Wow... that module sucks. No EVDO? At least it has audio, unlike MultiTech's CDMA module.
What we actually need is a reference dev kit for Qualcomm's MSM6500 or better to be able to make something. Or a module based off it. It'd be neat if we could create a CDMA module that would plug in place of the iPhone's baseband board.
Right now, I'm still thinking we're better off getting old EVDO phones, powering up just the board with no screen or keypad, and using that as a module through rs232 or usb.
Verizon is allowing the certification of devices (most likely not phones), to use their network.
There are three parties involved: 1) Network owner : Verizon. 2) Device manufacturer : 3) End user :
Just because #2 is now allowed to register stuff with #1's net, doesn't mean #3 can do whatever they want.
Example? Kindle uses Sprint's 3G EVDO network to provide connectivity without WiFi. Amazon (#2) got permission from Sprint to have these non-phone devices connect to their network. Amazon also foots the EVDO bill for #3. As end users (#3), we're not given any easy way to use Kindle for anything but buying books and wifi with a crappy browser.
I'm a total fan of open source projects. Except I don't like the GPL. I consider it too restrictive because realistically, I need to be able to work with different licenses and closed source projects.
If I make something derived from a GPL'd library, then yes, I'll GPL it only out of compliance.
But if I had a choice, my public works would be BSD-style because I'll put my money where my mouth is and share.
Yeah, ebay isn't the place to go for a 29 inch CCFL. There are companies online that specialize in selling these, so I'm sure they have a good way of shipping them.
For me, I just bought a broken panel when I needed the one for the 15 inch LCD:)
Which is exactly why I picked up a modern rear projection DLP TV from my friend. The projector unit is the same that they sell in the tiny projector boxes for presentations, and the failure mode is typically the light bulb. Mine has a hatch on the front to let you swap it out.
LCD backlights fail often, but they're not that difficult to swap. Most panels are designed with a tray at the top and bottom that slides out with the tube. The hard part is sourcing the tube or about the same length. I've done a fix where I installed the backlight CCFL from a broken 17inch desktop monitor into a smaller 15 inch desktop monitor. A little silicone later to block out the extra light and now I have a working monitor again.
It's most likely the connector cable that goes from the lcd panel to the logic board. You can't fix the cable. They use two types. One that's a plastic ribbon (less likely). One's a bundle of micro-coax wires which you can't easily solder to because you can barely see the connector end.
The good news is that you ought to be able to buy a cracked display unit or a bezel which still has the cable. The LCD panel itself has a connector. The logic board has another connector. So the cable can be unplugged from both ends.
(I've done this to a Titanium Powerbook G4 before.)
Hehehe. My friend gave me a broken 60 inch rear projection HDTV. $250 for a new bulb later, it's nice and bright in my living room.
Depending on the technology for your HDTV, it could be just as easily fixable. Just have to know what's the most common failure case and make sure it's one you can cover.
Just curious. I have two VCRs that aren't working. One plays too fast, the other plays then stops. (symphonics and sony) I opened them both up and couldn't find any mechanism that would cause either problem. Got any ideas?
I had one semester of free time between high school and college, and I spent it buying broken Playstation 1 units off ebay and fixing them for resale. Great source of income back then. It was enough to pay off all my PS2 games and accessories as the PS2 came out shortly after I got to college.
I then moved on to iPods for a little while. The price of broken iPods pushed me to Motorola E815 cell phones.
And now I'm doing nintendo DS's despite having a full time job.
Apple's not only providing the phone, but also providing the means for the iPhone users to handle activation, as well as the majority of customer support (warranty, assistance, marketing and training).
Yes, AT&T stores have marketing materials and people who can help you with the iPhone, but not much more than a normal phone. The idea is that you'll typically go to the Apple store for iPhone specific support.
This $18 isn't just a share of the plan. It's a share of what it would have cost AT&T to get and MAINTAIN the subscriber, seeing how for iPhones, Apple is handling a lot of it.
"Why did they wait for 3 months after the release to announce it?"
Because Apple announces only when they can deliver or can give a pretty close date for delivery.
