I think you are trying to be funny and failed (also I think you are mixing up chroot and/dev/null)
If you chroot, you can't see any files if the aren't in the chroot directory or below. So, if you want to back up/etc, your chroot better include/etc (either as the root or as a subdirectory).
This is why the OP said the backup finished faster...because after the chroot, none of the files that needed to be backed up were visible.
If you have to do things between computers in your local network (transfer a file, or execute a process, or whatever), doing it using an unencrypted protocol means that any computer doing sniffing in that network could get both content and access.
With switches, you can't see any packets unless they are destined for the Ethernet card in the machine doing the sniffing. So, it means that one of computers involved in the transfer is compromised, in which case there isn't any need to sniff the traffic...you just read the file from the local disk. It could be that your switch is compromised, and has been instructed to copy packets from other ports. In that case, encryption over the network would help.
Plus, if I did through some catastrophe (say, a housefire, which would also destroy your physical collection), re-torrenting is trivial.
If you don't mind 600x400 resolution with low bitrate MP3 audio, then you can find most everything, but if DVD quality is the worst you'll accept, then it's actually quite hard to find active torrents of older content, unless it's insanely popular (or at least very popular among computer geeks).
If you want HD, then it's even tougher, even with some content that is still actively for sale. Also, I have noticed that torrents seem to follow basic economics...once the original source is cheap enough, the torrent tends to become unseeded.
It tells you plainly how to get delisted for free. But that requires you to do some serious work and find out who you have spamming on your network.
It's worse than that, since there is no way to find the exact e-mail that was considered "spam" by the blacklist.
E-mail arrives to my inbox all the time that may have scored high enough on someone else's spam filter to be rejected. Some of it is spam, some of it is not. What UCEProtect uses to decide that an e-mail is spam is unknown, and thus it is impossible to stay off their list merely by adjusting behavior.
The worst part is that many spam filters use DNS block lists to either add score to the spam or to outright block. So, it's possible that UCEProtect is marking items as spam based on other DNS block lists, while those lists decided that an e-mail was spam because it was sent from an IP on the UCEProtect list. Basically, once you are on any poorly administered (and possibly extortionate) DNS block list, regardless of whether you sent anything that a reasonable human would consider spam, it's likely you will end up on all of them. Which, in turn, makes it much harder to get off those lists without paying the extortion fee.
My 2010 Honda's manual very specifically says not to use ethanol blends higher than 10%. I'll trust Honda's word over those of the corn lobby.
Most car manuals have that phrasing, but it really means "don't use the only other widely available ethanol blend, which is 85% ethanol", since E85 will do some serious damage in vehicles that aren't designed to use it.
How about considering the fact that 300 Euros is nothing to an ISP. But it's enough to make it infeasible for spammers to pay up.
A spammer with one IP address would be paying US$115 (I don't know why the summary lists the fees in Euros, as all of them are actually in US dollars) and US$345 for one "allocation". The de-listing has to come with a guarantee of not getting back on the list as easily (because the assumption is you're not going to pay to remove a real spammer). For those low prices, a real spammer would actually be glad to pay. And, without the guarantee, UCEPROTECT's unknown method of determining spam could put the IP right back on the list.
As a side note, in every other case when you see things like "Netzmask" on a web page or e-mail that is trying to get money out of you, you'd call it a scam, as would be anything that makes money with little to no work.
All they do is set up some mail servers that classify incoming e-mail as spam or not using an unknown algorithm and put the results into DNS entries. This is all automated, and takes just a few days to configure at most. After that, it just runs itself, and it might result in some cash coming their way. You can do this on a less than US$100/month Internet connection, so just one de-listing payment per month and they have a profit. So, why isn't UCEPROTECT a scam?
How did this make it to the front page, especially with SSD prices being what they are?
I have a 20TB RAID array that cost me about $0.10/GB, including controllers. If you can afford to build a 20TB array using SSD, you have far more money than I do. You will also need more controllers than I do (port multipliers divide the bandwidth, which you don't want to do for SSDs), since you'd need at least 20 SSDs (if you were willing to pay about $2.50/GB), but more likely more than 45 (at about $0.85/GB).
You also need special controllers that understand SSDs and can pass TRIM commands, and that will add about $0.15/GB. And, you'll need a much more expensive motherboard, since you need at least 24 PCIe lanes that can all be used for something other than video cards, but likely more than 40. Last, since this is Slashdot, you might not be able to use those special controllers, as not all of them have drivers for the kernel version you want to use.
