Posted January 12, 2009

Weclock
The Creeper
Registered: Sep 2008
From United States

hansschmucker
Micro-Shambler
Registered: Sep 2008
From Germany
Posted January 12, 2009
It happened with even smaller files; More like 10KB per file. I haven't tried it yet with files in the low MB ranges, but I doubt it'll happen for anything but files in the 4kb to 200kb range.

tomba4
resU weN
Registered: Sep 2008
From Sweden
Posted January 13, 2009

Josh04
New User
Registered: Sep 2008
From United Kingdom
Posted January 13, 2009

This was the same in Vista, but probably worse in XP, where they could have been in either "Application Data" or "Local Settings/Application Data". At least they're grouped.

hansschmucker
Micro-Shambler
Registered: Sep 2008
From Germany
Posted January 13, 2009


I'm not exactly sure how an overlay works... seems like colorkeying for later filling by the GPU. If that's the case, then it probably won't work in Vista or 7 as the compositor will likely get in the way. Besides, you'd get a lot of pink shadows :)
Yay, just noticed that they finally fixed one terribly annoying bug in fsutil, which made it pretty much impossible to get a drive listing in batch files on XP: Previously, the drive entries where separated by 0 chars. Sadly, a 0 char also terminates a string, so it was not possible to save the output of fsutil beyond the first drive letter to a shell variable. Now, it's finally a space. I know this isn't one of the most obvious problems of XP, but it is one that bothered me, as plenty of tasks are automated through batch files on my system.
And locate32 works.... I'm saved!
Post edited January 13, 2009 by hansschmucker

JudasIscariot
Thievin' Bastard
Registered: Oct 2008
From Poland
Posted January 25, 2009
Sorry for the idiot question here, but I don't have another PC to use for this beta. Can I just choose to upgrade with this beta and then roll back so to speak if I am not satisfied or is this a Full-will-wipe-everything-you-have-install??? I remember the other Windows versions having an uninstall option in case you didn't like the upgrade...

TapeWorm
Old bastard
Registered: Sep 2008
From Canada
Posted January 25, 2009

I don't know for certain, but I do recall reading on the download page that "you'll have to reinstall your previous operating system once the beta has expired (August)" or something to that effect, so I'm going to say no, it's unlikely you'll be able to just roll it back.
Regardless, I think you should either install it to another partition or drive and dual boot it if you're looking to try it (that's what I did). It'll save you a lot of grief. At least then you can just blow it away when you're done with it.

JudasIscariot
Thievin' Bastard
Registered: Oct 2008
From Poland
Posted January 25, 2009


Regardless, I think you should either install it to another partition or drive and dual boot it if you're looking to try it (that's what I did). It'll save you a lot of grief. At least then you can just blow it away when you're done with it.
Cool thanks for the heads up. Now I just have to clear a million gigs of space ...wheeee!!

Weclock
The Creeper
Registered: Sep 2008
From United States
Posted January 25, 2009

JudasIscariot
Thievin' Bastard
Registered: Oct 2008
From Poland
Posted January 25, 2009
I can play System Shock 2 on this OS without a hassle and thousands of hours scouring different forums to try and find a solution. Now it's click and play baby! Hell, even my virgin copy of Fallout 1 works fine after setting some compatibility mode settings. Goodbye Vista! Hello pretty OS! (Yes I am sure Linux is better as an OS but there's too many damn distros out there.....)

Weclock
The Creeper
Registered: Sep 2008
From United States
Posted January 25, 2009
there was a benchmark done to compare the 3 operating systems (vista, 7, xp) and they said XP still outperformed 7 and that 7 only performed marginally better over vista, but this was because vista and 7 both have special code (read: extra code) to work with multi core processors while XP does not.
this argument has sealed it for me, I'm going to put XP on my desktop.
this argument has sealed it for me, I'm going to put XP on my desktop.

TapeWorm
Old bastard
Registered: Sep 2008
From Canada
Posted January 25, 2009
(read: extra code)
You mean a code path? You'd be terribly surprised at how often this kind of thing is done to handle special cases in so many applications under so many different operating systems. It's really very normal. After all, I wouldn't want multi-core specific code running on a single core system, especially if that multi-core specific code could crash and/or slow things down on a single core system.
As for claims of speed:
http://blogs.zdnet.com/hardware/?p=3236
Also, did your benchmark measure the speed of a milestone version or the recent 1st Beta? There's a difference from what I've been reading.
That benchmark claims that 7 beats XP handily (which I think could be bull, but I'm not about to compare the 3 systems to find out). So who's right? At this point: no one. It's a beta, it'll likely have debug code in it which will slow it down. So don't make a final judgement until it's out in its finalized form and it's been used by people. There's bound to be lots of hype, both for and against and a lot of it is merely speculation and crap.
Anyway... Unless a user has a real twisted desire to mess up their system in the name of science or they get some sick perverse pleasure at installing operating systems, I don't think it's a good idea to just install it for the hell of it and I especially think it's a bad idea to use it for day to day operations - unless you're looking at porn. Enter at your own risk as they say...

