I've owned the Kyocera 6035 and now own the 7135. These are palm-based phones, and both sync with evolution as well as any other palm. (Which means, they sync, but you run into occasional wierdness like duplicate records.)
The 6035 is kind of clunky (it's a pretty big phone) but due to its size and simple black and white LCD, the battery life was incredible. I could take it on a business trip for a week and not begin to worry about depleting the battery.
The 7135 is cooler, it's a big (but much smaller than the 6035) flip phone with color backlit LCD. Battery life is pretty awful due to the small battery and backlight. Oh, and the batteries occasionally explode. That said, I love it, works well and is a great geek gadget (you can convert MP3s into ringtones, among other things, which with a star trek sound effects MP3 library allows you to attain maximum geekiness and preserve your virginity forever with zero effort.)
Consistently various governments seem to think this fusion of public and private databases is somehow going to answer a problem.
The problem is... the private databases they usually intend to integrate are crap. Some private audits of consumer credit databases turned up error rates of 70 percent. "Minor" errors like wrong country of birth, birthdates, identity, etc. One study had a identifing information error rate -- bad addresses, etc -- of 40%.
So completely separate from the debate of whether this is the right thing to do -- is the fact that what they're doing is braindead in the first place, creating a massive database with bad information.
On second thought... maybe it's better they do this, as it's going to almost immedicately kill the credibility of projects like this if it ever gets tested in court.
Very true. There was a story in Mechanical Design magazine about the fiero before the car came out, and it was a vette killer. A _cheap_ vette killer. So it had the prospect of killing a very profitable car, so GM didn't help it at all.
I had a later mode; (1986) and still regret giving it up 10 years later. The 1984s sucked beyond belief, but as the Forbes story mentions, by the end (1988) is was a really nice car.
The most amazing thing to me was the body. 10 years later, my car didn't have a single ding, due to the flexible plastic body panels.
I notice that a number of people knocking SPF are looking at it breaking some sort of standard, or that it's an exclusive, it's-the-only-answer technology, ie it's being proposed as a silver bullet.
It's not. SPF just provides one more bit of helpful information -- which IPs email from the sender's domain should really be coming from.
While someone could use SPF in a pure binary decision system that breaks SMTP, it's going to be an incomplete solution. Just like blacklists, whitelists, and bayesian filtering are also incomplete solutions.
However, you start using these things in combination and magic happens.
Example: I use ASSP for server-side spam filtering. ASSP uses bayesian filtering, but also whitelists people you email and uses blacklists.
The blacklist implementation is interesting, however, as when it determines an IP is blacklisted it simply starts off with a higher spam probability in the bayesian stage -- it's not truly blacklisted, just more suspicious.
You could do the same thing with SPF, initially giving a lower spam probability to mailservers with SPF, and when there's more infrastructure using SPF, switching to penalizing non-SPF servers.
Nice thing about this approach: it doesn't require everyone to convert their infrastructure, but it does incentivise legitimate servers to do so without penalty. It doesn't break any standards. Legitimate mail still gets through, but spam suffers.
Stop thinking that all spam solutions have to be single silver bullets. Anti-spam tools can be additive.
And since the 6502 only took a few thousand transistors, you could fit ~ 32,000 parallel 6502s on a single chip using today's processor manufacturing technology. Running at 2 GHz. Pretty impressive performance possibilities.
Of course, keeping a 256000 bit 6502 VLIW implementation fed with data would be next to impossible. But quite a few specialize components such as video processors do something like this.
While in high school and college, I was the beneficiary of an extreme version of this.
I'd routinely stumble on a hard calculus problem; after a bit I'd move on to other problems, come back to the problem one last time, and then go to bed.
I had a recurring dream where there was a book at my workspace I now jokingly refer to as the "Book of all knowledge." In the dream, I'd encounter the problem, open the book like any other reference book, and it would always open to the page with the answer to the problem on it, fully worked out step-by-step as good calculus answers should be.
When I awoke, I knew the answer and essentially copied down the result from the dream, understanding how I got there due to the step-by-step nature of the answer.
Obviously this didn't work all that great for tests.
Seriously, though, it's probably one of my fondest memories of advanced math, having the coolest book in the world in my dreams. I also suspect it points out that problem solving even in very technical domains is a creative, right-brain process and not nearly as logical as people would be led to believe.
As noted, not only are pop-up/under ads a problem, but flash ads are a serious issue in terms of sucking CPU cycles. If you're using older/intentionally slow (low power) hardware, flash ads can drive unusually high CPU utilizations and make pages unusable.