So, yeah, for a boot drive, SSDs kick ass, but for storing your movie collection, not only are they 10 times more expensive than magnetic disks, but they are way overkill as far as performance is concerned.
On the other hand, there's a reason Cisco gear is expensive: it's enterprise class. [anecdote snipped]
And yet, I can do everything you needed to do on the Netgear managed equipment I use at home, and it costs less than the "Cisco Small Business" equipment that couldn't do the job for you. Compared to "real" Cisco equipment, it may not be as powerful, but for the target market of "small business", it's more than enough.
Sometimes, too, less powerful is better. I can't tell you the number of times I watched Cisco-certified people take 20-30 minutes to make configuration changes to Cisco equipment that would take seconds on other network gear simply because there aren't nearly as many features that need to be dealt with.
I haven't checked into it as far as you, but if I leave GPS checked "on", even when I'm not using Latitude, my battery life is horrible.
Is the GPS icon (the satellite dish) visible on the notification bar? If so, then some task is running that is actively querying the GPS receiver, and that can kill your battery.
On my HTC Thunderbolt, I leave GPS "on" all the time so that apps don't have to ask if they can turn it on, and I get at least 36 hours between charges, and as much as 60 if I'm not making voice calls (which is the #1 battery suck for me). The "what used your battery" screen never shows the GPS as taking any significant portion...it's always "Phone standby", "Display", and a couple of commonly used apps that take up everything.
The shutter is typically open for much less than 1/24 of a second even on a film camera.
Yes, I know. This controls the exposure (and also affects how much blurring you get). But, the start of opening is every 1/24 of a second (form normal movies), and where the subject is located at each of those times compared to where they were located the time before is what controls whether the image has motion blur, along with the shutter speed.
The problem with interpolation is that it only knows the "pixel count" movement (and even then needs a powerful processor to get the movement correct), when it really needs to know the physical distance (both to the subject and how much it moved), the lens focal length and the iris size (which together with the distances give you knowledge of whether the subject should be in focus in the first place), the shutter speed, and the frame rate. Now, the camera does have most of this available, but I find it hard to believe that it also has the processing power to do full vector movement analysis and stores both a 48fps video along with the interpolated 24fps video.
And no, the digital sensor doesn't get "tired" and need to rest between frames.
In a sense, yes it does. If you keep the shutter open for an exposure that is long relative to the frame rate, then you get junk data from the leftover light that has still excited the sensor. Otherwise you wouldn't need a physical shutter to block the light...you'd just sample the sensor every 1/24 of a second. And, yes, I know that many cheap camcorders work without a physical shutter, and it shows in the image quality.
So, it's a an interpolation of some sort, and not an actual 24fps shutter, which means it won't look "right".
There has to be a physical shutter to make the sensor work right, as it needs time to reset to "all off", and this can't be done in the presence of light. And, if that physical shutter is running at 48fps, then any other frame rate will be an interpolation, with all the issues that I mention.
They're already making a 24fps theatrical version that has the appropriate amount of extra motion blur. Why would they drop every other frame?
I didn't know that this source was already made, but it will likely look "wrong" if the blur is added digitally. There's a lot going on with an analog shutter opening and closing, even if it is in front of a fixed digital sensor instead of a moving film.
For example, without knowing what the relative movement of camera vs. subject was in some form of exact measurement, you can't know what movement should blur and what should not. Assuming that an object is sharp in every frame of the 48fps source, knowing that it moved 27 pixels from frame 1 to frame 2 and 32 pixels from frame 2 to frame 3 doesn't tell you if that movement rate was "too fast" for the virtual 24fps shutter between frame 1 and frame 3 unless you know the exact distance moved in the real world. In other words, just because it was sharp at 48fps doesn't mean it should be blurry at 24fps...maybe it would only be blurry at 10fps.
Most BR players should be fine as we already have higher FPS video in the form of 30p 3D (which is 2 full 1080p frames stacked on top of each other, so the same decoding requirement as 2D 60p).
Actually, most Blu-Ray players are not 3D. Many (if not all) newly sold ones are, but the vast majority in homes are not. In addition, there are many 3D formats available to Blu-Ray, but the most popular is a full left-eye encode along with difference information for the right eye. This means that the 2nd "frame" isn't anywhere near a full frame, even in the busiest 3D scenes.