Most people don't realize that the iPhone is essentially the first release of Leopard. Articles talking about WWDC mention that Core Animation and a bunch of other stuff were in Leopard but were developed originally for the iPhone. The way I saw it, iPhone was an incomplete product. It did everything marketing required it do on release. (it's an iPod, it browses the web, and it is a phone. Can't deny it fit the criteria, ne?) But it could do more. Trouble is, when you also have the company's baseline platform to finish up, it's kinda more important to work on Leopard, isn't it?
Since Leopard wasn't done yet, and many bugs were being worked out on both Leopard and iPhone, it wasn't certain that they could bring out a native SDK and keep Leopard on track.
Now that Leopard's off to the CD presses (afterall, if they're shipping in less than a week, the final CD had to have been handed to the presses at least a week ago), it's perfect time to evaluate the progress of other internal projects. Now that the OSX team has probably gotten close to finishing up a native iPhone SDK, give them the 4 months to finish QA, write up documentation, translate it to all the languages, and release it in Feburary.
Market forces had nothing to do with this. PR outrage had nothing to do with this. All that just made everybody talk more about the iPhone. For every person who said "iPhone sucks" there were more who said "iPhone? What is this iPhone? Oh cool! *opens wallet*".
Not having enough engineers lined up to finish the OS had everything to do with this.
If you think I'm giving them too much credit, so be it. Based on Apple's behavior before, I guessed this would happen when they made the iPhone web app announcement. When hackers got access to iPhone 1.0.2's file system, I made a quick scan of the bluetooth manager and it pretty much assured me it would happen. Now that it's been announced officially, it feels like no big deal to me.
Sure, if MS had to develop their own TCP stack, they'd have less of an advantage in the OS market. Would this have translated to more BSD users? No. The market isn't a zero sum game. If Windows doesn't do as well, there's no reason BSD would do better. Afterall, they could have both sucked it up and OS/2 could have won.
In fact, if the BSD implementation was not as open as it were, every OS would have had to develop their own TCP stack or go GPL. Going GPL is obviously not in the interest of many businesses.
What does this imply? TCP would not have become the standard it is now. And we'd be back in the same situation as the 80s where every network device had to speak Arcnet, Appletalk, a version of TCP compatible for each OS' TCP stack bugs, Lantastic, IPX/SPX...etc...
Simply put, if the reference design for TCP was GPLed instead of BSD, we're all fucked. Only GPLed OSs, of which there are little compared to non-GPL OSs, would have that stack. Every other OS would have their own implementation complete with various bugs and inconsistencies.
If you think that "oh, that web page doesn't work with my version of firefox" is bad, then what about "oh, my linux box can't talk with any solaris hosts, unless it's got solaris TCP version 7.56.9-beta42 and my kernel is 2.8.666." or some crap like that.
The simple fact that because a BSD-licensed TCP stack allows any OS, regardless of size, intentions, or evilness as judged by us Slashdotters, to take the stack and integrate it into their OS demonstrates it's own ability to maintain freedom: If everybody's starting with this stack, you're starting with a standard version of which you need to maintain compatibility with or you have cut yourself out of the community to your own detriment. If you have some brilliant ideas to make it better, sure, you can let everybody else know and contribute back, or it could be a trade secret of yours that you'll have to maintain on your own. The fact that the original code is free and out there and enjoyed by as many people as possible has already been fulfilled even when you closed your fork. Besides, maintaining and significant changes and then merging in new developments from the mainline is still a pain in the ass no matter what. If it wern't, you probably didn't make any changes significant enough for people to care anyways, so it's not like we as a whole lose.
As far as the users of the code are concerned, in the case of TCP, they are enjoying one of the most important freedoms the writers intended: That their computer can communicate with the world. The majority of Windows users arn't programmers anyways. But no matter who you are, having a computer that can network with everybody else's is pretty important.
The way I see it. BSD helps proliferation. GPL encourages work to be consolidated. The question isn't what's "freer", it's who you want using it.
