Not to mention the Zilog Z8000! Imagine where we would be if it had won the 16-bit wars. We'd all be running TRSDOS 64-bit edition and SuperScripsit 2011.
The air is not and has never been a desktop replacement machine. I have an original model and it is a great laptop if you are on the go and money is not an option. My biggest complaint with the original Air is that battery life is abysmal. It sounds like they have focused most on improving that. The other problem with the original Air was the price. Now, you may complain about the price of the new version, but the fact is they have brought it down quite a bit.
Machines like this don't need the latest and greatest CPU. There were very few things that I do on my original air that tax its CPU. That's because I'm not editing video, etc on it. You basically have grown-up machine that you can do actual work on and it's not much bigger than an iPad.
Do I see a need to have both an ipad and an air? Probably not. If you have an ipad already, then chances are you would be better served with a larger laptop.
As far as the ethernet port. On a machine like this, it's not that important. I have a USB ethernet dongle and I've used it 2-3 times.
From my experience, I completely agree. My partner and I are raising a baby girl (we are both men) and 'ma' was the first sound she made. She often says ma-ma and doesn't know any mother. I have assumed that the word mama comes from the sounds that babies make.
The n32 ABI did have double the number of floating point registers available to it (32 vs 16).
Perhaps this is because mips 3000 had 32 32-bit fp registers, but half the registers had limited functionality. They could be used as 16 64-bit fp registers. The r4000 had 32 64-bit fp registers.
SGI did something similar when they brought out 64-bit MIPS chips and a 64-bit version of IRIX. They also doubled the number of registers. However, they permitted enabling those registers in 32-bit mode as well. Thus you could compile to old-32 ABI, or new-32 ABI (with 2x as many registers) or 64-bit ABI. So, most apps would be compiled for the new-32 bit ABI since few needed 64-bits but most could benefit from more registers. It's too bad AMD didn't do the same thing.
But under standard 32-bit XP, each process gets only a 2GB address space. That is much more of a limitation than 4GB total RAM for many applications, especially since graphics drivers, etc may eat into that 2GB.
Also, if you use an app that needs large contiguous chunks of memory, you may run into a situation where you have say 600MB of address space left, but no one chunk is larger than 200MB. Then if your app needs a 250MB chunk of memory, it will fail even though there is 600MB left.
Going to a 64-bit address space solves these issues.
I develop medical software for a living and this is the scariest thing I have ever read.
Your example is incomplete. Imagine that you type "rm -rf / junk" and the system responds "Delete/junk?", so you answer "Y" and it then deletes the whole filesystem.
It is most certainly a bug. First, there is a mismatch between what is shown on the screen and what the system is doing. That is a bug by any definition. Second, the system obviously had gaps in its validation of input. This makes it no less of a bug than many of the others listed (eg fingerd bug).
Furthermore, it is the responsibility of designers and developers of medical software to ensure that potential hazards are identified and mitigated. A hazard of "calculated dose does not match image shown on screen" is not some obscure hazard that no one would have thought of - it is the first that comes to mind!
Please tell me that these people are not involved in medical software anymore.
The medical market (at least that I am familiar with) has switched. LCD's are good enough for most applications and what is important is that users are demanding them.
We are in the same position as far as stereo displays, however we are already using autostereoscopic displays.
Don't be so sure. My mother used to work in a
small hospital and one time a robot dialer
worked its way through every phone in the hospital. I think they eventually got the phone company to intervene and block the calls.
When are the 64-bit benchmarks coming
on
AMD64 Preview
·
· Score: 1
I have read so many reviews of opteron/athlon64 running in 32-bit mode under windows.
There is good opportunity right now for some site to stand apart from the crowd by doing a real review on a 64-bit version of linux. Does anyone know of any out there?
We are using the FOX toolkit and we are happy.
It is a simple toolkit, designed and implemented mostly by one person. It uses the primitive drawing routines of the underlying platform (win32 or X) so looks the same on all platforms. It is written in C++, but there are ruby and python bindings.
The author has spent a lot of time looking at existing UI libraries and tried to learn from them before writing his own.
The privacy requirements put a crimp on research institutions. In order for research to proceed, information needs to flow freely. For example, you are at at research hospital developing software to diagnose some condition. You will not be able to do this because your software will not be hipaa compliant (how can it be, it is just a research project). I agree that privacy is important, but hipaa solves a problem that does not exist and the regulations were written by people (beaurocrats) who don't know anything about what they're regulating.
