CCleaner really likes to clean
-
I know CCleaner cleans up a lot of stuff, but 2,232,362,150% of the drive's total capacity (714752MB) is a bit too eager methinks...
-
Do you clean your pc for privacy reasons?
-
@XIU said:
Do you clean your pc for privacy reasons?
I guess you're trying to tell him, that it doesn't boost his pc performance, right?
My suggestion, which can help with both things, is ramdisk. Since I've put FF cache and profile on a ramdisk, It works better than ever... Also, in some ramdisk apps you can set it so it will revert to a specific image on boot, without saving it - it means, that your history and other activity will make changes only in ram :)
-
@Gerino said:
My suggestion, which can help with both things, is ramdisk.
I wanted to put my swap file on a ram disk back in the windows 98 days. Windows wouldn't let me. :(
Turning swap off completely would stop Quake 2 from starting up...
-
@Zemm said:
I wanted to put my swap file on a ram disk back in the windows 98 days. Windows wouldn't let me. :(
Turning swap off completely would stop Quake 2 from starting up...
I remember trying that too, though possibly under XP. I thought 1GB of ram was huge, IIRC, and there was no way I needed swap.
Is this still a problem? Will Windows emulate swap for old apps that explicitely request swap space these days?
EDIT: I ask because of my netbook with an SSD, where it is recommended that swap is disabled due to excessively poor performance and excessive wear it causes on the SSD.
-
Is there any other reason to do so?
Plus, I don't really like having loads of leftover cruft on my workstation.
Cue jokes about Windows bloat in 3..2..1..
-
@RaceProUK said:
Cue jokes about Windows bloat in 3..2..1..
Not a joke, but I think the bloat issues are fixed in Vista and Windows 7. I had my Vista install for 3 years, and it never ran any slower from having lots of apps installed, or due to age-- not in the way XP's login screen would take longer and longer to come up the longer it had been installed. (It did run slower when system updates wiped the DLL cache, but that's temporary.)
Anyway, you're running Windows 7, I think you'll find bloat isn't an issue. Except perhaps psychologically.
-
@blakeyrat said:
@RaceProUK said:
Cue jokes about Windows bloat in 3..2..1..
Not a joke, but I think the bloat issues are fixed in Vista and Windows 7. I had my Vista install for 3 years, and it never ran any slower from having lots of apps installed, or due to age-- not in the way XP's login screen would take longer and longer to come up the longer it had been installed. (It did run slower when system updates wiped the DLL cache, but that's temporary.)
Anyway, you're running Windows 7, I think you'll find bloat isn't an issue. Except perhaps psychologically.
I have to second this. Windows 7 is basically set it and forget it. On any decent modern system, nothing really needs to be done to the system to tweak performance, etc. Most old school tweaks that were good on XP are either not benificial at all or can screw performance in some cases.
When Vista first came out, my peers at my job were insistent on disabling superfetch so that there was more free RAM displayed in task manager. Stupid. They were also keen to disable indexing because they thought it slowed down the system. Stupid. It only runs when idle.
Typical benchmarks don't reflect real world usage. Measure how much time it takes for a user to search for a Word document from the search bar in the start menu and then open that document in Word. See if that real world process is faster with indexing and superfetch turned off rather than turned on.
/rant.
-
I think it's the constant defragmentation when the machine is idle that helps the most. The performance problems of fragmented NTFS drives are well known.
-
As a general rule of thumb, IO is *always* the bottleneck. Anything that improves that improves everything.
-
@pauly said:
When Vista first came out, my peers at my job were insistent on disabling superfetch so that there was more free RAM displayed in task manager. Stupid.
To be fair to them, the memory reporting in Vista's Task Manager was pretty confusing, just showing "Total/Cached/Free". (Windows 7 shows "Total/Cached/Available/Free", which makes a hell of a lot more sesne.) Some supposedly Enterprise-class software has made this error.
@pauly said:
They were also keen to disable indexing because they thought it slowed down the system. Stupid. It only runs when idle.
Ok, yeah, that one's just stupid.
