| ActiveMac | Anonymous | Create a User | Reviews | News | Forums | Advertise |
|
|
|
User Controls |
|
New User |
|
Login |
|
Edit/View My Profile |
|
|
|
|
|
|
|
ActiveMac |
|
Articles |
|
Forums |
|
Links |
|
News |
|
News Search |
|
Reviews |
|
|
|
|
|
|
|
News Centers |
|
Windows/Microsoft |
|
DVD |
|
ActiveHardware |
|
Xbox |
|
MaINTosh |
|
News Search |
|
|
|
|
|
|
|
ANet Chats |
|
The Lobby |
|
Special Events Room |
|
Developer's Lounge |
|
XBox Chat |
|
|
|
|
|
|
|
FAQ's |
|
Windows 98/98 SE |
|
Windows 2000 |
|
Windows Me |
|
Windows "Whistler" XP |
|
Windows CE |
|
Internet Explorer 6 |
|
Internet Explorer 5 |
|
Xbox |
|
DirectX |
|
DVD's |
|
|
|
|
|
|
|
TopTechTips |
|
Registry Tips |
|
Windows 95/98 |
|
Windows 2000 |
|
Internet Explorer 4 |
|
Internet Explorer 5 |
|
Windows NT Tips |
|
Program Tips |
|
Easter Eggs |
|
Hardware |
|
DVD |
|
|
|
|
|
|
|
Latest Reviews |
|
Applications |
|
Microsoft Windows XP Professional |
|
Norton SystemWorks 2002 |
|
|
|
Hardware |
|
Intel Personal Audio Player 3000 |
|
Microsoft Wireless IntelliMouse Explorer |
|
|
|
|
|
|
|
Site News/Info |
|
About This Site |
|
Affiliates |
|
ANet Forums |
|
Contact Us |
|
Default Home Page |
|
Link To Us |
|
Links |
|
Member Pages |
|
Site Search |
|
Awards |
|
|
|
|
|
|
|
Credits |
 |
Study: Windows 7 can boot more slowly than Vista | |
|
||
| Time: 09:22 EST/14:22 GMT | News Source: ZDNet | Posted By: John Quigley | ||
|
Although Windows 7 has been praised for loading and shutting down faster than prior versions, according to one software company, in many cases the new operating system can take longer to get started than Windows Vista. Iolo Technlogies, which sells PC tune-up software, said its lab unit found that a new machine running Windows 7 takes one minute and 34 seconds to become usable, as compared to one minute and six seconds for Windows Vista. Iolo said it measured not the time it takes for the desktop to appear — which can be as little as 40 seconds on a fresh installation of Windows 7 — but rather the time it takes to become fully usable, "with CPU cycles no longer significantly high and a true idle state achieved". | ||
|
Write Comment
Return to News |
Displaying 1 through 25 of 336 Last | Next |
The time now is
12:38:47 PM
ET.
Any comment problems? E-mail us |
| #1 By 11888 (198.103.167.20) at 10/8/2009 11:22:40 AM |
|
This is getting silly. Who is rebooting so often that a few seconds here or there matters?
Boot the bastard up and after a while try to remember the last time you had to reboot. |
| #2 By 2960 (72.196.201.130) at 10/8/2009 11:27:39 AM |
|
"Iolo Technlogies, which sells PC tune-up software..."
Well now. There ya go. |
| #3 By 243849 (216.171.102.38) at 10/8/2009 11:29:59 AM |
|
Obvious conflict of interest here. They sell tune up software and are likely going to see a drop in sales because Win7 performs better.
Also, Win7 has delayed service starting, so obviously "with CPU cycles no longer significantly high and a true idle state achieved" will give higher results. |
| #4 By 230538 (74.219.160.9) at 10/8/2009 12:37:09 PM |
|
Does anyone know if Win7 has parallel service start-up to quicken start-up times?
I know *nixes have this ability and this is how they have quickened their startup times. OSuSe 11.1 (x64) when compared to Vista (x64) (dual-boot) was A LOT quicker. Have been running Win7x64 for the last couple months and it's A LOT faster than Vista and about on par or faster than as OSuSe . Plus, Win7 is great. I hate coming to work because I have to work on a XP machine. ha-ha |
| #5 By 15406 (216.191.227.68) at 10/8/2009 12:48:25 PM |
|
#1: Boot the bastard up and after a while try to remember the last time you had to reboot.
