Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

Comment Re:IO limited? (Score 2, Informative) 208

Typically for graphics cards, the only data sent over PCIe is texture data, vertex lists, and commands. The bulk of the operations done by the card are running the commands over the vertex lists while bringing in texture data. The commands are almost always a multi-pass or pipeline so each vertex will be used in computations more than once. The result is the pushed to the monitor, not the PCIe. So, yes, in general, a graphics card will have more FLOPs than I/O bandwidth.

Comment Re:Sounds familiar. (Score 2, Insightful) 571

People who will try to cause fear and injury aren't new. There hasn't been any proof that all this legislation and fear mongering around curiosity has actually made us any safer. We live in an inherently dangerous world. It's time to realize that we can't baby-proof it. Then we can get back to doing research, having odd hobbies, and being generally curious without fear of being accosted.

Comment Re:So they broke it, and made it theirs. (Score 1) 163

Perhaps you should actually read up on the technology (link are already in other comments) and realize that the sleep proxy service handles some requests without waking the machine. For something like a ping, it doesn't wake the machine, but instead the proxy responds to the ping directly. Same for service advertisements. It only wakes the machine when the proxy can't handle the request on the machine's behalf.

So yeah, it does solve the problem. Now you've proven that not only are you unable to perform basic research, but that you ignore the facts presented and continue claim something entirely refuted by the facts.

Comment Re:Not necessarily fake (Score 5, Insightful) 511

If you've ever had a display calibrated, you'd know that even the existing RGB color space can't be completely recreated with existing RGB-based displays. The problem is in the inability of LEDs or LCD or plasma panels to produce light uniformly in the three color channels. If you can add a 4th channel that lets the RGB color space be more accurately produced by the display, then you will see an improvement. It won't make the source any better, but the output generated by the display for that input will be better.

Comment Re:I just don't see the issue (Score 1) 559

So in other words, it was OK for everyone to broadcast information that they don't really want to be public because they didn't expect anyone to actually make it public. Then, when someone does, it's the fault of the collector that everything was available? Huh? Perhaps it would be more prudent for individuals to consider what having something be made public means before deciding to do so. The options for not broadcasting SSIDs have been in APs since the beginning.

A probably poor analogy: When I'm visiting my parents, I tend to not bother locking my car doors since they live in the middle of nowhere. I don't expect anyone to steal my car because it is unlikely that someone would know that I leave it unlocked and would venture out that far to steal a car. Now, a company comes along and records locations and the number of cars that have unlocked doors. If it helps, consider that this can be determined for most cars without touching the car. If my car gets stolen, do I sue the company for making it known that my car was frequently unlocked in this area? No, I realize how dumb I was, file an insurance claim, and start shopping for a new car. I probably won't leave my car unlocked any more either.

Comment Re:Too many Linux-incompatible-with-Linux distros (Score 1) 148

Uh, no. OS X provides a rich set of libraries as part of the base OS. Apple goes to great lengths to ensure compatibility between OS versions (libSystem is compatible to version 1). The only time any software includes a library inside their app bundle is if they wrote it or it is an OSS library that isn't in the base OS. Most apps don't need to.

Comment Re:30ms? (Score 1) 334

That's why we have virtual address spaces. Each process gets a 4GB address space for a 32-bit OS. Only 2-3.5GB (depending on OS) will be available for actual program code (not shared libraries) and data. Of course, those virtual addresses get translated to physical addresses by the OS on a per-page (or at least per-range) basis, so all 4GB of physical RAM can be used assuming there is more than one process running.

That's all a simplification of how modern VM subsystems work, of course.

As to memory "stolen" by hardware, the big one there is video cards. A fair number of OSes still map the entire video memory into the physical memory address map. You might think this would be a problem, but on the x86 architecture at least, physical addresses have been 36 bits for a while. Even with a few video card with 1GB of video ram each, there is still enough space to address all 4GB of physical RAM.

Comment Re:I'm Not a Betting Man... (Score 1) 235

A controlling interest in Google is owned by the CEO and two founders. Their IPO stated that this would be the case and that public investors would be able to share in the financial gains, but not significantly in the direction or operation of the company. If those three have decided that China isn't worth it, there is little the investors can do to stop them.

Slashdot Top Deals