Anyways, I agree with the claim this doesn't harm BSD because this is exactly what they wanted. They want people using their code. Another way of thinking of it? You know OLPC right? And how Intel/MS wanted to subsidize cheap PCs in the OLPC market just to proliferate Windows? Read this and tell me what you think: http://wilshipley.com/blog/2006/11/youve-gotta-fig ht-for-your-right-to.html I see the similarity. When Windows adopted the BSD stack, the authors of the BSD stack basically won.
My roommate in college loved those eraserhead pointing devices. So that makes two.
On the other hand, I hate it. The tendency to overshoot is something I couldn't correct easily since you have less muscles involved to adjust accuracy with. My challenge to him was to dual me in Unreal Tournament. Me on a trackpad, him on a trackpoint. Unfortunately, he chickened out cuz I was sure I'd be able to make my point despite how crappy of an experience I can imagine UT on a trackpad to be.
Either you're trolling or you've simply missed his point completely. The poster above has it right.
The problem isn't freedom or lack of freedom. If you're given a linux-based device with the ability to install software of your choice, you have your freedom.
What the parent wants is uniformity and proper user interface design. A singular target for support. If you're of the minority who doesn't like that singular target, it's no problem. You can go take your hardware and compile your own distro. Nothing's going to stop you. Don't get me wrong, the standards you talk about for interoperability are important. But you could have some 50 different linux VOIP phones and it wouldn't help if every single one of them have horrible UIs, right? Same thing applies to computers.
For the less technical people, the "one true Linux distro" would help greatly in getting users adjusted to Linux. This is necessary for acceptance. If every single Linux box looks completely different, wouldn't the average user get confused? Heck, think about helping your mom's friend learn how to use their new Linux box. If they can't tell you what desktop enviroment they're running, how are you to know that they're using KDE, Gnome, fwm95, xfce, or blackbox? Even among techies like us, there's merit in having the "one true Linux distro" in that sometimes, we don't want to take the time to learn about all the different tools before making a choice. We just want one good enough until we feel it's not, and the freedom to change it.
A singular Linux platform benefits us all. One consistent UI for n00bs. One consistent UI for writers to write books about to help users. One consistent UI for us programmers to help the n00bs learn. One consistent target for writing apps in any language the programmer prefers. One consistent target for companies who just can't release their apps as open source, and one consistent platform to officially support. (Users of any other distros can support themselves as they do now) One consistent target for hardware manufacturers to write drivers for. (If we don't freeze a kernel interface for hardware drivers, we're doomed. Freeze one interface. If there comes a time where we must change it, we can always have a 2nd interface.) One starting point for users to learn with, but the flexibility for them to load anything else they want.
My distro of choice for making servers and stuff is Gentoo. But until there's a sign that Linux will become easier to support, my main computer will be my Mac.
From what I recall, there were discussions which mention that moving from well-optimized Altivec code to well-optimized SSE3 code will result in a significant performance drop.
Either, you're doing a mac specific app and use the Accelerate.framework which handles conversion to SSE3 or Altivec depending on the platform......or, you'll just pull in all the SSE work you did from the Windows Flash runtime since it's the same chip and these are all not OS dependent.
Same thing for Photoshop. The plugin architecture makes it hella easy since they should have started with plugins for all the heavy stuff anyways. Recycling! It's not just for cans.
Plus, it's a benefit to Windows users as well. A smaller render engine with less features = faster rendertime and easier to debug. This means that Outlook could theoritically improve in performance. Can you believe it?
USB2 is theoretically 480Mbps, but in reality it maxes out around 260Mbps. (33MB/sec) http://www.barefeats.com/usb2.html (read the "tested on Windows" part) Plus, if you look into it further, using USB devices saps CPU power. USB2 is a polled I/O bus meaning that the CPU must constantly do work monitoring each device for activity. Even empty USB connectors require constant CPU effort to scan for device insertion.
Next, 7200RPM drives max out around 60MB/sec. We really don't have a reason to be using anything past ATA/66 if you have just one drive on each channel. I mean, look at the max speed for the 160MB Seagate desktop drive and the fastest 7200rpm notebook drive ever. Not 60MB/sec. http://www.notebookreview.com/default.asp?newsID=2 840
Flash drives have the advantage in seek time, but the majority of time you're working with RAM on disk, it's going to be pages, which the bandwidth can easily negate the seek time advantage.
