My parents us some PoS they bought at best buy that runs Windows. I don't "do the support work". All I'm saying is if I did buy a computer for my parents [well actually I did buy them the PoS but I was not exactly rolling in the dough] it wouldn't be some overpriced shiny Mac.
I don't get all the pride and joy over Macs. Until very recently you couldn't effectively dual-boot them [it was a hack more than anything else]. They're designed not to be opened (e.g. mac mini) and they charge more for the same parts as anyone else without really much more success.
Sure they look nice but you know what, so does my Inspiron 630m and it didn't cost me nearly as much, is better equipped and hasn't had a problem since I first turned it on nearly six months ago.
If I were poor, forget in a third world nation, I would NOT be thinking of buying a Mac.
First, it's nicer to have the manufacturer install upgrades if they don't rape you for it. Specially since Dell sold me the 100GB disk for the price they charge for the 60GB drives.
Second, I like my family so I wouldn't subject them to Mac OS.
Third, not all AMD boxes are "noisy" or even bulky. Shuttles exist for AMD and Intel that take less room and can be very quiet.
Fourth, As for software, I only use Linux OSes on my boxes. So it's moot. If my parents want an "email box" they can go buy their own. I don't support monopolies.
Mac laptops are routinely way more expensive then offerings from Dell. Specially when Dell has their sales that include ram/HD upgrades.
When I bought my inspiron 630m it came with 1GB of ram, 100GB disk, 2Ghz Pentium M and the three year warranty for 2300$ [with taxes]. That was when the BASELINE Mac laptop was $2000 on its own, that is 1.5Ghz G4, 512MB of ram, 60GB disk, etc... Upgrading and warrantying the thing would have cost [iirc] about $2850 or so.
Granted they're not $5000 each they are routinely more expensive.
As for the Mac Mini, it's $699 [CDN] for a 1.5Ghz single core processor with 512MB of ram. Big deal. I could buy a 2Ghz dual-core AMD64, 1GB of ram and a proper case (e.g. one where I can install new stuff) for about the same price.
Linkworld 3230-02C2222U Black Steel ATX Mid Tower Computer Case 430W Power Supply
Also known as... "fire hazard".
Every single case I've seen that combined with a PSU costs less than $100 is a total piece of shit I wouldn't trust in my house. Out of spec PSUs and really cheap case construction (re: flimsy) just doesn't make it worth it.
Congrats, you bought used parts most likely. Or they've been severely discounted (e.g. discontinued). The actual retail cost... [sounds like price is right] is a bit higher than $200.
Here's a thought, maybe these "third world" nations should focus on getting technology in the classroom. When students evolve to handle real jobs (other than sorting beans or whatever) they can then afford computers in the home.
You think everyone had computers in their homes in 1960? 1970? hell even 1980 and 1990? I still remember going to a friends house in the early 90s because he had a Pentium.
The problem is we're violating the prime directive here. They wanna play catchup and have all the nice toys we have right now without developing a society to make it really feasiable.
I mean with my salary, I could live like a fucking King in most nations. But that is just because my job is worth something in the society I am (apparently). Right now there really isn't any African silicon valley if you know what I mean...
I work in the C world and my experience is the same. I often [like at least once per month] get a support request for my LibTom projects that involves the MSVC IDE, my code and "I globbed all the files into a project" despite the fact I have MSVC make files...
A lot of "programmers" out there don't know how building software works, e.g. compile, assemble, link, load, etc. They just think "I click build and an.exe pops out". Which may work well for limited windows applications, but when you start doing real work and have to pull in numerous libraries or have odd linker scripts it isn't as nice.
The latest release 0.8.1 of QEMU (along with 1.3.0 of KQEMU) achieves a surprisingly fast and stable emulation for a free (as in beer) tool.
You can save yourself the money and just use QEMU. It emulates a PC just fine and can run most anything as a guest. I use it for a Windows guest so I can write my book. Granted my workstation is a "bit" high end, but when I full screen it, it's just like running a real Windows box (shudder).
Trick is to make sure the KQEMU accelerator is loaded and running correctly. Which isn't really hard if you know how to run./configure && make install
It's interesting but you have to take things into context.
