"I was evaluating visual studio.net again this week and whilst it might technically do everything it needs to, its slower than VS 6 at most things."
Yes, well, it parses and creates a tree in the background while you are typing. You could switch that off to start with. Working with VS 6 feels like using the Kawa development tools for Java (may it rest in pieces). You get a lot of functionality, a better plugin structure and lots more. They did OK with the last version in my opinion, even if it is still *way* behind Eclipse for the language features.
For this to work, there is a need to set a common API for the hardware acceleration. Otherwise, everybody has to do the same thing over and over again, or just ignore the instructions altogether.
Take for instance the C3 processor from VIA. The latest already do SHA, AES, RSA and hardware random generation. Will AMD use similar instruction sets, or will they use completely different instructions or even processors? How am I, as a programmer, going to use these instructions? These instructions should also be very well specified. For instance, the current SHA implementation of VIA doesn't allow streaming (from Java at least), you first have to put everything ready in RAM. That's not a good thing (TM), and could have been avoided if they left the calculations to the last block to be hashed out of the hardware.
Another thing is software support. Buying a C3 will do nothing to your existing applications, unless they use a plugin structure for software. E.g. in Java you can put a specialized VIA "cryptographic provider" in front of all the other providers, and the C3 will be automatically used. It can also be used within Windows using the CSP software library from VIA. But from NSS, the Netscape Security Suite also used by Firefox, you are still using the normal CPU instructions. For things like compression, which is almost always directly programmed in software, it will be much harder to use the additional instructions. Applications, in other words, should not only use threads, they should also be pretty modular and plugable (the next software revolution I hope).
Although I am very much in favour of speciallized instructions, I think a good support structure would be very much needed. Fortunately, AMD (or VIA for that matter) does not have a very bad track record on standardization. Or Open Source for that matter.
"Are current CPUs optimised for physics simulations? No."
No, but some add on cards, and in not too long a time, video processors may have physics simulators. That does not really
"For image processing? No."
True. If you don't count the various multi-media instructions.
"For data compression? No."
True.
"For encryption? No."
Except the Sun T1 (aka Niagara) and VIA C3 processors that is. Especially the C3 (in the Epia range of Mini-ITX motherboards) do SHA hashing operations, AES operations, faster RSA operations and to top it all off have an extremely usefull random generator.
Maybe, but being seen as a monopoly, they should remove this restriction none-the-less if they plan to sell this within Europe. If I as a user think that a driver should install, it should install. That's all there is to it.
Yeah, and those law inforcement agencies have clamped down on the drugs problem in the same way as well. Like *that* is working. Wishfull thinking goes both ways, and neither one may even be correct in the real world. It need not be black and white - as some floundering sons of founding fathers are finding out at this time. Pfff.
"So if someone is fired from their job for being addicted to drugs or alcohol, for being obese, for watching porn or chatting online at the office, then I think it's perfectly acceptable. I say this even though I am guilty of two of these offenses myself. I'm not willing to defend my own behavior by calling it a disease."
I think it is pretty damn brave that you come out on slashdot to admit that you are on drugs and watch porn at the office.
I would not know about that. 5 years ago, a fast drive would do a seek in about 12 ms. Now it is 9 ms. That's not impressive. Density and streaming speed has increased manyfold within that time frame. Seek time has not. And it'll be some time before the random access speed of flash is met by spinning disks (and more importantly maybe, the drive heads). Especially if you read the article carefully, and see that the 4x speed up mentioned is the overall speed speedup.
From the faq:
Q: How fast is your current SSD and what performance improvements does it offer?
A: The streaming R/W speeds are 57 MB/s and 32 MB/s, respectively, but the most significant performance advantage comes from its latency feature - less than 1 millisecond; roughly 10-15x faster than a hard disk drive.
"Just add more ram to your email server, and the size of the disk cache should increase automatically:P"
Yeah, well, some people don't want to loose the mail after retrieving it from the SMTP server. Then again, you would need to put the flash replacement drive in a RAID configuration as well to be completely sure. Besides that, you would need a 64-bit CPU and application to use more than 2 GB of RAM for some systems. But that's probably why the ":P" is there.
Maybe in future versions, the flash will be part of the design of the laptop. I'm pretty sure it will. The only reason to put flash on an IDE interface, PATA or SATA is to accomodate for older laptops. If the BIOS support starting from flash, there is absolutely no reason to put an expensive and sluggish interface in between.
