Forgot your password?
typodupeerror

Comment: Re:Performance (Score 1) 183

Want to kill off the desktops? Find something with better display and user input.

How about the same display and user input. Have you not heard of USB and HDMI? A laptop can be easily connected to an external keyboard, mouse, and monitor. I do this all the time! Since even a relatively low-end computer is more than good enough for most tasks, there is really little down-side to this approach. The extra expense is justified because you can carry it with you.

Comment: Re:Not a barrier (Score 4, Insightful) 183

Hmmm.. I remember the Atari 1020ST was sold as the first computer ever to be under $1 per Kilobyte. It is true that $0.50 / gigabyte is nothing magical from a tech standpoint, but this is not about tech, it is about psychology. Human beings are not entirely logical, and emotions play a large part in decisions.

Comment: Re:Performance (Score 3, Insightful) 183

Bingo. Laptop users. Laptops are on the way up, desktops are dying. And since the higher-end laptops (ultrabooks) are even ditching optical drives to save size and weight, what do you think are the odds that they will make space for a 2nd drive. In fact, I would not be surprised of the 2.5" drive bays went away entirely in the next three years, to be replaced by slots (probably PCIe or something similar). Unless you are going for a larger device -- gaming or workstation laptop, you are not going to have the luxury of two drive bays.

Comment: Re:while nice... (Score 2) 136

by harrkev (#47626717) Attached to: Parallax Completes Open Hardware Vision With Open Source CPU

Well, I can see a use for this. If you HAVE an existing FPGA, you could throw a processor on there for free. Some FPGAs have a CPU built-in (such as an ARM), but those parts cost more. With this, if you need some processor, this is not a bad choice. You could go for something like an 8051, but more options are nice to have. This also apparently has a nice software chain (compilers, interpreters, etc.).

If you really need a well-supported embedded soft processor, your choices are OpenRISC, 8051, Z80, 6502, or this (off of the top of my head, let me know if I missed something). Xilinx makes a MicroBlaze, but they charge money to unlock it.

Comment: Re:Open FPGA? (Score 1) 136

by harrkev (#47626645) Attached to: Parallax Completes Open Hardware Vision With Open Source CPU

A few years ago we finally started to use VHDL '93 at the moment we expecting to be able to use VHDL2008 in 2028 and this is not a joke, that date is realistic based on historic glacial movement of the hardware industry.

Seriously? You have my condolences for using VHDL. You have my deepest sympathy. Second, why the glacial pace? SystemVerilog is supported by all major sim makers (at least to the extent needed to support UVM). Even synthesis tools are starting to support the SystemVerilog constructs that make sense in hardware (structs, unions, etc.).

Really, unless you are stuck using some specific tools that you can't upgrade or update, there is no reason that you can't switch to SystemVerilog today! Although, I admit that SV does not bring nearly as much to synthesis as it does to Simulation.

BTW: I come from the custom silicon world. I don't really use FPGAs much, so SystemVerilog may be beyond the capabilities of the free tools.

Comment: Re:Tek smeck (Score 3, Insightful) 273

by harrkev (#47615417) Attached to: Hack an Oscilloscope, Get a DMCA Take-Down Notice From Tektronix

You could say that offering all options at a discount costs them nothing. You could also argue that it does deprive them of revenues. There are arguments both ways.

It is sort of like Windows 7 home vs Windows 7 pro vs. WIndows Server. They all pretty much share the same code base (maybe less so for the Server version). The only difference is a switch or two.

If you argue that turning on the FFT and serial protocols costs them nothing, you are right! Once the scope is in your hands, it costs Agilent and Tek next to nothing to enable that feature. For Agilent, it is an unlock code. For Tek, it is a module that costs them only a buck or two to make.

