Linked by Thom Holwerda on Fri 12th Dec 2008 23:44 UTC, submitted by google_ninja
I'm sure you're all still (sadly) familiar with the recent 'debate' I had with InfoWorld's Randall Kennedy, which detailed a lot of silly things. The seed of that discussion was planted with Kennedy's first article which, among other things, claimed that Windows 7 performed similarly to Windows Vista (meaning, slower than XP). Leaving the thread count discussion behind, Kennedy did include a benchmark which showed that Windows 7 performed similar to Windows Vista. There's a new benchmark out now, comparing a slightly more recent build of Windows 7 to Vista RTM/SP1 and XP SP3, and in these tests, Windows 7 blows all of those out of the water.
Thread beginning with comment 340182
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[3]: Microsoft Windows and User accounts
by soonerproud on Sat 13th Dec 2008 21:57
in reply to "RE[2]: Microsoft Windows and User accounts"
Member since:2008-03-05
But, the most vital thing Windows NEEDS to do is to stop permitting root-kits to be installed. This requires a redesign.
You have a lot of valid points but I have to take exception with this one. Windows is not the only popular operating system that can be affected by rootkits. OSX and Linux are vulnerable to rootkits too. So are you also suggesting a major redesign for those operating systems to prevent rootkit infections too?
Edit:
Another thing, with UAC enabled Vista is very hardened against rootkits. UAC is not perfect, but there is no denying it is effective when a smart user is behind the keyboard. Even UAC can't fix stupid.
This report was released in May of this year.
http://www.pcworld.com/businesscenter/article/146256/vistas_despise...
Edited 2008-12-13 22:02 UTC
RE[4]: Microsoft Windows and User accounts
by looncraz on Sun 14th Dec 2008 03:16
in reply to "RE[3]: Microsoft Windows and User accounts"
Member since:2005-07-24
Yes, indeed, any OS supporting root-kits needs to find a way around that, entirely.
They have their uses, but they are a bigger problem than a benefit. Ideally the OS would provide hooks for scanning software, rather than blindly allowing what root-kits allow.
UAC doesn't help the idiots - and they need to be protected more so than anyone else - they outnumber us. The more infected machines there are, the larger the problem, obviously.
--The loon
Recent Original Stories
Recent Comments
Headlines
Random Comments
Random Stories
Random OS Link
Member since:
2005-07-24
You can blame the OS, I've seen it done many times. It just doesn't make you right :-)
Windows COULD make it possible to install apps without admin rights, restricting access to local effects only.
It COULD do what it SHOULD do, and ask for an admin password in order to install the app ( like Linux ), rather than requiring a round-a-bout method - but there is something to be said for the Windows method as well.
What Windows MUST do, is require an Administrator password. Simple. All accounts with admin rights should also require passwords.
What Windows SHOULD do, as well, is completely segregate the registry entries for each application. Store a file for any entry an app makes, so that uninstalling that app will uninstall that registry hive. Make it MORE work to effect the registry globally, and simple to effect the app's own store.
The only global entry an app would get free is its installation entry, which it could not manipulate. That would fix an incredible amount of problems, and even stop many viruses and spyware.
But, the most vital thing Windows NEEDS to do is to stop permitting root-kits to be installed. This requires a redesign.
Another thing of note, is that the UI can be made to respond differently based upon the threat level of what is happening. Users rarely read anything other than the bold text - or maybe look at the icon. A warning about not having a CD in the tray looks nearly identical to one about a program trying to modify the kernel directly.
Of course, I could go on, but I'm on my third glass of wine and I think my current torrent is done... maybe I'll reboot to Windows Vista ( gotta know all OSes, y'know ) and play some Crysis...
--The loon
-- and relating more so the article:
A 'jump' of .7% is hardly worth writing home about, but is a start. Thing is, though, that Vista requires more hardware to get what Windows XP gets from it, while it can do more with the RAM it has. Windows 7 hasn't corrected enough to merit placement as an 'improvement' in terms of performance. Maybe a jump of 2-2% would merit the lowest entry of 'improvement' without specifying 'minor.' Whereas, 5-6% would be safely called a 'good improvement' and 2000% would be in line with what Microsoft typically tries to convince has been accomplished.
In any event, my upgraded system does run Vista better than XP in *some* cases. BUT, BeOS, Haiku, & Ubuntu all have - at the minimum - a 50% performance benefit ( BeOS about 5000% ) in the most important metric - perceived performance.
Edited 2008-12-13 04:38 UTC