Xbench is giving confusing results. My inital figure was for 'reduced' performance. I just re-ran the bench on the highest and automatic settings, and got 44.29 and 30.65. However it seems to take very little for the final figure to waver dramatically - I don't understand the weightings here. My iBook is an 800mhz w/640MB Ram.... as the only other thing running at the time of my first bench was Mail.app, I'm inclined to put it down to bad benchmarking.
Odd; I'm using automatic CPU speed for on battery use and 'highest' for on mains use. Activity monitor doesn't show anything using the CPU in the background. I've rerun this particular version of Xbench (v1.1.3) several times on this machine with comparable results. Unfortunately the xbench results comparison site is down at the moment so I can't see what the average is.
The 3.x interface was the most hideous thing ever. People want semi-opaque square windows, not crazy translucent donut shaped monstrosities. This was the very core of my deep seated loathing of the 3.x release anyway.....
Having the same OS on newer and more optimal hardware (read x86) as on old overpriced one (Sparc)? Give me a break.
x86 is an also-ran in the world of REAL servers I'm afraid. It's optimal for tiny little databases and applications that don't mind being segmented into 2GB chunks, but it really doesn't scale at all, and certainly isn't in the same league as UltraSPARC III, SPARC64, POWER4 etc. It's getting there with Opteron, but Intel have decided that it's time to stop using it, and Intel usually win, so I don't think it'll ever get there in the long term.
I can see Solaris x86 being very useful for scripting and development work for proper big tin. It certainly runs admirably well on my PII-333/256MB system on my desk at work.
if he'd been paying attention, he'd have caught the SSH vulnerability the other month
maybe he realised that ssh, like many things on the mac, isn't enabled by default, and so wouldn't be a problem to a great many people. So we have two possibilities, he's either a bad researcher or has invalidated his article by being selective to a fault.
I can't see the point of this
on
Linux Power Tools
·
· Score: -1, Redundant
My pillar drill wouldn't benefit one iota from a modular kernel.
Picture this:
1) I'm in the shops anyway, because I need to buy things like food.
2) I pick up 3 CD's
3) I purchase them
Suddenly I have a hardcopy of ~35 songs of better quality than I'd ever find on kazaa, and a nice little book to boot! In less than 10 minutes!
Correct me if I'm wrong but management wouldn't care; they're not that close to the staff at the outsourcers that they need to worry about training and skills. They simply pay for the skills that require, and pay when deadlines are met, and get remunerated when deadlines are missed. This is why management likes it so much, it's like built in insurance against timescales slipping....
Saying downloading music is infinitely cheaper is false; as the original author stated, the amount of time taken to get the file required of sufficient quality can be quite high. It depends how much you value your free time. I work normal office hours like anyone else, and I think my free time (this is bringing back memories of backwards bending J-curves from economics) is normally better spent either checking my favourite online media vendors or just popping into the music shop while I'm in town anyway. YMMV.
This just shows in summary how poorly designed (aesthetically) most GNOME applications are. All the applications look the same! You can see the brilliance of Apples work with expose because each application has a unique appearance.
My only worry is that the already nervous companies involved in this will see it as another nail in the coffin of Internet distributed music, effectively bombing us back to the stone age of going down the record store.
Linux will get an opengl rendered desktop, most window managers will default to a brushed metal look, fast user switching will be implemented (looking much like a big spinning cube). Some sort of special hotkeys will allow the user to see and cycle through their windows graphically.
I'm sorry, I just don't subscribe to this particular example of FUD. There is too much diversity in OSS for their to be a single damaging attack of the like that cripples Windows box with laughable regularity. Windows virii are borne through poorly secured services, badly written (and unfortunately all to uniformly used) applications, and a poor security model.
The only real opportunity would be through some single flawed release of one certain distribution, but even this is far-fetched and questionable. Most distributions are now using sensible alternatives to traditionally flawed services (sendmail being replaced by postfix, exim or qmail for example, even diversity there) and a few are shipping with basic firewall functionality by default. Also bear in mind that servers (where Linux really figures in terms of installation counts) don't search Google....
Slashdot Mirror
Xbench is giving confusing results. My inital figure was for 'reduced' performance. I just re-ran the bench on the highest and automatic settings, and got 44.29 and 30.65. However it seems to take very little for the final figure to waver dramatically - I don't understand the weightings here. My iBook is an 800mhz w/640MB Ram.... as the only other thing running at the time of my first bench was Mail.app, I'm inclined to put it down to bad benchmarking.