My guess is that they will probably convert the film to 50p, much like how they often convert normal films to 25p for PAL markets. The change in playback speed in not noticeable and sound is corrected for pitch. All HDTVs support PAL frame rates so compatibility should be pretty much universal.
That's three wrong guesses in 3 sentences...not bad.
First, almost no TV sold in the US (or any other NTSC country) support PAL input rates (25fps or 50fps). Second, the change in playback speed is noticable to many, and it's impossible to 100% correct for pitch. Last, the only 50fps mode supported by Blu-Ray players is interlaced, not progressive.
i wonder how this affects BR releases... can consumer devices handle that much throughput in HD?
No.
Blu-Ray does not support 48fps at all. It's not part of the current standard, which means no current Blu-Ray player supports it. In addition, there is no current mode supported by Blu-Ray that has at least 48fps and 1920x1080 pixels. So, to see every frame as filmed, they will need to use 1280x720/60p (with 3:2 pulldown added).
Otherwise, there will have to be some sort of frame combining to allow 1920x1080 pixels.
Also, the way I understand conversion technology, when converting to a format for display on a TV it is much easier to use tricks like 3:2 pulldown and interlacing to convert from 24 or 48 fps to the NTSC (60 Hz) or PAL (50 Hz) standards (some sources are sped up from 24 to 25 fps during conversion to PAL) than it is to convert 50 fps to something that can be shown on NTSC equipment.
The only current 24fps sources are Blu-Ray and HD-DVD...no TV broadcasts are sent at that rate, but most Blu-Ray players can be sent to output at 24fps, and many new TVs support this.
So, the easiest way to convert from 24fps is to not convert at all.
As soon as technology allowed me, I produced and projected video animations in 60fps to make pans more fluid. Would I be producing a movie today, I would try to shoot in 60fps for the same reason, much more fluid motion on big screens.
If Jackson had chosen 60fps for The Hobbit, it would have been a much better choice, at least as far as home video is concerned.
With a choice of 48fps as the source, we are going to get stuck with a much lower quality home video release, because there is no current format that allows at least 48fps and 1920x1080 resolution. So, to convert to 24fps, either the original footage will have to be filmed at 24fps, or else some sort of digital interpolation will have to be done. Neither will give the same quality that we have come to expect from current media, as instead of 24 frames per second where scenes with little motion have very sharp frames, pretty much every frame will show some sort of motion.
Bought one Antec Earthwatts long time ago. The PSU was not much more expensive than the others (good brands) so the savings are obvious.
Another thing TFA doesn't take into account is that the 80-Plus certified supplies tend to have better components overall than non-certified supplies.
Read some of the reviews at Hardware Secrets and you'll see that it's not uncommon for a well-built "350W" power supply to be able to output 450W, while a crappy 350W supply can't even handle 300W.
Not only that, but if your software is smart it only greylists server that it hasn't ever sent an email to anyway.
Unfortunately, inbound and outbound SMTP servers often don't have the same IP address, so this doesn't work in practice.
So after the first week of being installed all your important customers are automatically whitelisted
But this is still very true. Only the first e-mail message is delayed, and that delay is mostly controlled by the retry time set on the sending machine. There are some really annoying ones that retry once a minute for a couple tries, then back off to an hour or more. This is the worst as far as greylisting is concerned.
Once a server is whitelisted, then my implementation allows 40 days of no activity from that server before it drops off the whitelist. This means that even a once-a-month mailing list from an obscure server doesn't see any delay after the first time. About 1/4 of the 380,000 IPs that have ever contacted my e-mail servers are currently whitelisted.
You can't acknowledge receipt until the message is written to durable storage.
Which is exactly what I said. If the message is rejected, everything has taken place in the RAM disk, and you don't care. If the message is accepted, then sendmail has written it to the queue at the very least. As long as the queue is not on the RAM disk, you're fine.
And you have never run a really large email system. Bandwidth isn't really a limitation, disk io is.
