It was test code, written as a throw away proof of concept, on prototype hardware, run on it's own without anything else on the system, that had been grabbed, copied, reused several times and then tried to push into a the real hardware with the full system running at the same time. I did more than one double take when we found it, and again when I saw the real code.
Moral of the story: all code is production code, there is no such thing as throw-away one-off test code. Although that doesn't seem to get through to program managers who write schedules so the cycle repeats. If someone gets a thing working, someone somewhere will think "it worked in the lab, ship it!"
Of course it's a problem. My point was that for all practical issues, engineering difficulties are what imposes the limits before you have to worry about quantum effects.
Yes, it's a property of the DFT. Or more generally Information uncertainty is a property of all discrete systems which sample at fixed intervals, that's what nyquist is all about. If you didn't sample it, you can't know it if there's data outside of your sampled signal's bandwidth it can alias down onto your desired data and cause all kinds of problems.
The post I replied to implied there was something more going on at some fundamental physical level. At the quantum level quantum stuff is happening, sure. In quantum mechanics the uncertainty principle is expressed with a wave function which states the probability of the state of a particle. We don't talk about quantum uncertainties in digital signal processing.
I don't have the textbook you mention so you might provide a better source or link. The stuff I find by googling are not things we deal with in the digital domain. When you start including quantum wave function probabilities in your signal processing algorithms, let me know.:)
Fair enough, you can also under sample, and then use the aliased image if you know that the frequency range the alias lands on does not contain data. As you say, a bandwidth problem. But in the context of the original statement, it isn't the time length of the sample set, so much as the number of samples per unit time that determines FFT fidelity, time is part of it, but it's not independent.
There are costs and consequences for any particular hardware choice. Modern MP3 players (and/or phones) need a general purpose CPU for many of the things they do and then use that to do decode since a separate extra part is the worse option. If the device didn't already HAVE a general purpose CPU, then an arm core is probably not the best option to solve an individual DSP problem.
I think the parent is getting at a problem I see at my job fairly frequently. Professional Firmware and DSP developers that have no knowledge about the effects or consequence of their code on the hardware itself. Ultimately re-implementing something that used to run fine in 1/10th the ram, and 1/40th the processor cycles (or less) in a brand new dual core DSP and still can't figure out why they aren't meeting performance requirements. I've seen these guys demand a bigger, faster, more power hungry CPU to run their code when a previous version that was literally a fraction of the capability ran the same algorithm.
True Story: I'm an FPGA dev, and I was helping out software guys debug why the DSP was completely missing messages and seemed to be dropping data. Turns out doing a thousand lines of heavy DSP INSIDE the interrupt handler is probably going to hang up the CPU and cause it to miss interrupts. The worst part was the blank stares we got when we told them this, and had to explain why that was a bad thing.
It seems the idea of actually writing good, fast, and power efficient code is a dying art because it's perceived as 'not necessary'. In some instances this thinking is correct (e.g. desktop computers), but in the embedded space where you're trying to keep the cost of final products down and minimize power usage, grabbing the fastest DSP part that TI makes is probably not always the best solution... but it's what you get when you think it's automatically OK/good to 'do everything in software'. There are costs and penalties to account for.
For pure hobbyists, none of this really applies, but I suspect that a good many of the people that do this stuff as a hobby are also at least tangentially involved in it in some way at their jobs, so it serves to bring up the considerations.
What? Engineering difficulties occur in the sampling rates, the accuracy and noise floor of the A/D, and the hardware WAY before any quantum effects enter the picture. For FFT accuracy, the fundamental math is information theory and Nyquist.
More accurately, the number of frequency bins an FFT has is based on the number of samples you start with, which is based on the sampling rate times time, which ultimately gets back to Nyquist. You can only represent frequencies in the FFT that are within the nyquist rate, and an FFT has a number of frequency bins related to the number of samples being processed. The actual physical time of the sample window is a fall out of that combination of factors. Crank up the sampling rate over they same time period and you get more frequency bins and more accurate FFTs, contrary to what some of the other posters are going on about.
You are right, not everything is ABOUT politics explicitly, but given the state of patents and copyright and all the laws in the works that may affect them, even a new device/chip announcement is heavily influenced by the politics of the day. Even corporations which develop our vaunted tech know this which is why so many are heavily into the lobbying business. Whether you choose to be informed about that is up to you, but it's hard to separate the water from the soup. As an engineer, the why is just as important to me as the what, and I appreciate the additional commentary.
