That's funny, because an md5 checksum is always 128 bits, i.e. 32 bytes. Read the bottom of the page: info hash: SHA1 hash of the "info" section of the metainfo (*.torrent) Try "sha1sum" next time, and on the correct file.
Am I the only one that notices that Apple's response to every problem is a swift "let's delete this topic and pretend the problem doesn't exist"?
From the forums Terms of Service: Post constructive comments and questions. Unless otherwise noted, your Submission should either be a technical support question or a technical support answer. Constructive feedback about product features is welcome as well. If your Submission contains the phrase "Im sorry for the rant, but" you are likely in violation of this policy.
We cannot see what the deleted discussion looked like, but I think a topic starter like "How do I secure my iDisk?" is much less likely to be deleted than "Apple's iDisk has horrible security", even if the former leads to a heated discussion.
So if you want someone using Gmail to delete an email exchange they had with you, send them an additional message in the same thread offering to sell them Viagra.
It's actually quite hard to write a message that will trip good spam filters (e.g. Spamassasin, Gmail). Just mentioning a couple of keywords such as "viagra penis enlargement nigeria" will usually not work (you need either a LOT of "bad words", or send from a blacklisted IP). There is a special string that is recognized by the major spam filters: XJS*C4JDBQADN1.NSBN3*2IDNEN*GTUBE-STANDARD-ANTI-UBE-TEST-EMAIL*C.34X . See Wikipedia GTUBE. The problem is, if you're caught pulling that trick, it'll be hard to explain how you accidentally put that string in your email.
The big issue I have with DC is that its a potential safety hazard. Unlike AC which you'll feel if you touch one of the wires,... And for many years that's how electricians figured out if a wire was hot, they'd touch it, ensuring that they weren't grounded,
I can assure you from plenty of personal experience that you don't feel it when you touch 230 VAC while not grounded. Wearing shoes with rubber soles or standing on a wooden floor is enough isolation. (Bare-feeted on ceramic tiles is quite painful, though) It always scares the crap out of people when I demonstrate how I stick paperclips into the wall sockets...
As for electricians, screwdrivers with small built-in neon lights have existed for as long as I can remember. The current they need to light up is too small to feel, except if you ground yourself (it tingles a bit and the light gets brighter when you do that).
I have my AP open too, but I don't want to risk trouble with my ISP if somebody connects with a zombied computer that's spamming or sending viruses. The AP is built into a Thomson/Speedtouch ADSL modem (very common in Netherlands) that has some configurable firewalling. So I only allow HTTP, HTTPS, POP, IMAP, VPN, SSH and block everything directed towards my home network. SMTP is not meaningful since the configured server of a AP user is most likely not that of my ISP, which is probably the only mailserver that they can reach anyway. Of course, a determined and knowledgeable person might be able to hijack an IP address that is supposed to be part of my internal network and circumvent the firewall, but I feel more comfortable this way.
Oh, and when I'm bored and notice that someone is using my wifi, I fire up a wireless packet sniffer and see what kind of websites and email they are dealing with.:-)
It is far more eco-efficient to keep on driving with the old car than to buy a new one. Read about MIPS (Material Input per Service Unit) from here and here.
Would you mind giving me a more specific reference that, like, gives a table with meaningful numbers?
It makes quite a difference whether it's $500 in electricity or $500 of coal in an iron smelter.
What if the electricity is generated at a coal burning facility?
The price of raw coal is a minute fraction of the price of electricity. Coal combustion is about 30 MJ/kg = 8 kWh, or 3 kWh if you count the efficiency of a plant. Coal costs about US$ 50 per tonne [ref], so about US$ 0.017 per kWh electricity out of the $0.10--$0.20 you pay for your electricity. To put it differently, $500 in electricity produces as much CO2 as $50 in raw coal.
But that new car, even if it gets twice the MPG, costs energy to make--would an extra 20mpg offset the energy cost of making the car, and if so, how long would it take? Money aside, I don't know whether to keep the beater (which gets about 20mpg) or get a newer car.
many phone charges are about.2. That means for every watt delivered to the phone, there line losses in the grid are at least 3 W if not more.
I don't think that's how it works. The power factor mainly acts on the transmission losses, which are about 10% overall. So 1 W @ pf=0.2 means that the transmission losses are 10% * 1 W / 0.2 = 0.5 W in the worst case. It does not mean that the generators have to produce 5 watts of which 4 watts turn into heat.