At least monthly via Windows Update, or if you install any software more complicated than a game. #2: Well now. There ya go. Are they offering something that will speed up the boot time? If not, the connection is irrelevant. #3: Obvious conflict of interest here. See #2 above. Also, Win7 has delayed service starting, so obviously "with CPU cycles no longer significantly high and a true idle state achieved" will give higher results. So you're in agreement with the article's assessment then that 7 boots slower than Vista? |
| #6 By 89249 (64.207.240.90) at 10/8/2009 1:38:03 PM |
|
#5 Man you're right. Monthly we're talking about lets say 20 seconds of reboot time for updates. That's out of 2592000 seconds in a month. But lets say that we're talking about 100,000 computers in a huge school district. That's 555 HOURS OF REBOOT TIME.
RUNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! See I multiplied a insignificant number by a bajillion to make it significant than yelled "FIRE!" I R SMRT This message brought to you by zombies for linux... maybe another decade will help. |
| #7 By 20505 (216.102.144.11) at 10/8/2009 2:03:59 PM |
|
I dunno I converted my sons computer with a 2.1GHz processor and an old P965 chip set and 2GB of RAM into the fastest booting computer in my house with Win7 RC. (It may be serendipitous, but it's nice).
By the way this does make a difference to me. I always shut down my computer when I'm done using it. I am trying to be responsible with energy usage. |
| #8 By 15406 (216.191.227.68) at 10/8/2009 2:04:45 PM |
|
#6: Man you're right.
I know, thank you. Monthly we're talking about lets say 20 seconds of reboot time for updates. Your system reboots Vista in 20 seconds? You've got the best PC on the planet, my friend, and for that I am jealous. See I multiplied a insignificant number by a bajillion to make it significant than yelled "FIRE!" Yes, I understand from that previous thread that you really need a remedial course in economics. The concept of compound interest, where a small number becomes a big number when factored over time, would blow your mind. This message brought to you by zombies for linux Must be a new fringe group. Usually zombies are Windows boxes. |
| #9 By 11888 (74.12.173.240) at 10/8/2009 4:14:47 PM |
| #7, how come you don't use hibernate or sleep mode? |
| #10 By 13997 (68.118.60.164) at 10/8/2009 4:32:36 PM |
|
Ok, so let's assume that there is some truth to this...
Is there possibly some weird HD/Controller/Driver combination that is slower? Maybe? Is this common? No... Is your computer going to be the 0.01% that is slower? Probably not... Just because a snake oil salesman is correct that you could get diease XYZ, doesn't mean that you have a chance of getting it. It would be like saying that Malaria is bad, and selling Eskimos a treatment for it at the same time, not probably going to be needed, and the snake oil probably isn't going to help anyway. There is a conflict of interest and there is also a borderline 'malware' trend with software optimization tools. It is like this, do you really think that if some magical piece of software could speed up Windows, that Microsoft with its 1,000s of engineers would not have thought of this 'magical' solution? Especially when software like this is just messing with registry settings or checking driver updates, stuff that will probably screw up your computer, and then they can sell you version 2.0 to speed up Windows again. Seriously, if there are tricks to speed up Windows, Microsoft has already added them. Win7, like Vista, has to optimize, and therefore benchmarking is hard to do, as the system could keep getting a little faster each time it is booted, and when you are not booting an OS very much, how can you be sure that all the drivers and optimizations are clicking along? Think of just these couple of things: 1) The prefetch has to start from scratch, and this means learning application load patterns and driver load patterns. This data is used for how to load faster to what to load and when, as Win7 has a parallel load process that can start several drivers/services at a time. 2) Defrag has to put files in optimal locations. (Automatically it runs once a week, Weds night) So your specific brand of HD will be faster in some places and you have drivers that should be in position XYZ on the HD so that the OS loads faster, but until Defrag runs, they won't get there, so a 'new' install will not be 'optimal'. 3) Defrag also needs the data from #1, the Prefetch so that it knows the boot history and where to put things so they can be available to load as fast as possible. However, again if you have only rebooted a couple of times, this 'history' is limited, and it won't be the 'optimal' until after the system has adjusted and been rebooted a few times. 4) People don't reboot often. Most Win7 users are only doing an intitial reboot for drivers and updates, and that is it. They use the Sleep and Hibernate features, so optimal boot time data can't even begin to be collected. (But this is also a good, thing, as people are not 'rebooting' their computers and there is not a need to reboot hardly EVER.) So what is the real world experience? Well it is going to be your own. Myself? I could show you cheap Acer Aspire netbooks with traditional HDs booting to the Win7 desktop in under 15secs. (Most people's cell phones take longer to turn on from a cold boot.) This same Netbooks resume from Sleep instantly and from Hibernate(Powered Off) in 4 seconds. However, it is more important that with things like Sleep and Hibernate, you will maybe reboot your computer once a month, as Win7 really is that stable and there really is no reason to be 'rebooting' nor should 'booting' be a measure of performance anymore. |
| #11 By 20505 (216.102.144.11) at 10/8/2009 5:01:46 PM |
|
MrR.
