVD is a valid benchmark but not when you don't say what you are measuring and/or what each OS scored - if one OS finished 0.00001 of a second later than the fastest OS, does that put it in 3rd place, for instance?), blatant sucking up.
If you're gonna claim to be a technology journalist and do such a comparison, at least do it vaguely correctly.
And, yeah, it's purely guesswork but the disclosures section on the author says nothing and yet everything I can find from him (even on Linux.com) is anti-Linux, pro-Microsoft and even the hint of possibly-future-pro-Apple stuff he mentions in passing never shows up as anything other than anti-Apple sentiment. We all have our opinions but this guy's just out to boost MS. Either he gets a lot of nice stuff in the post *cough* Windows 7 Beta's, Microsoft hardware to review *cough* or he's on the payroll.
Disregard.
Re:Still making 32 bit? (Score:4, Interesting)
What bothers me about Vista 64 is that Microsoft do not let you load unsigned drivers. Got a driver from a vendor that's not signed? You have to go through the trouble of signing it yourself and kicking your OS into test mode. The problem became worse with SP1 when MS made several known workarounds disappear.
I understand they're trying to work against root kits but I'd rather be able to easily install any drivers I choose on my own system then have Microsoft protecting me against myself and causing me all kinds of grief. I've also never been hit by a root kit and I would guess that regular viruses are just as problematic and more common for nearly everyone.
Let's wait for the final version (Score:2, Interesting)
Does Win7 seem faster than Vista or XP ?
Don't worry, Microsoft has still plenty of time to fix this behaviour !
BTW, the article is really lame, since there is absolutely no indication why Win7 is faster.
How much did the writer get paid by Microsoft for this advertisement ?
Re:Still making 32 bit? (Score:3, Interesting)
Re:win7 rocks (Score:2, Interesting)
Re:I question the results. (Score:3, Interesting)
Partitioning the drive won't make the test any more fair. It may lessen fragmentation between each "chunk" of the drive than an OS would ordinarily take (if you decided to falsely assume that you can put more than one copy of Windows on a single partition without it blowing up).
Hard drives are cheap, and quite re-usable. Get three identical ones. Do your testing, throw the results online, and then reformat the drives and throw them in the nearest fileserver.
Re:Still making 32 bit? (Score:5, Interesting)
I agree. Nobody is selling 32-bit processors anymore.
Intel's Atom processor is 32-bit.
Linux can handle 32-bit applications on 64-bit OSes. Surely MS can do the same?
It's the proprietary drivers that make it hard for MS to do the same. In Linux the vast majority of drivers are maintained in source, so this isn't as much of a problem.
Re:Microsoft has a good version of Vista! (Score:4, Interesting)
Re:I question the results. (Score:5, Interesting)
Here are some benchmarks right over at tomshardware [tomshardware.com] that show that the "SlashDot world" in this case is accurate (amazing!).
This was on a system with 2 GB of RAM, so according to you Vista should have been faster, but it wasn't. So your idea that it's the RAM that's the problem is bollocks.
Anecdotally, a colleague of mine was complaing her brand new lenovo thinkpad with Vista was slow compared to her imac -- she was kind of amazed that the they had the same processor and memory.
Re:Still making 32 bit? (Score:4, Interesting)
And in practice extra x86-64 registers are not that great improvement because modern CPUs got very good at pipelining and data prefetching.
Good point. However, those extra registers may matter quite a bit more for something like the 64-bit Atom processors, which deliberately forgo most speculative features that mitigate register pressure. It would be interesting to see whether it's a better use of silicon to make an out-of-order processor or a 64-bit in-order processor when you're operating under the power constraints of the Atom. The current existence of 64-bit in-order Atom processors suggests that the performance per watt impact of 64-bit is better than out-of-order execution. I suspect this is because 64-bit takes less silicon than OOE, in a similar manner to how useful a good implementation of simultaneous multithreading can be.
Re:I question the results. (Score:2, Interesting)
I did that too, but only after making floppy disk images of all of them.. Never know when MSDOS 3.0-3.2 and 6 might be useful.. not to mention you don't even need to write the images to a floppy disk, you just boot from the floppy image in either vmware or vbox...
Re:Completely useless (Score:1, Interesting)
yes. How dare I remain a skeptic. How dare I question something that none of us have anything to compare to.
Is that what you're saying? That we should just lap up zdnet which is known to basically love microsoft unconditionally in the first place? Or the "how can he possibly score windows 7 better in every single category" except for two, part? You don't find that suspicious before the OS has even been released?
When you see benchmarks from hardocp, even tomshardware (as much as they're biased sometimes), or any other reputable website then we have something to debate or believe. I don't doubt windows 7 will be an improvement but not only is the thing not out yet but basically they can't disclose a benchmark. You're saying that dancing the EULA must mean that his information is reputable.
Maybe next time, you should think before you post some completely inane crap.
Re:Two reasons for this (Score:3, Interesting)
Cool idea. Now let's take a step back and look at the realities:
Port VitrualPC to Win7: I suspect it'll already run on Win7, but even if it won't they'd do that anyhow.