You can't pull that much power in such a short time without blowing a fuse or other various consequences of a fairly large draw over a short time.
The GP mentioned equipment like phones and PDAs, which typically have a battery of about 700 mAh at 3.7 V = 9000 Ws. Household wiring in Europe can usually do 16 amps at 230 V = 3680 W. With a factor 2 for conversion losses and power factor it means you could recharge the battery in 5 seconds or so. The biggest problem is that you need some kind of switched power supply with 10 times more capacity than that in a desktop pc, so that would be a pretty big and expensive battery charger.
Re:You have no idea what you are talking about
on
GIMP 2.4 Released
·
· Score: 4, Informative
The reason for more than 8 bits is for processing in the digital realm. For instance if your picture is 1/4 as bright as you want it, and you multiply by 4, then you lose two bits of resolution (as the bottom 2 will be zero).
There are other advantages of 16+ bits. 8-bit RGB images are usually in sRGB space, which means that the luminance of a pixel is not proportional to the pixel value, but rather something like the 2.2'th power except for a small range near zero. That is convenient for encoding a large contrast range in just 256 values, but sucks for operations that are inherently linear operators on the luminance, such as background substraction and blurring. With 16+ bits, all operations can be done in linear space without loss of resolution at the darker colors.
When did Amazon "invent" it though? It was obviously invented before it was patented, but how long before?
The filing date was August 2004; in the American patent law, you have to file within one year after the first public disclosure. (In Europe, you can't file a patent at all after a public disclosure.) I think I've seen virtual URL spaces before August 2003, but I'm not sure about search requests. Note that the patent specifically claims having the search terms directly at the root of the web site, e.g. www.example.com/search+terms , unless it is specifically not a search term (e.g. www.example.com/welcome.html) For example, Wikipedia redirects URLs like http://en.wikipedia.org/url to the URL article, but I don't know when they started doing that. They were certainly already doing that in 2004.
MS paid it: http://www.news.com/2100-1014_3-5255715.html ; I read about the bank account in the newspaper (as in: the paper version), as a side note to an interview with EC commissioner Kroes. Sorry, no link.
"For evaluation purposes only"? Uh, does it imply that these documentations provided do no even need give enough information for a _functional_ implementation, or am I too cynical?
I didn't copy the whole text (as indicated by [...]), so read the original if you're unsure. I interpreted it as: MS should provide free excerpts to prospective buyers of the specs for evaluation purposes to make sure that the buyer knows what he is going to spend his money on.
1.What exactly does this cover? Which network protocols? Which data formats?
See the EC ruling (PDF), especially article 999 on page 277:
(999) Microsoft should be ordered to disclose complete and accurate specifications for the protocols used by Windows work group servers in order to provide file, print and group and user administration services to Windows work group networks. This includes both direct interconnection and interaction between a Windows work group server and a Windows client PC, as well as interconnection and interaction between a Windows work group server and a Windows client PC that is indirect and passes through another Windows work group server. The use of the term specifications makes clear that Microsoft should not be required to disclose its own implementation of these specifications, that is to say, its own source code. The term protocol relates to the rules of interconnection and interaction between instances of the Windows client PC operating system and the Windows work group server operating system.
Also interesting:
(1008) The requirement for the terms imposed by Microsoft to be reasonable and non- discriminatory applies in particular: [...] there is a need to ensure that potential beneficiaries will have the opportunity to review, themselves or through third parties designated by them, the specifications to be disclosed; Microsoft should be able to impose reasonable and non-discriminatory conditions to ensure that this access to the disclosed specifications is granted for evaluation purposes only;
[...] to any remuneration that Microsoft might charge for supply; such a remuneration should not reflect the strategic value stemming from Microsoft s market power in the client PC operating system market or in the work group server operating system market;
The decision does not seem to give a hard number for how much MS may charge for disclosure of the specs.
If they had started paying it initially, with the decrease of the dollar and increase of the euro, it would have saved them a lot of money.
They already paid directly after the EC decision three years ago. The money was placed on a special bank account where neither the EU nor MS could touch it until the decision by the European Court.
Patents are legal documents. That is why they are written in legalese.
Only the section with the claims, that exactly defines what type of device or process is covered by the patent, is supposed to be legalese. Most of the bulk of a patent is supposed to be readable for an engineer who deals with similar devices or processes, and that has indeed mostly been my experience with patents in a field that I know well (i.e., laser physics). But someone without an appropriate physics background should not expect to be able to understand the text, just like they would not be able to grasp what's in a typical paper in Physical Review or Journal of Applied Physics.