On the other hand, it actually DID cost something to include those features. A lot of serial decode stuff is done hardware and software. The software costs a lot of money to develop and test. The hardware part adds some cost to every single unit sold, plus the cost to develop that test that. So, imagine that all of these extra features (FFT, serial decode, etc.) were included standard with every scope. This means that the price would have to be raised to cover all of the NRE costs. So, the price of the scope rises for everybody. For those that need the extra features, they are getting a great bargain. For everybody else, they are paying more for something that they don't need.

So, by locking features that need to be unlocked, you piss off the people who feel like the features are already there, and they are being artificially prevented from doing something that they ought to be able to do. If you unlock everything, you raise the price for the very budget-conscious customers. There is no perfect answer.

Comment: Re:Tek smeck (Score 5, Informative) 273

by harrkev (#47613755) Attached to: Hack an Oscilloscope, Get a DMCA Take-Down Notice From Tektronix

In all fairness (and as a former Agilent employee), you would not believe the amount of work that goes into those things that you don't get with cheap PC-based scopes and low-end stand-along scopes. They do a LOT of work making sure that the front end (analog stuff between BNC and A/D converters) is correct. Also, lots of DSP-ish type stuff right after the A/D too. I am a digital designer, and I worked on some of the oscilloscope chips, and I don't even understand a lot of that of that stuff.

For a hobbyist working with bandwidth-limited signals, and everything is 5V or less, the cheaper brands are probably fine. However, how do you tell if your scope is lying to you? Do you know aliasing when you see it? I have seen some PC-based scopes do the voltage offset (where you twist the little knob to move the waveform up and down) all in software, and seen the clipping in the A/D -- nasty stuff. You really need do to that in the analog front end You also have how many waveforms per second that you can display. If you have a glitch that happens only rarely, if you are capturing only 30 or 100 waveforms per seconds, you might not see the glitch. On the other hand, if your scope is capturing 50,000 waveform/second, you stand a MUCH greater chance of seeing it.

I do admit that scopes are a pricey purchase, and part of that is due to the low volumes involved and the high amount of R&D. But, if you need something that you can trust (you make your living off design work and are not just a hobbyist), you really need to get something professional from a reputable company.

Comment: Re:USB DACs (Score 1) 502

Personally, I use built-in audio. It really IS good enough for most purposes - I have never been dissatisfied with the quality of my laptop DAC.

My original point was that cheap USB audio (those under $10) are crap, and most people who just want to improve the sound, and CAN tell the difference, don't need the fancy DSP stuff.

I want to Sweetwater's web site. They have a bunch of brands of USB audio interfaces in the $100 range from such brands as Alesis, PreSonus, Yamaha, and M-Audio. Behringer even makes $30 ones, but reviews are mixed. Still, if you need line-in on a laptop, that is the cheapest way. If you ARE into sound and music, you can get even mixers with audio interfaces built-in. Alesis even makes some rather nice studio monitors (speakers) with a USB interface.

Comment: Re:Creepy (Score 1) 188

by harrkev (#47436269) Attached to: DARPA Successfully Demonstrates Self-Guiding Bullets

I don't think the materials science is there to deal with forming/deforming a projectile on the order of 300,000 rpm (presuming a 1:7 twist & 3,000 fps).

Piezo actuators should have no problems working at up high tens of KHz, and even up in the hundreds. Peizo elements are used in tweeters, where they have to react up to at least 20 KHz, in the right range for this project. Piezos do not have much distance that they can travel, but at that speed, you might not need much distance. All you really need is a little paddle that can stick out an slow the bullet down on one side.

Comment: Re:Alternate use for this technology (Score 1) 188

by harrkev (#47436233) Attached to: DARPA Successfully Demonstrates Self-Guiding Bullets

And here's a clue for you. NO one over there wants peace at all, ever.

Sorry, but many Muslims are taught to hate Jews from a very young age:
https://en.wikipedia.org/wiki/...

There are other examples, but this is the most famous one that I could think of.

From what I understand (and I know many American Jews who have visited Israel), the Jews pretty much just want to be left alone.

