I had one JMicron drive that was silently failing to write one section of the drive (every other bit was always zero). I'd always heard that SSDs were supposed to check that, but I'm not entirely sure that was a JMicron specific issue.
If you send a write request and then a read request to the JMicron controller, you could very well wait a second before your first bit is returned. That is quite a bit of latency.
Incredibly, Microsoft had this working in Windows Vista. Granted, very few people know about it so they suffer through. I had a 24" LCD and a 17" laptop screen both running at 1900x1200 in Vista. Icons and windows had the exact same dimensions in centimeters between the two.
The downside is it depended on the application being built with support for the rendering, which I believe all.NET apps had by default. Older Windows applications that were hard coded for 96 DPI rendered as normal, and were zoomed by the OS, resulting in a slightly blurry look.
Agreed. If I knew what I was doing, I would fabricate a frame and adapter that could be used with a variety of aftermarket replacement laptop LCDs. Then I would sell them to people that want small high resolution monitors. I've seen people talk about this, but nobody has bothered to do much about it.
I'd like to add the ironic #4: Parents taking an interest in their child's education that have no perspective. In lower socio-economic areas uneducated/welfare parents decide to "do the right thing" and become heavily involved in their child's education. Unfortunately this invariable devolves into arguing with the teachers because they failed their son for not doing any of the work, or learning the material. Parents that never learned to succeed* often have a terribly warped view of how and why children should be rewarded in education. Their involvement in their child's education usually causes more trouble than good.
*Success isn't necessarily measured in monetary value.
Buying ATI was one of the only good things AMD corporate has done recently. Talk to engineers at AMD about the past several years and you will get an ear full about the mismanagement that has been going on.
Well, buying ATI was a potentially good move. It remains to be seen if they manage to leverage these incredible new resources they purchased. (And no, selling graphics cards labeled as "AMD" was not the point of purchasing ATI.)
I had a similar experience with a low card diet, except that I did exercise. I was still somewhat overweight, and a routine visit to the doctor with a bloodtest showed all of my blood levels to be good. The doctor said everything about me was healthy, but that I needed to lose weight...
but as it continues you feel better and at the end of your workout you feel great. This is true whether your fat or thin.
Sorry, but this is simply false. You're taking your own measure and generalize it over everyone.
After my body warms up and I finish the exercise, I feel aching, strained, soaked with sweat, and generally far worse off than before I started. Maybe some people find physical effort to be pleasurable, I don't. At best I don't mind it, at worst it feels horrible. I always wanted my PE classes to finish already, and when it was finally over, I was feeling crawling out of the gym, aching all over and not in a good way.
This mirrors my own experience. After years of working out (everything from only weights to only cardio), changing my diet (from high fiber to low carb), losing weight (with diet and exercise), gaining weight (off the diet, or exercise), I can conclusively say that I have never felt better during or after exercising. I don't know what it is that other people feel that makes them say they feel better after exercising, but I have never felt that. Really.
For some of us, working out is just plain crap, all the time, every time.
For those that haven't come across Starship Troopers 2, it was produced so cheaply that all of the effect were new actors green screened in front of scenes from the original movie, and not in a good way. Combine that with acting so bad they must have been paying in literal peanuts, and you have a movie so bad it could rival the Star Wars Holiday Special.
I was just trying to be helpful by pointing out an easier method for using the Kill-A-Watt, although I don't typically use it for extended periods of time. I usually just plug it in and test the various states.
Thanks for the link. Very interesting, especially "Write operations can take approximately four times longer than read operations." I'm guessing that the article is directed primarily at people that want users to store all email locally rather than simply archiving old email in a more read only fashion, although the arguments seem to hold either way.
The funny thing is our file server storage is sufficiently fast that using PST files over the gigabit LAN is substantially faster than on the local system. And the load is negligible, as was mentioned it's essentially read only storage for old data.
Currently users save emails as a year or two per PST file. Users have roaming profiles, and all folders are automatically redirected to file shares, including a share specifically for Outlook PST files. As the PST files don't change in any meaningful way, the daily backups are set to skip them. (PST files change each time Outlook touches them.)
One of my duties is to regularly scour through the backups and try to find ways of minimizing the amount of data that needs to be backed up each time, as well as reducing overall backup time. Moving old emails to PST files that are backed up less frequently helped, as well as reducing the load on the Exchange server. (A good deduplicating backup solution would help a lot here, if they ever decided the outrageous licensing fees were worth it.)
I know another slashdotter mentioned that hosting PST files on a Windows server could bring the server to its knees, but we've done testing and have never had an issue. Luck to us?
I know we still use PST files. People want to keep 10GB in emails, which just becomes ridiculous from an IT perspective. It adds a lot for storage and backup costs, plus backup times and space. What would be useful is if there was an Exchange archive system that were essentially read only. Move all emails older than a year to it so you only have to back it up once every blue moon and the storage could be slower.
What about collisions? Hopefully you aren't storing emails from more than one person, or that would be a very interesting potential security leak.