My question to you would be: How do you define quality? Optical lens quality and overall camera quality, auto-focus, shutter-lag, etc? And at such a low resolutions many of those factors are remarkably forgiving. You don't need that good a lens to get decent in focus pictures at (1152x864), it also depends heavily on what you're taking pictures of and where (indoors using lots of flash, and tungsten lighting, outdoors with natural light, etc.), and it was also only any good if you were printing no larger than 4x6. Even if it was the best digital point and shoot camera on the market at that time, the market moved forward and Kodak did not. I think one of my relatives had that camera actually, if I recall the shutter lag was horrendous making it virtually impossible to take anything but still photos. Granted most PHD digital cameras of the time had similar faults, but again that's why they progressed while Kodak lagged behind.
??? How's that happen, typically the lower the process size, the lower the energy use for the same design (and the higher possible clock speed before heat issues occur).
Not strictly true. The switching current is what goes down, but leakage current goes up. For older technologies that power threshold was recently crossed, that's why each new process step isn't just about feature size, but new tech to reduce leakage current so the power draw doesn't go insane. So reducing feature size, particularly at the scales we have now have real leakage current issues that need to be addressed at each step. High-K metal gates at 32nm was key to address this. And now Intel has the 3D Tri-Gate that helps reduce leakage current. I should note that these technologies also help increase/maintain switching speeds at the lower voltages/currents/feature sizes as well, which is a big selling point, but power is a huge part of that equation, particularly now in the mobile space.
ANY barriers to entry, legal, capital, physical, destroy the assumptions implicit in a free market, and thus turn it into a detriment to society. The whole system is flawed that way. I don't know why you get hung up on legal type barriers to entry into a market when there are so many others which are larger, more significant and more detrimental to a functioning (non-monopolistic) economy.
Free market capitalism isn't some some kind of ideal or philosophical paradigm to be pursued in it's own right. Looking at the US economy BEFORE the Sherman anti-trust act we see it's great for a few guys to get incomprehensibly wealthy (see: Robber Barons, followed by the fallout exposed by the 'muckrakers'). I'd like to point out this included deploying private armed men to shoot people (look at the early history of the labor movement). Power is power, whether it's the legal cudgel or the physical one doesn't matter. Monopolies make power easier to maintain so they are the natural result of any unrestricted economic system.
Free market is a tool society uses to make society better. Unrestricted free market doesn't do that, so we don't practice it. Whether you like or not, anarchy doesn't work. However, the free-ish market is a very useful tool for distribution of limited resources when certain assumptions are valid. Under those assumptions the market is a wonderful and powerful tool capable of frankly astonish feats. Without those assumptions it's an unstable system, and just like in control theory, a little negative feedback makes all the difference.
I've not heard any good arguments for the complete abolition of all IP laws. Instinctively I favor that ideal knowing the power of information flow, but then you have nothing protecting the little guys and history is replete with the consequences if we care to look. So a middle ground is sensible, there are some really good arguments for drastically limiting the scope of the artificial monopoly which is granted by IP laws.
Having lived in Rochester and worked at Kodak I'm quite saddened by the current state of the company, although it's a fate completely self inflicted. As for the 'government attacking business' line, Kodak deserved it's anti-trust come-uppance. You didn't mention Kodak also got barred from selling private label film for anti-trust reasons (1921). As well they were barred from mandating Kodak processing for Kodak films (1954). (just like Ford can't force you legally to go to for for repair service) a short summary of Kodak's history with the Sherman Antitrust Act
In short, Kodak did some illegal things and got busted. Stuff that's illegal for very very good reasons. However, Kodak ultimately failed because they lacked foresight and made bad business decisions. Here's a Forbes article citing a few cases: How Success Killed Eastman Kodak
The ultimate Irony that will probably be lost on you is that based on free market principles Kodak is a perfect example of the market doing it's job and punishing a company that has not kept up and is no longer producing valuable products services that people want to buy. By EVERY measure they deserve their fate. If they HAD NOT been 'attacked by the ebil gubment' this fate would probably have been avoided, and we probably wouldn't even have digital imaging now because they would have invented it, sat on it and/or prevented digital cameras from being imported. When I think about the amount of astronomy tech and science, defense tech, consumer imaging (camera phones, etc), professional imaging that would likely have been prevented or severely retarded (as in slowed down) by allowing Kodak to throw it's weight around owning all imaging/picture taking back in the 20's and 50's, I'm in fact quite grateful that we prevented a monopolist from abusing it's position.