Comment: Re:Creepy (Score 2) 188

by harrkev (#47434677) Attached to: DARPA Successfully Demonstrates Self-Guiding Bullets

You got it. Lasers are cheap... electronics are cheap... batteries are cheap... spoofing is cheap.

Really, as an engineer, I can imagine two ways for this thing to work, and it depends on if the projectile spins. Typically, bullets spin so that they act as gyroscopes -- always pointing the same direction (YouTube has videos of guys firing pistols into ice -- ice stops bullet which just sits there and spins like a top).

If the projectile spins, you can, in theory, guide it with a single fin that can extend or retract. You could not use a standard camera as such, because you are spinning wildly. Assume 2000 FPS bullets -- if you want to shoot a mile, you need at least this much. Also assume a 1-in-12 twist (real twists are in the range of 1-in-7 to 1-in-14, depending on shape and weights of bullet). That means that the bullet is spinning with a approximate rotation of 2 KHz. I doubt that you could have an effective regular camera spinning like that and still work. A better way would be to have a linear sensor (a line camera) that looks forward and to the side. This could operate. When you see a bright stop, see how far it is from the center. More off-center = kick your fin a bit more. This is simple and straightforward. However, since the bullet is spinning and you do not know when the camera will cross the laser, you probably need to keep the laser on full time. This is probably the easiest and cheapest way to accomplish this, but should be easily spoofed. You could maybe put a crypto on the laser signal by changing the intensity of the signal without turning it off, but it would have to be a much lower frequency than 2 KHz because that is your effective sample rate. If you assume 500 Hz signal (four-times oversampling), you would only get about 500 bits of data before you hit your target (assuming a target 2000 feet away). Is that enough to actually apply crypto? I am not sure...

On the other hand, if the bullet is NOT spinning, you can use a regular camera and regular fins to control it. In that case, it is entirely reasonable to embed some sort of cryptographic modulation on the signal. In any case, the existence of a 2-D sensor makes the bullet more expensive, and increases the amount of processing that needs to be done. It should, however, be more feasible to put crypto, but at greater cost.

Comment: Re:Alternate use for this technology (Score 3, Insightful) 188

by harrkev (#47434367) Attached to: DARPA Successfully Demonstrates Self-Guiding Bullets

I don't get the US. I mean, by now you should have noticed that the bigger and more complicated the technology, the more you play into your opponent's hands. First of all, you're using high tech weapons in a low tech war. You can't really fire any round anymore that doesn't cost you more than what your target cost your enemy.

Off topic, I admit, but this reminds me of the current Isreal/Hamaas conflict. Just launch simple, dumb, and cheap unguided rockets from the Gaza Strip. Isreal has an "Iron Dome" defense system that is supposedly pretty effective at stopping them -- at $1,000,000 per shot. Great way to bankrupt an enemy...

Comment: Re:Creepy (Score 1, Interesting) 188

by harrkev (#47434329) Attached to: DARPA Successfully Demonstrates Self-Guiding Bullets

Actually, this should not be scary at all! You just need to figure out the frequency and modulation of the laser used. Then, just make sure that you have such a laser pointed at the guy beside you. You are suddenly safe from snipers! Just make sure that you do not like the guy beside you.

Seriously, the only way this could be spoof-proof is to modulate the laser with some type of crypto.

Comment: Re:Why 80% (Score 1) 278

by harrkev (#47433621) Attached to: William Binney: NSA Records and Stores 80% of All US Audio Calls

Hey, Obama promised "Hope and Change." Isn't that what we have here? Admittedly, Bush started this -- probably. Or maybe he inherited the seeds from Clinton or earlier -- who knows how far back this trail goes? But Obama has had almost 6 years to fix things. Instead, under his watch, things have gotten worse.

In Obama's defense, I do not know if Romney would have done things any differently, but I suspect we would probably still be here even if he had won.

We are Microsoft. Unix is irrelevant. Openness is futile. Prepare to be assimilated.

Working...