The chance of a random collision for SHA-2/256 is 1 in 2^256, which is less probable than you winning every lottery for the rest of your life. Zero, for all practical purposes.
Yes, storing emails for multiple accounts. Users don't know the hash values of the files on the server, and they can't access the database directly. The files aren't overwritten, so they couldn't be corrupted. An attack would involve dumping the database to find relevant hash values. Then discovering an attack against SHA-2 (no practical attacks exist against SHA-1, so SHA-2 is pretty safe right now) to generate a new file with an identical hash value. Then email that file to themselves, and download the pre-existing hashed file.
So, no, not a security leak. If a weakness were discovered against SHA-2, then rehashing to a new method would be trivial. And if can already get a dump of the database, then access is probably compromised already, and it doesn't matter.
Would you provide links to the resellers you used for the parts? Also, I'm really curious what your total system power consumption is. I want to build a file server with several drives, and I've been looking for a way to justify spending a little more for a low powered Atom.
I'd be curious what your total idle power is because 40% of 400W would still be quite a bit. (I would expect 10W per drive max, but I'd still like to know.)
It wasn't clear to me in your original message if the "uplink" referred to was at the radio level, or the backhaul from the tower. I assume you were referring to the cell device's ability to send bits to the tower, and not the tower's ability to send bits to the (central office?). Is that correct?
The corruption isn't a fault of the spec, it's the implementation. A proper implementation wouldn't have this problem.
This is not necessarily true. Some designs are inherently more resilient to the natural random corruption which can occur on computers. The issues with PST files may be the result of poor implementation from Outlook, or it may be the design of PST files themselves. It would require close inspection of the specification to know for sure.
You can just drag those messages off to another machine running IMAP and then have google pop them off from there.
That's what I did. I had Google auto-label the emails it pulled out of the IMAP accounts. Sadly, I somehow lost a lot of my emails at one point, but I still have most of them going back to 2000 imported into my Gmail account.
Slashdot Mirror
I had one JMicron drive that was silently failing to write one section of the drive (every other bit was always zero). I'd always heard that SSDs were supposed to check that, but I'm not entirely sure that was a JMicron specific issue.
If you send a write request and then a read request to the JMicron controller, you could very well wait a second before your first bit is returned. That is quite a bit of latency.
Incredibly, Microsoft had this working in Windows Vista. Granted, very few people know about it so they suffer through. I had a 24" LCD and a 17" laptop screen both running at 1900x1200 in Vista. Icons and windows had the exact same dimensions in centimeters between the two.
The downside is it depended on the application being built with support for the rendering, which I believe all .NET apps had by default. Older Windows applications that were hard coded for 96 DPI rendered as normal, and were zoomed by the OS, resulting in a slightly blurry look.
Agreed. If I knew what I was doing, I would fabricate a frame and adapter that could be used with a variety of aftermarket replacement laptop LCDs. Then I would sell them to people that want small high resolution monitors. I've seen people talk about this, but nobody has bothered to do much about it.
http://www.google.com/cse?cx=008032414425079535247%3Akplxrakvu20&q=laptop+lcd&sa=Search
Here are some inverter boards, but they all seem to be relatively low resolution, and only supporting a VGA (DB15) connector. (Why exactly is it always cheaper to have a VGA port instead of a DVI port on a digital LCD panel?)
http://controllers-lcds.shopeio.com/inventory/catalog.asp?ACTION=2&cat=Lcds&sub=Controllers
So yeah, it's a mystery.
I'd like to add the ironic #4: Parents taking an interest in their child's education that have no perspective. In lower socio-economic areas uneducated/welfare parents decide to "do the right thing" and become heavily involved in their child's education. Unfortunately this invariable devolves into arguing with the teachers because they failed their son for not doing any of the work, or learning the material. Parents that never learned to succeed* often have a terribly warped view of how and why children should be rewarded in education. Their involvement in their child's education usually causes more trouble than good.
*Success isn't necessarily measured in monetary value.
Buying ATI was one of the only good things AMD corporate has done recently. Talk to engineers at AMD about the past several years and you will get an ear full about the mismanagement that has been going on.
Well, buying ATI was a potentially good move. It remains to be seen if they manage to leverage these incredible new resources they purchased. (And no, selling graphics cards labeled as "AMD" was not the point of purchasing ATI.)
Great co-workers, good projects, extremely poor corporate management.
So, just like AMD?
I had a similar experience with a low card diet, except that I did exercise. I was still somewhat overweight, and a routine visit to the doctor with a bloodtest showed all of my blood levels to be good. The doctor said everything about me was healthy, but that I needed to lose weight...
-Atamido
but as it continues you feel better and at the end of your workout you feel great. This is true whether your fat or thin.
Sorry, but this is simply false. You're taking your own measure and generalize it over everyone.