Again, I said this too: "CPU time....is insignificant compared to the bandwidth, disk writes". I guess Slashdot has gotten to the point where nobody reads anything before posting. But, having talked with people who run e-mail systems for millions of users, not having to receive the spam in the first place (unfortunately usually through blacklists, but often using greylisting) is the biggest win, as bandwidth is in reality far more of a limitation. It's quite easy to put together a disk array that gives you 500MB/sec throughput, while it's not easy to pay for a 4Gbps inbound line (which it what it would take to saturate those disks). Even assuming 8 bytes written per incoming byte, it's still pretty easy to spread load to effectively 2-4GB/sec worth of disk, while 2-4Gbps of Internet connectivity is pricey.
That's why most companies outsource their their mass-mailing to 3rd parties like MailJet, MailChimp, SendGrid...
As an e-mail server operator, I'm glad they do, as it makes it easy to block all the spam from those companies at the server level.
I've been added to e-mail lists without my permission quite often because I had to provide an e-mail address for the company to send me a bill or other actual important e-mail. They pass the e-mail address on to these third party companies without any confirmation on my part. Then, when you do go through the unsubscribe process, those companies claim that it might take 4-5 days to remove you from the list, when in reality we know it should happen instantly. This is so that when you keep getting the spam for the next week or so, you might not report it, or maybe if you are truly stupid, you might buy some of the junk that is advertised.
In addition, once you have any dealing with a company like MailChimp, they can play the "previous business relationship" card when sending you "something you might be interested in". And, since none of the companies you list are confirmed opt-in (which requires that they send you a first e-mail when you supposedly subscribe, and unless you click a link or reply to that e-mail, you do not get added to the list), they can play fast and loose with claims that you signed up for something that you did not.
Why does he need to send 400,000+ emails in the first place? If it's just a list of proxy domains, why not just have an RSS feed that people can subscribe to? No emails needed.
Because the people who want the proxy list would need to use a proxy to be able to read the RSS feed, as censorship in their country would block access once it was learned what the RSS feed contained.
On the other hand, incoming e-mail can be blocked as spam, but you can't decide to block an e-mail as spam without knowing something about it, and if it isn't coming from the same domain or IP address, and doesn't have the same content as previous spam, it's pretty tough to block. So, once the e-mail gets through, the "damage" is done (as far as the censors are concerned).
Hashcash distributes the load somewhat, in that it forces spammers to use more resources to send out their message and can slow them down somewhat.
Unfortunately, until you get to a significant number of bits, hashcash doesn't take all that long to compute, and you can pre-compute them.
I use 23-bit hashcash on all my outgoing e-mails, but if the address has been sent to before, there is likely a pre-computed 25-bit hashcash waiting. I use idle server time to pre-compute for any address that has been sent to from my servers. Since the hashcash expires in 25 days, I don't have to do this very often unless the recipient is a frequent one. Then, to keep the database small, I remove addresses from the "sent to" table unless they have been recently active, where "recent" depends on the total amount of activity to that address.
Blacklists are nice because they reduce server loads. Sure, running a statistical classifier for one user is not so hard, but if you have to process hundreds of millions of messages per day, that is a lot of CPU time spent on spam.
The CPU time spent on running something like SpamAssassin is insignificant compared to the bandwidth, disk writes, etc., caused by spam. Keeping the incoming e-mail in a RAM disk until you have truly accepted it for delivery (which isn't dangerous even if the server crashes hard) is the #1 thing that speeds up e-mail intake. At that point, scanning takes almost no time.
As you mention, though, greylisting does the best job of keeping your overall load down, since you don't even need to use network bandwidth on the body unless the sending server is known to retry, which basically eliminates every botnet member. Maybe this solution would work now that other more "instant" messaging systems are readily available, but 15 years ago when IM wasn't really a corporate thing, I couldn't use greylisting because "it slowed down e-mail too much", even though it didn't slow it down at all for the "important" clients, as their servers got whitelisted anyway.
From what I understand, the regulation uses something like replay gain to calculate the average volume. I don't know the exact formula being used, but it is designed to avoid the situation you refer to where both the quietest part and loudest part of the commercial is just as loud as the average volume of the show.
I don't know how they are going to enforce it, though, even with a lot of complaints, since you'd need a recording of the entire show plus commercials to know if it was a violation.
Yeah, most other 27" monitors have 13% more PPI. A panel with the same 2.37:1 aspect ratio but the same PPI as other monitors would be 2880x1216. As an upgrade to my 24" 1920x1200 monitor, that would be a reasonable option.
With other 27" monitors giving me 2560x1440, the LG in TFA is completely useless.