-
@Gerino said:
My suggestion, which can help with both things, is ramdisk. Since I've put FF cache and profile on a ramdisk, It works better than ever... Also, in some ramdisk apps you can set it so it will revert to a specific image on boot, without saving it - it means, that your history and other activity will make changes only in ram :)
Thank God I run Linux, where such a thing would be redundant and silly.
-
@morbiuswilters said:
@Gerino said:
My suggestion, which can help with both things, is ramdisk. Since I've put FF cache and profile on a ramdisk, It works better than ever... Also, in some ramdisk apps you can set it so it will revert to a specific image on boot, without saving it - it means, that your history and other activity will make changes only in ram :)
Thank God I run Linux, where such a thing would be redundant and silly.
Can't tell if serious...
-
@blakeyrat said:
@morbiuswilters said:
@Gerino said:
My suggestion, which can help with both things, is ramdisk. Since I've put FF cache and profile on a ramdisk, It works better than ever... Also, in some ramdisk apps you can set it so it will revert to a specific image on boot, without saving it - it means, that your history and other activity will make changes only in ram :)
Thank God I run Linux, where such a thing would be redundant and silly.
Can't tell if serious...
Well, serious as in "making a ramdisk for Firefox would be a waste of time in Linux". Come on, I'm always being fair and defending Windows against illegitimate sniping. But I'm a Linux user, so once in awhile I should be allowed to act like a snob. Otherwise, what is the point?
-
@morbiuswilters said:
@blakeyrat said:
I used to put some KDE stuff on a ramdisk, all so Kate would open up instantly.@morbiuswilters said:
Well, serious as in "making a ramdisk for Firefox would be a waste of time in Linux". Come on, I'm always being fair and defending Windows against illegitimate sniping. But I'm a Linux user, so once in awhile I should be allowed to act like a snob. Otherwise, what is the point?@Gerino said:
My suggestion, which can help with both things, is ramdisk. Since I've put FF cache and profile on a ramdisk, It works better than ever... Also, in some ramdisk apps you can set it so it will revert to a specific image on boot, without saving it - it means, that your history and other activity will make changes only in ram :)
Thank God I run Linux, where such a thing would be redundant and silly.Can't tell if serious...
-
@blakeyrat said:
@pauly said:
It would be stupid if it actually only ran on idle. I'm not sure about W7, but Vista and XP have serious problems with their definition of 'idle'. Indexing's not particularly processor intensive, but it does bugger up disk I/O - so of course Windows judges idleness based on CPU usage. Most people search so rarely that an index is entirely pointless.They were also keen to disable indexing because they thought it slowed down the system. Stupid. It only runs when idle.
Ok, yeah, that one's just stupid.
-
@davedavenotdavemaybedave said:
It would be stupid if it actually only ran on idle.
It does.
@davedavenotdavemaybedave said:
I'm not sure about W7, but Vista and XP have serious problems with their definition of 'idle'. Indexing's not particularly processor intensive, but it does bugger up disk I/O - so of course Windows judges idleness based on CPU usage.
Vista does not have that issue. (XP did, and had the nasty tendency to start indexing while playing full-screen games. But now you're going way back.) In any case, Vista doesn't judge idleness based (solely) on CPU usage, but user interaction... otherwise, if you keep a DVD ripper going all the time (like I do), your computer would never actually be "idle" and indexing/backups/etc would never run. CPU usage might be a factor, but user interaction is a bigger one.
I will hand you this, though: Vista has a tendency to attempt to index huge files, and once its started it won't suspend it. For example, if you have a 5 GB .zip archive with a hundred thousand files, Vista's indexer treats that as a "single file" for the idleness purpose. Windows 7 fixed that, as far as I can tell.
@davedavenotdavemaybedave said:
Most people search so rarely that an index is entirely pointless.
If they're using Vista or Windows 7, and not searching a dozen times a day, they're doing something wrong.
In any case, "most" people will never need a lot of things they have anyway, since the index doesn't hurt anything, and could potentially save you hours later on, why would you go out of your way to turn it off? I hope you don't treat (for example) medical insurance the same way you treat disk indexing.
-
@morbiuswilters said:
Well, serious as in "making a ramdisk for Firefox would be a waste of time in Linux".
