The decision long ago to not support a default ai means you are at the mercy of the community. This means that each ai available (you have to download them sight-unseen if you want to play single player) ranges in quality from crap to passable. Each one was lovely crafted until their creator got bored, so you can imagine that many of them don't support later added features like load/save, which are essential to long-running games.

The ais are also typically balanced to blow you out of the water in temperate medium difficulty but then fall when you go north/south and up the difficulty. For years NoCAB has been the ONLY ai to even come close in exotic locals, and development on that beauty has been dead since 2009.

Every year I download and try out.the latest ai versions and am disappointed. Then I play a round with NoCAB for old times sake and forget about it for another year.

I believe you two are talking about different sorts of noise.

I believe GP was referring to SIGNAL noise on the USB line playing havoc with his RS-232 adapter. It does not mean that the laptop FAN is loud. It's also not certain that you would have issues with normal USB devices, give what a corner case RS-232 adapters are.

It's intended to reduce EMI induced on other devices, not listen in on your conversations :D

In the bad old days of Betamax, if your local video store did not have what you wanted, you didn't get it. And shelf space was at a premium, so if a movie was carried at all, it was more likely to be VHS.

Today Apple's digital distribution of content solves that problem. They've proven that they can make a massive profit off a small segment of the market, and that their customer base is loyal enough to keep on trucking (much like loyal Betamax users only switched due to lack of content).

So yes, while it's an accurate connection between iOS and Betamax on the surface, it's a completely different world today. Apple will command a smaller-and-smaller share of the mobile market, but it will reach a certain percentage and stay there.

A transmitter with an eight mile range would be extremely easy to detect.

I have to disagree with this. You could hide in the noise if you used Frequency Hopping Spread Spectrum. It's not perfect at preventing direction finding, but it's very hard for your average enthusiast to track.

Well a 486 with a larger cache pretty much WAS the Pentium I dude.

Not quite.

The Pentium had dual integer pipes (with some limitations), and a fully-pipelined FPU unit with full hardware support for FP add/mult/div. Double the bus width (to help feed the thirsty FPU and dual int pipes). Branch prediction (4 state). Really, the larger, better-architected cache were on the low-order of importance.

The 486 had just one fully-pipelined integer unit, and a limited NON-PIPLEINED FPU. The FPU hardware was cut-down compared to the 80387 it replaced (due to limitations imposed by the desired die size). This was more than made up for by the removal of the communication overhead between the two chips (15+ cycles on 386/387), and higher clock speeds. So overall performance increased over the 80387, but it was absolutely DESTROYED by the Pentium FPU.

It's not a "nice:" tablet, it's a cut-rate excuse to get Tegra 4 in the news. If they had properly outfitted this thing, it would be the same price as the Nexus 7 2013 model (and exactly the same performance, which a shittier screen).

Any tablet that ships in this day-and-age with just 1GB of ram is not "nice." You just try and load more than 4 tabs in your web browser before running out of ram. And while benchmarks don't tend to care about memory capacity, it will certainly make a difference in games (which is the primary focus of this SoC).

Yup, Anand made the observation here that 4k @ 60Hz = over 14 Gbps of bandwidth. Since Thunderbolt 2 is not actually any faster than version 1, (just allows channel bonding), 20Gbps is a real limit! And you only get three of those.

So if you want to run 4K you have a measly 400MB/s available on a channel, which means you will need to dedicate 1 of the three per-display. And the HDMI port is attached to the third channel, which means you'll have almost nothing left when running triple display!

Apple's insistence on combining DisplayPort with Thunderbolt has come back to bite them in the ass. The very REASON the industry developed DisplayPort 1.2 (and soon 2.0) is because they need MORE BANDWIDTH! And on any other SANE architecture, the display outputs are all provided by the graphics card, instead of wasting bandwidth being shuttled across the PCIe bus to fight for bandwidth with what should have been a dedicated storage bus!

.

The Neural Network that creates prose!

The pig go.
Go is to the fountain.
The pig put foot. Grunt.
Foot in what? ketchup.

The dove fly.
Fly is in sky.
The dove drop something.
The something on the pig.

The pig disgusting. The pig rattle.
Rattle with dove.
The dove angry.
The pig leave.

The dove produce.
Produce is chicken wing.
With wing bark.
No Quack.

That and the placing of full responsibility upon those you caught tends to give them more reason to rat out their friends (they usually know at least one of the other perpetrators).

They can spread the restitution costs if they implicate others, and possibly get a more lenient jail sentence. Going easy on them will just give them all the more reason to keep their stupid mouth shut.

The book Starship Troopers was *not* critically attached to the concept of mobile infantry suits, it was simply there to hook the anti-social young adults attracted to Sci-Fi books in the 1950s. If Heinlein had written it more like a popularity contest, it would have (1) felt more like real war, something the kids were trying desperately to escape, and (2) it would have felt more like high school, something the kids were trying desperately to escape.

