Parent meant Single-frequency network with single channel multicast - several transmitters simultaneously send the same signal over the same frequency channel. It's way more efficient than what we have now - adjacent senders cannot use the same frequency as it causes interference. For example, on FM here, 101.20 and 101.50 is the same channel, but from different towers - one covers North-Holland, other South.
This really makes me doubt their ability to benchmark / scale things properly. In the article, he sounds like facebook is completely CPU bound, and yet he's slamming the latest generation server processors by Intel / AMD?
From all the benchmarks I've seen, like Anandtech's and from personal experience, web servers scale pretty much linearly with clock speed * IPC and the amount of cores present in the system. The addition of HT is good for another 20% throughput.
What they need to do is to look at their setup, and make sure there isn't another bottleneck - have you spawned enough threads and processes to utilize the system completely? PHP may be "thread safe", but that usually means that there's a huge lock around everything that could be dangerous and one process refuses to use more than 100% cpu on 1 core, so serve it with apache-prefork + load balancer + separate static file server. Same thing for Python - fork off more copies via mod_wsgi even in threaded mode, as many as you can afford within the available RAM, or the Global Interpreter Lock will limit the CPU usage to 1 core.
If you have setup the environment well and there are no other bottlenecks, web services scale perfectly with the available CPU power. And that has increased by an insane amount for the Xeon 54xx to 55xx, it's almost doubled the performance in most server apps (OLTP, VM), but even the PHP test case which failed to scale to 16 cores in a single process was good for +39%.
It's an older chipset, but the Intel X-38 based motherboards accept ECC Dimm's. The price premium? 5 euros per 2GB, I've installed 4 of them in this system. As for something more recent, the X-58 Nehalem boards are ready for ECC, but to use it you need to install a Xeon.
This is a classic case of bad defaults. Yes, you will always have a trade off between performance and security, but going for either extreme is bad usability!
People expect that, without explicit syncing, the data is safe after a short period of time, measure in seconds. The old defaults were: 5 seconds in ext3, in NTFS metadata is always and data flushed asap with but no guarantees. In practice, people don't lose huge amount of work.
What happened is that the ext4 team thought waiting up to a *minute* to reorder writes is a good idea - choosing for the extreme end of performance.
My question is: WHY? Does it really matters to home users that KDE or Firefox starts 0.005 seconds faster? Apparently, the wait period is long enough to have real life consequences even with limited amount of testers, imaging what happens when it gets rolled out to everyone. On servers, it's redundant. Data is worth much, much more than anything you hope to gain and SSD's, battery backed write cache on controllers and SAN's have taken care of fsync's() already. If you run databases, those sync their disks anyway, so you just traded a huge chunk of reliability for "performance" on stuff like/home,/var/mail and/etc.
The "solution" of mounting the volume with the sync everything flag is just stupid. Yay, lets go for the other extreme - sync every bit moving to the disk. Isn't it already obvious that either extreme is silly?
Just set innodb^W ext4_flush_log_at_trx_commit on something less stupid already, flushing once every second shouldn't kill any disk. Copy Microsoft for config options: * Disable flush metadata on write -> "This setting improves disk performance, but a power outage or equipment failure might result in data loss". * Enable "advanced performance" disk write cache -> "Recommended only for disks with a battery backup power supply" etc etc. * Enable cache stuff in RAM for 60s -> "Just don't do it okay, it's stupid."
And here's another happy user of Vista. Sure, it's not perfect, but given enough spare power, you will appreciate the new features built in.
Currently, I'm running a 64 bit setup with 8GB of ram. Windows never felt faster before as everything is prefetched and you don't have to worry about closing programs anymore (eclipse for coding, netbeans for profiling & packaging, plus WoW for distraction and all the standard background services like apache + mysql). The last combo can bring a non 64bit box to it's knees with it's 3G ram limit, not to mention the slowness of swapping in firefox again after a long idle period. Vista x64 is lightyears ahead on XP x64 on driver support and usability, this single feature is enough to convince me. With DIMM's on 140 euro for 8G, why settle for anything less? Almost all developers will appreciate the ability to run everything at the same time without touching the swap file.
