I think you're right, and I would guess that the startup costs are much cheaper with this technology. But I wonder how useful it is when we can already print thousands transistors for pennies after the initial cost of a fab. Maybe it will allow for easier tinkering for people sitting in their garage? Would be pretty cool to build your own diode.
I think the 5830 is actually just a little better than the 5770 in real world performance but uses a significantly more power. I think the 4870 is actually also pretty close to the 5770 in performance, might actually be faster in fact. But the 5770 is based on the 5800 series core I think, just a little cut down.
Heck, now you can get e-ink ebook readers for $150. The Sony Pocket Reader is discounted to that price in a lot of places, and the Kobo reader apparently also is at that price too. I just got a Pocket Reader off woot and it's pretty awesome so far, besides the small screen, but I think any ebook reader that's less than a full size letter paper is going to have problems with viewing PDFs without reflowing. As a note, my Pocket Reader actually reflows the text in a PDF pretty well for the most part, but it just involves a lot of page turning (4ish page turns for each real page). I actually am finding that while most devices are moving towards convergence, dedicated ebook readers are actually pretty nice. I agree, I'm not sure who this is targetted at, especially since it doesn't have the brand appeal of someone like Apple. Maybe if Apple made something like this, people would buy it for $600-800, but anyone else would be SOL (Apple would probably also make it more usable, RDF notwithstanding, but that's another story).
Well, that is true. But depending on who you talk to, some people will tell you that it's not unlikely that we'll find a way around the wall when the time comes. I myself don't know much about process technology, but one of my electronics professors would occasionally mention that we've thought that there were dead ends in the technology before, until someone made a breakthrough discovery and found a way around the roadblocks. However, we are indeed getting to a point where we're limited by the size of the atoms themselves, which is problematic. But the future is exciting and unknown, so who knows what will happen next.
Actually, since semiconductor manufacturing gets cheaper practically exponentially (yay Moore's law) eventually SSD prices should catch up and undercut mechanical disk prices, just because of the manufacturing process. But that's a long ways off yet.
Hmm, I think Flash usually zooms too. Sometimes I use that as a hackish replacement for fullscreen on my second monitor because flash exits full screen if it doesn't have focus.
Oh right, within tolerances. But you know what I mean. When you ask for a 1k resistor within 5%, you don't want something within 5% of 1024, you want it within 5% of 1000.
Yes, that makes sense; that's why the prefix should be different to distinguish that we're not talking about math, physics, etc. By the way, prefixes are very important in electrical engineering, and 1 kilo ohm is definitely not 1024 ohms, so your point about kilo's meaning in computer engineering is kind of invalid. Using kilo to mean 1024 is just an abuse of notation. It's kind of silly to assume that you can hijack prefixes to mean whatever you want, and you're kind of arguing against yourself when you talk about things like 10 bits in a byte - of course we don't have 10 bits in a byte, but then we don't call it a decabit, we call it a byte. Of course, all that being said, kibibyte and mebibyte are really awkward to say so I do say kilobyte and megabyte because that's what the OS has always said and that's what people understand, but you have to admit that it is imprecise to just go around hijacking prefixes to mean something different.
Eh, Avast always had the registration, 5.0 just made it easier to do. Although I guess the only reason I use Avast instead of MSE is momentum; I've been using it for several years and it just has an untangible familiarity (silly, I know). Otherwise I would have switched to MSE a long time ago. On a related note, it would be nice if MS would allow third parties to tap into Windows Update and make it more like Linux package managers.
I don't own a smartphone, but I'm about to buy one and I agree for the same reasons. There's no way I'm going to be running emacs over ssh on an on-screen keyboard, it's just not going to work. I'd like to be able to type without covering up half my screen, thanks.
Duh, I know what instruction sets are. I don't expect MIPS and x86 to have the same operand order, but I'm just surprised that there are assemblers that use different operand order. I guess the Intel style is closer to what I expect from MIPS I guess.
So, we only learned MIPS in school. Why does x86 have different syntaxes? Do different assemblers just have different psuedoinstructions or something? That seems a bit kludgy to me...
No problem. It's a far cry from being able to run normal Linux userland programs on Android (which may be possible if you have root or something? at the least you can apparently install Debian in a chroot jail. I dunno, don't have an Android device yet) but it is a nice and necessary addition to Android.
The funniest part is that much of the 2xx series cards are just renamed 9000 series cards, and much of those are renamed or die-shrunk 8000 series cards. That said, Charlie Demerjian is hardly an unbiased source of reporting on nVidia, although I think he does have good reasons for his "grudge."
I guess the 5670 has lower power consumption than the cards it replaces. As an owner of a Radeon 2900, I can appreciate what that would be useful for, like not sounding like a vacuum cleaner while playing games. But I don't know whether I'd want to upgrade to the 5670, since as you said it seems to not be significantly faster than a 3870, but it costs a fair bit more.
Nah, you can set minimum and maximum clock speeds in Windows without difficulty, and third party apps like RMClock give you a lot of fine control including undervolting. This is really nothing new except for dynamic GPU switching.
Slashdot Mirror
What about functional programming then? Java kind of blows if you want to use that programming paradigm.
I think you're right, and I would guess that the startup costs are much cheaper with this technology. But I wonder how useful it is when we can already print thousands transistors for pennies after the initial cost of a fab. Maybe it will allow for easier tinkering for people sitting in their garage? Would be pretty cool to build your own diode.
I think the 5830 is actually just a little better than the 5770 in real world performance but uses a significantly more power. I think the 4870 is actually also pretty close to the 5770 in performance, might actually be faster in fact. But the 5770 is based on the 5800 series core I think, just a little cut down.