Yeah, but it's a waste of time in Windows too. That's what was throwing me.
@morbiuswilters said:
Come on, I'm always being fair and defending Windows against illegitimate sniping. But I'm a Linux user, so once in awhile I should be allowed to act like a snob. Otherwise, what is the point?
Ok, fair enough.
-
@Xyro said:
@morbiuswilters said:
@blakeyrat said:
I used to put some KDE stuff on a ramdisk, all so Kate would open up instantly.@morbiuswilters said:
Well, serious as in "making a ramdisk for Firefox would be a waste of time in Linux". Come on, I'm always being fair and defending Windows against illegitimate sniping. But I'm a Linux user, so once in awhile I should be allowed to act like a snob. Otherwise, what is the point?@Gerino said:
My suggestion, which can help with both things, is ramdisk. Since I've put FF cache and profile on a ramdisk, It works better than ever... Also, in some ramdisk apps you can set it so it will revert to a specific image on boot, without saving it - it means, that your history and other activity will make changes only in ram :)
Thank God I run Linux, where such a thing would be redundant and silly.Can't tell if serious...
How long ago was this? This would be largely useless in any recent version of Linux. If you have RAM to spare, then all of the files will end up in kernel disk cache, without the need for a ramdisk. If you don't have RAM to spare, Linux will just swap out your ramdisk. Either way, the kernel generally knows better than you how to manage RAM and disk cache.
-
@blakeyrat said:
@davedavenotdavemaybedave said:
It would be stupid if it actually only ran on idle.
It does.
@davedavenotdavemaybedave said:
I'm not sure about W7, but Vista and XP have serious problems with their definition of 'idle'. Indexing's not particularly processor intensive, but it does bugger up disk I/O - so of course Windows judges idleness based on CPU usage.
Vista does not have that issue. (XP did, and had the nasty tendency to start indexing while playing full-screen games. But now you're going way back.) In any case, Vista doesn't judge idleness based (solely) on CPU usage, but user interaction... otherwise, if you keep a DVD ripper going all the time (like I do), your computer would never actually be "idle" and indexing/backups/etc would never run. CPU usage might be a factor, but user interaction is a bigger one.
You're right, in that I was oversimplifying for comic effect. You're wrong inasmuch as that's not the point. The point is that disk-indexing doesn't only run on idle, and in fact many business PCs are never properly idle - they're either in use, or off. I've unfortunately done enough tech support to see that this causes a lot of problems when you have things like disk indexing, virus scans, and so-on all waiting for an idle period that never happens. Oh, and just to help matters along, a lot of mainstream corporate AV packages will do on-access scanning of every file indexed.
I tested the effects scientifically on a large(ish) sample of users complaining of performance issues. Half the group had indexing turned off, half were just told their PC had 'had a tune-up'. There was a massive difference in user response, measured by the number of new PCs sold to the control group. (WTF special: the company I worked for at the time then instigated a process of ensuring indexing was turned on for all clients by default. Got to help those sales guys...)
That, by the way, is excluding all the PCs I touched where an immediate and obvious improvement was gained by turning off indexing. They generally had things conflicting with the indexing that shouldn't - but those things were often business-critical. There was a time when putting Google Desktop Search and Windows Search on the same PC would be beautiful to watch - GDS would notice a change, re-index the disk, and write its changed index to disk; Windows Search would see GDS's large changed file, re-index it, and then write its own index to disk; rinse and repeat. My experience is that indexing often causes show-stopping slowdowns in performance.
@blakeyrat said:
If they're using Vista or Windows 7, and not searching a dozen times a day, they're doing something wrong.
Or they know where things are? Even so, what's the performance benefit? Either indexing is costly in terms of performance, or searching is so cheap indexing is pointless. I don't know, I'm still using XP myself. Maybe I'm missing a big shift like the difference always-on broadband brought.
@blakeyrat said:
In any case, "most" people will never need a lot of things they have anyway, since the index doesn't hurt anything, and could potentially save you hours later on, why would you go out of your way to turn it off? I hope you don't treat (for example) medical insurance the same way you treat disk indexing.
