I'd like a display with the readability of paper. This means at least 300 dpi, preferably closer to 600, so you wouldn't need the blurring we fancily call anti-aliasing. I won't even start with the lighting/contrast issues...
If you think of RAM as mostly a mechanism to hide the latency of the disk then, in the not so distant future, it could become redundant (and even a performance bottleneck).
I also think of RAM as a way to avoid wearing out SSDs.
USB 3.1 and the USB-C connector fixed a lot of these problems which basically came down to, IMHO, reinventing FireWire.
So in essence, the current spec is Firewire + USB 2.0 bolted together, because it still needs the old set of pins to stay compatible. It's "universal" because when you glue enough different physical standards together, there's a chance that one of them will fit.
Slashdot isn't really social media. It's more an anti-social network.
Facebook is called "social media". I don't think Slashdot is any less social than that. In fact, it's even better because it doesn't pretend to be anything "social". Having "friends" here is a bit of a stretch, though.
It's a distributed trust network, right? Why would banks that survive on trust want that distributed?
Apparently, they don't. Each institution is building its own, private blockchain to stay buzzword compliant, not because it makes any technical sense.
It's like hearing about the Internet for the first time, and proceeding to build a private, closed version -- which really happened several times, but eventually people realized that the whole point about the Internet is not being private or closed.
Monero (http://getmonero.org) is the most private existing cryptocurrency.
Not quite. It's one of the most private ones.
It is similar to Bitcoin but based on an entirely distinct codebase (CryptoNote).
This should already ring a bell -- it's based on something else. There are a few other CryptoNote coins, which are no less private. Some of them have technical advantages, like being much faster to update the local blockchain. In addition, there are other privacy-oriented cryptocoins which have nothing to do with CryptoNote.
Monero's advantage is having a large developer/support base. IMHO, Monero is like the Microsoft of privacy-oriented cryptocoins, with lots of money and men in suits behind it. While they can deliver their product, they're not exactly the most exciting tech.
Instead we get a chunk of silicon that can't be used for anything except x265
That is not at all the case, for instance all of Apples advanced math libraries (under the umbrella "Accelerate") will make use of a GPU if one is present and faster than at the CPU. They handle things like linear algebra stuff (linpack) and all sorts of vector operations, including FFT...
This sounds very much like OpenCL or CUDA, or whatever the Applets call when you do general computing on GPU shaders. The video decoding bit is a different ASIC altogether.
That said, it would be nice to do more of these common operations like video decoding in GPGPU software instead of all these different bits for different purposes. I understand that some of the open-source GPU drivers already do some video decoding in the shaders, but that's about it. I guess something like ffmpeg/libva in OpenCL is too much to ask, since many people already have strong enough CPUs for those things, and there are no obvious efficiency gains compared to dedicated decoding hardware.
we get a chunk of silicon that can't be used for anything except x265, which I'm sure will have Windows drivers and be supported in Flash player, and maybe will get experimental Linux support 2 years after.
I watch H.265 videos with MPlayer on Linux, using an Nvidia GPU for the decoding. Admittedly, it's with the closed Nvidia drivers, but apparently the free drivers also handle the video decoder. It's dead easy since the VDPAU interface is fully open -- doing the same on Intel or AMD GPUs needs some extra work, and the results are not as nice. Of course, not all GPUs have the decoders in the first place.
... who laments the loss of the humble 2.5" HDD [or, at the very least, a removable 2.5" SSD] from a laptop or netbook these days?
I understand why companies are moving away from the idea of removable, normalized-size drives [it's cheaper to make the machine when it's just a daughter-board] but the loss of flexibility really chafes.
There are standards for smaller flash drives, but the situation isn't looking quite universal. One thing that bugs me is the smaller version of SATA in 1.8'' drives; the shared connector between 3.5'' and 2.5'' drives was an incredible improvement over the PATA situation.
Generally, I find it dumb when "desktop" and "mobile" components are artificially segregated by the use of different connectors etc., while it's really the same computer tech underneath. I like to use quiet and low-power stuff for my non-mobile needs, basically because I have to pay for the electricity and listen to the fans myself. Also, you'd think it were cheaper to build fewer different models of everything. SATA for 2.5'' and 3.5'' was a brief moment of sanity, and now we're in the same mess again.