Before you get one of those laptops, put your OS, applications and documents on the flash, and all the media on a 1.8" or 2.5" HDD drive on a (well-powered) USB port. If you use development on a laptop, with meriads of applications and documents, you'll probably well off with a desktop replacement anyway, and lots of those have room for an additional drive already.
And how were you planning to do that? Having the OS authenticate itself? No, on a common PC, the OS is the factor that should protect you. The drive is just what it always was, a drive.
Flash has a maximum number of *write* cycles. Not read/write cycles. The trick is to only store the data that does not change readily (hah, or should that be "writely"?), which I think was what MS had in mind. Of course, using the hdd as backup and a checksum over the sectors would be a good idea none-the-less, if only as protection against malicious programs re-writing the entire contents of the drive until it fails. A bit of paranoia is needed if you consider security.
Don't forget spinup time. Spinup time is pretty important, especially for notebooks. Notebooks are bound to suspend the disk a lot more than desktop drives, so to safe power.
This one "http://www.thinfilm.se/default.aspx" has been around much longer and has much better graphics. Actually, the contents have not changed since I first looked at it...so many years ago that I cannot even find the article from the register anymore. It's too long ago. Also, Intel used to invest in this technology. Just to give you an idea on how much vaporware this could be.
That's a simple and great idea, even though it is not original by far. But with the current memory prices, it's probably pretty feasable for many systems. But why: did they not put a fast (fake) S-ATA 300 controller on it - or even just a simple bootable Windows storage device with driver on CD? Why isn't it available for the superior (16x) PCI-x interfaces? And last of all, unbuffered non-ECC memory? Count me out for any trusted systems as well. Good idea, *horrible* implementation.
Bah, they can always build in a "press F8 to skip hardware detection", or have a few interesting options for this in the control panel, for the next boot. Skip legacy detection would be a favourite. USB/PCI detection should be very quick anyway, due to the identification strings on the bus. But Windows does not even let you *view* the PCI strings. Instead, it shows you you've got an "unknown device" in the device manager. Talk about stupid...
The developers won't make those decissions: you'll have to pay through your nose to get on that list of trusted CA's. So by definition, the first station is not the developer. And once the payment is there, do you think they can say: "well, you've payed, but you won't get on the list"? Nope, that is indeed a pretty serious limitation of the current technology.
At least there should be different levels of trust. At the moment, you can choose different kind of levels of certificates, but the browsers will accept anyone of them - without displaying that trust level. So what's the point?
At a former job, we had a complete conversation on voicemail between two guys of a supporting company (technical support). It went like "... so how do end this voicemail?" "I think you should just hang up." "No that could not be right." "Yes, sure *message broken off*. I am *not* kidding. So hey, it could be worst. Another supporting company in the Netherlands actually called us up, *after* having gone through their senior technician, to as what the difference was between "enabled" and "disabled". I more or less explained that they were "disabled".
What they really were saying was: "I am not a magician" because that's what mathematicians are to them. These people have left the schools wondering what happened, and were never told *how to think*. And now they are too afraid to get it wrong to actually even try it out. What they fall back on is something called common sense. They get it right most of the time by because their brain sub-consiously translates it for them, but that's not like *consiously* doing the math. That's why all this reasoning for 20 minutes was not going to work. He could as easily have spent the rest of his time talking to a wall. Anyway, that's my opinion, only based on experience.
First of all, don't forget that the inexpensive USB-flash cards that are sold are mostly 6 MB read / 3MB write devices, while the more expensive ones are about 25/20 MB respectively. If you want to fill 1/2 GB on a regular basis, I would opt for the faster one (the so called 100X-150X speeds).
The quality of the card readers does not make any difference, it's the flash that stores the backup. The readers can be bought anywhere for about 10 dollars, so what's the problem?
As said, USB-drives are just another backup medium. Since they are random read/write, I would use any backup application that can handle removable disks. I would not care too much about the limited number of writes, since it is very unlikely that this would matter if you back up only dayly or weekly.
Slashdot Mirror
This must be the truth; it's almost always an Anonymous Coward that does the killing.
"I was evaluating visual studio .net again this week and whilst it might technically do everything it needs to, its slower than VS 6 at most things."