If you want your work to be popular, you target what your audience wants to hear, not what you think is cool. Heinlein did the right thing in his time targeting introverts, and Paul Verhoeven did the right thing making PEOPLE AND THEIR RELATIONSHIPS (and not technology/effects) the initial focus.

It would have been much harder to entice general movie audiences with a character who is a somewhat anti-social "god made by technology." Imagine selling people on a movie where Loki is the hero instead of Thor - Thor himself is already a tougher sell than most comic book heroes, and if you used Loki instead it would be impossible!

The story works surprisingly well without the mobile suits, and it NEEDED to if it was ever going to be put on film.

...and how does region locking play in to that? Best I can tell, you could prevent multiplayer piracy without locking people out of the game in central ohio because you think their located in London for some idiot reason.

Region locking allows them to sell the game for different prices to different regions. YES, you pay more for video games then someone in Nigeria does - get over yourself!

Also, the biggest benefit you get from region locking is the ability to do a staggered release. This is CRITICAL for a primarily multiplayer title BF4, because they don't have to deal with the server load spikes (and associated technical support peaks) you get with a simultaneous release. Or have you already forgotten what happened when Rockstar let everyone rush online at the same time?

This is a BUG - the players here were supposed to be included with the rest of the US block, but were not. But this is a bug that will fix itself, and rather quickly - they will not die from waiting another day.

Interesting. I'm surprised that they managed to pull that off over a PCI bus.

They were able to do it because the PowerVR chip processed a tile-at-once and only wrote the final rendered frame over PCI. Since you never have to read the frame buffer for more complex multi-pass effects (unlike all other cards at the time), you could get by with PCI throughput. The card had local memory to store textures and the scene draw ordering buffer.

Out of curiosity, did the PowerVR cards manage to behave well in that regard, or could you induce situations where firing up the 3d cratered throughput on any IDE/ethernet/whatever peripherals on the same PCI bus, or where frames dropped all over the place because your IDE controller decided that something needed to Get Written NOW, and grabbed mastery of the bus at the wrong moment?

This was a problem experienced on ALL early video cards, thanks to the mess that was early PCI busmastering. Most people just ignored it, just like tearing. The only time you really cared was if the video card caused your sound to stutter.

AMD designed this chip for maximum performance in minimum die space. They managed to cram Titan-level performance in 435mm^2. That's including a 512-bit memory bus AND 64 ROPs, so they're not exactly cutting corners!

The Titan uses a 551mm^2 die-size, and although some of that is fused-off, the majority of the difference is because Nvidia designed it wide and slow for power first, performance second. This is because the part was targeted first-and-foremost at professionals, where performance/watt and cooler noise is actually a concern.

By prioritizing die space over efficiency, AMD were able to offer their card at the $550 launch price-point. AMD is betting that hardcore enthusiasts won't care about power, especially when the card destroys the Titan at 4k resolutions. I guarantee you Nvidia could not make a profit at the same price, and that's why their reaction part (GTX 780 Ti) will be priced at $650.

You can't expect companies to work miracles when all they have is the same old 28mn process. You can emphasize efficiency or die size, but you can't do both!

Expensive pain in the ass? as in REAL hardware processing using a chipset instead of spare IO pins on the chipset and let the processor waste time doing the job?

Yeah, you needed that on a DV camera because transferring was a real-time thing thanks to the TAPE FORMAT. You had similar sync issues back when people had to load programs from cassettes. Buffers were expensive, so the channel had to be high-bandwidth and low-latency.

It was also a hard problem to solve because the format purposefully used a LOWER COMPRESSION RATIO than was possible at the time (roughly 30 Mbits/s for DVD-quality video). This was required to reduce the overall cost of the on-board processing hardware, and took advantage of the gargantuan size of the DV tape.

So, for an isolated corner-case, the Firewire system was developed. It had enough hardware support so that systems could maintain transfer rates with minimum latency, and added a reasonable amount of cost versus going with a higher video compression ratio. It was also forward-specced for the eventual release of HDV, which required roughly 100 Mbit.

Yeah... And explain why it was and still is the standard in pro video and audio? Oh it's because USB is crap for transferring huge amount of of data.

Momentum. USB 2 had more than enough bandwidth even for HDV. The worst-case read/write speeds I have seen have been 30/20 MB/s, and that's over twice the read speed you need for HDV.

USB 3.0 requires the "pain in the ass" that yu complain about as it requires a chipset to do the processing instead of being cheap crap that requires the processor to waste cycles on it.

And it introduced all those fancy features a decade after Firewire did. NOW it's incredibly cheap to package that extra processing power for accessories, because there's all sorts of free space in modern southbridge designs (they are usually pad-limited). And it's a whole lot easier to fit a complex feature in a 32nm process than it ever was inside of a 500nm process!