HD4770 dates back to 2008. It's a 5 year old card. 5 years is an eternity in the IT industry. All the driver updates in the world aren't going to help that.
Interesting. However, how "reusable" is it? I'm guessing it's not very which will just add to the expense, meaning that ultimately this is a pretty niche product.
My gaming machine is actually 5 years old - built in 2008, it originally housed a Geforce 9800GTX. The CPU is an intel Quad core Q9300 - pretty low end at the time, plus 4GB of DDR2 RAM - very old, very out of date.
On that machine, I could play the likes of BF3 on low settings reasonably well. I swapped the graphics card for a Geforce 560 Ti and now I can play BF3 on med/high at 1920x1200. Nothing else has changed, same old CPU, same DDR2 RAM.
To be fair, people said the same things about 720p versus 1080p - that you couldn't see the difference, that it was negligible, etc. Hell, people had the same arguments about HD in general - DVD was plenty good for most uses and HD was actually slow to catch on:
DVD sales represent more than half of the revenue studios generate from most of their movies. But those sales are expected to grow just 2 percent this year, a far cry from the double-digit growth the industry enjoyed just two years ago. High-definition DVD's were supposed to pick up the slack, but technical delays and a thorny format war between camps led by Sony and Toshiba have dampened expectations.
So I would expect 4K to have a similar uphill battle in this regard but eventually it'll just be more economical to switch to 4k-only panel production for the major manufacturers. By 2020, we'll probably be on 4K as standard.
Exactly, it's just early days and all this doom and gloom about the format is ridiculous. I remember when 1080p video started appearing and I tried playing a sample on my AMD 3400+ processor - of course, it skipped and jumped all over the place. The file was also huge, my paltry 2Mbit internet connection took ages downloading it and my monitor was too small to display it properly. It was excessive, took up a lot of space and required fast, new hardware. How could such a thing ever catch on?!
Oh easy, this is technology and technology constantly moves forward. If you're the kind of person that has ever complained about having a quad-core processor in your phone as "unnecessary", then please hand in your geek card and get off slashdot. Technology always moves forward, things always get better and nothing will stop that. There is never a "Good enough", things can always be done faster or with less power.
Whelp, I've went and made a bit of an ass of myself. The actual bandwidth requirements for uncompressed 1080p (24FPS) is 95MB/s, not 190MB/s as I originally stated. The 190MB/s figure was for the RGB 4:4:4 format. The rest of my point still stands, though.
I'm calling bullshit. The summary talks about uncompressed video, glancing over the fact that even 1080p uncompressed requires a bitrate of 190 MBytes/s (at 24FPS) - faster than most HDD's can handle. Storage space required? 667Gb per HOUR, or a solid TB for a 90min film. Do you need a massive SSD to play 1080p files? Do you even need an SSD to edit and encode them? No, you don't.
Compression has always been essential, even for DVD. Uncompressed SD videos would still fill a dual-layer blu-ray (50GB) after about 30mins. Yes, we'll need better CODECS to handle 4k and yes a lot of work needs to be done, but the size of uncompressed video isn't and never has been the issue. At worst, it'll take a slightly newer disk format (I don't see why a 4-layer Blu-ray disk - which exists today, couldn't do the job) and better internet connections to stream.
Devices will be released THIS YEAR capable of outputting 4k - just look at Tegra 4, or nVidia's SHIELD, which has been demonstrated live supplying a 4K TV (Admittedly, probably up-scaled content, but obviously the technology is there today).
The real issue is content, in that nothing is really in 4K right now. The transportation and storage method is not.
Errr....no, I don't think you understand. The SDK has nothing to do with the Android OS other than it's use to develop apps that run on it. There's nothing to "take" from the SDK that you could put into a ROM. The only thing the SDK has from the AOSP codebase is the Android Virtual Machine, however this is still based on the same AOSP code that gets released. Very occasionally the SDK gets updated and released before the AOSP code does, but ROMs like Cyangoen don't bother with the AVMs as they're useless on real devices. The best you can do is look at the undocumented API's within the SDK but this doesn't affect ROM development, rather it affects app development.
No they don't. They just state that you can't take Google's SDK, change some bits and release it as your own "SDK++!". This more affects anything that tries to bundle the SDK (May affect something like Unity? I don't know how that compiles Android apps).
I've already commented so I can't, but this post needs more mod points - people are missing the difference between the SDK and the Android source code. They're not related.
Yeah. SDK. Not the AOSP code. The SDK has nothing to do with Cyanogen or any other custom ROM, they're based off of the same code, code that's DESIGNED to work with what the SDK produces. This does not affect CM.