Consider that you are a patient with a problem. Hipaa now prevents other doctors from examining your case without permission. How many problems get diagnosed accidently by doctors checking on patients that are not theirs? This will no longer happen because they don't have permission to check.
First, CVS is built on top of RCS and, as such, doesn't handle binary files. Okay, that's a fib; it sorta kinda does, but it's very klunky, and easily prone to errors. Further, it's easy for the "binary-ness" of a file to be lost (i.e. be treated as text), resulting in all kinds of nasty corruption. Best Practices will avoid this, but everyone has to be on their toes all the time.
Hmm. I have never seen this happen. I can't imagine how it could unless you use something like the cvs admin command. However, I'll take your word that it can happen, but to me seems unlikely.
Second, CVS has no notion of "transactions". Let's say you check in a bugfix/new feature to the kernel. The change involves modifying six different files. CVS does not see this checkin as a single transaction, but six completely separate ones. So a lot of information about the scope of a given change is not easily found. The only way you can know a particular change affected multiple files is by noticing that their checkin comments are identical.
You are certainly right about CVS not having transactions. However, in the case you bring up, if you are concerned that the change may need to be backed out, etc, you can create tags before and after committing to get a "poor man's transaction."
Further, if you perform a checkin against multiple files and one or more of them has a conflict (someone else checked in a change before you did), CVS will simply halt at the conflicting file; earlier files successfully checked in up to that point are not backed out.
This is wrong and has been as long as I can remember. Maybe you have not used cvs for a long time? If the files you are trying to check in have changed, cvs will prevent the commit from starting. You will have to merge the changes to your local copy via cvs update before it will let you commit. Perhaps they added this feature because of complaints like yours.
The only time I have seen a commit get partially committed was when I was committing over a network and the network connection died in the middle. I admit that it is a pain if it happens.
I think that many times when people create things - whether paintings, novels, or software - one of the motivating factors is to get recognition from others. That does not taint the works that are created - it is just one of the reasons that people create things. Would it be better if they were creating them for money?
Using the konquerer example, I am not saying that they should have preferential treatment. However, if you work on konquerer knowing that it is the official browser of KDE and then find that one of the major distributions will include a "KDE" that has mozilla as its primary browser, then you will feel that your work is being treated unfairly.
How can you call the resulting desktop KDE when some of the major components have been replaced?
Perhaps the best solution is for Red Hat to supply 3 desktops: standard KDE, standard gnome, and a mish-mash "Red-Hat" desktop.
As a KDE user, I am not that bothered by what red hat is doing. I think that most KDE users use some KDE apps, some gnome apps, some apps that are neither. It is my right to choose which software I run. So, for example, when I started using KDE, I used konquerer because that was the browser for KDE. Eventually, I decided that I liked mozilla better and switched to that.
However, if you look at it from the KDE developers' point of view, it is different. Imagine that you are a konquerer developer. Until now, you know that if someone tries out KDE, they will try out konquerer and that makes you happy. They may not stick with it, but that is ok. However, with the new redhat, even if a user chooses KDE, they may never see konquerer! Now you begin to wonder what are you writing all this open source code for if no one will see it. I think that is why the KDE developers are so upset.
My daughter & I had just finished a salad at Neiman-Marcus Cafe in Dallas
and decided to stuff our fat asses with cookies. Because our family of fat
asses are such cookie lovers, we decided to try the "Neiman-Marcus chicken
meat Cookie". It was so excellent that I asked if they would give me the
recipe. With a sort of twisted smile, she said, "Yes." I asked how much,
and she responded, "Two fifty K." I said with approval, just add it to my tab.
one day later, I received my VISA statement from Neiman-Marcus and it was
$250,029.95. I looked again and I remembered I had only spent $9.95 for two
salads and about $20.00 for cookies. As I glanced at the bottom of the
statement, it said, "Cookie-Recipe - $250,000.00." Boy, was I pissed!! I
called Neiman's Accounting Dept. and told them the waitress said it was "two
fifty K," and I did not realize she meant $250,000.00 for a cookie recipe.
I tried to play ignorant but they didn't fall for it.
I asked them to take back the recipe and reduce my bill and they said
they were sorry, but because all the recipes were this expensive so not
just everyone could duplicate any of our bakery recipes...the bill would
stand. Two days later some folks from the department store showed up and gave
me the ass beating of my life. They went insane on my fat ass. I didn't
think this act of senseless violence would ever end. My fat ass felt a
kind of pain I could never describe. The beating grew more sickening
by the minute. My pig like screams could be heard around the neighborhood.