I think I have to stand up for my accomplishments this time. Regarding a BA in CS versus a BS CS&E, they're not that different in some schools.
I got a BA from the CS program. And by the time I graduated in my 8 semesters, I had implemented a compiler, a stupid OS simulation, a disassembler, and my personal favorite: an actual working CPU on a FPGA.
Lemme tell you what the difference between properly doing a BA and a BS is: 1) The BA takes more classes with live girls in them. But doesn't take general physics. (not taking Electricity and Magnetism doesn't hurt you if you take Intro to Microdevices where you learn all of it anyways in the context of gate logic) 2) The BS gives you more time to work on your projects by consolidating like 4 GE requirements into a class known as English 135AC, which is just like any other class except now you don't have to take 3-4 of them. 3) I had to get good grades in college to make it into the BA major. My roommates just had to get good grades in high school to get a BS.
Assuming you discount high school work, getting the BA in CS took more effort than the BS (with a CS focus), and was the only thing that didn't make sense at school. On the other hand, if you were talking about getting a BS with a EE focus (everything below mosfet level), those guys are hardcore.
Like digicams and video cards, there's multiple factors in determining "goodness".
Digicams: Megapixels get you detail at the expense of noise and sensitivity. Color accuracy gets you prettiness. Color accuracy goes DOWN as megapixels goes up because of noise and sensitivity. Finding the sweet spot for a consumer camera becomes key.
Mice: Camera resolution (DPI) versus framerate (FPS).
To track at a higher speed, you need higher framerates. This is so that at the higher speed, it can actually get the delta between two frames. If your movement speed surpasses what your framerate gets you, it can't tell how far you moved because the two pictures look like it got one frame, and then instantly teleported somewhere else the next frame. No delta. Think about playing quake at 1 frame per second. You see the guy shooting you with a rocket launcher one frame, you see blackness the next. What happened?:)
Multi-layered boards have been used everywhere for the last couple decades. But in this case, the reason they're able to achieve the cleanliness you speak of is simply because most of the stuff they need is integrated into modules or chips, and they're able to be more flexible in routing because of the lower clock speeds.
Consider the two wireless modules, all the messy-looking stuff is actually hidden under the shielding. Additionally, the compactness and the reduced heat generation is simply because they're not pushing Ghz's like the PS3 and Xbox360. For the same reason a PentiumPro/P2/P3/PentiumM/Core/Core2 PC can be compacted down easier than a P4, the Wii runs cooler than the competition. (In fact, the GameCube and the Wii both use a chip very similar to the PPC750 aka the PowerPC G3, which is very efficient and cool)
Another benefit of using lower clocked circuits is that routing is easier. If you've got lines that have high frequencies, they actually cause lines running in parallel to have fluctuations in voltage (whether by inductance or capacitance, I forget the details), so you have to do weird-looking routing patterns and computer analysis to make sure your signals actually get there without being jumbled up completely. They don't even have to distance them as much, so they could indeed hide it inside the inner layers, which helps make it harder for people who are trying to hack it and make mod chips.
There's a lot of factors in designing a board. From my perspective, cleanliness is just a side effect.
Slashdot Mirror
Wow... that module sucks. No EVDO? At least it has audio, unlike MultiTech's CDMA module.
What we actually need is a reference dev kit for Qualcomm's MSM6500 or better to be able to make something. Or a module based off it. It'd be neat if we could create a CDMA module that would plug in place of the iPhone's baseband board.
Right now, I'm still thinking we're better off getting old EVDO phones, powering up just the board with no screen or keypad, and using that as a module through rs232 or usb.
No way.
Verizon is allowing the certification of devices (most likely not phones), to use their network.
There are three parties involved:
1) Network owner : Verizon.
2) Device manufacturer :
3) End user :
Just because #2 is now allowed to register stuff with #1's net, doesn't mean #3 can do whatever they want.