I have used sleep with Vista and it works great but I have 5 computers in the house. I also run a NAS box. I hazard to guess that the energy consumption of all of these devices left on sleep mode 24/7 is not inconsequential. I also have taken up a "jihad" against capacitance on/off switches. For some reason every maker of electronic equipment thinks that their product is so important that it can't be shut off, only placed in sleep mode. This is the same ridiculous thinking that leads software makers to load an applet every time Windows boots. I have a hard wall switch for my personal electronics that allows me to shut it off. Does it save a lot of energy? I don't know because I haven't measured it but I suspect it is significant. End of rant. |
| #12 By 29664 (97.118.132.126) at 10/8/2009 10:51:00 PM |
| Crap I was totally considering going to 7 but, not anymore. This is a deal breaker. Back to OS/2 for me! |
| #13 By 81201 (79.33.55.219) at 10/9/2009 1:55:36 AM |
|
@wannabehaxor
services loading is performed on multi-threads since Windows 2000+ |
| #14 By 143 (96.28.65.208) at 10/9/2009 4:12:35 AM |
|
#12
Crap I was totally considering going to 7 but, not anymore. This is a deal breaker. Back to OS/2 for me! LOL |
| #15 By 113862 (74.204.153.254) at 10/9/2009 4:39:38 AM |
|
"...sells PC tune-up software..."
Hmmmmm...... |
| #16 By 89249 (64.207.240.90) at 10/9/2009 12:41:01 PM |
|
#8 Boy would I love to read a book you've written about economics. Please get published I will buy 10 copies to try and understand the innerworkings of your mind.
Oh and a better description of the principles discussed in previous threads would be depreciation. GG though. |
| #17 By 15406 (216.191.227.68) at 10/9/2009 1:35:38 PM |
| #16: You'd love my book; it's only $1 per page. Cheap, eh? Don't ask how many pages it has as that's irrelevant to the fact that it's $1 per page. |
| #18 By 21912 (71.83.109.74) at 10/9/2009 2:48:31 PM |
|
It's a little-known fact that Santa Claus used Iolo's "System Mechanic" software to tune up the navigational computer in his sleigh for the last several years, enabling him to continue completing his rounds in a single night despite the increased workload. Those results speak for themselves.
Some people may have missed the real news here -- Iolo hasn't yet gone out of business ;-) |
| #19 By 89249 (64.207.240.90) at 10/9/2009 4:08:05 PM |
|
#17 While I doubt it possible. If the value of having the knowledge of the information of each page would be useful for 10 years. I would be paying $0.10/yr for that knowledge. Lets say from the knowledge of each page I could increase my productivity by $0.30/yr. At that point your book would be a great deal. If the value of each page is only going to yield me $0.05/yr in extra productivity the price of your book would be too much and I'd find a cheaper alternative.