There are a few other features would probably made me upgrade too: Cleartype over RDP, the new start menu (no more program hunting) and much improved administrative tools like the extended eventlog for built in features and services, kernel tracers in performance tools. Also, almost every feature was updated. Minor changes like interruptable io in explorer, ctrl-scroll for changing view mode (larger - small icon - details - small list), DWM (live preview in alt-tab + hover over taskbar is nice, too bad [win]+tab is mostly a toy / tech demo) are quite nice to have.
Overall, I don't see why people are so negative over Vista. Yes, it waste much more RAM compared to XP, don't try to run it with 512MB and some reserved for onboard VGA, but with current dimm prices that really isn't a problem. If you give it enough memory, it's not noticeable slower at all, if not much faster. I've never seen eclipse reboot in less than 4 seconds, but that's normal here after the first launch after boot, it stays file-cached after that.
For a very large site I would recommend C/C++ and maybe java if they have the budget for hardware.
And throw away the near perfect scalability of the J2EE stack so you can reinvent the wheel again? I'd love to see someone try to do something like JBoss AS + clustering in plain C/C++, there's no way that you can get a decent ROI from such a project with the current hardware prices.
Webkit nightly (r30881) with Safari 3 on Windows XP = 90%, same as OS X. Too bad the text rendering is kinda broken, the anti-aliasing looks horrible, or it could become one of the major players on this platform.
With only 3 buttons, no way I'll be using this mouse in WoW. I ran out of buttons here on a Logitech G5: left and right for the usual select and action, thumb is an instant cast I use often so you can hold your hand on WSAD, tilt left is autorun, tilt right is the "oh shit" button, desperate prayer on my priest, ice block on my mage.
It may seem an overkill at first, but it really is much easier to keep moving / casting with your left hand and not have to worry about the actions mapped on the mouse.
Small batch problem? I think OpenGL solved that one with display lists - basically, you create a list with commands once, then you can execute the whole batch with a single call, instead of calling glVertex3f() for every single vertex.
Fixed pipeline isn't really an api problem either, the gpu's added a function to allow a programmer to change pipeline type, from vertex to pixel and the other way around. It doesn't look like it's hard to implement in OpenGL either, it's just a setPipelines(int, int, int) call.
From the article: To put this in real terms this new shader can take triangles and treat them like objects controlling them at their vertices. These primitives are then passed on to the pixel shader.
The source of all incompatibility - the geometry shader. But it looks like only another pass of vertex manipulation after the vertex shaders...
Conclusion: DX10 is just another step in the evolution of 3D graphics, nothing spectacular, and definately not fixing the problems with PC games. It will take a lot more than new api's to fix the gameplay (shooter #2674) and qa = release and patch afterwards.
This isn't AJAX - It's java, embedded in a web browser. They are creating 'light' versions with AJAX, but what you saw is the full version, created with swing.
Trying to create anything as complex as an office suite with a clientside interpreted language, and html + browser as graphical toolkit is just plain stupid, imo. ThinkFree got the message, and used the right tools to do the job.
There are a few ways to do something...
on
Source Code & Copyright
·
· Score: 5, Interesting
For example, there aren't much variation in ways to code a doubly linked list. If a project in java needs one, you need to write it yourself, because it isn't in java.util.* yet. With a standard coding style in that language, I've seen quite a few near identical looking implementations for an assignment.
It's about time to stop suing over one snippet of code in a project - there are only so many ways to do the basic tasks. It's how you use the individual lego blocks to build something that counts - if you copy the whole design and claim it as your own, then you deserve to be sued, not for using five white ones to build a wall, as everyone does that.
Doh. It's a *multiplayer* game, and people make appointments to get 40 friends together to do the challeging parts of the game. And generally, it's planned on Sat / Sunday, because people don't have to work.
After the little crash, most of us went away for food, shopping etc, be back at 21:00 and hope that someone noticed that the server needed a little restart. It's a good thing you have voice comm instead of only the ingame chat.