Example? Kindle uses Sprint's 3G EVDO network to provide connectivity without WiFi. Amazon (#2) got permission from Sprint to have these non-phone devices connect to their network. Amazon also foots the EVDO bill for #3. As end users (#3), we're not given any easy way to use Kindle for anything but buying books and wifi with a crappy browser.
Square that? Many of us don't need to.
I'm a total fan of open source projects. Except I don't like the GPL. I consider it too restrictive because realistically, I need to be able to work with different licenses and closed source projects.
If I make something derived from a GPL'd library, then yes, I'll GPL it only out of compliance.
But if I had a choice, my public works would be BSD-style because I'll put my money where my mouth is and share.
Yeah, ebay isn't the place to go for a 29 inch CCFL.
:)
There are companies online that specialize in selling these, so I'm sure they have a good way of shipping them.
For me, I just bought a broken panel when I needed the one for the 15 inch LCD
Which is exactly why I picked up a modern rear projection DLP TV from my friend.
The projector unit is the same that they sell in the tiny projector boxes for presentations, and the failure mode is typically the light bulb. Mine has a hatch on the front to let you swap it out.
LCD backlights fail often, but they're not that difficult to swap. Most panels are designed with a tray at the top and bottom that slides out with the tube. The hard part is sourcing the tube or about the same length. I've done a fix where I installed the backlight CCFL from a broken 17inch desktop monitor into a smaller 15 inch desktop monitor. A little silicone later to block out the extra light and now I have a working monitor again.
It's most likely the connector cable that goes from the lcd panel to the logic board.
You can't fix the cable. They use two types. One that's a plastic ribbon (less likely). One's a bundle of micro-coax wires which you can't easily solder to because you can barely see the connector end.
The good news is that you ought to be able to buy a cracked display unit or a bezel which still has the cable. The LCD panel itself has a connector. The logic board has another connector. So the cable can be unplugged from both ends.
(I've done this to a Titanium Powerbook G4 before.)
Hehehe. My friend gave me a broken 60 inch rear projection HDTV.
$250 for a new bulb later, it's nice and bright in my living room.
Depending on the technology for your HDTV, it could be just as easily fixable. Just have to know what's the most common failure case and make sure it's one you can cover.
Just curious. I have two VCRs that aren't working. One plays too fast, the other plays then stops. (symphonics and sony) I opened them both up and couldn't find any mechanism that would cause either problem. Got any ideas?
I had one semester of free time between high school and college, and I spent it buying broken Playstation 1 units off ebay and fixing them for resale. Great source of income back then. It was enough to pay off all my PS2 games and accessories as the PS2 came out shortly after I got to college.
I then moved on to iPods for a little while. The price of broken iPods pushed me to Motorola E815 cell phones.
And now I'm doing nintendo DS's despite having a full time job.
Actually it isn't hard to believe.
Apple's not only providing the phone, but also providing the means for the iPhone users to handle activation, as well as the majority of customer support (warranty, assistance, marketing and training).
Yes, AT&T stores have marketing materials and people who can help you with the iPhone, but not much more than a normal phone. The idea is that you'll typically go to the Apple store for iPhone specific support.
This $18 isn't just a share of the plan. It's a share of what it would have cost AT&T to get and MAINTAIN the subscriber, seeing how for iPhones, Apple is handling a lot of it.
I'd so mod you up if i had mod points.
The only good thing that has come out of the intel switch is the employee stock purchase plan.
"Why did they wait for 3 months after the release to announce it?"
Because Apple announces only when they can deliver or can give a pretty close date for delivery.
Most people don't realize that the iPhone is essentially the first release of Leopard. Articles talking about WWDC mention that Core Animation and a bunch of other stuff were in Leopard but were developed originally for the iPhone. The way I saw it, iPhone was an incomplete product. It did everything marketing required it do on release. (it's an iPod, it browses the web, and it is a phone. Can't deny it fit the criteria, ne?) But it could do more. Trouble is, when you also have the company's baseline platform to finish up, it's kinda more important to work on Leopard, isn't it?
Since Leopard wasn't done yet, and many bugs were being worked out on both Leopard and iPhone, it wasn't certain that they could bring out a native SDK and keep Leopard on track.