After my body warms up and I finish the exercise, I feel aching, strained, soaked with sweat, and generally far worse off than before I started. Maybe some people find physical effort to be pleasurable, I don't. At best I don't mind it, at worst it feels horrible. I always wanted my PE classes to finish already, and when it was finally over, I was feeling crawling out of the gym, aching all over and not in a good way.
This mirrors my own experience. After years of working out (everything from only weights to only cardio), changing my diet (from high fiber to low carb), losing weight (with diet and exercise), gaining weight (off the diet, or exercise), I can conclusively say that I have never felt better during or after exercising. I don't know what it is that other people feel that makes them say they feel better after exercising, but I have never felt that. Really.
For some of us, working out is just plain crap, all the time, every time.
-Atamido
I was under the impression that an 80 IQ bordered on mentally deficient. Is this not the case?
Are you saying the emergency brake lever doesn't work in a Prius if there is no power?
For those that haven't come across Starship Troopers 2, it was produced so cheaply that all of the effect were new actors green screened in front of scenes from the original movie, and not in a good way. Combine that with acting so bad they must have been paying in literal peanuts, and you have a movie so bad it could rival the Star Wars Holiday Special.
I was just trying to be helpful by pointing out an easier method for using the Kill-A-Watt, although I don't typically use it for extended periods of time. I usually just plug it in and test the various states.
Thanks for the link. Very interesting, especially "Write operations can take approximately four times longer than read operations." I'm guessing that the article is directed primarily at people that want users to store all email locally rather than simply archiving old email in a more read only fashion, although the arguments seem to hold either way.
The funny thing is our file server storage is sufficiently fast that using PST files over the gigabit LAN is substantially faster than on the local system. And the load is negligible, as was mentioned it's essentially read only storage for old data.
Currently users save emails as a year or two per PST file. Users have roaming profiles, and all folders are automatically redirected to file shares, including a share specifically for Outlook PST files. As the PST files don't change in any meaningful way, the daily backups are set to skip them. (PST files change each time Outlook touches them.)
One of my duties is to regularly scour through the backups and try to find ways of minimizing the amount of data that needs to be backed up each time, as well as reducing overall backup time. Moving old emails to PST files that are backed up less frequently helped, as well as reducing the load on the Exchange server. (A good deduplicating backup solution would help a lot here, if they ever decided the outrageous licensing fees were worth it.)
I know another slashdotter mentioned that hosting PST files on a Windows server could bring the server to its knees, but we've done testing and have never had an issue. Luck to us?
I know we still use PST files. People want to keep 10GB in emails, which just becomes ridiculous from an IT perspective. It adds a lot for storage and backup costs, plus backup times and space. What would be useful is if there was an Exchange archive system that were essentially read only. Move all emails older than a year to it so you only have to back it up once every blue moon and the storage could be slower.
I just put together a SuperMicro 5014A-H 1U server
So you have a link to the product page? I can't seem to track it down.
What about collisions? Hopefully you aren't storing emails from more than one person, or that would be a very interesting potential security leak.
The chance of a random collision for SHA-2/256 is 1 in 2^256, which is less probable than you winning every lottery for the rest of your life. Zero, for all practical purposes.
Yes, storing emails for multiple accounts. Users don't know the hash values of the files on the server, and they can't access the database directly. The files aren't overwritten, so they couldn't be corrupted. An attack would involve dumping the database to find relevant hash values. Then discovering an attack against SHA-2 (no practical attacks exist against SHA-1, so SHA-2 is pretty safe right now) to generate a new file with an identical hash value. Then email that file to themselves, and download the pre-existing hashed file.
So, no, not a security leak. If a weakness were discovered against SHA-2, then rehashing to a new method would be trivial. And if can already get a dump of the database, then access is probably compromised already, and it doesn't matter.
Would you provide links to the resellers you used for the parts? Also, I'm really curious what your total system power consumption is. I want to build a file server with several drives, and I've been looking for a way to justify spending a little more for a low powered Atom.
I use extension cords with my Kill-A-Watt to make it easier to view and plug in.
I'd be curious what your total idle power is because 40% of 400W would still be quite a bit. (I would expect 10W per drive max, but I'd still like to know.)
It wasn't clear to me in your original message if the "uplink" referred to was at the radio level, or the backhaul from the tower. I assume you were referring to the cell device's ability to send bits to the tower, and not the tower's ability to send bits to the (central office?). Is that correct?
The corruption isn't a fault of the spec, it's the implementation. A proper implementation wouldn't have this problem.
This is not necessarily true. Some designs are inherently more resilient to the natural random corruption which can occur on computers. The issues with PST files may be the result of poor implementation from Outlook, or it may be the design of PST files themselves. It would require close inspection of the specification to know for sure.
I thought TBird used a different file for each folder. Is that not correct?
You can just drag those messages off to another machine running IMAP and then have google pop them off from there.
That's what I did. I had Google auto-label the emails it pulled out of the IMAP accounts. Sadly, I somehow lost a lot of my emails at one point, but I still have most of them going back to 2000 imported into my Gmail account.