I've stopped using fox news in particular as they seem to have a couple flash advertisers that drive my CPU utilization from about 1-2% normal to 75% when the ad appears -- and in a tabbed-style browser they continue to suck CPU cycles even when I'm not looking at them.
Clearly the default assumption of the people writing the ads in flash is that the installed base of PCs is as fast as the workstation they're writing the ad on, instead of the much slower reality.
Because it adapts, each new technique the spammers try ends up diluting the effect and ruining it for all spammers. And because they're greedy and will sell each other out without hesitation, it's basically using their own motivations against themselves.
Might as well put in a plug for my favorite bayesian filter: ASSP
As others have pointed out, it's most likely a flat phased array antenna.
There's a couple attributes that would make it attractive for a extraterrestrial application. They're very compact for the gain they provide, and within the limits of the design they can be electronically steered (that is, no moving parts). I would imagine they probably have a mechanical coarse steering mechanism and electronic fine steering.
Sadly I can't seem to find any confirmation of this, just a few mentions of other spacecraft such as MESSENGER using phased array antennas.
If you're really a radio newbie you should know that gain is how well the antenna concentrates the signal. An isotropic radiator basically receives/transmits signals in a perfectly spherical manner. By sacrificing the directional coverage you can increase the gain. A great example is a flashlight bulb -- uncovered it radiates almost everywhere; with a parabolic reflector it radiates a beam. When they talk about using the low gain and high gain antennas they're basically talking about the radiation pattern.
You generally use low gain antennas for signal acquisition when you don't have control over where the antennas are going to be pointed. Once you know where everything is, you can point the high-gain antenna at the target. With more gain you have a better signal-to-noise ratio and can then crank up the data rates.
Phased array antennas work essentially by combining a large number (an array) of simple low-gain antennas such that they add their signals together (in phase) in a particular direction. In other directions the signals don't add the same way and there's much less gain. At microwave frequencies like X-band (about 8 GHz), a simple dipole antenna is only about an inch long, so it's easy to put a bunch of dipole-equivalents in a small space to make an array.
If the IRS were answering, the answer would be yes.
You would be guilty of "structuring," which is organizing financial transactions in such a way as to conceal the true transaction. This would be interesting to them. Presumably to the FBI as well. As far as if this would be noticed statistically, who know what "profile" they're looking for.
IIRC, there was a landmark case in NY about 10 years ago where someone structured a transaction to hide the source of income, but paid taxes on the total amount, and they were found guilty of breaking the structuring law.
No doubt there are numerous replies about it being in Feynman's book.... however:
There are a number of anecdotal references of the exact same story being attributed to Joseph Campbell (of Power of Myth fame)
and more to the point:
A number of native american tribes (particularly the northeastern tribes such as the Iroquois) have creation myths where the world/universe is supported on a turtle's back.
Interestingly enough, in hindu creation myth Vishnu takes the form of a turtle which supports the world's creation on its back.
There are others. The universe being on a turtle's back is a very pervasive myth.
So where it's REALLY came from it likely a religious archetype that predates writing.
In general, yes, shared memory sucks bandwidth. As others pointed out, the calculations are pretty straightforward (X * Y * #bytes/pixel * refresh rate = Bandwidth).
However, in today's systems it's FAR more complicated that this.
First, some older implementations, particularly the Intel 810, used a 4MB display cache. The net of this is that the display refresh was generally served from a secondary memory and didn't interfere with main memory bandwidth. As well, Intel used some technology Chips & Tech developed that basically did run-length encoded compression on the display refresh data (look right at your screen now, there's a LOT of white space, and RLL will shrink that substantially.)
Today most chip sets incorporate a small buffer for the graphics data and compression techniques to minimize the impact of display refresh on bandwidth.
But wait -- it gets even MORE complicated. With integrated graphics on the north bridge of the chip set, the memory controller in the chip set knows both what the CPU and what the graphics core want to access. So the chip set actually does creative scheduling of the memory accesses so that the CPU doesn't get blocked unless absolutely necessary. So most of the time the CPU is either getting its memory needs services by its own cache, or it's getting (apparently) un-blocked access to memory. So the impact of graphics is much less than the simple equation above would suggest.
Finally... we now have dual-channel memory systems. Even more tricks to keep the graphics and CPU memory accesses separate come into play here.