Darwin states survival of the fitest.
This experiment didn't say the ideal mutation was the ONLY mutation that occurred.
That the same mutation occurred is particularly interesting but you also have to look at the amount of variation available. I mean, just how long is the DNA of a microbe anyways? Laws of probability state that if the available mutation pool is small enough you should see a good overlap.
It's like putting six playing cards in a bag and pulling 3 of them, placing them back, pulling 3 more and being amazed that you had pulled at least one of them twice.
I'll bet there WERE other mutations but they died off and the one with the heat resistance won out twice in a row.
Big deal. That doesn't fly in the face of Darwinism.
That the same mutation eventually occurred twice given a limited genome doesn't surprise me.
Tom
Re:So it almost seems evolution follows a... desig
on
Is Evolution Predictable?
·
· Score: 3, Insightful
Wrong. You're looking at it from the wrong angle.
Darwin stated that the mutations were RANDOM and that those that led to better survivors would out live those without.
E.g., if my kids can cope with the sun better than your kids then there is a chance they won't die off from cancer. If all their kids and grandkids and so on, I'd have a family tree that tends to live longer and reproduce more, etc.
It has nothing to do with the mutations being guided.
Um I'm not confused at all. Addressing more than 4GB requires PAE in 32-bit mode which is a lame hack.
And yes, there is 5GB of addresses in my system as the BIOS moves the last GB of DDR memory to the 4GB-5GB slot. That is if I want to talk to my last GB of memory I can't use a 32-bit address.
And yes, I do occasionally have processes that take nearly 4GB of ram (random experiments).
And no, I don't write gigabyte source files. But even a rather small file can take 100-250MB of ram to compile with full optimizations turned on. 4GB gives me a nice ceiling with room for cache which is important (I do a lot of builds). Also it's split as 2GB per node (this is a 2P box). So even though I could get along with 2GB of memory it would only be 1GB per zone which really isn't enough anyways (did I mention they're dual core processors?).
Other than that, with all else equal, 32bit machine should be faster, at least in theory.
The problem is this isn't true. Look up what unit of memory K8 processors read. Hint: it isn't 32-bits. The only time it is an issue is if your pointers flood your cache. As I said in another post, what percentage of your data space do you suppose is pointers anyways?
I don't get the pushback on this. Anything you compile that is non-trivial most likely uses the extra GPRs. So you know the 84 shared objects you have loaded to make your KDE or Gnome desktop work? Yeah, they can use them. Between me hitting a key and it showing up in mozilla are probably three or four libraries. The more efficient they are the better.
I view not being able to use Flash as a feature...:-)
The only package that I've had serious problems with was glibc 2.4. I still haven't upgraded from 2.3.6 but eventually when things are slower [e.g. not writing a book due on Tuesday] i'll backup the system and try it out...
2.4.21 is old even in the 2.4 series specially given that 2.4.32 came out in NOVEMBER of 2005!!!
Personally barring a technological reason I'd just migrate to 2.6 and be done with. You get the latest and greatest fixes/updates/upgrades and the nerd-pride of running newer code.
Slashdot Mirror
Chances are if you can't spend $400 a year on a PC you also can't afford $30/month on net access to make the computer ACTUALLY USEFUL.
Tom
My parents us some PoS they bought at best buy that runs Windows. I don't "do the support work". All I'm saying is if I did buy a computer for my parents [well actually I did buy them the PoS but I was not exactly rolling in the dough] it wouldn't be some overpriced shiny Mac.
I don't get all the pride and joy over Macs. Until very recently you couldn't effectively dual-boot them [it was a hack more than anything else]. They're designed not to be opened (e.g. mac mini) and they charge more for the same parts as anyone else without really much more success.
Sure they look nice but you know what, so does my Inspiron 630m and it didn't cost me nearly as much, is better equipped and hasn't had a problem since I first turned it on nearly six months ago.
If I were poor, forget in a third world nation, I would NOT be thinking of buying a Mac.
Tom
First, it's nicer to have the manufacturer install upgrades if they don't rape you for it. Specially since Dell sold me the 100GB disk for the price they charge for the 60GB drives.