I'm a bit surprised that nobody mentioned this yet, but IIRC the main difference between a tube and a transistor amplifier when it is not driven into saturation is that a tube amp needs an output transformer that has ohmic resistance in its primary/secondary windings. I can't recall where I read it, but apparently adding a 0.5 to 1 ohm resistance in series with the loudspeaker cable can make a regular transistor amplifier (output resistance less than 0.04 ohm) sound like a tube amp. Obviously a series resistance is a bad think from a technical point of view, since the filter circuits in the loudspeaker were designed for an ideal 0-ohm amplifier.
UNLESS OTHERWISE MUTUALLY AGREED TO BY THE PARTIES IN WRITING, LICENSOR OFFERS THE WORK AS-IS AND ONLY TO THE EXTENT OF ANY RIGHTS HELD IN THE LICENSED WORK BY THE LICENSOR. THE LICENSOR MAKES NO
A bit offtopic, but could anyone explain to me why English-language license texts and Terms of Service always are written in obfuscated language that is full of capitals? What's wrong with bold face or italic for emphasis? I recently looked at the judgment by the European court on the Microsoft antitrust case, and I was surprised by how comprehensible it was. Apparently, being legally clear and being human-readible are not mutually exclusive.
No article I've read on the subject so far has made any mention of how the specifications need to be licensed, however. If anyone is in the know, please share that information.
5a)... make the Interoperability Information available to any undertaking having an interest in
developing and distributing work group server operating system products and shall, on
reasonable and non-discriminatory terms, allow the use of the Interoperability Information...
What defines "reasonable" is defined in articles 1005 and further on page 280:
1008-ii) to any remuneration that Microsoft might charge for supply; such a
remuneration should not reflect the strategic value stemming from
Microsofts market power in the client PC operating system market or in the
work group server operating system market;
1008-iii) to restrictions that Microsoft may impose as to the type of products in which
the specifications may be implemented; such restrictions should not create disincentives to compete with Microsoft, or unnecessarily restrain the ability
of the beneficiaries to innovate;
The appeal decision is here, but since it basically concludes with "the 2004 ruling is mostly upheld", it is not so interesting to read.
Alexa.com is a website that aggregates data from their spy ware tool. So it heavily skews their information to the technically incompetent/ windows pc users.
That effect is offset by technically oriented website owners who are interested in how their site is doing compared to the competition, and technically oriented people who are more interested in statistics anyway. My own website has a part that attracts mostly nontechnical people (85/11% IE vs FF browsers) and a part aimed at technical people (48/41% IE). The nontechnical site gets 10x more page views than the technical site, but according to Alexa, the ratio is more like 2:1.
Time dilation due to speed? Pft hardly. Maybe there are subatomic particles such as electronics whose movement becomes dampened when approach speed because they have a fixed absolute speed, or even slow downs at the quantum level [maybe], but that doesn't mean time actually moves at a different speed.
I think you're talking about special relativity, not general relativity, and that you have never studied it in more depth than at a lay man level. The whole point of special relativity is that time is just another dimension in addition to the three space dimensions we already have, and that you need to use the correct coordinate transformation if you switch to a different basis set. If you express a point (x,y) in terms of axes that are at 45 deg angles with the original x and y axes, you end up with sqrt(0.5)*(x+y, x-y) in the new coordinate system, coupling the x and y coordinates with each other. In special relativity, there is a similar transformation when one coordinate system moves w.r.t. the other one. Rather than coupling just x with y, it couples all of x, y, z, and time. Now indeed this leads to counter-intuitive effects for humans that aren't used to moving at close to the speed of light, which you might describe as "time dilation" when you try to map the observations to the concepts of our non-relativistic everyday world. An example of why time dilation is not the right way to look at it is the case that one observer is standing at a fixed position, while the other one is moving. Both of them will think the time of the other observer is dilated, while their own time is normal. You can't point out for which of the two observers the time is dilated, simply because the concept of time dilation is inconsistent. (However, when one of the observers turns around and comes back, his clock will appear to be behind, but that has to do with the change in speed he underwent when turning around, not the speed on its own. Speed changes, i.e., acceleration, are described in general relativity, not in special relativity).
but one day it'll be shown that it's just plain silly. I don't have a better idea but do not prescribe to relativity.