One problem is that a lot of FPGA hackers write their functionality in Verilog or VHDL. Those provide a fairly high level syntax that makes it easy enough to design the hardware.
The original chips, however, were designed using schematics.
I'd like to elaborate on this one: if you're using an FPGA to begin with, you don't really have the choice of using anything lower-level. First, the FPGA equivalents of machine language and compilers are usually closed systems, so Verilog and VHDL are often all you can do.
Second, the FPGA hardware itself is not infinitely malleable. You have a limited set of circuit elements and interconnects, and this coarseness is also a limiting factor on clock speeds. When implementing something new, decent hackers will use the FPGA quirks to their advantage, instead of trying to ape the features of hand-drawn circuits. This is of course a problem if you're trying to emulate a given piece of hardware exactly. OTOH, in the case of NES and its few MHz, you can probably make up for it in raw speed and software-like solutions.
FPGA design isn't a magic wand that just makes emulation perfect. You still need to figure out how the chips work internally.
Apparently, you don't need a Cyclone V to implement NES on FPGA. If you're happy with VGA output, a measly Spartan 3 will do:
https://danstrother.com/fpga-n...
7, is just silly, blockchains dont in any way suddenly revolutionise anything - they are one cute solution to one specific problem.
You forgot the usual "it's a pyramid scheme, and besides I'm bitter for not being an early adopter, waaaaaaaa!". Also, what new technology _suddenly_ revolutionizes anything?
Also, don't name your product "android" unless it's an actual humanlike robot. Ditto for "hoverboard" etc. And don't even get me started about "bandwidth".
Slashdot Mirror
I'd like a display with the readability of paper. This means at least 300 dpi, preferably closer to 600, so you wouldn't need the blurring we fancily call anti-aliasing. I won't even start with the lighting/contrast issues...
If you think of RAM as mostly a mechanism to hide the latency of the disk then, in the not so distant future, it could become redundant (and even a performance bottleneck).
I also think of RAM as a way to avoid wearing out SSDs.
USB 3.1 and the USB-C connector fixed a lot of these problems which basically came down to, IMHO, reinventing FireWire.
So in essence, the current spec is Firewire + USB 2.0 bolted together, because it still needs the old set of pins to stay compatible. It's "universal" because when you glue enough different physical standards together, there's a chance that one of them will fit.
Slashdot isn't really social media. It's more an anti-social network.
Facebook is called "social media". I don't think Slashdot is any less social than that. In fact, it's even better because it doesn't pretend to be anything "social". Having "friends" here is a bit of a stretch, though.
It's a distributed trust network, right? Why would banks that survive on trust want that distributed?
Apparently, they don't. Each institution is building its own, private blockchain to stay buzzword compliant, not because it makes any technical sense.
It's like hearing about the Internet for the first time, and proceeding to build a private, closed version -- which really happened several times, but eventually people realized that the whole point about the Internet is not being private or closed.
Sounds a bit like my home stdio.
Except that Linus is the first name, while Putin is the family name. So it should be more like Vlad the Compiler.
All of our phones and digital cameras have a maximum SD card limit, most 64Gb.
SD is limited to 2 GB, SDHC to 32 GB, and SDXC to 2 TB. Aren't standards great?
Monero (http://getmonero.org) is the most private existing cryptocurrency.
Not quite. It's one of the most private ones.
It is similar to Bitcoin but based on an entirely distinct codebase (CryptoNote).
This should already ring a bell -- it's based on something else. There are a few other CryptoNote coins, which are no less private. Some of them have technical advantages, like being much faster to update the local blockchain. In addition, there are other privacy-oriented cryptocoins which have nothing to do with CryptoNote.
Monero's advantage is having a large developer/support base. IMHO, Monero is like the Microsoft of privacy-oriented cryptocoins, with lots of money and men in suits behind it. While they can deliver their product, they're not exactly the most exciting tech.
2 of my home computers could not playback the video. H.265 requires quite a bit more CPU/GPU processing power for both encoding and decoding.
Your GPU probably has a hardware decoder for H.264, but not for H.265.