Never cross a big department store.
It's been a long time since I have programmed
directly to Xlib. Can you tell me where the
limitation is? I am not disagreeing, just curious. Glancing
at XCreateColormap, for example, there
is nothing obvious there (colors use shorts).
Regardless, I would expect many visual apps
on linux to use OpenGL and I know that you
can use more than 8 bits per channel with OpenGL as I have done that on SGI's.
The nvidia drivers have their bugs, sure, but in my experience (graphics programmer), blue screens are very rare. Perhaps something is wrong with your hardware.
Also, OpenGL has no scene graph. I think you don't know what you're talking about.
The impression I got from the story is that Intel put this in now so that they can figure out what bottlenecks they face in turning hyperthreading into an advantage. Also, it gives compiler writer, OS writers, and application writers some exposure to the technology. I would not be surprised if a few years down the road this is a big win for some environments (ie, not office2.005k)
Slashdot Mirror
Not to mention the Zilog Z8000! Imagine where we would be if it had won the 16-bit wars. We'd all be running TRSDOS 64-bit edition and SuperScripsit 2011.
Machines like this don't need the latest and greatest CPU. There were very few things that I do on my original air that tax its CPU. That's because I'm not editing video, etc on it. You basically have grown-up machine that you can do actual work on and it's not much bigger than an iPad.
Do I see a need to have both an ipad and an air? Probably not. If you have an ipad already, then chances are you would be better served with a larger laptop.
As far as the ethernet port. On a machine like this, it's not that important. I have a USB ethernet dongle and I've used it 2-3 times.
From my experience, I completely agree. My partner and I are raising a baby girl (we are both men) and 'ma' was the first sound she made. She often says ma-ma and doesn't know any mother. I have assumed that the word mama comes from the sounds that babies make.
Perhaps this is because mips 3000 had 32 32-bit fp registers, but half the registers had limited functionality. They could be used as 16 64-bit fp registers. The r4000 had 32 64-bit fp registers.
SGI did something similar when they brought out 64-bit MIPS chips and a 64-bit version of IRIX. They also doubled the number of registers. However, they permitted enabling those registers in 32-bit mode as well. Thus you could compile to old-32 ABI, or new-32 ABI (with 2x as many registers) or 64-bit ABI. So, most apps would be compiled for the new-32 bit ABI since few needed 64-bits but most could benefit from more registers. It's too bad AMD didn't do the same thing.
Also, if you use an app that needs large contiguous chunks of memory, you may run into a situation where you have say 600MB of address space left, but no one chunk is larger than 200MB. Then if your app needs a 250MB chunk of memory, it will fail even though there is 600MB left.
Going to a 64-bit address space solves these issues.
Your example is incomplete. Imagine that you type "rm -rf / junk" and the system responds "Delete /junk?", so you answer "Y" and it then deletes the whole filesystem.
It is most certainly a bug. First, there is a mismatch between what is shown on the screen and what the system is doing. That is a bug by any definition. Second, the system obviously had gaps in its validation of input. This makes it no less of a bug than many of the others listed (eg fingerd bug).
Furthermore, it is the responsibility of designers and developers of medical software to ensure that potential hazards are identified and mitigated. A hazard of "calculated dose does not match image shown on screen" is not some obscure hazard that no one would have thought of - it is the first that comes to mind!
Please tell me that these people are not involved in medical software anymore.
FEMA is not technically a TLA.
We are in the same position as far as stereo displays, however we are already using autostereoscopic displays.
Here's an example where the government is trying to deport a canadian who was caught with marijuana when he was 15 years old. http://www.atlanta.creativeloafing.com/2005-03-03/ news_feature.html
Don't be so sure. My mother used to work in a small hospital and one time a robot dialer worked its way through every phone in the hospital. I think they eventually got the phone company to intervene and block the calls.
There is good opportunity right now for some site to stand apart from the crowd by doing a real review on a 64-bit version of linux. Does anyone know of any out there?
The author has spent a lot of time looking at existing UI libraries and tried to learn from them before writing his own.
See the website for more info.
Consider that you are a patient with a problem. Hipaa now prevents other doctors from examining your case without permission. How many problems get diagnosed accidently by doctors checking on patients that are not theirs? This will no longer happen because they don't have permission to check.