Slashdot Mirror
I think you are trying to be funny and failed (also I think you are mixing up chroot and /dev/null)
If you chroot, you can't see any files if the aren't in the chroot directory or below. So, if you want to back up /etc, your chroot better include /etc (either as the root or as a subdirectory).
This is why the OP said the backup finished faster...because after the chroot, none of the files that needed to be backed up were visible.
If you have to do things between computers in your local network (transfer a file, or execute a process, or whatever), doing it using an unencrypted protocol means that any computer doing sniffing in that network could get both content and access.
With switches, you can't see any packets unless they are destined for the Ethernet card in the machine doing the sniffing. So, it means that one of computers involved in the transfer is compromised, in which case there isn't any need to sniff the traffic...you just read the file from the local disk. It could be that your switch is compromised, and has been instructed to copy packets from other ports. In that case, encryption over the network would help.
Plus, if I did through some catastrophe (say, a housefire, which would also destroy your physical collection), re-torrenting is trivial.
If you don't mind 600x400 resolution with low bitrate MP3 audio, then you can find most everything, but if DVD quality is the worst you'll accept, then it's actually quite hard to find active torrents of older content, unless it's insanely popular (or at least very popular among computer geeks).
If you want HD, then it's even tougher, even with some content that is still actively for sale. Also, I have noticed that torrents seem to follow basic economics...once the original source is cheap enough, the torrent tends to become unseeded.
It tells you plainly how to get delisted for free. But that requires you to do some serious work and find out who you have spamming on your network.
It's worse than that, since there is no way to find the exact e-mail that was considered "spam" by the blacklist.
E-mail arrives to my inbox all the time that may have scored high enough on someone else's spam filter to be rejected. Some of it is spam, some of it is not. What UCEProtect uses to decide that an e-mail is spam is unknown, and thus it is impossible to stay off their list merely by adjusting behavior.
The worst part is that many spam filters use DNS block lists to either add score to the spam or to outright block. So, it's possible that UCEProtect is marking items as spam based on other DNS block lists, while those lists decided that an e-mail was spam because it was sent from an IP on the UCEProtect list. Basically, once you are on any poorly administered (and possibly extortionate) DNS block list, regardless of whether you sent anything that a reasonable human would consider spam, it's likely you will end up on all of them. Which, in turn, makes it much harder to get off those lists without paying the extortion fee.
My 2010 Honda's manual very specifically says not to use ethanol blends higher than 10%. I'll trust Honda's word over those of the corn lobby.
Most car manuals have that phrasing, but it really means "don't use the only other widely available ethanol blend, which is 85% ethanol", since E85 will do some serious damage in vehicles that aren't designed to use it.
How about considering the fact that 300 Euros is nothing to an ISP. But it's enough to make it infeasible for spammers to pay up.
A spammer with one IP address would be paying US$115 (I don't know why the summary lists the fees in Euros, as all of them are actually in US dollars) and US$345 for one "allocation". The de-listing has to come with a guarantee of not getting back on the list as easily (because the assumption is you're not going to pay to remove a real spammer). For those low prices, a real spammer would actually be glad to pay. And, without the guarantee, UCEPROTECT's unknown method of determining spam could put the IP right back on the list.
As a side note, in every other case when you see things like "Netzmask" on a web page or e-mail that is trying to get money out of you, you'd call it a scam, as would be anything that makes money with little to no work.
All they do is set up some mail servers that classify incoming e-mail as spam or not using an unknown algorithm and put the results into DNS entries. This is all automated, and takes just a few days to configure at most. After that, it just runs itself, and it might result in some cash coming their way. You can do this on a less than US$100/month Internet connection, so just one de-listing payment per month and they have a profit. So, why isn't UCEPROTECT a scam?
How did this make it to the front page, especially with SSD prices being what they are?
I have a 20TB RAID array that cost me about $0.10/GB, including controllers. If you can afford to build a 20TB array using SSD, you have far more money than I do. You will also need more controllers than I do (port multipliers divide the bandwidth, which you don't want to do for SSDs), since you'd need at least 20 SSDs (if you were willing to pay about $2.50/GB), but more likely more than 45 (at about $0.85/GB).
You also need special controllers that understand SSDs and can pass TRIM commands, and that will add about $0.15/GB. And, you'll need a much more expensive motherboard, since you need at least 24 PCIe lanes that can all be used for something other than video cards, but likely more than 40. Last, since this is Slashdot, you might not be able to use those special controllers, as not all of them have drivers for the kernel version you want to use.