Samsung already has their own mobile OS, though: Bada. That's mostly why this move is being questioned as a large "..er?". That and there's actually several existing OS's that are free and open source, so why make yet another one? There has to be more benefit than just "It's not Android" for it to be worthwhile. I don't think anyone thinks Samsung is stupid, I think it's more a case of we don't seem to know a lot about Tizen and what makes it so special.
Skype in itself is incredibly dodgy at the moment. There's an annoying bug with current clients whereby messages can arrive in the wrong order. This causes you to not get notified about new messages on Android and even on Windows the messages appear half way up the conversation.
The solution? Install an older version of the client. Been having this "bug" for weeks now.
The A6 is said to use a 1.3 GHz custom Apple-designed ARMv7 based dual-core CPU, called Swift, rather than a licensed CPU from ARM like in previous designs
Sure, you can argue that the A6 is based on previous designs that go all the way back to Samsung+Intrinsity, but the point is the CURRENT chips are Apple's design, from an in-house Apple team that's just as happy to work with TSMC as it is to work with Samsung.
I don't think anyone's debating TSMC's track record or competency here. Like I said, I'm sure apple is well aware of all this and will have a contract that's favourable to them.
I don't think this move is just to screw over Samsung (although that's no doubt a happy coincidence for apple). I think someone at Apple has realised that Samsung could decide not to renew the contract and just as easily screw over Apple. As someone else has already pointed out, Apple is still keeping Samsung as a manufacturer in the meantime so even if TSMC does fuck up horribly, Apple won't be in too much trouble.
There's nothing wrong with ensuring you have more than one supplier for a critical component, especially one that only a handful of companies can produce.
This is slightly different. The same chip is being produced (it's Apple's design), it's just a different manufacturer. No doubt Apple will be paying per chip and not per wafer, so if anything does fuck up it'll be on TSMC's head. Plus it's not like TSMC doesn't know a thing or two about producing chips.
Slashdot Mirror
"It obviously gets far worse" is referring to "how bad people are at estimating", not the lifespan of the Flash Memory.
Had an SSD in my laptop for just over a year and a half now, no issues what so ever. Daily use as well.
I'm not sure if people like Stallman are helping the scene at all.
Easy, because it's bashing Windows and this is Slashdot.
HD4770 dates back to 2008. It's a 5 year old card. 5 years is an eternity in the IT industry. All the driver updates in the world aren't going to help that.
Interesting. However, how "reusable" is it? I'm guessing it's not very which will just add to the expense, meaning that ultimately this is a pretty niche product.
My gaming machine is actually 5 years old - built in 2008, it originally housed a Geforce 9800GTX. The CPU is an intel Quad core Q9300 - pretty low end at the time, plus 4GB of DDR2 RAM - very old, very out of date.
On that machine, I could play the likes of BF3 on low settings reasonably well. I swapped the graphics card for a Geforce 560 Ti and now I can play BF3 on med/high at 1920x1200. Nothing else has changed, same old CPU, same DDR2 RAM.
To be fair, people said the same things about 720p versus 1080p - that you couldn't see the difference, that it was negligible, etc.
Hell, people had the same arguments about HD in general - DVD was plenty good for most uses and HD was actually slow to catch on:
http://www.nytimes.com/2006/06/13/technology/13disc.html?pagewanted=all&_r=0
So I would expect 4K to have a similar uphill battle in this regard but eventually it'll just be more economical to switch to 4k-only panel production for the major manufacturers. By 2020, we'll probably be on 4K as standard.
Exactly, it's just early days and all this doom and gloom about the format is ridiculous. I remember when 1080p video started appearing and I tried playing a sample on my AMD 3400+ processor - of course, it skipped and jumped all over the place. The file was also huge, my paltry 2Mbit internet connection took ages downloading it and my monitor was too small to display it properly. It was excessive, took up a lot of space and required fast, new hardware. How could such a thing ever catch on?!
Oh easy, this is technology and technology constantly moves forward. If you're the kind of person that has ever complained about having a quad-core processor in your phone as "unnecessary", then please hand in your geek card and get off slashdot. Technology always moves forward, things always get better and nothing will stop that. There is never a "Good enough", things can always be done faster or with less power.
Whelp, I've went and made a bit of an ass of myself. The actual bandwidth requirements for uncompressed 1080p (24FPS) is 95MB/s, not 190MB/s as I originally stated. The 190MB/s figure was for the RGB 4:4:4 format. The rest of my point still stands, though.