Drivers to share the clipboard: Sure. Of course, you'd also need drivers to handle OLE stuff (drag-and-drop, for example). I'm sure it could be done, but don't make the mistake of assuming it's trivial. It takes a bit of work (made utterly painless, but still required) just to allow near-seamless mouse movement in and out of the virtual window.
Launching virtual machine now launches the application rather than the desktop: this is pretty easy. Of course, you still need to account for the bootup time of XP. Even with hardware virtualization, this is at least tens of seconds. I'd really rather not wait that long every time.
Virtually 100% all current software would work: Except, you know, anything that needs 3d hardware acceleration. Or direct driver access. Or more than two COM ports (yes, such programs exist, and VPC's limitation to 2 COM ports is an issue for the one program we have that won't quite work right in Vista. The problem could be worked around, but it's indicative of the greater issue).
Sandboxed by default. How sandboxed? Windows supports an incredible number of forms of inter-process communication. Some programs rely quite heavily on such things. You could allow the VPC to run one process and all the programs that it spawns, perhaps, but there would still be problems. Hell, this sort of excessive sandboxing is supposedly the reason the iPhone can't even handle simple cut/copy/paste!
How much RAM do these virtual systems have? Each virtual machine would need a good chunk of RAM, especially with the overhead of running all those excess copies of Windows. However, they would also compete with native apps for physical RAM. What do you do when some process that runs on Windows 2000 starts demanding 2GB of working set? Is VPC supposed to automatically enlarge the physical RAM allocated to that machine? Is it supposed to use its own pagefile? Perhaps you'd like to somehow get it to use the global pagefile instead?
I hope this is enough to help you realize that, noble though your end goal is, your method simply would not work.
Re:I question the results. (Score:3, Interesting)
It's not pointless.
Re:I question the results. (Score:3, Interesting)
Re:I question the results. (Score:5, Interesting)
Beyond that, I have yet to see any conclusive benchmarks posted by the defenders of Vista on this thread showing any proof that Vista is faster than XP, just empty assertions. What I do see is a bunch of Microsoft fanboys comforting themselves that their favorite brand released an OS that has turned out to be a flop.
Let me qualify my positions here though. I have Vista installed on an old hard drive on a brand new PC -- my own conclusion is that Vista is not as bad as everyone makes out, but you all need to stop pretending that Vista is fast. It isn't. It's not terribly slow on nice hardware, and it looks very nice and it has some nice features, e.g., the DX10 features on new games, but it's not fast.
Re:I question the results. (Score:5, Interesting)
True. I wasn't exactly clear. I'm talking OS audio subsystem for delivering audio from apps to the hardware. Not apps.
Right. Zero latency. Talk about lies. It establishes callbacks in the apps, writing into shared memory segments which are then mixed and delivered to the standard linux audio device. Yeah. Zero latency as long as you stay ahead of the playback. Just like pretty much every sound system since the days of the original Soundblaster Pro using DMA. Where's the signal processing layer in there? Oh, it's third party. Where's the channel synchronization? Can't find it. And awesome how it punts sample rate changes back to the apps. And it uses floats as the sample format? Talk about a really bad design decision. I mean you get three of four apps going in hi definition audio (96/24/7.1) and you're going to be seeing twenty or thirty percent of your system going down the shit hole just to do sample format conversions. And what is the upside? Nothing. For every 32 bits of sample data you get 24 bits of mantissa and a useless exponent. And shockingly enough it's all software. Where's that hardware acceleration you're so fond of?
And what happens under load and the realtime scheduler can't quite keep up? Ah, I see, you get drop outs. What happens on Vista? Nothing, they hook into the scheduler to guarantee that their audio paths get time on the CPU.
It's not a matter of delaying individual streams. It's a matter of delaying individual channels from the same stream. So that your rear speakers sitting against the far wall behind you play just a bit earlier.
Re:Two reasons for this (Score:3, Interesting)
Drivers to share the clipboard: Sure. Of course, you'd also need drivers to handle OLE stuff (drag-and-drop, for example). I'm sure it could be done, but don't make the mistake of assuming it's trivial. It takes a bit of work (made utterly painless, but still required) just to allow near-seamless mouse movement in and out of the virtual window.
This stuff is already done by their competitors. So, even if it isn't 'trivial', it is something they will have to do anyway.
Virtually 100% all current software would work: Except, you know, anything that needs 3d hardware acceleration. Or direct driver access. Or more than two COM ports (yes, such programs exist, and VPC's limitation to 2 COM ports is an issue for the one program we have that won't quite work right in Vista. The problem could be worked around, but it's indicative of the greater issue).
So, your saying that MS sucks too much and does not have the resources to bring their product up the match their competitors who are already on the market? I'm not buying it.
Sandboxed by default. How sandboxed? Windows supports an incredible number of forms of inter-process communication. Some programs rely quite heavily on such things. You could allow the VPC to run one process and all the programs that it spawns, perhaps, but there would still be problems.