So, yeah, for a boot drive, SSDs kick ass, but for storing your movie collection, not only are they 10 times more expensive than magnetic disks, but they are way overkill as far as performance is concerned.
On the other hand, there's a reason Cisco gear is expensive: it's enterprise class. [anecdote snipped]
And yet, I can do everything you needed to do on the Netgear managed equipment I use at home, and it costs less than the "Cisco Small Business" equipment that couldn't do the job for you. Compared to "real" Cisco equipment, it may not be as powerful, but for the target market of "small business", it's more than enough.
Sometimes, too, less powerful is better. I can't tell you the number of times I watched Cisco-certified people take 20-30 minutes to make configuration changes to Cisco equipment that would take seconds on other network gear simply because there aren't nearly as many features that need to be dealt with.
I haven't checked into it as far as you, but if I leave GPS checked "on", even when I'm not using Latitude, my battery life is horrible.
Is the GPS icon (the satellite dish) visible on the notification bar? If so, then some task is running that is actively querying the GPS receiver, and that can kill your battery.
On my HTC Thunderbolt, I leave GPS "on" all the time so that apps don't have to ask if they can turn it on, and I get at least 36 hours between charges, and as much as 60 if I'm not making voice calls (which is the #1 battery suck for me). The "what used your battery" screen never shows the GPS as taking any significant portion...it's always "Phone standby", "Display", and a couple of commonly used apps that take up everything.
The shutter is typically open for much less than 1/24 of a second even on a film camera.
Yes, I know. This controls the exposure (and also affects how much blurring you get). But, the start of opening is every 1/24 of a second (form normal movies), and where the subject is located at each of those times compared to where they were located the time before is what controls whether the image has motion blur, along with the shutter speed.
The problem with interpolation is that it only knows the "pixel count" movement (and even then needs a powerful processor to get the movement correct), when it really needs to know the physical distance (both to the subject and how much it moved), the lens focal length and the iris size (which together with the distances give you knowledge of whether the subject should be in focus in the first place), the shutter speed, and the frame rate. Now, the camera does have most of this available, but I find it hard to believe that it also has the processing power to do full vector movement analysis and stores both a 48fps video along with the interpolated 24fps video.
And no, the digital sensor doesn't get "tired" and need to rest between frames.
In a sense, yes it does. If you keep the shutter open for an exposure that is long relative to the frame rate, then you get junk data from the leftover light that has still excited the sensor. Otherwise you wouldn't need a physical shutter to block the light...you'd just sample the sensor every 1/24 of a second. And, yes, I know that many cheap camcorders work without a physical shutter, and it shows in the image quality.
So, it's a an interpolation of some sort, and not an actual 24fps shutter, which means it won't look "right".
There has to be a physical shutter to make the sensor work right, as it needs time to reset to "all off", and this can't be done in the presence of light. And, if that physical shutter is running at 48fps, then any other frame rate will be an interpolation, with all the issues that I mention.
They're already making a 24fps theatrical version that has the appropriate amount of extra motion blur. Why would they drop every other frame?
I didn't know that this source was already made, but it will likely look "wrong" if the blur is added digitally. There's a lot going on with an analog shutter opening and closing, even if it is in front of a fixed digital sensor instead of a moving film.
For example, without knowing what the relative movement of camera vs. subject was in some form of exact measurement, you can't know what movement should blur and what should not. Assuming that an object is sharp in every frame of the 48fps source, knowing that it moved 27 pixels from frame 1 to frame 2 and 32 pixels from frame 2 to frame 3 doesn't tell you if that movement rate was "too fast" for the virtual 24fps shutter between frame 1 and frame 3 unless you know the exact distance moved in the real world. In other words, just because it was sharp at 48fps doesn't mean it should be blurry at 24fps...maybe it would only be blurry at 10fps.
Most BR players should be fine as we already have higher FPS video in the form of 30p 3D (which is 2 full 1080p frames stacked on top of each other, so the same decoding requirement as 2D 60p).
Actually, most Blu-Ray players are not 3D. Many (if not all) newly sold ones are, but the vast majority in homes are not. In addition, there are many 3D formats available to Blu-Ray, but the most popular is a full left-eye encode along with difference information for the right eye. This means that the 2nd "frame" isn't anywhere near a full frame, even in the busiest 3D scenes.