A better theory will probably be developed at some time. But that will not degrade the current, established relativity theory to be "plain silly", just like Newton's laws didn't become silly after the development of quantum mechanics and relativity. The equations for the unified theory would look just like those of relativity if the length scale isn't too small, just like classical mechanics if in addition the energy is small, and just like quantum mechanics if both the length scale and energy are small. Physics is about describing nature quantitatively, not about attaching a deeper meaning to it or answering the question "why" nature is the way it is.
the formerly-clipped flat edges will have had the edges rounded off [after lossy encoding] and may bulge slightly higher as they more closely resemble sinusoids. This can actually sound better than the original clipped signal (as clipping is highly audible in double-blind tests and strains the ear)
I doubt it. The clipping strains the ear because it introduces loads of high-frequency energy, but the encoder will nicely preserve most of the energy content, just redistribute it a bit over time. Rather than have it all concentrated on two samples (50 usec), it gets distributed over one encoded block (4096 samples = 185 msec). But it sounds just as horrible.
I used to make live recordings of (classical concerts) with a minidisc player (ATRAC encoding), and sometimes I had the recording level a bit too high which caused occasional clipping. Indeed, the waveform gets rounder around the clipped parts, but manually drawing smooth curves to undo the clipping wouldn't help much because the high-frequencies were everywhere, not just at the clipped parts of the waveforms.
A hot side for steam in a typical BWR of 1000K (1300F) seems kinda high-- I'd say more realistic is ~600F.
OK, I admit, I didn't have time to look up the exact numbers. According to wikipedia it's 540 C (810 K, 1000 F) and 38 C (310 K) for a fossil-fuel plant, which gives 62% theoretical maximum efficiency, which happens to be the same as my first estimate.
You still have to turn that into eletrical which is going to ding you again, maybe another 25%
From mechanical to electrical power can be made very efficient, more than 90%. The biggest problem is that the turbines and various heat exchangers are not operating at near-equilibrium conditions that are necessary to reach the theoretically possible efficiency. In a nuclear plant, there are extra heat exchangers to prevent the steam from becoming radioactive, which gives an additional loss. According to the Wikipedia article, nuclear plants have a 34-38% efficiency, and fossil-fuel plants between 36% and 48%.
Slashdot Mirror
That's funny, because an md5 checksum is always 128 bits, i.e. 32 bytes. Read the bottom of the page: info hash: SHA1 hash of the "info" section of the metainfo (*.torrent) Try "sha1sum" next time, and on the correct file.
From the forums Terms of Service: Post constructive comments and questions. Unless otherwise noted, your Submission should either be a technical support question or a technical support answer. Constructive feedback about product features is welcome as well. If your Submission contains the phrase "Im sorry for the rant, but" you are likely in violation of this policy.
We cannot see what the deleted discussion looked like, but I think a topic starter like "How do I secure my iDisk?" is much less likely to be deleted than "Apple's iDisk has horrible security", even if the former leads to a heated discussion.
It's actually quite hard to write a message that will trip good spam filters (e.g. Spamassasin, Gmail). Just mentioning a couple of keywords such as "viagra penis enlargement nigeria" will usually not work (you need either a LOT of "bad words", or send from a blacklisted IP). There is a special string that is recognized by the major spam filters: XJS*C4JDBQADN1.NSBN3*2IDNEN*GTUBE-STANDARD-ANTI-UBE-TEST-EMAIL*C.34X . See Wikipedia GTUBE. The problem is, if you're caught pulling that trick, it'll be hard to explain how you accidentally put that string in your email.
I can assure you from plenty of personal experience that you don't feel it when you touch 230 VAC while not grounded. Wearing shoes with rubber soles or standing on a wooden floor is enough isolation. (Bare-feeted on ceramic tiles is quite painful, though) It always scares the crap out of people when I demonstrate how I stick paperclips into the wall sockets...
As for electricians, screwdrivers with small built-in neon lights have existed for as long as I can remember. The current they need to light up is too small to feel, except if you ground yourself (it tingles a bit and the light gets brighter when you do that).
I have my AP open too, but I don't want to risk trouble with my ISP if somebody connects with a zombied computer that's spamming or sending viruses. The AP is built into a Thomson/Speedtouch ADSL modem (very common in Netherlands) that has some configurable firewalling. So I only allow HTTP, HTTPS, POP, IMAP, VPN, SSH and block everything directed towards my home network. SMTP is not meaningful since the configured server of a AP user is most likely not that of my ISP, which is probably the only mailserver that they can reach anyway. Of course, a determined and knowledgeable person might be able to hijack an IP address that is supposed to be part of my internal network and circumvent the firewall, but I feel more comfortable this way.