So, the short answer is yes, there's an impact, but it used to be much worse. Innovative design techniques have greatly reduced the impact so that in non-degenerate cases it doesn't affect the system too much. In a degenerate case of your app never getting cached and doing nothing but pound on the memory system with accesses, however, then you'll see the impact in line with the bandwidth equation above.
Given google's IPO situation, and someone I've never heard of touting themselves as being better than google. Maybe this somone is looking for a bit of capitalization themselves?
As another poster mentioned above, ohphone (the OpenH323 app) has been ported as part of the core of XMeeting. Check here for the ohphonex application.
Like several of the others, I would recommend OpenH323 . It's fairly easy to set up.
You also have the option of setting up a gateway (hardware isn't cheap, however.) I've used the voicetronix board. This would give the option of calling locally if your or she had other friends in the area.
Also, I've found Voip far easier to use with real (or semi-real) phone hardware. Cisco has the ATA-186, though it operates best with proprietary codecs. I've ended up using a VoipVoice handset -- it's USB and appears as a USB audio device under linux so it's pretty easy to configure, though it's a little weak on the volume. I got mine through IPblue though I am sure there are other sources/options for open-source IP phones.
Creative labs used to sell the voipblaster, and this is still available from ebay, and that lets you use conventional phone hardware -- it's basically a USB-to-phone-jack converter with a hardware codec, and would likely be idea if you're working with anyone who is technology-challenged.
With the few tech support disasters I've had to deal with, by phone usually has worked best -- but there are tricks to getting it to work, like making sure you take notes of who you talk to so you can reach them again. With some companies there's a decided change in quality of tech support depending on the time of day. (Strangely enough for me, it worked out that calling from 2am-4am has netted the best tech support.)
I strong concur with the recommendation to Google first. I've also found that email tech support pretty much sucks -- I've had an email request to download updated firmware for a 3Com wireless router that's on week 3 of waiting for approval at the moment, and all too many similar experiences.
Like it or not, self-help (google, mailing lists) is pretty much the way to go these days.
As someone who intentionally runs a low-performance box as a primary system (VIA Epia 533) I'd be pretty unhappy with some snarfing up a few cycles. Junked-up web sites with flash and excessive java/javascript are REALLY noticable when you're browsing at the low end of the power curve.
I run a cpu monitor in the background and when a site wants to run one of the more annoying classes of advertisements, utilization usually pegs... I can't imagine what something that intentionally sucked cycles would do.
Well, since I'm not finished yet, cooking hasn't been a big deal. I mostly use a microwave, microwave power consumption "averages to zero" over the course of the day (you maybe run it five minutes at 1 KW, so actually watt-hours are very low.) I've been living in a "simulated" solar mode at my primary house to test stuff out, though.
Hot water is handled by solar. I've got two 8' x 10' solar thermal panels feeding an 80 gallon hot water heater. Presently there's no back-up, but I'll probably swap out the heater I have (electric, but not connected, it's just a useful tank) with a gas model for backup. BTW, single-biggest payoff from a power consumption angle is solar hot water (at least in the desert southwest.) I have to cover one panel in the summer as it heats 80 gallons to better than 180 degrees. In the winter I'm currently seeing temps around 90 when running on both panels (outside air temp ~ 40-45 for the high.) I may add another thermal panel to boost that temp some.
I don't watch much TV there, but I have a portable unit. Even a big-screen TV isn't much of a problem, I have a rear-projection Hitachi unit at my real home that measures 170 watts when on. What's a killer that I haven't dealt with yet but will have to when I stay there and DO watch TV is that satellite recievers are about 35 watts, and expect to be on 24 hours a day. That's bad.
Ironically, the boondocks area is better served with cellular and wireless data services than my house on the edge of the city (Phoenix) is. 1 Mbps wireless is available (it's high plains desert, so one tower covers hundreds of square miles.) They also upgraded the cells in the area to 3G this past summer, so service isn't a problem.
Anyway, moving off-grid isn't cost effective if grid power is available, but the biggest things I've done have been 1) solar hot water (probably a 5 KWh savings per day) 2) compact flourescent lighting (saving 1 KWh or better per day) and moved my servers from ~150 W units to one 50 W unit (fileserver) and one 15 W unit (I use a Epia 533 for a firewall/web/mail server). Over 24 hours the power savings adds up on PCs.
I'm in the process of building (year #5, woo-hoo!) a totally off-grid home in (almost) the middle of nowhere. I've gone through this already.