Now that Leopard's off to the CD presses (afterall, if they're shipping in less than a week, the final CD had to have been handed to the presses at least a week ago), it's perfect time to evaluate the progress of other internal projects. Now that the OSX team has probably gotten close to finishing up a native iPhone SDK, give them the 4 months to finish QA, write up documentation, translate it to all the languages, and release it in Feburary.
Market forces had nothing to do with this. PR outrage had nothing to do with this. All that just made everybody talk more about the iPhone. For every person who said "iPhone sucks" there were more who said "iPhone? What is this iPhone? Oh cool! *opens wallet*".
Not having enough engineers lined up to finish the OS had everything to do with this.
If you think I'm giving them too much credit, so be it. Based on Apple's behavior before, I guessed this would happen when they made the iPhone web app announcement. When hackers got access to iPhone 1.0.2's file system, I made a quick scan of the bluetooth manager and it pretty much assured me it would happen. Now that it's been announced officially, it feels like no big deal to me.
Sure, if MS had to develop their own TCP stack, they'd have less of an advantage in the OS market. Would this have translated to more BSD users?
g ht-for-your-right-to.html
No. The market isn't a zero sum game. If Windows doesn't do as well, there's no reason BSD would do better. Afterall, they could have both sucked it up and OS/2 could have won.
In fact, if the BSD implementation was not as open as it were, every OS would have had to develop their own TCP stack or go GPL.
Going GPL is obviously not in the interest of many businesses.
What does this imply? TCP would not have become the standard it is now. And we'd be back in the same situation as the 80s where every network device had to speak Arcnet, Appletalk, a version of TCP compatible for each OS' TCP stack bugs, Lantastic, IPX/SPX...etc...
Simply put, if the reference design for TCP was GPLed instead of BSD, we're all fucked. Only GPLed OSs, of which there are little compared to non-GPL OSs, would have that stack. Every other OS would have their own implementation complete with various bugs and inconsistencies.
If you think that "oh, that web page doesn't work with my version of firefox" is bad, then what about "oh, my linux box can't talk with any solaris hosts, unless it's got solaris TCP version 7.56.9-beta42 and my kernel is 2.8.666." or some crap like that.
The simple fact that because a BSD-licensed TCP stack allows any OS, regardless of size, intentions, or evilness as judged by us Slashdotters, to take the stack and integrate it into their OS demonstrates it's own ability to maintain freedom:
If everybody's starting with this stack, you're starting with a standard version of which you need to maintain compatibility with or you have cut yourself out of the community to your own detriment.
If you have some brilliant ideas to make it better, sure, you can let everybody else know and contribute back, or it could be a trade secret of yours that you'll have to maintain on your own. The fact that the original code is free and out there and enjoyed by as many people as possible has already been fulfilled even when you closed your fork. Besides, maintaining and significant changes and then merging in new developments from the mainline is still a pain in the ass no matter what. If it wern't, you probably didn't make any changes significant enough for people to care anyways, so it's not like we as a whole lose.
As far as the users of the code are concerned, in the case of TCP, they are enjoying one of the most important freedoms the writers intended: That their computer can communicate with the world. The majority of Windows users arn't programmers anyways. But no matter who you are, having a computer that can network with everybody else's is pretty important.
The way I see it. BSD helps proliferation. GPL encourages work to be consolidated. The question isn't what's "freer", it's who you want using it.
Anyways, I agree with the claim this doesn't harm BSD because this is exactly what they wanted. They want people using their code.
Another way of thinking of it? You know OLPC right? And how Intel/MS wanted to subsidize cheap PCs in the OLPC market just to proliferate Windows? Read this and tell me what you think: http://wilshipley.com/blog/2006/11/youve-gotta-fi
I see the similarity. When Windows adopted the BSD stack, the authors of the BSD stack basically won.
My roommate in college loved those eraserhead pointing devices. So that makes two.
On the other hand, I hate it. The tendency to overshoot is something I couldn't correct easily since you have less muscles involved to adjust accuracy with. My challenge to him was to dual me in Unreal Tournament. Me on a trackpad, him on a trackpoint. Unfortunately, he chickened out cuz I was sure I'd be able to make my point despite how crappy of an experience I can imagine UT on a trackpad to be.