Imagine where we would be now if we continued with that policy instead of abandoning it. *sigh*
If I recall correctly it had more to do with some arbitrary and insane insistence on 'Consumer Imaging' being the business focus, which is why you got cheap consumer cameras (easy share), printer docs (with attempts to cash in on printer paper consumables), but little pro-sumer stuff, and the occasional/rare super high-end imagers/gear (like those used in telescopes, etc).
This is also why they sold off/spun off their profitable medical imaging groups, chemicals group, and they've tried to get rid of their profitable Document Imaging group (high-end, high-speed document scanners) several times. They've been constantly trying to push themselves into the most difficult and price-competitive market possible, cheapo consumer cameras. I think the ultimate goal was to maintain some kind of grasp of the photo printing business as their cash cow with consumable manufacturing/selling. To be fair, they still do a good job printing pictures, but people don't really want/need to do that anymore with rare exceptions. And people that still do prints do it in-house or have local labs that do the work.
Ok, pet peave of mine. That's a terrible phrase, and Cecil will tell you why. Basically it boils down to this example: If you see a no U-Turn sign that means that everywhere you do NOT see explicitly prohibiting you from making U-turns, U-turns are legal/OK. The exception proves the counter-rule.
Are any researchers really looking for skilled FPGA designers? It's a pretty dedicated skillset, but work on research would be more interesting than my current job. Also, it should be noted that the devices themselves (FPGAs), and the tools needed for the design flow (particularly synthesis tools) are expensive, and computationally intensive in and of themselves. Unfortunately we don't have the open source tools of the software world available to us.
Sure there's a reasonable alternative. Nobody is allowed to donate. Campaign sizes are fixed, and provided for. Fixed and equal amounts of airtime/debate time for everyone who gets enough signatures. Equating monetary donations to speech is where the problem starts. I don't necessarily think that's wrong, but it opens too many floodgates that you can't really close in an equitable manner.
If they were being designed, built, and marketed to be affordable by individuals, why yes, yes you should. That's a big if, but it serves to point out the false equivalency in your statement.
Machines need to be cleaned, fed, maintained, grounds need to be patrolled, locked, cleaned, gardened. Those folks need cars, which need washing. They need food which needs preparing. The factory machines need upgrading/designing/replacing/installing. New buildings need building, their driveways need paving/repaving/repainting. I think you understimate the value of having the manufacturing here. All the profit dollars that company's product makes that go to maintaining the plant and it's surrounds get directly fed to the local economy... vs someplace else. Those people can then go buy the products they participate in making. (Hat tip to good ol' Henry Ford who at the very least understood that basic principle)
I was more addressing the implication in the OP that the PC sim market was faltering/missing in general rather than just stereoscopic 3D. I did miss that though, I read 3D and think, well we've had 3D since what.. quake3d? or earlier.:)
I wonder what nVidia/AMD's stereoscopic 3D extrapolation tech would do with something like the ED flight sims.
Check out Eagle Dynamics (http://www.digitalcombatsimulator.com). Full KA-50 Blackshark (attack helicopter) and A-10C Warthog study sims. Absolutely mind-boggling levels of fidelity, (which you can tone down by turning on various assists, at least in blackshark). They also have a lower fidelity multi-aircraft package called Lock On: Modern Air Combat (with a Flaming Expansion). Currently Blackshark and Flaming Cliffs share multiplayer, and can play online together. There's some more expansions coming to bring all their sims into the multiplayer space, as well as an as yet unspecified jet fighter study sim.
All in all, if you're a modern flight sim fan of any kind, you owe it to yourself to check them out.
A) many diseases aren't vaccinated against so hardly the point
B) Vaccines don't have 100% efficacy, and there is some small portion of the population which can't get vaccinated (not to mention children and pregnant women) who compose the high risk group. This is is why herd immunity is important.
Slashdot Mirror
hehe, indeed. Cut'n'pasty code reuse no-design fail.