I can't imagine the indexing saving hours - it shouldn't even take hours, so at worst I could always index as needed. I certainly can't imagine a search taking hours, particularly one I'm likely to repeat. As I've explained, I don't agree that it hurts nothing. As for med insurance, I apply exactly the same criterion - cost to benefit ratio. For what it's worth, I don't have medical insurance because I live in a Euro-pansy socialist featherbed and have sufficient liquidity to pay to go private as-and-when. As the Heinlein character said "I don't hedge my health, I'm healthy." I only bring the quote up because it came to mind, so I went and searched for it, specifying directory but not file. The directory is not indexed, but the search was instantaneous.
-
@davedavenotdavemaybedave said:
The point is that disk-indexing doesn't only run on idle,
Yes it does.
@davedavenotdavemaybedave said:
and in fact many business PCs are never properly idle - they're either in use, or off.
Wow, where do you work? "No Lunch Ever, Inc." "Guards Will Shoot You If You Go To An Offsite Meeting Products, Ltd." "Turn Off the Computer When Leaving Or Rectal Inversion.com."
@davedavenotdavemaybedave said:
I tested the effects scientifically on a large(ish) sample of users complaining of performance issues. Half the group had indexing turned off, half were just told their PC had 'had a tune-up'. There was a massive difference in user response, measured by the number of new PCs sold to the control group.
Wait, what? Not only do you work at a company that doesn't let employees leave their desk for any reason, but they also have to buy their PCs? Man!
@davedavenotdavemaybedave said:
There was a time when putting Google Desktop Search and Windows Search on the same PC would be beautiful to watch
Yeah, well, if you do retarded things you get retarded results. So your company doesn't let people leave their desks, makes employees buy their own PCs, but is perfectly fine installing Google Desktop on company computers, even though Google Desktop stores your search index in the cloud!
@davedavenotdavemaybedave said:
Or they know where things are? Even so, what's the performance benefit? Either indexing is costly in terms of performance, or searching is so cheap indexing is pointless. I don't know, I'm still using XP myself.
Well, then why the fuck are you even posting this? We're not talking about XP, we're talking about Vista and Windows 7. I said in my previous post that indexing sucked in XP, this is already covered ground.
Anyway, my point is everything in Vista/Windows 7 is based around searching. Finding a program in the Start menu is done 50 times quicker by typing its name in the search field. Finding what setting you want in the Control Panel? Same. Finding an old email? Ditto. If someone's using Vista/Windows 7 and not making use of these features, then they might as well just have thrown their money in the toilet.
@davedavenotdavemaybedave said:
I can't imagine the indexing saving hours - it shouldn't even take hours, so at worst I could always index as needed. I certainly can't imagine a search taking hours, particularly one I'm likely to repeat.
First of all, why is it relevant whether or not you'd be likely to repeat the search?
Secondly, you must have better bosses than I do at your WTF company, if you never get asked things like, "hey what was the total on that hardware purchase we made 11 months ago? You know the one with the monitor stands."
@davedavenotdavemaybedave said:
For what it's worth, I don't have medical insurance because I live in a Euro-pansy socialist featherbed and have sufficient liquidity to pay to go private as-and-when. As the Heinlein character said "I don't hedge my health, I'm healthy." I only bring the quote up because it came to mind, so I went and searched for it, specifying directory but not file. The directory is not indexed, but the search was instantaneous.
You have a directory on your computer named "I don't hedge my health, I'm healthy?" Your posting is full of WTFs today.
-
@Xyro said:
I used to put some KDE stuff on a ramdisk, all so Kate would open up instantly.
A girl with a tech fetish? Lucky b*st*rd!
-
@blakeyrat said:
If someone's using Vista/Windows 7 and not making use of these features, then they might as well just have thrown their money in the toilet.
Soo... by your logic, if someone just happens to know where the stuff they need is, they still have to use the search feature to access it? Because otherwise they would just have wasted their money upgrading toVista/7, because indexing/searching is the one and only worthwhile feature of those OSes?
-
@Anonymouse said:
Soo... by your logic, if someone just happens to know where the stuff they need is, they still have to use the search feature to access it?