Either you're trolling or you've simply missed his point completely.
The poster above has it right.
The problem isn't freedom or lack of freedom. If you're given a linux-based device with the ability to install software of your choice, you have your freedom.
What the parent wants is uniformity and proper user interface design. A singular target for support.
If you're of the minority who doesn't like that singular target, it's no problem. You can go take your hardware and compile your own distro. Nothing's going to stop you. Don't get me wrong, the standards you talk about for interoperability are important. But you could have some 50 different linux VOIP phones and it wouldn't help if every single one of them have horrible UIs, right? Same thing applies to computers.
For the less technical people, the "one true Linux distro" would help greatly in getting users adjusted to Linux. This is necessary for acceptance. If every single Linux box looks completely different, wouldn't the average user get confused? Heck, think about helping your mom's friend learn how to use their new Linux box. If they can't tell you what desktop enviroment they're running, how are you to know that they're using KDE, Gnome, fwm95, xfce, or blackbox? Even among techies like us, there's merit in having the "one true Linux distro" in that sometimes, we don't want to take the time to learn about all the different tools before making a choice. We just want one good enough until we feel it's not, and the freedom to change it.
A singular Linux platform benefits us all.
One consistent UI for n00bs.
One consistent UI for writers to write books about to help users.
One consistent UI for us programmers to help the n00bs learn.
One consistent target for writing apps in any language the programmer prefers.
One consistent target for companies who just can't release their apps as open source, and one consistent platform to officially support. (Users of any other distros can support themselves as they do now)
One consistent target for hardware manufacturers to write drivers for. (If we don't freeze a kernel interface for hardware drivers, we're doomed. Freeze one interface. If there comes a time where we must change it, we can always have a 2nd interface.)
One starting point for users to learn with, but the flexibility for them to load anything else they want.
My distro of choice for making servers and stuff is Gentoo. But until there's a sign that Linux will become easier to support, my main computer will be my Mac.
Cuz we can't cool them off fast enough without a refrigerant or peltier stacks?
Seriously, it's like high rpm engines and friction. Clock it higher, heat it faster, wear it out sooner.
Torque is good. It's all about the IPC.
Too bad for you, it's easier to lube a engine than a toggle a transistor with less waste heat.
SSE3 damn fast?n ce/Conceptual/Accelerate_sse_migration/migration_s se_translation/chapter_4_section_8.html
Even Apple's dev docs mention to developers who are planning on using SSE that there will be plenty of problems:
http://developer.apple.com/documentation/Performa
From what I recall, there were discussions which mention that moving from well-optimized Altivec code to well-optimized SSE3 code will result in a significant performance drop.
Dude, all work's been done.
...or, you'll just pull in all the SSE work you did from the Windows Flash runtime since it's the same chip and these are all not OS dependent.
Either, you're doing a mac specific app and use the Accelerate.framework which handles conversion to SSE3 or Altivec depending on the platform...
Same thing for Photoshop. The plugin architecture makes it hella easy since they should have started with plugins for all the heavy stuff anyways. Recycling! It's not just for cans.
SSDs arn't using static ram. They use flash or DRAM which is much slower than SRAM
Plus, it's a benefit to Windows users as well. A smaller render engine with less features = faster rendertime and easier to debug. This means that Outlook could theoritically improve in performance. Can you believe it?
Not even close.
2 840
s /ocz_rally_high_performance_usb_2_0-dual_channel-f lash_memory_drive
USB2 is theoretically 480Mbps, but in reality it maxes out around 260Mbps. (33MB/sec)
http://www.barefeats.com/usb2.html (read the "tested on Windows" part)
Plus, if you look into it further, using USB devices saps CPU power. USB2 is a polled I/O bus meaning that the CPU must constantly do work monitoring each device for activity. Even empty USB connectors require constant CPU effort to scan for device insertion.