Oh, and when I'm bored and notice that someone is using my wifi, I fire up a wireless packet sniffer and see what kind of websites and email they are dealing with. :-)
Would you mind giving me a more specific reference that, like, gives a table with meaningful numbers?
The price of raw coal is a minute fraction of the price of electricity. Coal combustion is about 30 MJ/kg = 8 kWh, or 3 kWh if you count the efficiency of a plant. Coal costs about US$ 50 per tonne [ref], so about US$ 0.017 per kWh electricity out of the $0.10--$0.20 you pay for your electricity. To put it differently, $500 in electricity produces as much CO2 as $50 in raw coal.
It makes quite a difference whether it's $500 in electricity or $500 of coal in an iron smelter.
Get the newer car. The CO2 emission for manufacturing a new car in the UK is 0.7 tonnes as of 2006, which is roughly 250 kg (300 liters = 75 gallons) of fuel. This is all thanks to the extensive recycling of cars. I don't know about the situation in the US, though.
I don't think that's how it works. The power factor mainly acts on the transmission losses, which are about 10% overall. So 1 W @ pf=0.2 means that the transmission losses are 10% * 1 W / 0.2 = 0.5 W in the worst case. It does not mean that the generators have to produce 5 watts of which 4 watts turn into heat.
The GP mentioned equipment like phones and PDAs, which typically have a battery of about 700 mAh at 3.7 V = 9000 Ws. Household wiring in Europe can usually do 16 amps at 230 V = 3680 W. With a factor 2 for conversion losses and power factor it means you could recharge the battery in 5 seconds or so. The biggest problem is that you need some kind of switched power supply with 10 times more capacity than that in a desktop pc, so that would be a pretty big and expensive battery charger.
There are other advantages of 16+ bits. 8-bit RGB images are usually in sRGB space, which means that the luminance of a pixel is not proportional to the pixel value, but rather something like the 2.2'th power except for a small range near zero. That is convenient for encoding a large contrast range in just 256 values, but sucks for operations that are inherently linear operators on the luminance, such as background substraction and blurring. With 16+ bits, all operations can be done in linear space without loss of resolution at the darker colors.
The filing date was August 2004; in the American patent law, you have to file within one year after the first public disclosure. (In Europe, you can't file a patent at all after a public disclosure.) I think I've seen virtual URL spaces before August 2003, but I'm not sure about search requests. Note that the patent specifically claims having the search terms directly at the root of the web site, e.g. www.example.com/search+terms , unless it is specifically not a search term (e.g. www.example.com/welcome.html) For example, Wikipedia redirects URLs like http://en.wikipedia.org/url to the URL article, but I don't know when they started doing that. They were certainly already doing that in 2004.
MS paid it: http://www.news.com/2100-1014_3-5255715.html ; I read about the bank account in the newspaper (as in: the paper version), as a side note to an interview with EC commissioner Kroes. Sorry, no link.
I didn't copy the whole text (as indicated by [...]), so read the original if you're unsure. I interpreted it as: MS should provide free excerpts to prospective buyers of the specs for evaluation purposes to make sure that the buyer knows what he is going to spend his money on.
See the EC ruling (PDF), especially article 999 on page 277:
Also interesting:
The decision does not seem to give a hard number for how much MS may charge for disclosure of the specs.
They already paid directly after the EC decision three years ago. The money was placed on a special bank account where neither the EU nor MS could touch it until the decision by the European Court.
Only the section with the claims, that exactly defines what type of device or process is covered by the patent, is supposed to be legalese. Most of the bulk of a patent is supposed to be readable for an engineer who deals with similar devices or processes, and that has indeed mostly been my experience with patents in a field that I know well (i.e., laser physics). But someone without an appropriate physics background should not expect to be able to understand the text, just like they would not be able to grasp what's in a typical paper in Physical Review or Journal of Applied Physics.
I'm a bit surprised that nobody mentioned this yet, but IIRC the main difference between a tube and a transistor amplifier when it is not driven into saturation is that a tube amp needs an output transformer that has ohmic resistance in its primary/secondary windings. I can't recall where I read it, but apparently adding a 0.5 to 1 ohm resistance in series with the loudspeaker cable can make a regular transistor amplifier (output resistance less than 0.04 ohm) sound like a tube amp. Obviously a series resistance is a bad think from a technical point of view, since the filter circuits in the loudspeaker were designed for an ideal 0-ohm amplifier.
