I find that taking notes takes away some of the focus that's required to actually understand what's being said. Everything worth writing down should already be in the textbook, or in supplementary materials provided by the teacher.
I never understood why students need to take notes. When I was in college I never took notes, instead I tried to listen and understand what was being said. The rest of the required information I got from the course material that was prepared by the teacher.
Knives and small firearms are no serious threat on a plane, when the cabin doors are locked. Bigger firearms and/or ammo can be detected with a metal detector.
It's not harder if you know electronics. A PID op amp needs only 4 op amps, 9 resistors and 2 capacitors. No need to debug
Obviously, for someone experienced in electronics, a PID op amp circuits is easy, and needs no debugging.
For someone experienced in microcontroller based digital control, a firmware PID control is easy too, and has no need for debugging either.
Also, this project isn't just a simple PID controller. They also have filters, mixing of accelerometer and gyroscope data, steering, and PWM motor control.
I'd add a second (or even third) gyroscope and accelerometer, and have the controller compare the inputs. If they are too far apart, the controller goes into "failure mode" where it will cut power to the motor. When they are consistent, you can average the values for lower noise.
Even on the existing design, you can compare the gyroscope and accelerometer.
The higher part count is surely on the side of the digital controller. Just look at the diagrams offered: analog means direct processing of signals, no A/D. Just some op-amps, pwm, done.
How does adding some op-amps and discrete PWM reduce part count ? Have you counted all the resistors and capacitors for the filtering ?
With a digital controller, you can implement the A/D, filtering, control loops, and the motor PWM all inside the same device. You can even avoid some of the A/D stuff by using a accelerometers and gyroscopes with a digital interface.
and filters are implemented easier with some RCs around an op-amp
An RC filter can be implemented in a single line of code on a controller.
Not exactly. I was talking about fitting a smooth curve through the points, like a higher order polynomial, and integrating that.
Anyway, once you put the data points in a computer, it's fairly trivial to use any number of methods to get good results. It's amazing that anybody would use inferior techniques.
Depends on what you use it for, for a portable camera, 2TB is quite large. But some other manufacturer may decide these cards are perfect to replace that noisy hard drive in a PVR, and then the 2TB may fill up fairly quickly.
So sending the full address block for each 512 byte sector requested would slow all storage transfers down by about half a percent
At these high bit rates, the round trip latency for a single sector request is going to be dominating anyway. The obvious solution is to transfer multiple blocks in long bursts whenever possible.
Of course, the Linux kernel code is different because it has been modified and expanded in the last 20 years, and every change to the code has a new copyright. This is different than redistributing the exact same material in different forms.
Of course, it all depends, and I've not seen your design, so it could be that a 555 is simpler. In other cases, a microcontroller may be simpler. What's cheaper also depends on the circumstances. Price at digi-key for a NE555 from TI is 29 cents, while their cheapest microcontroller (PIC10F200) is only 34 cents (both @ 100 pieces). The microcontroller also comes in SOT23-6, so it takes up less board space, not even including the C and Rs. And maybe the microcontroller, in the same circuit, can also replace some other functions as well.
If you already have microcontroller, and know how to use it, it may be simpler than a 555. With most microcontrollers you won't need any external parts, and you can easily adjust the timing.
Slashdot Mirror
My thoughts exactly.
I find that taking notes takes away some of the focus that's required to actually understand what's being said. Everything worth writing down should already be in the textbook, or in supplementary materials provided by the teacher.
I never understood why students need to take notes. When I was in college I never took notes, instead I tried to listen and understand what was being said. The rest of the required information I got from the course material that was prepared by the teacher.
For the little amount of sheet music that gets distributed to preschoolers, they could have turned a blind eye.
Knives and small firearms are no serious threat on a plane, when the cabin doors are locked. Bigger firearms and/or ammo can be detected with a metal detector.
That's not exactly brain surgery, is it ?
http://www.youtube.com/watch?v=THNPmhBl-8I
You might as well print out the levels on several pieces of paper, stack them all up, and put the correct one on top.
And Google has shown no willingness to filter Groups spam. I used to read Usenet through Google Groups, but it's now totally unusable.
A lot of PC motherboards fail because the electrolytic caps wear out, not because the CPU dies.
Obviously, for someone experienced in electronics, a PID op amp circuits is easy, and needs no debugging.
For someone experienced in microcontroller based digital control, a firmware PID control is easy too, and has no need for debugging either.
Also, this project isn't just a simple PID controller. They also have filters, mixing of accelerometer and gyroscope data, steering, and PWM motor control.
True, you can't model a realistic RC filter in one line of code, but you can model a perfect one, which is exactly what you'd want in this case.
I'd add a second (or even third) gyroscope and accelerometer, and have the controller compare the inputs. If they are too far apart, the controller goes into "failure mode" where it will cut power to the motor. When they are consistent, you can average the values for lower noise.
Even on the existing design, you can compare the gyroscope and accelerometer.
How does adding some op-amps and discrete PWM reduce part count ? Have you counted all the resistors and capacitors for the filtering ?
With a digital controller, you can implement the A/D, filtering, control loops, and the motor PWM all inside the same device. You can even avoid some of the A/D stuff by using a accelerometers and gyroscopes with a digital interface.
An RC filter can be implemented in a single line of code on a controller.
Instead of spaghetti code, we now have spaghetti hardware :)
http://www.etotheipiplusone.net/pics/seg/seg_102.jpg
With core memory, a read is destructive, so it's not truly static.
My pet peeve is the unnecessary use of the underscore in struct _FOO. I prefer this:
typedef struct foo { int blah; } foo;
I hate caps too, except for constants.
Not exactly. I was talking about fitting a smooth curve through the points, like a higher order polynomial, and integrating that.
Anyway, once you put the data points in a computer, it's fairly trivial to use any number of methods to get good results. It's amazing that anybody would use inferior techniques.
Just fit a curve through the data points (or several piecewise connected curves if necessary), and integrate the results the standard way.
Depends on what you use it for, for a portable camera, 2TB is quite large. But some other manufacturer may decide these cards are perfect to replace that noisy hard drive in a PVR, and then the 2TB may fill up fairly quickly.
At these high bit rates, the round trip latency for a single sector request is going to be dominating anyway. The obvious solution is to transfer multiple blocks in long bursts whenever possible.
Of course, the Linux kernel code is different because it has been modified and expanded in the last 20 years, and every change to the code has a new copyright. This is different than redistributing the exact same material in different forms.
Of course, it all depends, and I've not seen your design, so it could be that a 555 is simpler. In other cases, a microcontroller may be simpler. What's cheaper also depends on the circumstances. Price at digi-key for a NE555 from TI is 29 cents, while their cheapest microcontroller (PIC10F200) is only 34 cents (both @ 100 pieces). The microcontroller also comes in SOT23-6, so it takes up less board space, not even including the C and Rs. And maybe the microcontroller, in the same circuit, can also replace some other functions as well.
If you already have microcontroller, and know how to use it, it may be simpler than a 555. With most microcontrollers you won't need any external parts, and you can easily adjust the timing.
I'm not really in the mood to prep a roast in the morning before I leave for work. I'll just make one in the weekend.
Sounds like a lot of work, even when it works perfectly. I just scan articles with my nose whenever I doubt the freshness.
I highly doubt Intel is interested in competing in the low-volume, low-performance markets.