Next, 7200RPM drives max out around 60MB/sec. We really don't have a reason to be using anything past ATA/66 if you have just one drive on each channel. I mean, look at the max speed for the 160MB Seagate desktop drive and the fastest 7200rpm notebook drive ever. Not 60MB/sec.
http://www.notebookreview.com/default.asp?newsID=
Finally, the fastest class of USB2 flash drives is the dual channel ones. Essentially RAID stripe across two chips. It puts out data at 28MB/sec. http://www.ocztechnology.com/products/flash_drive
Flash drives have the advantage in seek time, but the majority of time you're working with RAM on disk, it's going to be pages, which the bandwidth can easily negate the seek time advantage.
I think I have to stand up for my accomplishments this time. Regarding a BA in CS versus a BS CS&E, they're not that different in some schools.
I got a BA from the CS program.
And by the time I graduated in my 8 semesters, I had implemented a compiler, a stupid OS simulation, a disassembler, and my personal favorite: an actual working CPU on a FPGA.
Lemme tell you what the difference between properly doing a BA and a BS is:
1) The BA takes more classes with live girls in them. But doesn't take general physics. (not taking Electricity and Magnetism doesn't hurt you if you take Intro to Microdevices where you learn all of it anyways in the context of gate logic)
2) The BS gives you more time to work on your projects by consolidating like 4 GE requirements into a class known as English 135AC, which is just like any other class except now you don't have to take 3-4 of them.
3) I had to get good grades in college to make it into the BA major. My roommates just had to get good grades in high school to get a BS.
Assuming you discount high school work, getting the BA in CS took more effort than the BS (with a CS focus), and was the only thing that didn't make sense at school. On the other hand, if you were talking about getting a BS with a EE focus (everything below mosfet level), those guys are hardcore.
Where did I go? I went to UC Berkeley.
Like digicams and video cards, there's multiple factors in determining "goodness".
:)
Digicams: Megapixels get you detail at the expense of noise and sensitivity. Color accuracy gets you prettiness. Color accuracy goes DOWN as megapixels goes up because of noise and sensitivity. Finding the sweet spot for a consumer camera becomes key.
Mice: Camera resolution (DPI) versus framerate (FPS).
To track at a higher speed, you need higher framerates. This is so that at the higher speed, it can actually get the delta between two frames. If your movement speed surpasses what your framerate gets you, it can't tell how far you moved because the two pictures look like it got one frame, and then instantly teleported somewhere else the next frame. No delta. Think about playing quake at 1 frame per second. You see the guy shooting you with a rocket launcher one frame, you see blackness the next. What happened?
Multi-layered boards have been used everywhere for the last couple decades. But in this case, the reason they're able to achieve the cleanliness you speak of is simply because most of the stuff they need is integrated into modules or chips, and they're able to be more flexible in routing because of the lower clock speeds.
Consider the two wireless modules, all the messy-looking stuff is actually hidden under the shielding.
Additionally, the compactness and the reduced heat generation is simply because they're not pushing Ghz's like the PS3 and Xbox360. For the same reason a PentiumPro/P2/P3/PentiumM/Core/Core2 PC can be compacted down easier than a P4, the Wii runs cooler than the competition. (In fact, the GameCube and the Wii both use a chip very similar to the PPC750 aka the PowerPC G3, which is very efficient and cool)
Another benefit of using lower clocked circuits is that routing is easier. If you've got lines that have high frequencies, they actually cause lines running in parallel to have fluctuations in voltage (whether by inductance or capacitance, I forget the details), so you have to do weird-looking routing patterns and computer analysis to make sure your signals actually get there without being jumbled up completely. They don't even have to distance them as much, so they could indeed hide it inside the inner layers, which helps make it harder for people who are trying to hack it and make mod chips.
There's a lot of factors in designing a board. From my perspective, cleanliness is just a side effect.
Gotta give some love for RAID 0.
:)
:P
I got myself my old Athlon 2000 with a triple channel Mylex DAC960 hardware raid controller.
Two 9gb 15kRPM drives per channel for a 120MB/sec sustained transfer rate.
Sure, if one drive dies, it's a goner, but hey, all this PC does is run warcraft 3 and give me bragging rights.