Odd; I'm using automatic CPU speed for on battery use and 'highest' for on mains use. Activity monitor doesn't show anything using the CPU in the background. I've rerun this particular version of Xbench (v1.1.3) several times on this machine with comparable results. Unfortunately the xbench results comparison site is down at the moment so I can't see what the average is.
I did some preliminary xbench's on my machines, and my G4's all got quicker, yet my G3 iBook (800mhz) got slower:
Powerbook: 84 -> 87
eMac: 105 -> 112
iBook: 34 -> 26
Trolly as hell... Don't mod up...
Darwin powerbook 7.2.0 Darwin Kernel Version 7.2.0: Thu Dec 11 16:20:23 PST 2003; root:xnu/xnu-517.3.7.obj~1/RELEASE_PPC Power Macintosh powerpc
The 3.x interface was the most hideous thing ever. People want semi-opaque square windows, not crazy translucent donut shaped monstrosities. This was the very core of my deep seated loathing of the 3.x release anyway.....
Not that there's very much in /etc anyway.
x86 is an also-ran in the world of REAL servers I'm afraid. It's optimal for tiny little databases and applications that don't mind being segmented into 2GB chunks, but it really doesn't scale at all, and certainly isn't in the same league as UltraSPARC III, SPARC64, POWER4 etc. It's getting there with Opteron, but Intel have decided that it's time to stop using it, and Intel usually win, so I don't think it'll ever get there in the long term.
I can see Solaris x86 being very useful for scripting and development work for proper big tin. It certainly runs admirably well on my PII-333/256MB system on my desk at work.
maybe he realised that ssh, like many things on the mac, isn't enabled by default, and so wouldn't be a problem to a great many people. So we have two possibilities, he's either a bad researcher or has invalidated his article by being selective to a fault.
My pillar drill wouldn't benefit one iota from a modular kernel.
Where does this leave WPA security then? My Airport Extreme base station just let me start using it and I feel more secure already!
I thought windows 9x was limited to a 47 day uptime?
Picture this: 1) I'm in the shops anyway, because I need to buy things like food. 2) I pick up 3 CD's 3) I purchase them Suddenly I have a hardcopy of ~35 songs of better quality than I'd ever find on kazaa, and a nice little book to boot! In less than 10 minutes!
Everyone being blacklisted for using this might have the nice side effect of making more effective blacklists :)
but then you have to sift through a bunch of crap afterwards, so it really makes no difference how you do it.
Correct me if I'm wrong but management wouldn't care; they're not that close to the staff at the outsourcers that they need to worry about training and skills. They simply pay for the skills that require, and pay when deadlines are met, and get remunerated when deadlines are missed. This is why management likes it so much, it's like built in insurance against timescales slipping....
Saying downloading music is infinitely cheaper is false; as the original author stated, the amount of time taken to get the file required of sufficient quality can be quite high. It depends how much you value your free time. I work normal office hours like anyone else, and I think my free time (this is bringing back memories of backwards bending J-curves from economics) is normally better spent either checking my favourite online media vendors or just popping into the music shop while I'm in town anyway. YMMV.
Seriously though, that site is so misinformed it's untrue. Was it written in response to a 10.1 beta?
This just shows in summary how poorly designed (aesthetically) most GNOME applications are. All the applications look the same! You can see the brilliance of Apples work with expose because each application has a unique appearance.
My only worry is that the already nervous companies involved in this will see it as another nail in the coffin of Internet distributed music, effectively bombing us back to the stone age of going down the record store.
Oh no wait - Windows has to do all this first.
Do the bandwidth saving maths examples include the fact that slashdot uses mod_gzip?
Welcome our new resource-rich overlord.
I for one welcome our new Cocoa overlords. No seriously, OS 9 sucked big style.
The only real opportunity would be through some single flawed release of one certain distribution, but even this is far-fetched and questionable. Most distributions are now using sensible alternatives to traditionally flawed services (sendmail being replaced by postfix, exim or qmail for example, even diversity there) and a few are shipping with basic firewall functionality by default. Also bear in mind that servers (where Linux really figures in terms of installation counts) don't search Google....