It was test code, written as a throw away proof of concept, on prototype hardware, run on it's own without anything else on the system, that had been grabbed, copied, reused several times and then tried to push into a the real hardware with the full system running at the same time. I did more than one double take when we found it, and again when I saw the real code.
Moral of the story: all code is production code, there is no such thing as throw-away one-off test code. Although that doesn't seem to get through to program managers who write schedules so the cycle repeats. If someone gets a thing working, someone somewhere will think "it worked in the lab, ship it!"
Of course it's a problem. My point was that for all practical issues, engineering difficulties are what imposes the limits before you have to worry about quantum effects.
Yes, it's a property of the DFT. Or more generally Information uncertainty is a property of all discrete systems which sample at fixed intervals, that's what nyquist is all about. If you didn't sample it, you can't know it if there's data outside of your sampled signal's bandwidth it can alias down onto your desired data and cause all kinds of problems.
The post I replied to implied there was something more going on at some fundamental physical level. At the quantum level quantum stuff is happening, sure. In quantum mechanics the uncertainty principle is expressed with a wave function which states the probability of the state of a particle. We don't talk about quantum uncertainties in digital signal processing.
I don't have the textbook you mention so you might provide a better source or link. The stuff I find by googling are not things we deal with in the digital domain. When you start including quantum wave function probabilities in your signal processing algorithms, let me know. :)
Fair enough, you can also under sample, and then use the aliased image if you know that the frequency range the alias lands on does not contain data. As you say, a bandwidth problem. But in the context of the original statement, it isn't the time length of the sample set, so much as the number of samples per unit time that determines FFT fidelity, time is part of it, but it's not independent.
There are costs and consequences for any particular hardware choice. Modern MP3 players (and/or phones) need a general purpose CPU for many of the things they do and then use that to do decode since a separate extra part is the worse option. If the device didn't already HAVE a general purpose CPU, then an arm core is probably not the best option to solve an individual DSP problem.
I think the parent is getting at a problem I see at my job fairly frequently. Professional Firmware and DSP developers that have no knowledge about the effects or consequence of their code on the hardware itself. Ultimately re-implementing something that used to run fine in 1/10th the ram, and 1/40th the processor cycles (or less) in a brand new dual core DSP and still can't figure out why they aren't meeting performance requirements. I've seen these guys demand a bigger, faster, more power hungry CPU to run their code when a previous version that was literally a fraction of the capability ran the same algorithm.
True Story: I'm an FPGA dev, and I was helping out software guys debug why the DSP was completely missing messages and seemed to be dropping data. Turns out doing a thousand lines of heavy DSP INSIDE the interrupt handler is probably going to hang up the CPU and cause it to miss interrupts. The worst part was the blank stares we got when we told them this, and had to explain why that was a bad thing.
It seems the idea of actually writing good, fast, and power efficient code is a dying art because it's perceived as 'not necessary'. In some instances this thinking is correct (e.g. desktop computers), but in the embedded space where you're trying to keep the cost of final products down and minimize power usage, grabbing the fastest DSP part that TI makes is probably not always the best solution... but it's what you get when you think it's automatically OK/good to 'do everything in software'. There are costs and penalties to account for.
For pure hobbyists, none of this really applies, but I suspect that a good many of the people that do this stuff as a hobby are also at least tangentially involved in it in some way at their jobs, so it serves to bring up the considerations.
What? Engineering difficulties occur in the sampling rates, the accuracy and noise floor of the A/D, and the hardware WAY before any quantum effects enter the picture. For FFT accuracy, the fundamental math is information theory and Nyquist.
More accurately, the number of frequency bins an FFT has is based on the number of samples you start with, which is based on the sampling rate times time, which ultimately gets back to Nyquist. You can only represent frequencies in the FFT that are within the nyquist rate, and an FFT has a number of frequency bins related to the number of samples being processed. The actual physical time of the sample window is a fall out of that combination of factors. Crank up the sampling rate over they same time period and you get more frequency bins and more accurate FFTs, contrary to what some of the other posters are going on about.
Aliasing can be avoided by having a high enough sampling frequency.
You are right, not everything is ABOUT politics explicitly, but given the state of patents and copyright and all the laws in the works that may affect them, even a new device/chip announcement is heavily influenced by the politics of the day. Even corporations which develop our vaunted tech know this which is why so many are heavily into the lobbying business. Whether you choose to be informed about that is up to you, but it's hard to separate the water from the soup. As an engineer, the why is just as important to me as the what, and I appreciate the additional commentary.