Second, I like my family so I wouldn't subject them to Mac OS.
Third, not all AMD boxes are "noisy" or even bulky. Shuttles exist for AMD and Intel that take less room and can be very quiet.
Fourth, As for software, I only use Linux OSes on my boxes. So it's moot. If my parents want an "email box" they can go buy their own. I don't support monopolies.
Tom
Mac laptops are routinely way more expensive then offerings from Dell. Specially when Dell has their sales that include ram/HD upgrades.
When I bought my inspiron 630m it came with 1GB of ram, 100GB disk, 2Ghz Pentium M and the three year warranty for 2300$ [with taxes]. That was when the BASELINE Mac laptop was $2000 on its own, that is 1.5Ghz G4, 512MB of ram, 60GB disk, etc... Upgrading and warrantying the thing would have cost [iirc] about $2850 or so.
Granted they're not $5000 each they are routinely more expensive.
As for the Mac Mini, it's $699 [CDN] for a 1.5Ghz single core processor with 512MB of ram. Big deal. I could buy a 2Ghz dual-core AMD64, 1GB of ram and a proper case (e.g. one where I can install new stuff) for about the same price.
Tom
Just wait for adware to issue print commands with coupons...
Oh christ, I just gave them an idea...
Tom
I said low end, not cheap.
$30 case == crap, flimsy, low spec PSU, etc...
$30 CRT == crap, noises, power, can't do decent refresh rates, etc..
etc, etc, etc.
So what you're saying is you can build a $200 PC, it's just a complete piece of crap.
Tom
Linkworld 3230-02C2222U Black Steel ATX Mid Tower Computer Case 430W Power Supply
... "fire hazard".
Also known as
Every single case I've seen that combined with a PSU costs less than $100 is a total piece of shit I wouldn't trust in my house. Out of spec PSUs and really cheap case construction (re: flimsy) just doesn't make it worth it.
You'd be better off air boarding it.
Tom
Awww, an upset troll, do you need help removing the latest ad infested never go away must be on startup application from your PeeCee?
:-)
I can help you install Gentoo.
Tom
where you pay $5000 for a PC that costs $500
:-)
*cough* Apple *cough*.
Oh you meant over time in installments... hehehe.
The renewable fee for net access is partially justified (maybe not the $30 figure but the fact there IS a figure...).
It takes power, space, staff and equipment to run an ISP. It isn't like all the customers could just pay $29.95 once and have net for life.
On the other hand, a Windows install takes none of Microsofts time and shouldn't be forced into a renewable fee schedule.
Tom
Congrats, you bought used parts most likely. Or they've been severely discounted (e.g. discontinued). The actual retail cost ... [sounds like price is right] is a bit higher than $200.
Tom
Here's a thought, maybe these "third world" nations should focus on getting technology in the classroom. When students evolve to handle real jobs (other than sorting beans or whatever) they can then afford computers in the home.
You think everyone had computers in their homes in 1960? 1970? hell even 1980 and 1990? I still remember going to a friends house in the early 90s because he had a Pentium.
The problem is we're violating the prime directive here. They wanna play catchup and have all the nice toys we have right now without developing a society to make it really feasiable.
I mean with my salary, I could live like a fucking King in most nations. But that is just because my job is worth something in the society I am (apparently). Right now there really isn't any African silicon valley if you know what I mean...
Tom
Or a PC and install any BSD or Linux distro variant.
Yes, that's what the poor in third world nations require, overpriced "premium" computers that run a commercial OS where upgrades cost money.
Well played poster, you are a tool.
Tom
You can build a box for less than 200? Mobo: ~$70, Processor: ~$80, Case+PSU: $120, 512MB of memory: $50, 80GB disk: $60, cdrom drive: $20
That's $400 and gonna be the very low end of the lowest.
Where do you shop?
Tom
It was a joke. The idea was you could snub the other OSes out of spite.
I wouldn't buy a Mac, pay the premium and then not run an OS that gave me the most lattitude to do my work.
MacOS may look perdy but USE flags are god. XP may play games but PocketNES rules them all.
Just MHO.