The short answer is if power is all you care about, a notebook is better. My Sony TR1A consumes 13 watts of AC when plugged in at full-tilt according to my power meter, and display-off sucks just 5 watts. However, I care about more than power, like uptime/durability, ability to use off-the-shelf components, and being able to support some modest external hardware, so the actual system that stays on 24/7 is a VIA Epia 533 MHz box. It takes about 11-12 watts but can peak around 24 watts or so, plus a 15-inch LCD monitor which is pretty much never on, but consumes 16 watts when it is; so you're looking at about 30 watts with one of the lowest power desktop configurations possible.
Other odds and ends consume power as well. Inkjet printers are great (Epson C82 sucks 1 watt in standby, about 5 when printing.) Larger ethernet hub-like things are about 5 watts as well. (That'd be hubs, terminal adapters, wireless routers. nearly everything in my 3com officeconnect stack at my real house is 5W per unit.)
As far as power budgets go, it's the time the devices on that really gets you. An 18W average setup running 24 hours a day is 432 watt-hours, almost half a kilowatt hour. While I have a pretty substantial solar array (1 KW at the moment) during the winter I'm going to see maybe 4Kw-hours per day, so I'm burning nearly 10% of my capacity on just one PC.
So.... you can use your kick-ass system if you want, and if you limit it to one hour a day v. 24 hours a day, you'd be ahead.
BTW, my losses on the total system aren't bad at all. Inverters and batteries are pretty efficient these days. I'd gladly accept a 50% loss on storage and conversion, however, if I got 50% efficiency out of the panels instead of the sub-20% that's typical of solar today.
Apparently the strained silicon technology came about due to research related to mechanical stress problems they were encountering across the entire chip -- so it already was an issue. Their research solved the mechanical stress problem, and they later realized by intentionally localizing the effect they could basically place the strain at individual transistors to improve performance.
Because the effect is localized and controlled it's no longer an issue of concern, AFAIK.
Heat sinks, etc, shouldn't alter the strain at the transistor level. Remember, we are talking about this at the atomic level, so any macro-level strain like a heat sink would have to be pretty substantial to work its way down into the crystal lattice structure to the point of affecting performance. (Sort of humorous if it did, though, as it would imply microprocessors would go faster if you squeezed them. In reality Intel is actually stretching the size of the normal silicon lattice structure, so heat sink stress (compression) would actually be working against you, but it's also occuring in the wrong axis (the lattice stretching is 2D X-Y, not Z-axis.)
My first PC... TRS-80 Model I, Level I. 4K RAM, Cassette storage.
I worked all summer on a paper route in the late 70s to gather the $379 it took to upgrade to 16K of RAM and Level II, so I could use PEEK and POKE commands to stuff Z-80 machine language into the RAM. Oh, the deal with the upgrade is that they kept the old memory, too.
To put that in perspective, $379 for 4K bytes is 2.3 cents per byte. Many of today's sub-$1,000 PCs have 512MB of RAM. The equivalent cost of that RAM in the late 70s would be $12.3 million dollars. I think that's about what the A-10 aircraft cost at the time.
I usually drag that out when people say PC components are too expensive.
For what it's worth, many years ago when I crossed paths with some cell-phone product design types, there was a hybrid product concieved, originally to improve service and battery life -- a pager/cell phone. (We're not talking SMS here, but plain old POCSAG paging.)
Anyway, with this approach you could work if you wished to retain positional anonymity -- have a conventional pager (which is just a reciever) notify you of calls, then choose to power up the cell or not.
As practically every other post has pointed out, positioning by radio has no requirement of GPS being present. Any transmitter can be position located. Amateur radio opertators actually have contests to do this -- foxhunts -- and the equipment to do position finding of non-spread-spectrum tranmitters is pretty trivial to make or buy.
If you want your whereabouts to remain unknown, don't transmit. Simple as that.
I had always wondered if the spammers actually kept track of whether the mail was actually sent successfully or not. My guess was they didn't, because I have users on my domain that have been gone five years, returning 550 responses the entire time, still getting the same spam they attempt to send to me.
So now we know they're checking this. Obviously they're not using it to change their lists -- so possibly it must be part of the billing side (ie only pay for mails that get successfully delivered.)
Semi-related unsolicited free plug for open-source spam control:
I recently switched from agressive DNSBL spam control to ASSP, a bayesian filter that proxies your mail server. It stopped all those virus-infected zombie mails that were getting through the DNSBLs dead in their tracks (pardon the pun). I can't recommend ASSP highly enough. http://assp.sourceforge.net/
Slashdot Mirror
I've owned the Kyocera 6035 and now own the 7135. These are palm-based phones, and both sync with evolution as well as any other palm. (Which means, they sync, but you run into occasional wierdness like duplicate records.)