Money argument aside, because it has nothing to do with anything -- YES it is far fucking better to tap the WinKey and type some shit, than it is to hit up the Explorer and click your way to the file you want.
BUT
That only goes if you need one or two specific files (or items!). If you need to manage a collection of files, such as a website's html, graphics, css, etc, then you use the file manager.
As it turns out, people need specific files and items pretty often.
-
@blakeyrat said:
Finding an old email? Ditto.
Oh shit, you can serach for emails in the start menu.
+11111 wba
I used it naturally to search the start menu and all is subsidiaries, and also for recently opened files, but beyond that, no so much. I did search a newwork drive for 2 year old photos that nobody could remember and found 'em in half a second, so that's good, but I'd have gotten the same results with XP, which I'm on at home still.
-
@davedavenotdavemaybedave said:
I tested the effects scientifically on a large(ish) sample of users complaining of performance issues. Half the group had indexing turned off, half were just told their PC had 'had a tune-up'. There was a massive difference in user response, measured by the number of new PCs sold to the control group.
I find your experiment description insufficient. Half the group had indexing turned off and were told this, and half the group had indexing still on and were told a lie? Or not? But good on you to Use Science, obviously.
@davedavenotdavemaybedave said:
a lot of problems when you have things like disk indexing, virus scans, and so-on all waiting for an idle period that never happens (...) My experience is that indexing often causes show-stopping slowdowns in performance.
Anectodal: I never have any problems with my work Vista machine, which sports Symantec/Norton enterprise/whatever.
@davedavenotdavemaybedave said:
...business-critical. There was a time when putting Google Desktop Search...
Are these things related or do they suffer from a missing linebreak? Google Desktop Search can't be business critical, can it?
@davedavenotdavemaybedave said:
The directory is not indexed, but the search was instantaneous.
I too don't really see a grand practical improvement in speed comparing XP search to Vista/7 search, other than that Vista/7 results show up in real-time as one types without any delay. For me, it's more a matter of interface: in Vista/7, It's Just BetterTM
-
@blakeyrat said:
Anyway, my point is everything in SSDS is based around searching. Finding a program in the SSDS menu is done 50 times quicker by typing its name in the search field. Finding what setting you want in the Control Panel? Same. Finding an old email? Ditto. If someone's using SSDS and not making use of these features, then they might as well just have thrown their money in the toilet.
FTFY
-
@dhromed said:
@davedavenotdavemaybedave said:
...business-critical. There was a time when putting Google Desktop Search...
Are these things related or do they suffer from a missing linebreak? Google Desktop Search can't be business critical, can it?
No missing line-break, but a separate point. GDS was not one of the business-critical applications I was thinking of :)@dhromed said:
I find your experiment description insufficient. Half
Sorry, unclear. All users were told the same thing, but only half actually had anything changed. I should probably also have noted that the majority of the difference in perceived speed was seen in the control group as well, so plainly telling someone their PC is faster is the best way to make them think so :)
the group had indexing turned off and were told this, and half
the group had indexing still on and were told a lie? Or not? But good on
you to Use Science, obviously.
-
@davedavenotdavemaybedave said:
so telling someone their PC is faster is the best way to make them think so :)
Yea, hence the success of pointless PC Cleaner apps.
@davedavenotdavemaybedave said:
All users were told the same thing, but only half actually had anything changed.
Okay, so the experiment showed that turning off indexing doesn't actually do anything noticeable?
-
@blakeyrat said:
@davedavenotdavemaybedave said:
No, really, it doesn't. Or at least, it does, but the definition of 'idle' is poor. For a practical example, consider that reading a long email may well seem like idleness, so the PC starts indexing. User finishes reading the email, tries to open the attachment, and because the disk is busy, it takes a few seconds instead of being instant.The point is that disk-indexing doesn't only run on idle,
Yes it does.