MOST applications don't share data via inter-process with other applications beyond a simple clip board. But even for the ones that do, you would at least know what applications were trying to access what. Even if you had the simple choices of "No Access", "Clipboard Access", and "Full Access", you would be head and shoulders above what we have now because even if some users just always said "Full Access", they wouldn't be worse off, and anybody that has any concern for security isn't going to do that. I know that I would certainly click "No Access" when my freeware Falangy Counting software asked for access to Quicken.
Hell, this sort of excessive sandboxing is supposedly the reason the iPhone can't even handle simple cut/copy/paste!
That is total BS. Cut/Copy/Paste does not work because while the iPhone interface is neat, it has some serious problems, and Apple simply chose not to implement cut/copy/paste. Likely because they felt it would make the interface too complicated.
How much RAM do these virtual systems have? Each virtual machine would need a good chunk of RAM, especially with the overhead of running all those excess copies of Windows. However, they would also compete with native apps for physical RAM. What do you do when some process that runs on Windows 2000 starts demanding 2GB of working set? Is VPC supposed to automatically enlarge the physical RAM allocated to that machine? Is it supposed to use its own pagefile? Perhaps you'd like to somehow get it to use the global pagefile instead?
Memory actually gets BETTER if you had them running in a virtual machine. Currently an XP application can only access 3 gigs of ram, and anything beyond that must swap to a physical disk. You see, back in the days before dirt was invented and the memory that an OS would use could be counted in bytes, we had this thing called a "RAM Disk". It was a chunk of memory that the OS saw as a physical disk, even though it was in RAM. Since a 64-bit OS could access 16 exabytes of RAM, the RAM disk starts to make sense again. If the machine were virtualized over a 64 bit OS, the swap file could be set to run from the disk, or it could be set to run from a ram disk which would mean that you could get far more real memory to the 32 bit application virtualized than you could from running it natively on a 32-bit OS.
I hope this is enough to help you realize that, noble though your end goal is, your method simply would not work.
No, this does not help me realize that the method would not work. It just shows that you are being short sighted about OSes, and don't really understand how computers work. Or have just drank the "MS said so, so it must be true" Cool-Aid concerning backward compatibility.
Speed: XP Windows 7 Vista (Score:1, Interesting)
Here are my benchmarks using SETI@Home on an Intel Quad Core 9660 with 4 GB RAM. All are fresh installs of the 32 bit versions of each OS on their own SATA2 drives with all updates installed. YMMV
Windows XP SP3
Float Pt (MOPS) 2999.2
Integer Spd (MOPS) 6238.63
Turnaround (days) 0.11
Windows Vista SP1
Float Pt (MOPS) 2935.12
Integer Spd (MOPS) 6076.71
Turnaround (days) 0.20
Windows 7
Float Pt (MOPS) 2910.61
Integer Spd (MOPS) 5714.78
Turnaround (days) 0.19
So Windows gets slightly less performance from the same hardware with each new version but Windows 7 does get slightly more SETI@Home results done in a day than Vista though still not as many as XP.
Re:I question the results. (Score:3, Interesting)
If I didn't have compatibility issues with Windows 2000, I would still be using it (for games, that is). It is the only Microsoft product I have ever been content with.
Re:I question the results. (Score:1, Interesting)
So what you're saying, is that Microsoft deliberately crippled SP3 to make Vista look better?
Re:Still making 32 bit? (Score:3, Interesting)
"Rootkits? You wish. Microsoft was trying to prevent people from doing sneaky things like rip DVDs and record the audio output from programs like Windows Media Player to "rip" DRMed files."
Vista x64 does not prevent people from ripping audio streams, DVD disks or Blu-Ray disks. Almost all the tools that allow this on 32 bit windows work on 64 bit windows. Even the Slysoft people don't have a problem with 64 bit windows.
Re:Still making 32 bit? (Score:3, Interesting)
PowerPC is freaking ancient and were supported for 7-8 years, which is all you can reasonably expect
The last PowerPC Mac I bought was less than 8 years ago. They only started selling Intel Macs in 2006 - about three years ago from the launch of Snow Leopard - and they didn't transition all of their product lines until some time in 2007. Leopard was the only version of the OS to be released after the switch. Not supporting three-year-old hardware is pretty poor, even by Apple's standards.
Re:I question the results. (Score:3, Interesting)
The installer will tell you to reboot anyway, but the driver has been updated.
This is possibly the best feature in Vista
Just how badly does XP SP3 hurt performance? (Score:3, Interesting)
> The test you link to used SP2, while the new tests use SP3. XP SP2 and SP3 aren't the same thing. In fact, most benchmarks put Vista SP1 ahead of XP SP3 or at least within spitting distance of each other.
I think that's pretty telling, actually, assuming it's the reason. Did Microsoft manage to destroy XP's performance with SP3 enough that it's now below Vista? Did their software department design that "upgrade" or did marketing? (Assuming the two departments haven't been unified this whole time...)
Or do we have some bad benchmark data here?