Yes, well, it parses and creates a tree in the background while you are typing. You could switch that off to start with. Working with VS 6 feels like using the Kawa development tools for Java (may it rest in pieces). You get a lot of functionality, a better plugin structure and lots more. They did OK with the last version in my opinion, even if it is still *way* behind Eclipse for the language features.
For this to work, there is a need to set a common API for the hardware acceleration. Otherwise, everybody has to do the same thing over and over again, or just ignore the instructions altogether.
Take for instance the C3 processor from VIA. The latest already do SHA, AES, RSA and hardware random generation. Will AMD use similar instruction sets, or will they use completely different instructions or even processors? How am I, as a programmer, going to use these instructions? These instructions should also be very well specified. For instance, the current SHA implementation of VIA doesn't allow streaming (from Java at least), you first have to put everything ready in RAM. That's not a good thing (TM), and could have been avoided if they left the calculations to the last block to be hashed out of the hardware.
Another thing is software support. Buying a C3 will do nothing to your existing applications, unless they use a plugin structure for software. E.g. in Java you can put a specialized VIA "cryptographic provider" in front of all the other providers, and the C3 will be automatically used. It can also be used within Windows using the CSP software library from VIA. But from NSS, the Netscape Security Suite also used by Firefox, you are still using the normal CPU instructions. For things like compression, which is almost always directly programmed in software, it will be much harder to use the additional instructions. Applications, in other words, should not only use threads, they should also be pretty modular and plugable (the next software revolution I hope).
Although I am very much in favour of speciallized instructions, I think a good support structure would be very much needed. Fortunately, AMD (or VIA for that matter) does not have a very bad track record on standardization. Or Open Source for that matter.
"Are current CPUs optimised for physics simulations? No."
No, but some add on cards, and in not too long a time, video processors may have physics simulators. That does not really
"For image processing? No."
True. If you don't count the various multi-media instructions.
"For data compression? No."
True.
"For encryption? No."
Except the Sun T1 (aka Niagara) and VIA C3 processors that is. Especially the C3 (in the Epia range of Mini-ITX motherboards) do SHA hashing operations, AES operations, faster RSA operations and to top it all off have an extremely usefull random generator.
Maybe, but being seen as a monopoly, they should remove this restriction none-the-less if they plan to sell this within Europe. If I as a user think that a driver should install, it should install. That's all there is to it.
It's only the most well known Belgium statue (I actually wouldn't know any others anyway):
http://en.wikipedia.org/wiki/Manneken_Pis
Yeah, lack of money most probably. Porn sells.
Yeah, and those law inforcement agencies have clamped down on the drugs problem in the same way as well. Like *that* is working. Wishfull thinking goes both ways, and neither one may even be correct in the real world. It need not be black and white - as some floundering sons of founding fathers are finding out at this time. Pfff.
"So if someone is fired from their job for being addicted to drugs or alcohol, for being obese, for watching porn or chatting online at the office, then I think it's perfectly acceptable. I say this even though I am guilty of two of these offenses myself. I'm not willing to defend my own behavior by calling it a disease."
I think it is pretty damn brave that you come out on slashdot to admit that you are on drugs and watch porn at the office.
I would not know about that. 5 years ago, a fast drive would do a seek in about 12 ms. Now it is 9 ms. That's not impressive. Density and streaming speed has increased manyfold within that time frame. Seek time has not. And it'll be some time before the random access speed of flash is met by spinning disks (and more importantly maybe, the drive heads). Especially if you read the article carefully, and see that the 4x speed up mentioned is the overall speed speedup.
From the faq:
Q: How fast is your current SSD and what performance improvements does it offer?
A: The streaming R/W speeds are 57 MB/s and 32 MB/s, respectively, but the most significant performance advantage comes from its latency feature - less than 1 millisecond; roughly 10-15x faster than a hard disk drive.
"Just add more ram to your email server, and the size of the disk cache should increase automatically :P"
Yeah, well, some people don't want to loose the mail after retrieving it from the SMTP server. Then again, you would need to put the flash replacement drive in a RAID configuration as well to be completely sure. Besides that, you would need a 64-bit CPU and application to use more than 2 GB of RAM for some systems. But that's probably why the ":P" is there.
Maybe in future versions, the flash will be part of the design of the laptop. I'm pretty sure it will. The only reason to put flash on an IDE interface, PATA or SATA is to accomodate for older laptops. If the BIOS support starting from flash, there is absolutely no reason to put an expensive and sluggish interface in between.