Tom
I work in the C world and my experience is the same. I often [like at least once per month] get a support request for my LibTom projects that involves the MSVC IDE, my code and "I globbed all the files into a project" despite the fact I have MSVC make files...
.exe pops out". Which may work well for limited windows applications, but when you start doing real work and have to pull in numerous libraries or have odd linker scripts it isn't as nice.
A lot of "programmers" out there don't know how building software works, e.g. compile, assemble, link, load, etc. They just think "I click build and an
Tom
The latest release 0.8.1 of QEMU (along with 1.3.0 of KQEMU) achieves a surprisingly fast and stable emulation for a free (as in beer) tool.
./configure && make install
You can save yourself the money and just use QEMU. It emulates a PC just fine and can run most anything as a guest. I use it for a Windows guest so I can write my book. Granted my workstation is a "bit" high end, but when I full screen it, it's just like running a real Windows box (shudder).
Trick is to make sure the KQEMU accelerator is loaded and running correctly. Which isn't really hard if you know how to run
Tom
The only good thing about having mac OS, gentoo and XP is that when you boot Gentoo you can say "I could boot XP or MacOS, I just choose not to." :-)
Tom
It's interesting but you have to take things into context.
Darwin states survival of the fitest.
This experiment didn't say the ideal mutation was the ONLY mutation that occurred.
That the same mutation occurred is particularly interesting but you also have to look at the amount of variation available. I mean, just how long is the DNA of a microbe anyways? Laws of probability state that if the available mutation pool is small enough you should see a good overlap.
It's like putting six playing cards in a bag and pulling 3 of them, placing them back, pulling 3 more and being amazed that you had pulled at least one of them twice.
Tom
I'll bet there WERE other mutations but they died off and the one with the heat resistance won out twice in a row.
Big deal. That doesn't fly in the face of Darwinism.
That the same mutation eventually occurred twice given a limited genome doesn't surprise me.
Tom
Wrong. You're looking at it from the wrong angle.
Darwin stated that the mutations were RANDOM and that those that led to better survivors would out live those without.
E.g., if my kids can cope with the sun better than your kids then there is a chance they won't die off from cancer. If all their kids and grandkids and so on, I'd have a family tree that tends to live longer and reproduce more, etc.
It has nothing to do with the mutations being guided.
Tom
Um I'm not confused at all. Addressing more than 4GB requires PAE in 32-bit mode which is a lame hack.
And yes, there is 5GB of addresses in my system as the BIOS moves the last GB of DDR memory to the 4GB-5GB slot. That is if I want to talk to my last GB of memory I can't use a 32-bit address.
And yes, I do occasionally have processes that take nearly 4GB of ram (random experiments).
And no, I don't write gigabyte source files. But even a rather small file can take 100-250MB of ram to compile with full optimizations turned on. 4GB gives me a nice ceiling with room for cache which is important (I do a lot of builds). Also it's split as 2GB per node (this is a 2P box). So even though I could get along with 2GB of memory it would only be 1GB per zone which really isn't enough anyways (did I mention they're dual core processors?).
Other than that, with all else equal, 32bit machine should be faster, at least in theory.
The problem is this isn't true. Look up what unit of memory K8 processors read. Hint: it isn't 32-bits. The only time it is an issue is if your pointers flood your cache. As I said in another post, what percentage of your data space do you suppose is pointers anyways?
Tom
yes I do use the extra registers. :-)
I don't get the pushback on this. Anything you compile that is non-trivial most likely uses the extra GPRs. So you know the 84 shared objects you have loaded to make your KDE or Gnome desktop work? Yeah, they can use them. Between me hitting a key and it showing up in mozilla are probably three or four libraries. The more efficient they are the better.
Tom
I view not being able to use Flash as a feature... :-)
The only package that I've had serious problems with was glibc 2.4. I still haven't upgraded from 2.3.6 but eventually when things are slower [e.g. not writing a book due on Tuesday] i'll backup the system and try it out...
Tom
2.4.21 is old even in the 2.4 series specially given that 2.4.32 came out in NOVEMBER of 2005!!!
Personally barring a technological reason I'd just migrate to 2.6 and be done with. You get the latest and greatest fixes/updates/upgrades and the nerd-pride of running newer code.
Tom