I don't know if you can do it from Explorer, but you can create symlinks from the command line, at least in Vista and 7. https://secure.wikimedia.org/wikipedia/en/wiki/NTFS_symbolic_link
So does NTFS.
Oh, didn't realize that, that makes my post invalid.
Also, it's a capacitive screen and they didn't mention an active digitizer so I don't know how well scribbling would work anyway.
Heck, now you can get e-ink ebook readers for $150. The Sony Pocket Reader is discounted to that price in a lot of places, and the Kobo reader apparently also is at that price too. I just got a Pocket Reader off woot and it's pretty awesome so far, besides the small screen, but I think any ebook reader that's less than a full size letter paper is going to have problems with viewing PDFs without reflowing. As a note, my Pocket Reader actually reflows the text in a PDF pretty well for the most part, but it just involves a lot of page turning (4ish page turns for each real page). I actually am finding that while most devices are moving towards convergence, dedicated ebook readers are actually pretty nice. I agree, I'm not sure who this is targetted at, especially since it doesn't have the brand appeal of someone like Apple. Maybe if Apple made something like this, people would buy it for $600-800, but anyone else would be SOL (Apple would probably also make it more usable, RDF notwithstanding, but that's another story).
Well, that is true. But depending on who you talk to, some people will tell you that it's not unlikely that we'll find a way around the wall when the time comes. I myself don't know much about process technology, but one of my electronics professors would occasionally mention that we've thought that there were dead ends in the technology before, until someone made a breakthrough discovery and found a way around the roadblocks. However, we are indeed getting to a point where we're limited by the size of the atoms themselves, which is problematic. But the future is exciting and unknown, so who knows what will happen next.
Actually, since semiconductor manufacturing gets cheaper practically exponentially (yay Moore's law) eventually SSD prices should catch up and undercut mechanical disk prices, just because of the manufacturing process. But that's a long ways off yet.
Hmm, I think Flash usually zooms too. Sometimes I use that as a hackish replacement for fullscreen on my second monitor because flash exits full screen if it doesn't have focus.
Probably because his name is Ma Jae Yoon?
Yeah, sure, that is true. I was more speaking in ideal terms. A 1k resistor within 5% is within 5% of 1000, not within 5% of 1024.
Oh right, within tolerances. But you know what I mean. When you ask for a 1k resistor within 5%, you don't want something within 5% of 1024, you want it within 5% of 1000.
Yes, that makes sense; that's why the prefix should be different to distinguish that we're not talking about math, physics, etc. By the way, prefixes are very important in electrical engineering, and 1 kilo ohm is definitely not 1024 ohms, so your point about kilo's meaning in computer engineering is kind of invalid. Using kilo to mean 1024 is just an abuse of notation. It's kind of silly to assume that you can hijack prefixes to mean whatever you want, and you're kind of arguing against yourself when you talk about things like 10 bits in a byte - of course we don't have 10 bits in a byte, but then we don't call it a decabit, we call it a byte. Of course, all that being said, kibibyte and mebibyte are really awkward to say so I do say kilobyte and megabyte because that's what the OS has always said and that's what people understand, but you have to admit that it is imprecise to just go around hijacking prefixes to mean something different.
Eh, Avast always had the registration, 5.0 just made it easier to do. Although I guess the only reason I use Avast instead of MSE is momentum; I've been using it for several years and it just has an untangible familiarity (silly, I know). Otherwise I would have switched to MSE a long time ago. On a related note, it would be nice if MS would allow third parties to tap into Windows Update and make it more like Linux package managers.
I don't own a smartphone, but I'm about to buy one and I agree for the same reasons. There's no way I'm going to be running emacs over ssh on an on-screen keyboard, it's just not going to work. I'd like to be able to type without covering up half my screen, thanks.
Duh, I know what instruction sets are. I don't expect MIPS and x86 to have the same operand order, but I'm just surprised that there are assemblers that use different operand order. I guess the Intel style is closer to what I expect from MIPS I guess.
So, we only learned MIPS in school. Why does x86 have different syntaxes? Do different assemblers just have different psuedoinstructions or something? That seems a bit kludgy to me...
No problem. It's a far cry from being able to run normal Linux userland programs on Android (which may be possible if you have root or something? at the least you can apparently install Debian in a chroot jail. I dunno, don't have an Android device yet) but it is a nice and necessary addition to Android.
Actually, they somewhat recently released a Native Development Kit which allows JNI style access to native code.
The funniest part is that much of the 2xx series cards are just renamed 9000 series cards, and much of those are renamed or die-shrunk 8000 series cards. That said, Charlie Demerjian is hardly an unbiased source of reporting on nVidia, although I think he does have good reasons for his "grudge."
I guess the 5670 has lower power consumption than the cards it replaces. As an owner of a Radeon 2900, I can appreciate what that would be useful for, like not sounding like a vacuum cleaner while playing games. But I don't know whether I'd want to upgrade to the 5670, since as you said it seems to not be significantly faster than a 3870, but it costs a fair bit more.
Not quite accurate. While GPGPU != CPU, there are things that GPGPUs can do far better than CPUs, and those things are more common than you'd think.
I agree completely, for example, video encoding is pretty common these days and can be GPU accelerated for massive gains in speed.
Nah, you can set minimum and maximum clock speeds in Windows without difficulty, and third party apps like RMClock give you a lot of fine control including undervolting. This is really nothing new except for dynamic GPU switching.