Weclock
The Creeper
Registered: Sep 2008
From United States
Posted January 25, 2009
that thing i was talking about previously
someone broke the goddamn hotlinks.
interesting, if for more than technical reasons
edit:
meh, i'm willing to goof off with my macbook. 32bit versions only, I'm running xp prof sp3 right now tell me what tests you want to see performed (i'm hoping they have the ability to automatically save the scores) then I'll install win7 beta and post the results of the same test.
if only because i want to know the answer, and because I want to try the damn thing anyway.
someone broke the goddamn hotlinks.
interesting, if for more than technical reasons
edit:
meh, i'm willing to goof off with my macbook. 32bit versions only, I'm running xp prof sp3 right now tell me what tests you want to see performed (i'm hoping they have the ability to automatically save the scores) then I'll install win7 beta and post the results of the same test.
if only because i want to know the answer, and because I want to try the damn thing anyway.
Post edited January 25, 2009 by Weclock

Gambler
Registered: Nov 2008
From United States
Posted January 25, 2009
I didn't do any file operation or gaming tests yet, but Windows 7 loads a minute 40 seconds on the machine where XP loads just 40 seconds. And it has 70MB registry out of the box against 35MB for XP.
That's on my home computer. I also installed it at an older workstation at work. While it's usable, it's definitely slower than Windows 2000 there.
All the praise in the media looks very, very fishy.
Disk throughoutput tests (write, sequential read, random read):
XP: 23.4, 447.3, 445.5
Windows 7: 26.0, 318.5, 342.1
Fater writes, slowe reads. Strange.
That's on my home computer. I also installed it at an older workstation at work. While it's usable, it's definitely slower than Windows 2000 there.
All the praise in the media looks very, very fishy.
Disk throughoutput tests (write, sequential read, random read):
XP: 23.4, 447.3, 445.5
Windows 7: 26.0, 318.5, 342.1
Fater writes, slowe reads. Strange.
Post edited January 25, 2009 by Gambler

TapeWorm
Old bastard
Registered: Sep 2008
From Canada
Posted January 26, 2009

if only because i want to know the answer, and because I want to try the damn thing anyway.
Oh fantastic. Fyi, IE8 was terribly buggy. That was -my- experience with it, so YMMV. And VPC wasn't entirely stable either if you're planning on using that. With that said, here's my recommendations:
Obviously some games would be a good place to start (I'd choose a cross section of games from varying eras and genres: Unreal 1, Doom 3, Crysis, Mass Effect, Warcraft or WOW, etc..). That's the easiest and they tend to stress hardware pretty good in real world situations.
But since they're stressing about database efficency in that article, I'd like you to pop on MS-SQL Express(no, I'm not being sarcastic, I think you can download it for free from MS). I want to see if what they're talking about is bullshit or not. I work with databases via Vista every day (client side only though, keep in mind that MSSQL is tuned for server environments, so it's not ideal on a workstation) and I've yet to see any performance differences. Sometimes my coworkers XP machine is slower than mine, other times it's the reverse given any number of random factors, and -most- of the time it's the same speed. I often confirm this since we run timing tests on our queries (using high performance counters if anyone is questioning how, the same timing used to calculate timings in games). We need to ensure our Vista performance is decent because our client is doing an upgrade to Vista soon (they're still on Win 2k for the majority of their workstations).
Anyway, go get MSSQL Express, start adding/updating/reading data using standard SQL queries and stored procedures.
I put very little stock in these benchmarks, especially concerning beta software. Often times they test for limited edge cases that either never come up or come up rarely. The important thing is every day performance, and even perceived performance. This is especially important if it means the difference between your clients/users destroying their hardware in frustration or just being reassured it's -doing something-.
I also installed it at an older workstation at work. While it's usable, it's definitely slower than Windows 2000 there.
All the praise in the media looks very, very fishy.
Disk throughoutput tests (write, sequential read, random read):
XP: 23.4, 447.3, 445.5
Windows 7: 26.0, 318.5, 342.
My XP box at work boots in about 1 minute and change, and my vista laptop boots at about 40-50 seconds. Win 7 on my home machine (which, admittedly, is quite beastly and I'm running the x64 version of Win7) booted in 20 seconds.
Compared to Win 2000 even XP runs like a drunken turtle. So I'm not surprised that Win 7 runs slower compared to that.
Benchmark tests are really meaningless in my opinion. Hardware gets better, software gets better and better optimized. Including operating systems. For example, I see a big difference between Vista post SP1 and pre SP1 both in speed and stability. Sadly that gets reset when they come out with a new version it seems. I remember when XP came out, Christ it was so slow booting and just slow in general, and it crashed every day on me. After SP2 it really got a lot better (SP1 was an improvement, but there were still issues). But compared to the Win 2k machines I used at that time for work, it was still a pig.
Post edited January 26, 2009 by TapeWorm