A bit offtopic, but could anyone explain to me why English-language license texts and Terms of Service always are written in obfuscated language that is full of capitals? What's wrong with bold face or italic for emphasis? I recently looked at the judgment by the European court on the Microsoft antitrust case, and I was surprised by how comprehensible it was. Apparently, being legally clear and being human-readible are not mutually exclusive.
Here is the full EC ruling of 2004: Commission Decision of 24.03.2004 relating to a proceeding under Article 82 of the EC Treaty (Case COMP/C-3/37.792 Microsoft). It contains all the reasoning behind the decision in surprisingly non-legalese English, as well as the decision itself on page 299--301:
What defines "reasonable" is defined in articles 1005 and further on page 280: The appeal decision is here, but since it basically concludes with "the 2004 ruling is mostly upheld", it is not so interesting to read.That effect is offset by technically oriented website owners who are interested in how their site is doing compared to the competition, and technically oriented people who are more interested in statistics anyway. My own website has a part that attracts mostly nontechnical people (85/11% IE vs FF browsers) and a part aimed at technical people (48/41% IE). The nontechnical site gets 10x more page views than the technical site, but according to Alexa, the ratio is more like 2:1.
I think you're talking about special relativity, not general relativity, and that you have never studied it in more depth than at a lay man level. The whole point of special relativity is that time is just another dimension in addition to the three space dimensions we already have, and that you need to use the correct coordinate transformation if you switch to a different basis set. If you express a point (x,y) in terms of axes that are at 45 deg angles with the original x and y axes, you end up with sqrt(0.5)*(x+y, x-y) in the new coordinate system, coupling the x and y coordinates with each other. In special relativity, there is a similar transformation when one coordinate system moves w.r.t. the other one. Rather than coupling just x with y, it couples all of x, y, z, and time. Now indeed this leads to counter-intuitive effects for humans that aren't used to moving at close to the speed of light, which you might describe as "time dilation" when you try to map the observations to the concepts of our non-relativistic everyday world. An example of why time dilation is not the right way to look at it is the case that one observer is standing at a fixed position, while the other one is moving. Both of them will think the time of the other observer is dilated, while their own time is normal. You can't point out for which of the two observers the time is dilated, simply because the concept of time dilation is inconsistent. (However, when one of the observers turns around and comes back, his clock will appear to be behind, but that has to do with the change in speed he underwent when turning around, not the speed on its own. Speed changes, i.e., acceleration, are described in general relativity, not in special relativity).
A better theory will probably be developed at some time. But that will not degrade the current, established relativity theory to be "plain silly", just like Newton's laws didn't become silly after the development of quantum mechanics and relativity. The equations for the unified theory would look just like those of relativity if the length scale isn't too small, just like classical mechanics if in addition the energy is small, and just like quantum mechanics if both the length scale and energy are small. Physics is about describing nature quantitatively, not about attaching a deeper meaning to it or answering the question "why" nature is the way it is.
I doubt it. The clipping strains the ear because it introduces loads of high-frequency energy, but the encoder will nicely preserve most of the energy content, just redistribute it a bit over time. Rather than have it all concentrated on two samples (50 usec), it gets distributed over one encoded block (4096 samples = 185 msec). But it sounds just as horrible.
I used to make live recordings of (classical concerts) with a minidisc player (ATRAC encoding), and sometimes I had the recording level a bit too high which caused occasional clipping. Indeed, the waveform gets rounder around the clipped parts, but manually drawing smooth curves to undo the clipping wouldn't help much because the high-frequencies were everywhere, not just at the clipped parts of the waveforms.
OK, I admit, I didn't have time to look up the exact numbers. According to wikipedia it's 540 C (810 K, 1000 F) and 38 C (310 K) for a fossil-fuel plant, which gives 62% theoretical maximum efficiency, which happens to be the same as my first estimate.
From mechanical to electrical power can be made very efficient, more than 90%. The biggest problem is that the turbines and various heat exchangers are not operating at near-equilibrium conditions that are necessary to reach the theoretically possible efficiency. In a nuclear plant, there are extra heat exchangers to prevent the steam from becoming radioactive, which gives an additional loss. According to the Wikipedia article, nuclear plants have a 34-38% efficiency, and fossil-fuel plants between 36% and 48%.