I'm calling bullshit. The summary talks about uncompressed video, glancing over the fact that even 1080p uncompressed requires a bitrate of 190 MBytes/s (at 24FPS) - faster than most HDD's can handle. Storage space required? 667Gb per HOUR, or a solid TB for a 90min film. Do you need a massive SSD to play 1080p files? Do you even need an SSD to edit and encode them? No, you don't.
Source: http://en.wikipedia.org/wiki/Uncompressed_video#1080i_and_1080p_HDTV_RGB_.284:4:4.29_uncompressed
Compression has always been essential, even for DVD. Uncompressed SD videos would still fill a dual-layer blu-ray (50GB) after about 30mins. Yes, we'll need better CODECS to handle 4k and yes a lot of work needs to be done, but the size of uncompressed video isn't and never has been the issue. At worst, it'll take a slightly newer disk format (I don't see why a 4-layer Blu-ray disk - which exists today, couldn't do the job) and better internet connections to stream.
Devices will be released THIS YEAR capable of outputting 4k - just look at Tegra 4, or nVidia's SHIELD, which has been demonstrated live supplying a 4K TV (Admittedly, probably up-scaled content, but obviously the technology is there today).
The real issue is content, in that nothing is really in 4K right now. The transportation and storage method is not.
Errr....no, I don't think you understand. The SDK has nothing to do with the Android OS other than it's use to develop apps that run on it. There's nothing to "take" from the SDK that you could put into a ROM. The only thing the SDK has from the AOSP codebase is the Android Virtual Machine, however this is still based on the same AOSP code that gets released. Very occasionally the SDK gets updated and released before the AOSP code does, but ROMs like Cyangoen don't bother with the AVMs as they're useless on real devices.
The best you can do is look at the undocumented API's within the SDK but this doesn't affect ROM development, rather it affects app development.
Ahhhhh yes, that does actually make more sense. "Auto" apparently means "6" in most routers.
You adjusted the power before changing the channels? That's like...wireless 101! (However I may be biased, having worked for an ISP).
No they don't. They just state that you can't take Google's SDK, change some bits and release it as your own "SDK++!". This more affects anything that tries to bundle the SDK (May affect something like Unity? I don't know how that compiles Android apps).
I've already commented so I can't, but this post needs more mod points - people are missing the difference between the SDK and the Android source code. They're not related.
Yeah. SDK. Not the AOSP code. The SDK has nothing to do with Cyanogen or any other custom ROM, they're based off of the same code, code that's DESIGNED to work with what the SDK produces.
This does not affect CM.
I'm pretty sure that depends on your definition of "Evil".
Samsung already has their own mobile OS, though: Bada. That's mostly why this move is being questioned as a large "..er?". That and there's actually several existing OS's that are free and open source, so why make yet another one? There has to be more benefit than just "It's not Android" for it to be worthwhile. I don't think anyone thinks Samsung is stupid, I think it's more a case of we don't seem to know a lot about Tizen and what makes it so special.
Skype in itself is incredibly dodgy at the moment. There's an annoying bug with current clients whereby messages can arrive in the wrong order. This causes you to not get notified about new messages on Android and even on Windows the messages appear half way up the conversation.
The solution? Install an older version of the client. Been having this "bug" for weeks now.
I notice that you've linked specifically to the 2 year old A4, not the currently-used A6:
http://en.wikipedia.org/wiki/Apple_A6
Sure, you can argue that the A6 is based on previous designs that go all the way back to Samsung+Intrinsity, but the point is the CURRENT chips are Apple's design, from an in-house Apple team that's just as happy to work with TSMC as it is to work with Samsung.
I don't think anyone's debating TSMC's track record or competency here. Like I said, I'm sure apple is well aware of all this and will have a contract that's favourable to them.
I don't think this move is just to screw over Samsung (although that's no doubt a happy coincidence for apple). I think someone at Apple has realised that Samsung could decide not to renew the contract and just as easily screw over Apple. As someone else has already pointed out, Apple is still keeping Samsung as a manufacturer in the meantime so even if TSMC does fuck up horribly, Apple won't be in too much trouble.
There's nothing wrong with ensuring you have more than one supplier for a critical component, especially one that only a handful of companies can produce.
You posted the same thing twice. Well done on being a competent internet user.
This is slightly different. The same chip is being produced (it's Apple's design), it's just a different manufacturer. No doubt Apple will be paying per chip and not per wafer, so if anything does fuck up it'll be on TSMC's head. Plus it's not like TSMC doesn't know a thing or two about producing chips.