Ha, I tried to contact them to let them know they should use html and got an error from their MS server. They are hopeless
Hmm. I have never seen this happen. I can't imagine how it could unless you use something like the cvs admin command. However, I'll take your word that it can happen, but to me seems unlikely.
Second, CVS has no notion of "transactions". Let's say you check in a bugfix/new feature to the kernel. The change involves modifying six different files. CVS does not see this checkin as a single transaction, but six completely separate ones. So a lot of information about the scope of a given change is not easily found. The only way you can know a particular change affected multiple files is by noticing that their checkin comments are identical.
You are certainly right about CVS not having transactions. However, in the case you bring up, if you are concerned that the change may need to be backed out, etc, you can create tags before and after committing to get a "poor man's transaction."
Further, if you perform a checkin against multiple files and one or more of them has a conflict (someone else checked in a change before you did), CVS will simply halt at the conflicting file; earlier files successfully checked in up to that point are not backed out.
This is wrong and has been as long as I can remember. Maybe you have not used cvs for a long time? If the files you are trying to check in have changed, cvs will prevent the commit from starting. You will have to merge the changes to your local copy via cvs update before it will let you commit. Perhaps they added this feature because of complaints like yours.
The only time I have seen a commit get partially committed was when I was committing over a network and the network connection died in the middle. I admit that it is a pain if it happens.
Using the konquerer example, I am not saying that they should have preferential treatment. However, if you work on konquerer knowing that it is the official browser of KDE and then find that one of the major distributions will include a "KDE" that has mozilla as its primary browser, then you will feel that your work is being treated unfairly.
How can you call the resulting desktop KDE when some of the major components have been replaced?
Perhaps the best solution is for Red Hat to supply 3 desktops: standard KDE, standard gnome, and a mish-mash "Red-Hat" desktop.
However, if you look at it from the KDE developers' point of view, it is different. Imagine that you are a konquerer developer. Until now, you know that if someone tries out KDE, they will try out konquerer and that makes you happy. They may not stick with it, but that is ok. However, with the new redhat, even if a user chooses KDE, they may never see konquerer! Now you begin to wonder what are you writing all this open source code for if no one will see it. I think that is why the KDE developers are so upset.
That is why I refuse to run real player any more. Content that comes only in real media format is not worth my time.
My daughter & I had just finished a salad at Neiman-Marcus Cafe in Dallas and decided to stuff our fat asses with cookies. Because our family of fat asses are such cookie lovers, we decided to try the "Neiman-Marcus chicken meat Cookie". It was so excellent that I asked if they would give me the recipe. With a sort of twisted smile, she said, "Yes." I asked how much, and she responded, "Two fifty K." I said with approval, just add it to my tab. one day later, I received my VISA statement from Neiman-Marcus and it was $250,029.95. I looked again and I remembered I had only spent $9.95 for two salads and about $20.00 for cookies. As I glanced at the bottom of the statement, it said, "Cookie-Recipe - $250,000.00." Boy, was I pissed!! I called Neiman's Accounting Dept. and told them the waitress said it was "two fifty K," and I did not realize she meant $250,000.00 for a cookie recipe. I tried to play ignorant but they didn't fall for it.
I asked them to take back the recipe and reduce my bill and they said they were sorry, but because all the recipes were this expensive so not just everyone could duplicate any of our bakery recipes...the bill would stand. Two days later some folks from the department store showed up and gave me the ass beating of my life. They went insane on my fat ass. I didn't think this act of senseless violence would ever end. My fat ass felt a kind of pain I could never describe. The beating grew more sickening by the minute. My pig like screams could be heard around the neighborhood. Never cross a big department store.
Regardless, I would expect many visual apps on linux to use OpenGL and I know that you can use more than 8 bits per channel with OpenGL as I have done that on SGI's.
SGI's have supported more than 24-bit color for years.
Click on the link and read the story. Step two completed!
Read the article from start to end. Step three completed!
Click on the link to read the slashdot comments. Step four completed!
See that noone else made the same remark as me. Step five completed!
Post sardonic comment about authors writing style. Step six completed!
The nvidia drivers have their bugs, sure, but in my experience (graphics programmer), blue screens are very rare. Perhaps something is wrong with your hardware.
Also, OpenGL has no scene graph. I think you don't know what you're talking about.
The impression I got from the story is that Intel put this in now so that they can figure out what bottlenecks they face in turning hyperthreading into an advantage. Also, it gives compiler writer, OS writers, and application writers some exposure to the technology. I would not be surprised if a few years down the road this is a big win for some environments (ie, not office2.005k)