Before you get one of those laptops, put your OS, applications and documents on the flash, and all the media on a 1.8" or 2.5" HDD drive on a (well-powered) USB port. If you use development on a laptop, with meriads of applications and documents, you'll probably well off with a desktop replacement anyway, and lots of those have room for an additional drive already.
And how were you planning to do that? Having the OS authenticate itself? No, on a common PC, the OS is the factor that should protect you. The drive is just what it always was, a drive.
Flash has a maximum number of *write* cycles. Not read/write cycles. The trick is to only store the data that does not change readily (hah, or should that be "writely"?), which I think was what MS had in mind. Of course, using the hdd as backup and a checksum over the sectors would be a good idea none-the-less, if only as protection against malicious programs re-writing the entire contents of the drive until it fails. A bit of paranoia is needed if you consider security.
Don't forget spinup time. Spinup time is pretty important, especially for notebooks. Notebooks are bound to suspend the disk a lot more than desktop drives, so to safe power.
Hah, you'll first have to get that name off of her, mr nerdy nerd :)
This one "http://www.thinfilm.se/default.aspx" has been around much longer and has much better graphics. Actually, the contents have not changed since I first looked at it...so many years ago that I cannot even find the article from the register anymore. It's too long ago. Also, Intel used to invest in this technology. Just to give you an idea on how much vaporware this could be.
That's a simple and great idea, even though it is not original by far. But with the current memory prices, it's probably pretty feasable for many systems. But why: did they not put a fast (fake) S-ATA 300 controller on it - or even just a simple bootable Windows storage device with driver on CD? Why isn't it available for the superior (16x) PCI-x interfaces? And last of all, unbuffered non-ECC memory? Count me out for any trusted systems as well. Good idea, *horrible* implementation.
Bah, they can always build in a "press F8 to skip hardware detection", or have a few interesting options for this in the control panel, for the next boot. Skip legacy detection would be a favourite. USB/PCI detection should be very quick anyway, due to the identification strings on the bus. But Windows does not even let you *view* the PCI strings. Instead, it shows you you've got an "unknown device" in the device manager. Talk about stupid...
The developers won't make those decissions: you'll have to pay through your nose to get on that list of trusted CA's. So by definition, the first station is not the developer. And once the payment is there, do you think they can say: "well, you've payed, but you won't get on the list"? Nope, that is indeed a pretty serious limitation of the current technology.
At least there should be different levels of trust. At the moment, you can choose different kind of levels of certificates, but the browsers will accept anyone of them - without displaying that trust level. So what's the point?
At a former job, we had a complete conversation on voicemail between two guys of a supporting company (technical support). It went like "... so how do end this voicemail?" "I think you should just hang up." "No that could not be right." "Yes, sure *message broken off*. I am *not* kidding. So hey, it could be worst. Another supporting company in the Netherlands actually called us up, *after* having gone through their senior technician, to as what the difference was between "enabled" and "disabled". I more or less explained that they were "disabled".
What they really were saying was: "I am not a magician" because that's what mathematicians are to them. These people have left the schools wondering what happened, and were never told *how to think*. And now they are too afraid to get it wrong to actually even try it out. What they fall back on is something called common sense. They get it right most of the time by because their brain sub-consiously translates it for them, but that's not like *consiously* doing the math. That's why all this reasoning for 20 minutes was not going to work. He could as easily have spent the rest of his time talking to a wall. Anyway, that's my opinion, only based on experience.
It's a bit sad that you are explaining this same thing now here on slashdot. The question is: are you prepared to do up to 20 minutes of this?
Yup, about 6.83 dollars the gallon at the moment (hi, welcome in the Netherlands).
First of all, don't forget that the inexpensive USB-flash cards that are sold are mostly 6 MB read / 3MB write devices, while the more expensive ones are about 25/20 MB respectively. If you want to fill 1/2 GB on a regular basis, I would opt for the faster one (the so called 100X-150X speeds).
The quality of the card readers does not make any difference, it's the flash that stores the backup. The readers can be bought anywhere for about 10 dollars, so what's the problem?
As said, USB-drives are just another backup medium. Since they are random read/write, I would use any backup application that can handle removable disks. I would not care too much about the limited number of writes, since it is very unlikely that this would matter if you back up only dayly or weekly.