This is why textbooks can charge more than the romance novels you read. The value is is more than justifable. Not to mention they are subsidized by government loans. See, no matter the amount of money. No matter how many times multiplied. My logic works :) |
| #20 By 13997 (68.118.60.164) at 10/9/2009 7:52:59 PM |
|
@13 suy
Yes NT has had multi-threading loading for a long time. However there have been changes in Vista and even more in Win7 that optimize this further. On multi-core processors (even HT) the secondary CPU cores are not extensively used for parallel process loading prior to Windows7. This is one reason a non-single core CPU will boot faster under Win7 than Vista everytime, as most of the parallel threads produced in Vista stay on the main single CPU core. There are also other optimizations of the order and what is allowed and not allowed and what times they load. There was some regression with XP SP2 as to the Network stack coming online, so that it would not be available until after protective services like the firewall, etc. There are also changes in drivers and services, having drivers loading parallel and services load parallel are also tweaked between and improved from the Win2k days to Win7. Anyone that is truly interesting in what is happening should check out the technet papers on this or articles by some of the NT kernel bloggers that have talked about these aspects since around the time Vista was released. A few searches will find the articles. @ #4 - Win7 is doing a lot more than just parallel loading services, and is several generations beyond what you find in *nixes, as drivers and other core aspects can be brought online faster than what the Linux and most *nix kernel models allow for. Just booting the kernel, Linux and NT would be a tight race, but considering NT is doing this with independant non-linked binaries, it is a testament to what NT can do. In Linux everything at the low kernel driver level has to be recompiled to the kernel version, in NT they are abstracted above and the kernel versioning is irrelevant.) - Yes these are very layman terms of what is happening.) |
| #21 By 15406 (99.240.77.173) at 10/9/2009 8:21:02 PM |
|
#20: You're Ketchum and I claim my $5.
|
| #22 By 23275 (75.146.73.217) at 10/10/2009 4:35:54 AM |
|
#21, No, he's not. avenger is the lead in his own story and the man is obviously as skilled as he is accurate - he even gets the order (priority) right and he's trying to make you/others familiar with some of the technologies in Windows 7 and why it is more usable than any previous version of Windows.
For those that are interested, use the words 'scheduler' and 'scheduler optimizations' to understand how and why Windows 7 is judged to be so great. I regret for Microsoft's engineers that this grossly over simplified explanation does not do all the hard work around the new scheduler justice. Getting so many related/dependent things right under so many varrying circumstances required that they address not only the science, but human perceptions about performance. They have played both like a harp and reconciled the needs of business/sales as well as their own science and engineering. Efficient is the best word to describe Windows 7 technically. Windows 7 carries forward the work done in Vista, but enhances perceptions around how soon a computer may be used after it starts. Scheduling is dynamic and depending upon what a user is doing, it adjusts what it does. If a user is not doing anything other than watching the system with a stop watch in hand, Windows 7 takes care of house cleaning chores like indexing and managing its prefetch caches. When a user begins to work and depending upon what they are doing, Windows 7 schedules resources and prioritizes them around what the user is doing, and after a time, against what a user has done. There is no easy way to validate performance against any set of tests, because by design, performance is subjective - relative to how a system is being tasked. Windows Vista/7 Ultimate and Enterprise will be slightly slower than down-level versions of the OS, because these versions feature secure startup that locks a machine until ALL protections are loaded - Avenger touches on how XP SP2 provided similar things (but not the same things by any means). |
| #23 By 15406 (99.240.77.173) at 10/10/2009 2:48:33 PM |
| #22: It wasn't what he said but the way that he said which read just like one of your typical posts. You're telling me that he's not you and not one of your kids? I would believe you if you denied it, but it did strike me that he sounds just like you. |
| #24 By 23275 (24.239.223.126) at 10/10/2009 4:24:26 PM |
| #23, No, I think he is smarter, actually <tips hat> |
| #25 By 37 (192.251.125.85) at 10/12/2009 11:13:57 AM |
|
"This is getting silly. Who is rebooting so often that a few seconds here or there matters?"
I have to restart my workstation at LEAST once a day, if not 2 times. Not to mention, I have to shut it down every day because it's frozen SOLID the next morning if I don't. Same thing for the other 16 workstations here. Windows garbage. (XP SP2) |
|
Write Comment
Return to News |
Displaying 1 through 25 of 336 Last | Next |
The time now is
12:38:47 PM
ET.
Any comment problems? E-mail us |