@blakeyrat said:
@davedavenotdavemaybedave said:
I used to work for a company that did outsourced support for a large number of rather varied clients. Working patterns are not as fixed as you imply. Several of their clients don't take longer than 15 minutes away from their desks all day (or spend lunchtime surfing the web), and turn the PC off when they're not there. Others only work on a PC/on-site/at all for a few hours a week, and have a dedicated PC that is shut-down the rest of the time. Still others have laptops that are only on when in use, and that have power management shutting them down when idle. I'm not saying that every PC suffers from issues related to never being able to do any maintenance tasks, but there are plenty. It's a significant issue, and it's another example of a situation where indexing is a significant cost. If the virus definitions aren't updating and scans don't have idle time to run, you don't want sodding indexing to jump in on the rare occasions when they do have an opportunity.and in fact many business PCs are never properly idle - they're either in use, or off.
Wow, where do you work? "No Lunch Ever, Inc." "Guards Will Shoot You If You Go To An Offsite Meeting Products, Ltd." "Turn Off the Computer When Leaving Or Rectal Inversion.com."
-
@morbiuswilters said:
@Xyro said:
It was many years ago, I think maybe 4 or 5... Actually, I can't quite recall if I was using a proper ramdisk or just tmpfs... But it did work. The initial load of Kate was always fast (subsequent loads would be fast anyway). The way I used Kate back then was to open a file, edit it, then close Kate down, then a few minutes later open it again. I hated waiting a few seconds for it to load, so I figured I'd just force it in RAM. I stopped bothering with it after the upgrade to ... 3.4? 3.5? because it was too painful to move all the libraries and icons and stuff to the ramdisk and have them load from there. And these days, Kate and/or my computer is fast enough to load Kate nearly instantly so it doesn't really matter anyway. (Although I have the same amount of RAM; I had 1 gig when it was a lot, and now I have 1 gig when it's not.)@Anonymouse said:I used to put some KDE stuff on a ramdisk, all so Kate would open up instantly.
How long ago was this? This would be largely useless in any recent version of Linux. If you have RAM to spare, then all of the files will end up in kernel disk cache, without the need for a ramdisk. If you don't have RAM to spare, Linux will just swap out your ramdisk. Either way, the kernel generally knows better than you how to manage RAM and disk cache.A girl with a tech fetish? Lucky b*st*rd!
Back then we also had Kassie, who would keep Kate warm and launch off our sessions...
-
@dhromed said:
Okay, so the experiment showed that turning off indexing doesn't actually do anything noticeable?
No. What did I write? I feel like I must be speaking a different language to everyone else today. And badly, at that. I'll start again.
- Group A - half the sample
- Group B - half the sample
- Both groups told the same thing
- Group A had indexing turned off, Group B did not - no other changes made
- Users in Group A reported themselves as happier with the results, users in Group B bought lots of new PCs relatively soon after.
Clear now?
-
@Thief^ said:
Is this still a problem? Will Windows emulate swap for old apps that explicitely request swap space these days?
EDIT: I ask because of my netbook with an SSD, where it is recommended that swap is disabled due to excessively poor performance and excessive wear it causes on the SSD.
I'm pretty certain that this is almost 100% to do with the wear-leveling on the SSD and magically killing off sectors as you go. Most likely the netbook is formatted NTFS and the logging alone will kill it. Have been up and down this problem with SSDs for years and the only way to get rid of it is to turn off paging. Back in the NT 4 days we were working on some things that required us to kill the pagefile, it worked and killing the pagefile completely became something that you could do with XP - unless you're running SSDs though, I really wouldn't suggest turning off paging otherwise, even my attempts on Win7 have had some problems...
-
@davedavenotdavemaybedave said:
No. What did I write?
You referred to two groups, both of which were happy in the end, and only one of which had undergone an actual change.@davedavenotdavemaybedave said:
users in Group B bought lots of new PCs relatively soon after.
Neither group A nor group B is the control group (C)?
In other words:
A - turned off -> happier
B - placebo -> still not happy -> new PCs
C - placebo -> happierConclusion: turning off indexing may or may not work?
I'm assuming B is not the control group, as you mentioned that the control group was happy with a placebo. And before you pull a blakey on me: I can quote you on that.I'm going to pick apart your words when I get home when I have some quiet time, because from where I'm standing now, it's a mess.