My question to you would be: How do you define quality? Optical lens quality and overall camera quality, auto-focus, shutter-lag, etc? And at such a low resolutions many of those factors are remarkably forgiving. You don't need that good a lens to get decent in focus pictures at (1152x864), it also depends heavily on what you're taking pictures of and where (indoors using lots of flash, and tungsten lighting, outdoors with natural light, etc.), and it was also only any good if you were printing no larger than 4x6. Even if it was the best digital point and shoot camera on the market at that time, the market moved forward and Kodak did not. I think one of my relatives had that camera actually, if I recall the shutter lag was horrendous making it virtually impossible to take anything but still photos. Granted most PHD digital cameras of the time had similar faults, but again that's why they progressed while Kodak lagged behind.
Thanks. :) It took me a while, but I think I caught on eventually.
??? How's that happen, typically the lower the process size, the lower the energy use for the same design (and the higher possible clock speed before heat issues occur).
Not strictly true. The switching current is what goes down, but leakage current goes up. For older technologies that power threshold was recently crossed, that's why each new process step isn't just about feature size, but new tech to reduce leakage current so the power draw doesn't go insane. So reducing feature size, particularly at the scales we have now have real leakage current issues that need to be addressed at each step. High-K metal gates at 32nm was key to address this. And now Intel has the 3D Tri-Gate that helps reduce leakage current. I should note that these technologies also help increase/maintain switching speeds at the lower voltages/currents/feature sizes as well, which is a big selling point, but power is a huge part of that equation, particularly now in the mobile space.
ANY barriers to entry, legal, capital, physical, destroy the assumptions implicit in a free market, and thus turn it into a detriment to society. The whole system is flawed that way. I don't know why you get hung up on legal type barriers to entry into a market when there are so many others which are larger, more significant and more detrimental to a functioning (non-monopolistic) economy.
*sigh*
Free market capitalism isn't some some kind of ideal or philosophical paradigm to be pursued in it's own right. Looking at the US economy BEFORE the Sherman anti-trust act we see it's great for a few guys to get incomprehensibly wealthy (see: Robber Barons, followed by the fallout exposed by the 'muckrakers'). I'd like to point out this included deploying private armed men to shoot people (look at the early history of the labor movement). Power is power, whether it's the legal cudgel or the physical one doesn't matter. Monopolies make power easier to maintain so they are the natural result of any unrestricted economic system.
Free market is a tool society uses to make society better. Unrestricted free market doesn't do that, so we don't practice it. Whether you like or not, anarchy doesn't work. However, the free-ish market is a very useful tool for distribution of limited resources when certain assumptions are valid. Under those assumptions the market is a wonderful and powerful tool capable of frankly astonish feats. Without those assumptions it's an unstable system, and just like in control theory, a little negative feedback makes all the difference.
I've not heard any good arguments for the complete abolition of all IP laws. Instinctively I favor that ideal knowing the power of information flow, but then you have nothing protecting the little guys and history is replete with the consequences if we care to look. So a middle ground is sensible, there are some really good arguments for drastically limiting the scope of the artificial monopoly which is granted by IP laws.
Having lived in Rochester and worked at Kodak I'm quite saddened by the current state of the company, although it's a fate completely self inflicted. As for the 'government attacking business' line, Kodak deserved it's anti-trust come-uppance. You didn't mention Kodak also got barred from selling private label film for anti-trust reasons (1921). As well they were barred from mandating Kodak processing for Kodak films (1954). (just like Ford can't force you legally to go to for for repair service) a short summary of Kodak's history with the Sherman Antitrust Act
In short, Kodak did some illegal things and got busted. Stuff that's illegal for very very good reasons. However, Kodak ultimately failed because they lacked foresight and made bad business decisions. Here's a Forbes article citing a few cases: How Success Killed Eastman Kodak
The ultimate Irony that will probably be lost on you is that based on free market principles Kodak is a perfect example of the market doing it's job and punishing a company that has not kept up and is no longer producing valuable products services that people want to buy. By EVERY measure they deserve their fate. If they HAD NOT been 'attacked by the ebil gubment' this fate would probably have been avoided, and we probably wouldn't even have digital imaging now because they would have invented it, sat on it and/or prevented digital cameras from being imported. When I think about the amount of astronomy tech and science, defense tech, consumer imaging (camera phones, etc), professional imaging that would likely have been prevented or severely retarded (as in slowed down) by allowing Kodak to throw it's weight around owning all imaging/picture taking back in the 20's and 50's, I'm in fact quite grateful that we prevented a monopolist from abusing it's position.