My guess is that they will probably convert the film to 50p, much like how they often convert normal films to 25p for PAL markets. The change in playback speed in not noticeable and sound is corrected for pitch. All HDTVs support PAL frame rates so compatibility should be pretty much universal.
That's three wrong guesses in 3 sentences...not bad.
First, almost no TV sold in the US (or any other NTSC country) support PAL input rates (25fps or 50fps). Second, the change in playback speed is noticable to many, and it's impossible to 100% correct for pitch. Last, the only 50fps mode supported by Blu-Ray players is interlaced, not progressive.
i wonder how this affects BR releases... can consumer devices handle that much throughput in HD?
No.
Blu-Ray does not support 48fps at all. It's not part of the current standard, which means no current Blu-Ray player supports it. In addition, there is no current mode supported by Blu-Ray that has at least 48fps and 1920x1080 pixels. So, to see every frame as filmed, they will need to use 1280x720/60p (with 3:2 pulldown added).
Otherwise, there will have to be some sort of frame combining to allow 1920x1080 pixels.
Also, the way I understand conversion technology, when converting to a format for display on a TV it is much easier to use tricks like 3:2 pulldown and interlacing to convert from 24 or 48 fps to the NTSC (60 Hz) or PAL (50 Hz) standards (some sources are sped up from 24 to 25 fps during conversion to PAL) than it is to convert 50 fps to something that can be shown on NTSC equipment.
The only current 24fps sources are Blu-Ray and HD-DVD...no TV broadcasts are sent at that rate, but most Blu-Ray players can be sent to output at 24fps, and many new TVs support this.
So, the easiest way to convert from 24fps is to not convert at all.
As soon as technology allowed me, I produced and projected video animations in 60fps to make pans more fluid. Would I be producing a movie today, I would try to shoot in 60fps for the same reason, much more fluid motion on big screens.
If Jackson had chosen 60fps for The Hobbit, it would have been a much better choice, at least as far as home video is concerned.
With a choice of 48fps as the source, we are going to get stuck with a much lower quality home video release, because there is no current format that allows at least 48fps and 1920x1080 resolution. So, to convert to 24fps, either the original footage will have to be filmed at 24fps, or else some sort of digital interpolation will have to be done. Neither will give the same quality that we have come to expect from current media, as instead of 24 frames per second where scenes with little motion have very sharp frames, pretty much every frame will show some sort of motion.
Bought one Antec Earthwatts long time ago. The PSU was not much more expensive than the others (good brands) so the savings are obvious.
Another thing TFA doesn't take into account is that the 80-Plus certified supplies tend to have better components overall than non-certified supplies.
Read some of the reviews at Hardware Secrets and you'll see that it's not uncommon for a well-built "350W" power supply to be able to output 450W, while a crappy 350W supply can't even handle 300W.
Not only that, but if your software is smart it only greylists server that it hasn't ever sent an email to anyway.
Unfortunately, inbound and outbound SMTP servers often don't have the same IP address, so this doesn't work in practice.
So after the first week of being installed all your important customers are automatically whitelisted
But this is still very true. Only the first e-mail message is delayed, and that delay is mostly controlled by the retry time set on the sending machine. There are some really annoying ones that retry once a minute for a couple tries, then back off to an hour or more. This is the worst as far as greylisting is concerned.
Once a server is whitelisted, then my implementation allows 40 days of no activity from that server before it drops off the whitelist. This means that even a once-a-month mailing list from an obscure server doesn't see any delay after the first time. About 1/4 of the 380,000 IPs that have ever contacted my e-mail servers are currently whitelisted.
You can't acknowledge receipt until the message is written to durable storage.
Which is exactly what I said. If the message is rejected, everything has taken place in the RAM disk, and you don't care. If the message is accepted, then sendmail has written it to the queue at the very least. As long as the queue is not on the RAM disk, you're fine.
And you have never run a really large email system. Bandwidth isn't really a limitation, disk io is.