The 6035 is kind of clunky (it's a pretty big phone) but due to its size and simple black and white LCD, the battery life was incredible. I could take it on a business trip for a week and not begin to worry about depleting the battery.
The 7135 is cooler, it's a big (but much smaller than the 6035) flip phone with color backlit LCD. Battery life is pretty awful due to the small battery and backlight. Oh, and the batteries occasionally explode. That said, I love it, works well and is a great geek gadget (you can convert MP3s into ringtones, among other things, which with a star trek sound effects MP3 library allows you to attain maximum geekiness and preserve your virginity forever with zero effort.)
Doesn't mean it's right.
Consistently various governments seem to think this fusion of public and private databases is somehow going to answer a problem.
The problem is... the private databases they usually intend to integrate are crap. Some private audits of consumer credit databases turned up error rates of 70 percent. "Minor" errors like wrong country of birth, birthdates, identity, etc. One study had a identifing information error rate -- bad addresses, etc -- of 40%.
So completely separate from the debate of whether this is the right thing to do -- is the fact that what they're doing is braindead in the first place, creating a massive database with bad information.
On second thought... maybe it's better they do this, as it's going to almost immedicately kill the credibility of projects like this if it ever gets tested in court.
Very true. There was a story in Mechanical Design magazine about the fiero before the car came out, and it was a vette killer. A _cheap_ vette killer. So it had the prospect of killing a very profitable car, so GM didn't help it at all.
I had a later mode; (1986) and still regret giving it up 10 years later. The 1984s sucked beyond belief, but as the Forbes story mentions, by the end (1988) is was a really nice car.
The most amazing thing to me was the body. 10 years later, my car didn't have a single ding, due to the flexible plastic body panels.
I notice that a number of people knocking SPF are looking at it breaking some sort of standard, or that it's an exclusive, it's-the-only-answer technology, ie it's being proposed as a silver bullet.
It's not. SPF just provides one more bit of helpful information -- which IPs email from the sender's domain should really be coming from.
While someone could use SPF in a pure binary decision system that breaks SMTP, it's going to be an incomplete solution. Just like blacklists, whitelists, and bayesian filtering are also incomplete solutions.
However, you start using these things in combination and magic happens.
Example: I use ASSP for server-side spam filtering. ASSP uses bayesian filtering, but also whitelists people you email and uses blacklists.
The blacklist implementation is interesting, however, as when it determines an IP is blacklisted it simply starts off with a higher spam probability in the bayesian stage -- it's not truly blacklisted, just more suspicious.
You could do the same thing with SPF, initially giving a lower spam probability to mailservers with SPF, and when there's more infrastructure using SPF, switching to penalizing non-SPF servers.
Nice thing about this approach: it doesn't require everyone to convert their infrastructure, but it does incentivise legitimate servers to do so without penalty. It doesn't break any standards. Legitimate mail still gets through, but spam suffers.
Stop thinking that all spam solutions have to be single silver bullets. Anti-spam tools can be additive.
One more tool against spam == a good thing.
Actually, yes, a 8-bit would scream.
And since the 6502 only took a few thousand transistors, you could fit ~ 32,000 parallel 6502s on a single chip using today's processor manufacturing technology. Running at 2 GHz. Pretty impressive performance possibilities.
Of course, keeping a 256000 bit 6502 VLIW implementation fed with data would be next to impossible. But quite a few specialize components such as video processors do something like this.
Have you considered a postive/negative ground converter like this and a conventional radio?
While in high school and college, I was the beneficiary of an extreme version of this.
I'd routinely stumble on a hard calculus problem; after a bit I'd move on to other problems, come back to the problem one last time, and then go to bed.
I had a recurring dream where there was a book at my workspace I now jokingly refer to as the "Book of all knowledge." In the dream, I'd encounter the problem, open the book like any other reference book, and it would always open to the page with the answer to the problem on it, fully worked out step-by-step as good calculus answers should be.
When I awoke, I knew the answer and essentially copied down the result from the dream, understanding how I got there due to the step-by-step nature of the answer.
Obviously this didn't work all that great for tests.
Seriously, though, it's probably one of my fondest memories of advanced math, having the coolest book in the world in my dreams. I also suspect it points out that problem solving even in very technical domains is a creative, right-brain process and not nearly as logical as people would be led to believe.