I'm currently trying to make front-end changes to a system comprising xsl, generated xml, xml settings, php-pear, pear-propel, a fucking Yggdrasil of a class inheritcance tree, half the code accessed via putty+vi, development source code, but a live database, and no documentation whatsoever. I spent the better part of the day adding a fucking checkbox to a form, and still haven't figured out how to fucking render it. I'm not really in the best of moods and clearest of minds right now.
-
I only posted the image to show CCleaner is way too eager :)
Wasn't expecting a massive Slashdot-scale debate about the merits of indexing :P
-
@RaceProUK said:
I only posted the image to show CCleaner is way too eager :)
Wasn't expecting a massive Slashdot-scale debate about the merits of indexing :P
LURK MOAR.
-
@RaceProUK said:
I only posted the image to show CCleaner is way too eager :)
Wasn't expecting a massive Slashdot-scale debate about the merits of indexing :P
NOBODY expects the Slashdot inquisition!
-
@Xyro said:
Actually, I can't quite recall if I was using a proper ramdisk or just tmpfs...
What would you consider the difference between "proper ramdisk" and tmpfs?
@Xyro said:
But it did work. The initial load of Kate was always fast (subsequent loads would be fast anyway). The way I used Kate back then was to open a file, edit it, then close Kate down, then a few minutes later open it again. I hated waiting a few seconds for it to load, so I figured I'd just force it in RAM.
That's a bit odd. Once you loaded it the first time it should have stuck around in dick cache, even after it closed (unless you were doing lots of i/o or using lots of RAM in between causing it to be flushed from disk cache).
-
@morbiuswilters said:
@Xyro said:
I would say a "proper ramdisk" is always in RAM and never swapped (like ramfs), whereas tmpfs has the option of being swapped to disk.Actually, I can't quite recall if I was using a proper ramdisk or just tmpfs...
What would you consider the difference between "proper ramdisk" and tmpfs?@morbiuswilters said:
@Xyro said:
I don't see why that should be so odd, how big do you think disk caches are? With web browsers or compilers running around in between text editing sessions, it's pretty easy to wipe out whatever cache you had of a previously run application.But it did work. The initial load of Kate was always fast (subsequent loads would be fast anyway). The way I used Kate back then was to open a file, edit it, then close Kate down, then a few minutes later open it again. I hated waiting a few seconds for it to load, so I figured I'd just force it in RAM.
That's a bit odd. Once you loaded it the first time it should have stuck around in dick cache, even after it closed (unless you were doing lots of i/o or using lots of RAM in between causing it to be flushed from disk cache).
-
@Gerino said:
My suggestion, which can help with both things, is
ramdisk.Just run lynx. Then there's no need for a ramdisk.
Or, you could run elinks; I use that for some sites, and it's just about as fast.
-
@davedavenotdavemaybedave said:
@dhromed said:
Okay, so the experiment showed that turning off indexing doesn't actually do anything noticeable?
No. What did I write? I feel like I must be speaking a different language to everyone else today. And badly, at that. I'll start again.
- Group A - half the sample
- Group B - half the sample
- Both groups told the same thing
- Group A had indexing turned off, Group B did not - no other changes made
- Users in Group A reported themselves as happier with the results, users in Group B bought lots of new PCs relatively soon after.
Clear now?
Let's assume you actually performed this test and didn't just make it up yesterday as some sort of debating tactic (which is my suspicion), it's still completely meaningless unless we know how large these groups are. Pro-tip: since you're making this all up anyway, go big and say you hired a few dozen grad students and did it on thousands of computers. While surfing a tsunami on a pink '57 Chevy in a post-apocalyptic sunken LA being used as a prison. (Visual aid.) Go big or go home! Woo!
Anyway, here are two points:
1) Your brain lies to you. Not only is there the aforementioned placebo effect ("rebuild your desktop file!"), there's a special brand of nostalgia specific to computers that tells your brain that the older a computer was, the more responsive it was. I've had people come to be and seriously claim that Word opened quicker on their Windows 95-running 486 than it does on their modern PCs. Bullshit. Word took like 2 solid minutes just to start up, and once it was started it didn't have a third of the features that modern Word does. Where does this belief come from? I have no idea.