Instead we get a chunk of silicon that can't be used for anything except x265
That is not at all the case, for instance all of Apples advanced math libraries (under the umbrella "Accelerate") will make use of a GPU if one is present and faster than at the CPU. They handle things like linear algebra stuff (linpack) and all sorts of vector operations, including FFT...
This sounds very much like OpenCL or CUDA, or whatever the Applets call when you do general computing on GPU shaders. The video decoding bit is a different ASIC altogether.
That said, it would be nice to do more of these common operations like video decoding in GPGPU software instead of all these different bits for different purposes. I understand that some of the open-source GPU drivers already do some video decoding in the shaders, but that's about it. I guess something like ffmpeg/libva in OpenCL is too much to ask, since many people already have strong enough CPUs for those things, and there are no obvious efficiency gains compared to dedicated decoding hardware.
we get a chunk of silicon that can't be used for anything except x265, which I'm sure will have Windows drivers and be supported in Flash player, and maybe will get experimental Linux support 2 years after.
I watch H.265 videos with MPlayer on Linux, using an Nvidia GPU for the decoding. Admittedly, it's with the closed Nvidia drivers, but apparently the free drivers also handle the video decoder. It's dead easy since the VDPAU interface is fully open -- doing the same on Intel or AMD GPUs needs some extra work, and the results are not as nice. Of course, not all GPUs have the decoders in the first place.
But when a bank gets robbed (or more likely, when a bank loses money on investments) then every taxpayer loses.
Also, pro tip: don't keep your life savings in an exchange.
So much so, they said it twice!
I understand why companies are moving away from the idea of removable, normalized-size drives [it's cheaper to make the machine when it's just a daughter-board] but the loss of flexibility really chafes.
There are standards for smaller flash drives, but the situation isn't looking quite universal. One thing that bugs me is the smaller version of SATA in 1.8'' drives; the shared connector between 3.5'' and 2.5'' drives was an incredible improvement over the PATA situation.
Generally, I find it dumb when "desktop" and "mobile" components are artificially segregated by the use of different connectors etc., while it's really the same computer tech underneath. I like to use quiet and low-power stuff for my non-mobile needs, basically because I have to pay for the electricity and listen to the fans myself. Also, you'd think it were cheaper to build fewer different models of everything. SATA for 2.5'' and 3.5'' was a brief moment of sanity, and now we're in the same mess again.
obligatory http://iki.fi/teknohog/music/c...
Me too. The crystallized leftovers of LSD in my spinal fluid are starting to wear off.
I use FB under a separate user account that has no access to the audio device (not to mention the files of my primary account).
One problem is that a lot of FPGA hackers write their functionality in Verilog or VHDL. Those provide a fairly high level syntax that makes it easy enough to design the hardware. The original chips, however, were designed using schematics.
I'd like to elaborate on this one: if you're using an FPGA to begin with, you don't really have the choice of using anything lower-level. First, the FPGA equivalents of machine language and compilers are usually closed systems, so Verilog and VHDL are often all you can do.
Second, the FPGA hardware itself is not infinitely malleable. You have a limited set of circuit elements and interconnects, and this coarseness is also a limiting factor on clock speeds. When implementing something new, decent hackers will use the FPGA quirks to their advantage, instead of trying to ape the features of hand-drawn circuits. This is of course a problem if you're trying to emulate a given piece of hardware exactly. OTOH, in the case of NES and its few MHz, you can probably make up for it in raw speed and software-like solutions.
FPGA design isn't a magic wand that just makes emulation perfect. You still need to figure out how the chips work internally.
Yes, the FPGA chips in particular.
Apparently, you don't need a Cyclone V to implement NES on FPGA. If you're happy with VGA output, a measly Spartan 3 will do: https://danstrother.com/fpga-n...
7, is just silly, blockchains dont in any way suddenly revolutionise anything - they are one cute solution to one specific problem.
You forgot the usual "it's a pyramid scheme, and besides I'm bitter for not being an early adopter, waaaaaaaa!". Also, what new technology _suddenly_ revolutionizes anything?
Also, don't name your product "android" unless it's an actual humanlike robot. Ditto for "hoverboard" etc. And don't even get me started about "bandwidth".
It's the only way to be sure.
Is there supposed to be a correlation between these two things? I'm not sure, though I do go to the Church of Discordia quite regularly.
...is there an app for that?