As noted, not only are pop-up/under ads a problem, but flash ads are a serious issue in terms of sucking CPU cycles. If you're using older/intentionally slow (low power) hardware, flash ads can drive unusually high CPU utilizations and make pages unusable.
I've stopped using fox news in particular as they seem to have a couple flash advertisers that drive my CPU utilization from about 1-2% normal to 75% when the ad appears -- and in a tabbed-style browser they continue to suck CPU cycles even when I'm not looking at them.
Clearly the default assumption of the people writing the ads in flash is that the installed base of PCs is as fast as the workstation they're writing the ad on, instead of the much slower reality.
recognizing this as a dupe indicates that you have a well-developed social life!
This is what I love about bayesian filtering.
Because it adapts, each new technique the spammers try ends up diluting the effect and ruining it for all spammers. And because they're greedy and will sell each other out without hesitation, it's basically using their own motivations against themselves.
Might as well put in a plug for my favorite bayesian filter: ASSP
As others have pointed out, it's most likely a flat phased array antenna.
There's a couple attributes that would make it attractive for a extraterrestrial application. They're very compact for the gain they provide, and within the limits of the design they can be electronically steered (that is, no moving parts). I would imagine they probably have a mechanical coarse steering mechanism and electronic fine steering.
Sadly I can't seem to find any confirmation of this, just a few mentions of other spacecraft such as MESSENGER using phased array antennas.
If you're really a radio newbie you should know that gain is how well the antenna concentrates the signal. An isotropic radiator basically receives/transmits signals in a perfectly spherical manner. By sacrificing the directional coverage you can increase the gain. A great example is a flashlight bulb -- uncovered it radiates almost everywhere; with a parabolic reflector it radiates a beam. When they talk about using the low gain and high gain antennas they're basically talking about the radiation pattern.
You generally use low gain antennas for signal acquisition when you don't have control over where the antennas are going to be pointed. Once you know where everything is, you can point the high-gain antenna at the target. With more gain you have a better signal-to-noise ratio and can then crank up the data rates.
Phased array antennas work essentially by combining a large number (an array) of simple low-gain antennas such that they add their signals together (in phase) in a particular direction. In other directions the signals don't add the same way and there's much less gain. At microwave frequencies like X-band (about 8 GHz), a simple dipole antenna is only about an inch long, so it's easy to put a bunch of dipole-equivalents in a small space to make an array.
If the IRS were answering, the answer would be yes.
You would be guilty of "structuring," which is organizing financial transactions in such a way as to conceal the true transaction. This would be interesting to them. Presumably to the FBI as well. As far as if this would be noticed statistically, who know what "profile" they're looking for.
IIRC, there was a landmark case in NY about 10 years ago where someone structured a transaction to hide the source of income, but paid taxes on the total amount, and they were found guilty of breaking the structuring law.
No doubt there are numerous replies about it being in Feynman's book.... however:
There are a number of anecdotal references of the exact same story being attributed to Joseph Campbell (of Power of Myth fame)
and more to the point:
A number of native american tribes (particularly the northeastern tribes such as the Iroquois) have creation myths where the world/universe is supported on a turtle's back.
Interestingly enough, in hindu creation myth Vishnu takes the form of a turtle which supports the world's creation on its back.
There are others. The universe being on a turtle's back is a very pervasive myth.
So where it's REALLY came from it likely a religious archetype that predates writing.
In general, yes, shared memory sucks bandwidth. As others pointed out, the calculations are pretty straightforward (X * Y * #bytes/pixel * refresh rate = Bandwidth).
However, in today's systems it's FAR more complicated that this.
First, some older implementations, particularly the Intel 810, used a 4MB display cache. The net of this is that the display refresh was generally served from a secondary memory and didn't interfere with main memory bandwidth. As well, Intel used some technology Chips & Tech developed that basically did run-length encoded compression on the display refresh data (look right at your screen now, there's a LOT of white space, and RLL will shrink that substantially.)
Today most chip sets incorporate a small buffer for the graphics data and compression techniques to minimize the impact of display refresh on bandwidth.
But wait -- it gets even MORE complicated. With integrated graphics on the north bridge of the chip set, the memory controller in the chip set knows both what the CPU and what the graphics core want to access. So the chip set actually does creative scheduling of the memory accesses so that the CPU doesn't get blocked unless absolutely necessary. So most of the time the CPU is either getting its memory needs services by its own cache, or it's getting (apparently) un-blocked access to memory. So the impact of graphics is much less than the simple equation above would suggest.
Finally... we now have dual-channel memory systems. Even more tricks to keep the graphics and CPU memory accesses separate come into play here.