Imagine where we would be now if we continued with that policy instead of abandoning it. *sigh*
If I recall correctly it had more to do with some arbitrary and insane insistence on 'Consumer Imaging' being the business focus, which is why you got cheap consumer cameras (easy share), printer docs (with attempts to cash in on printer paper consumables), but little pro-sumer stuff, and the occasional/rare super high-end imagers/gear (like those used in telescopes, etc).
This is also why they sold off/spun off their profitable medical imaging groups, chemicals group, and they've tried to get rid of their profitable Document Imaging group (high-end, high-speed document scanners) several times. They've been constantly trying to push themselves into the most difficult and price-competitive market possible, cheapo consumer cameras. I think the ultimate goal was to maintain some kind of grasp of the photo printing business as their cash cow with consumable manufacturing/selling. To be fair, they still do a good job printing pictures, but people don't really want/need to do that anymore with rare exceptions. And people that still do prints do it in-house or have local labs that do the work.
the exceptions that prove the rule
Ok, pet peave of mine. That's a terrible phrase, and Cecil will tell you why. Basically it boils down to this example: If you see a no U-Turn sign that means that everywhere you do NOT see explicitly prohibiting you from making U-turns, U-turns are legal/OK. The exception proves the counter-rule.
Are any researchers really looking for skilled FPGA designers? It's a pretty dedicated skillset, but work on research would be more interesting than my current job. Also, it should be noted that the devices themselves (FPGAs), and the tools needed for the design flow (particularly synthesis tools) are expensive, and computationally intensive in and of themselves. Unfortunately we don't have the open source tools of the software world available to us.
Indeed, and more importantly, preventing it from happening again.
Sure there's a reasonable alternative. Nobody is allowed to donate. Campaign sizes are fixed, and provided for. Fixed and equal amounts of airtime/debate time for everyone who gets enough signatures. Equating monetary donations to speech is where the problem starts. I don't necessarily think that's wrong, but it opens too many floodgates that you can't really close in an equitable manner.
If they were being designed, built, and marketed to be affordable by individuals, why yes, yes you should. That's a big if, but it serves to point out the false equivalency in your statement.
Machines need to be cleaned, fed, maintained, grounds need to be patrolled, locked, cleaned, gardened. Those folks need cars, which need washing. They need food which needs preparing. The factory machines need upgrading/designing/replacing/installing. New buildings need building, their driveways need paving/repaving/repainting. I think you understimate the value of having the manufacturing here. All the profit dollars that company's product makes that go to maintaining the plant and it's surrounds get directly fed to the local economy... vs someplace else. Those people can then go buy the products they participate in making. (Hat tip to good ol' Henry Ford who at the very least understood that basic principle)
I was more addressing the implication in the OP that the PC sim market was faltering/missing in general rather than just stereoscopic 3D. I did miss that though, I read 3D and think, well we've had 3D since what.. quake3d? or earlier. :)
I wonder what nVidia/AMD's stereoscopic 3D extrapolation tech would do with something like the ED flight sims.
Check out Eagle Dynamics (http://www.digitalcombatsimulator.com). Full KA-50 Blackshark (attack helicopter) and A-10C Warthog study sims. Absolutely mind-boggling levels of fidelity, (which you can tone down by turning on various assists, at least in blackshark). They also have a lower fidelity multi-aircraft package called Lock On: Modern Air Combat (with a Flaming Expansion). Currently Blackshark and Flaming Cliffs share multiplayer, and can play online together. There's some more expansions coming to bring all their sims into the multiplayer space, as well as an as yet unspecified jet fighter study sim.
All in all, if you're a modern flight sim fan of any kind, you owe it to yourself to check them out.
A) many diseases aren't vaccinated against so hardly the point
B) Vaccines don't have 100% efficacy, and there is some small portion of the population which can't get vaccinated (not to mention children and pregnant women) who compose the high risk group. This is is why herd immunity is important.