Again, I said this too: "CPU time....is insignificant compared to the bandwidth, disk writes". I guess Slashdot has gotten to the point where nobody reads anything before posting. But, having talked with people who run e-mail systems for millions of users, not having to receive the spam in the first place (unfortunately usually through blacklists, but often using greylisting) is the biggest win, as bandwidth is in reality far more of a limitation. It's quite easy to put together a disk array that gives you 500MB/sec throughput, while it's not easy to pay for a 4Gbps inbound line (which it what it would take to saturate those disks). Even assuming 8 bytes written per incoming byte, it's still pretty easy to spread load to effectively 2-4GB/sec worth of disk, while 2-4Gbps of Internet connectivity is pricey.
That's why most companies outsource their their mass-mailing to 3rd parties like MailJet, MailChimp, SendGrid...
As an e-mail server operator, I'm glad they do, as it makes it easy to block all the spam from those companies at the server level.
I've been added to e-mail lists without my permission quite often because I had to provide an e-mail address for the company to send me a bill or other actual important e-mail. They pass the e-mail address on to these third party companies without any confirmation on my part. Then, when you do go through the unsubscribe process, those companies claim that it might take 4-5 days to remove you from the list, when in reality we know it should happen instantly. This is so that when you keep getting the spam for the next week or so, you might not report it, or maybe if you are truly stupid, you might buy some of the junk that is advertised.
In addition, once you have any dealing with a company like MailChimp, they can play the "previous business relationship" card when sending you "something you might be interested in". And, since none of the companies you list are confirmed opt-in (which requires that they send you a first e-mail when you supposedly subscribe, and unless you click a link or reply to that e-mail, you do not get added to the list), they can play fast and loose with claims that you signed up for something that you did not.
Why does he need to send 400,000+ emails in the first place? If it's just a list of proxy domains, why not just have an RSS feed that people can subscribe to? No emails needed.
Because the people who want the proxy list would need to use a proxy to be able to read the RSS feed, as censorship in their country would block access once it was learned what the RSS feed contained.
On the other hand, incoming e-mail can be blocked as spam, but you can't decide to block an e-mail as spam without knowing something about it, and if it isn't coming from the same domain or IP address, and doesn't have the same content as previous spam, it's pretty tough to block. So, once the e-mail gets through, the "damage" is done (as far as the censors are concerned).
Hashcash distributes the load somewhat, in that it forces spammers to use more resources to send out their message and can slow them down somewhat.
Unfortunately, until you get to a significant number of bits, hashcash doesn't take all that long to compute, and you can pre-compute them.
I use 23-bit hashcash on all my outgoing e-mails, but if the address has been sent to before, there is likely a pre-computed 25-bit hashcash waiting. I use idle server time to pre-compute for any address that has been sent to from my servers. Since the hashcash expires in 25 days, I don't have to do this very often unless the recipient is a frequent one. Then, to keep the database small, I remove addresses from the "sent to" table unless they have been recently active, where "recent" depends on the total amount of activity to that address.
Blacklists are nice because they reduce server loads. Sure, running a statistical classifier for one user is not so hard, but if you have to process hundreds of millions of messages per day, that is a lot of CPU time spent on spam.
The CPU time spent on running something like SpamAssassin is insignificant compared to the bandwidth, disk writes, etc., caused by spam. Keeping the incoming e-mail in a RAM disk until you have truly accepted it for delivery (which isn't dangerous even if the server crashes hard) is the #1 thing that speeds up e-mail intake. At that point, scanning takes almost no time.
As you mention, though, greylisting does the best job of keeping your overall load down, since you don't even need to use network bandwidth on the body unless the sending server is known to retry, which basically eliminates every botnet member. Maybe this solution would work now that other more "instant" messaging systems are readily available, but 15 years ago when IM wasn't really a corporate thing, I couldn't use greylisting because "it slowed down e-mail too much", even though it didn't slow it down at all for the "important" clients, as their servers got whitelisted anyway.
From what I understand, the regulation uses something like replay gain to calculate the average volume. I don't know the exact formula being used, but it is designed to avoid the situation you refer to where both the quietest part and loudest part of the commercial is just as loud as the average volume of the show.
I don't know how they are going to enforce it, though, even with a lot of complaints, since you'd need a recording of the entire show plus commercials to know if it was a violation.
Yeah, most other 27" monitors have 13% more PPI. A panel with the same 2.37:1 aspect ratio but the same PPI as other monitors would be 2880x1216. As an upgrade to my 24" 1920x1200 monitor, that would be a reasonable option.
With other 27" monitors giving me 2560x1440, the LG in TFA is completely useless.