So, the short answer is yes, there's an impact, but it used to be much worse. Innovative design techniques have greatly reduced the impact so that in non-degenerate cases it doesn't affect the system too much. In a degenerate case of your app never getting cached and doing nothing but pound on the memory system with accesses, however, then you'll see the impact in line with the bandwidth equation above.
Given google's IPO situation, and someone I've never heard of touting themselves as being better than google. Maybe this somone is looking for a bit of capitalization themselves?
As another poster mentioned above, ohphone (the OpenH323 app) has been ported as part of the core of XMeeting. Check here for the ohphonex application.
Like several of the others, I would recommend OpenH323 . It's fairly easy to set up.
You also have the option of setting up a gateway (hardware isn't cheap, however.) I've used the voicetronix board. This would give the option of calling locally if your or she had other friends in the area.
Also, I've found Voip far easier to use with real (or semi-real) phone hardware. Cisco has the ATA-186, though it operates best with proprietary codecs. I've ended up using a VoipVoice handset -- it's USB and appears as a USB audio device under linux so it's pretty easy to configure, though it's a little weak on the volume. I got mine through IPblue though I am sure there are other sources/options for open-source IP phones.
Creative labs used to sell the voipblaster, and this is still available from ebay, and that lets you use conventional phone hardware -- it's basically a USB-to-phone-jack converter with a hardware codec, and would likely be idea if you're working with anyone who is technology-challenged.
With the few tech support disasters I've had to deal with, by phone usually has worked best -- but there are tricks to getting it to work, like making sure you take notes of who you talk to so you can reach them again. With some companies there's a decided change in quality of tech support depending on the time of day. (Strangely enough for me, it worked out that calling from 2am-4am has netted the best tech support.)
I strong concur with the recommendation to Google first. I've also found that email tech support pretty much sucks -- I've had an email request to download updated firmware for a 3Com wireless router that's on week 3 of waiting for approval at the moment, and all too many similar experiences.
Like it or not, self-help (google, mailing lists) is pretty much the way to go these days.
As someone who intentionally runs a low-performance box as a primary system (VIA Epia 533) I'd be pretty unhappy with some snarfing up a few cycles. Junked-up web sites with flash and excessive java/javascript are REALLY noticable when you're browsing at the low end of the power curve.
I run a cpu monitor in the background and when a site wants to run one of the more annoying classes of advertisements, utilization usually pegs... I can't imagine what something that intentionally sucked cycles would do.
Well, since I'm not finished yet, cooking hasn't been a big deal. I mostly use a microwave, microwave power consumption "averages to zero" over the course of the day (you maybe run it five minutes at 1 KW, so actually watt-hours are very low.) I've been living in a "simulated" solar mode at my primary house to test stuff out, though.
Hot water is handled by solar. I've got two 8' x 10' solar thermal panels feeding an 80 gallon hot water heater. Presently there's no back-up, but I'll probably swap out the heater I have (electric, but not connected, it's just a useful tank) with a gas model for backup. BTW, single-biggest payoff from a power consumption angle is solar hot water (at least in the desert southwest.) I have to cover one panel in the summer as it heats 80 gallons to better than 180 degrees. In the winter I'm currently seeing temps around 90 when running on both panels (outside air temp ~ 40-45 for the high.) I may add another thermal panel to boost that temp some.
I don't watch much TV there, but I have a portable unit. Even a big-screen TV isn't much of a problem, I have a rear-projection Hitachi unit at my real home that measures 170 watts when on. What's a killer that I haven't dealt with yet but will have to when I stay there and DO watch TV is that satellite recievers are about 35 watts, and expect to be on 24 hours a day. That's bad.
Ironically, the boondocks area is better served with cellular and wireless data services than my house on the edge of the city (Phoenix) is. 1 Mbps wireless is available (it's high plains desert, so one tower covers hundreds of square miles.) They also upgraded the cells in the area to 3G this past summer, so service isn't a problem.
Anyway, moving off-grid isn't cost effective if grid power is available, but the biggest things I've done have been 1) solar hot water (probably a 5 KWh savings per day) 2) compact flourescent lighting (saving 1 KWh or better per day) and moved my servers from ~150 W units to one 50 W unit (fileserver) and one 15 W unit (I use a Epia 533 for a firewall/web/mail server). Over 24 hours the power savings adds up on PCs.
I'm in the process of building (year #5, woo-hoo!) a totally off-grid home in (almost) the middle of nowhere. I've gone through this already.