(Normal nostalgia comes from people only remembering good moments and forgetting mediocre or bad moments. Computer nostalgia is different, because there was nothing good about running software on 486s, but for some reason people claim it was some kind of utopia compared to modern computers.)
2) You, davedavedavedavewhothefuckisdave, don't even use Vista or Windows 7. I think all you've been doing this entire thread is taking the admittedly shitty search indexing from XP, and just assuming that it hasn't been improved. Here's a shocking idea:
Why don't you actually use the products before talking about them? Especially when talking about them to people who have been using them for three solid years.
Debating with you is like debating with a blind man about what color Godzilla is.
-
-
@davedavenotdavemaybedave said:
No. What did I write? I feel like I
must be speaking a different language to everyone else today. And badly, at
that. I'll start again.- Group A - half the sample
- Group B - half the sample
- Both groups told the same thing
- Group A had indexing turned off, Group B did not - no other changes made
- Users in Group A reported themselves as happier with the results, users in Group B bought lots of new PCs relatively soon after.
Clear now?
Now I'm confused. I had previously understood what you'd written to be:
- Group A - 25% of the sample, indexing turned off, told nothing.
- Group B - 25% of the sample, indexing turned off, informed of tuneup.
- Group C - 25% of the sample, indexing left on, told nothing
- Group D - 25% of the sample, indexing left on, informed of tuneup.
- Users in Groups A and B reported themselves as happier with the results, users in Groups C and D bought lots of new PCs relatively soon after.
Of course, from the beginning, I'm wondering where were the ICFs, and how you managed to maintain experimental integrity after having every participant sign one.
-
-
@Xyro said:
@blakeyrat said:
Where does this belief come from?
Lower expectations.
Filed under: (or was that a rhetorical question?)Believe it or not, no. I honestly don't get it. I guess I can understand that, because people expect modern computers to be 500% faster (or whatever), they're disappointed when the modern computers are only 250% faster-- but they're still faster!
-
@blakeyrat said:
1) Your brain lies to you. Not only is there the
aforementioned placebo effect ("rebuild your desktop file!"), there's a
special brand of nostalgia specific to computers that tells your brain
that the older a computer was, the more responsive it was. I've had
people come to be and seriously claim that Word opened quicker on their
Windows 95-running 486 than it does on their modern PCs. Bullshit. Word
took like 2 solid minutes just to start up, and once it was started it
didn't have a third of the features that modern Word does. Where does
this belief come from? I have no idea.(Normal nostalgia comes from people only remembering good moments and forgetting mediocre or bad moments. Computer nostalgia is different, because there was nothing good about running software on 486s, but for some reason people claim it was some kind of utopia compared to modern computers.)
I may not have much flame clout, having created my account more recently than someone who still only has one post to its name. However, you seem like one who does not think too much.
At the time of Windows 95, the "good" moments were after Word *finally* loaded. The mediocre moments were while it was loading. As such, the good moments were remembered, and the bad moments were not. Also note, the "good" moments were when they found the controls they were wanting to use, through the sparse interface that was enabled by having so few features. The bad moments were not finding the needed features, followed by mediocre moments of finding a workaround, followed by the "good" moments of using the workaround.
The fact that none of the "good" moments were really good, either by todays standards, or by the standards of the enlightened few at the time using emacs (or, I guess, vi - I don't want to start a flame war here), does not really matter. What matters is that they were the happiest computer usage memories that these people had at the time, so they were the memories that were retained. This is basically the same technique used by Brits during WWII to enable nostalgia of the Blitz.
As such, computer nostalgia is the same as any other nostalgia. You merely don't understand nostalgia, while somehow comprehending all of the necessary components to understand it.
-
Who The Fuck is a joke account, right? Has to be.
-
@Zecc said:
@morbiuswilters said:
dick cache
Freudian slip or on purpose? We may never know.I guess I'm just so used to typing "dick" here that it slipped out.
-
@blakeyrat said:
Who The Fuck is a joke account, right? Has to be.
*checks through his list of sockpuppets* bstorer... dhromed.. belgariontheking.. davedavenotdavemaybedave... Nope, it's not any of mine.
-
i use it for porn