The short answer is if power is all you care about, a notebook is better. My Sony TR1A consumes 13 watts of AC when plugged in at full-tilt according to my power meter, and display-off sucks just 5 watts. However, I care about more than power, like uptime/durability, ability to use off-the-shelf components, and being able to support some modest external hardware, so the actual system that stays on 24/7 is a VIA Epia 533 MHz box. It takes about 11-12 watts but can peak around 24 watts or so, plus a 15-inch LCD monitor which is pretty much never on, but consumes 16 watts when it is; so you're looking at about 30 watts with one of the lowest power desktop configurations possible.
Other odds and ends consume power as well. Inkjet printers are great (Epson C82 sucks 1 watt in standby, about 5 when printing.) Larger ethernet hub-like things are about 5 watts as well. (That'd be hubs, terminal adapters, wireless routers. nearly everything in my 3com officeconnect stack at my real house is 5W per unit.)
As far as power budgets go, it's the time the devices on that really gets you. An 18W average setup running 24 hours a day is 432 watt-hours, almost half a kilowatt hour. While I have a pretty substantial solar array (1 KW at the moment) during the winter I'm going to see maybe 4Kw-hours per day, so I'm burning nearly 10% of my capacity on just one PC.
So.... you can use your kick-ass system if you want, and if you limit it to one hour a day v. 24 hours a day, you'd be ahead.
BTW, my losses on the total system aren't bad at all. Inverters and batteries are pretty efficient these days. I'd gladly accept a 50% loss on storage and conversion, however, if I got 50% efficiency out of the panels instead of the sub-20% that's typical of solar today.
I spoke with Intel about this in the spring.
Apparently the strained silicon technology came about due to research related to mechanical stress problems they were encountering across the entire chip -- so it already was an issue. Their research solved the mechanical stress problem, and they later realized by intentionally localizing the effect they could basically place the strain at individual transistors to improve performance.
Because the effect is localized and controlled it's no longer an issue of concern, AFAIK.
Heat sinks, etc, shouldn't alter the strain at the transistor level. Remember, we are talking about this at the atomic level, so any macro-level strain like a heat sink would have to be pretty substantial to work its way down into the crystal lattice structure to the point of affecting performance. (Sort of humorous if it did, though, as it would imply microprocessors would go faster if you squeezed them. In reality Intel is actually stretching the size of the normal silicon lattice structure, so heat sink stress (compression) would actually be working against you, but it's also occuring in the wrong axis (the lattice stretching is 2D X-Y, not Z-axis.)
I worked all summer on a paper route in the late 70s to gather the $379 it took to upgrade to 16K of RAM and Level II, so I could use PEEK and POKE commands to stuff Z-80 machine language into the RAM. Oh, the deal with the upgrade is that they kept the old memory, too.
To put that in perspective, $379 for 4K bytes is 2.3 cents per byte. Many of today's sub-$1,000 PCs have 512MB of RAM. The equivalent cost of that RAM in the late 70s would be $12.3 million dollars. I think that's about what the A-10 aircraft cost at the time.
I usually drag that out when people say PC components are too expensive.
For what it's worth, many years ago when I crossed paths with some cell-phone product design types, there was a hybrid product concieved, originally to improve service and battery life -- a pager/cell phone. (We're not talking SMS here, but plain old POCSAG paging.)
Anyway, with this approach you could work if you wished to retain positional anonymity -- have a conventional pager (which is just a reciever) notify you of calls, then choose to power up the cell or not.
As practically every other post has pointed out, positioning by radio has no requirement of GPS being present. Any transmitter can be position located. Amateur radio opertators actually have contests to do this -- foxhunts -- and the equipment to do position finding of non-spread-spectrum tranmitters is pretty trivial to make or buy.
If you want your whereabouts to remain unknown, don't transmit. Simple as that.
I had always wondered if the spammers actually kept track of whether the mail was actually sent successfully or not. My guess was they didn't, because I have users on my domain that have been gone five years, returning 550 responses the entire time, still getting the same spam they attempt to send to me.
So now we know they're checking this. Obviously they're not using it to change their lists -- so possibly it must be part of the billing side (ie only pay for mails that get successfully delivered.)
Semi-related unsolicited free plug for open-source spam control:
I recently switched from agressive DNSBL spam control to ASSP, a bayesian filter that proxies your mail server. It stopped all those virus-infected zombie mails that were getting through the DNSBLs dead in their tracks (pardon the pun). I can't recommend ASSP highly enough. http://assp.sourceforge.net/