I think that anyone that plays this game thinks the same at the moment... With the world servers overloaded all the times (it's already fragmented in 50+ small ones in europe), and recently even crashing every other day (EU-Neptulon, thu 2100, today at 1700 till now), Blizzard really need to start fixing stuff instead of dreaming about expansions and new features.
I wonder if they are paying any attention, because today, it's one of the two continents that have crashed on my realm. About everyone and his dog has paged a gamemaster and the forum is full of it, but there was no reaction from Blizzard till 3 hours later - [SERVER] Restart in 15:00. I guess it's too much to give an ingame reply, or post it in the status page or forum...
most undergraduate degrees in computer science these days are basically Java vocational training
I don't know how it is elsewhere, but here (in the netherlands), we are taught to use and understand all kinds of programming languages, the usual c/c++/java, scripting languages and functional/logical ones, haskell and prolog, and even a 100 hours assignment about powerpc asm.
Re:The java generics system is a joke
on
Java 1.5 vs C#
·
· Score: 1
Comparing ArrayList in java to List in C# is kinda nonsense, as you're always open to write your own classes implementing the java.util.List interface.
Depending where you're using the huge list of ints for, you can write your own implementation, if it ever becomes the bottleneck. While I haven't tested c#, in java (1.4.2b5), it spent 0.4 sec (-Xprof, so that's including some vm startup/gc) adding a million Integers(i) using the plain ArrayList, so it's not gonna be the bottleneck anytime soon.
The overall summary, which you may or may not agree with is: Most LCDs are laggier than CRTs (I'd be jumped in an alley if I went as far as to say *all* LCDs are, but I try to avoid sweeping generalizations). Do your own tests, and come to your own conclusions. If you're a gamer, be careful. And lastly, my Dell 2001FP may in fact be one of the laggiest LCDs in existence, *or* I just received a defective unit.
Well, as far as I know, it isn't your lcd. Looking at your video, i'd say the lag is *at least* 100 ms, but the image is crystal clear, no ghosting at all. @ 60 hz, that's 6+ images buffered, meaning the lcd has a (fifo) buffer of 33 MB?
But lcd's don't work that way. lcd's sync on the signal, and redirects the current line/pixel to the correct location on the screen, most of them without any sort of buffering. The no ghosting part means that the response time is not the problem, in fact, this monitor has been nominated the best gamer's lcd a while ago by anandtech.
Try another (few) pc's, video cards (you got an spare GF2 mx?), laptops etc. For sure, you'll see soon enough for yourself that it isn't the display that's faulty.
BTW, I'm working on one right now, I use it for games, console games (via composite in, en picture-in-picture while gamefaq-ing), and I've never seen anything like that special effect in your video. Oh, and try disable the secondary display, it might help (here, the secondary gone blurry, guess club3d (rad 9600xt) don't have the quality to use both)
Yesterday I was helping my friend get set up with a Java application she needs. She's a fairly experienced Windows user. She needed to install the Sun JRE 1.4 to get this thing to work. She needed help from me, and a bit of exploring on java.sun.com, to find out which is the right file for her to download to get this thing working. If Sun wants MS Windows users to upgrade to be able to run Java apps easily and painlessly, there needs to be a big link right on java.sun.com saying "Windows users, click here to download Java for your system." Such a link does not exist and the user has to figure out "do I want NetBeans (no), do I need a SDK (no), do I need J2ME (no), do I need the JRE (yes)".
You realized that java.sun.com is the developers site? There's a huge button on sun.com saying "FREE JAVA DOWNLOAD" pointing to java.com, follow another "FREE DOWNLOAD" link and you're at the JRE.
Slashdot Mirror
Parent meant Single-frequency network with single channel multicast - several transmitters simultaneously send the same signal over the same frequency channel. It's way more efficient than what we have now - adjacent senders cannot use the same frequency as it causes interference. For example, on FM here, 101.20 and 101.50 is the same channel, but from different towers - one covers North-Holland, other South.
This really makes me doubt their ability to benchmark / scale things properly. In the article, he sounds like facebook is completely CPU bound, and yet he's slamming the latest generation server processors by Intel / AMD?
From all the benchmarks I've seen, like Anandtech's and from personal experience, web servers scale pretty much linearly with clock speed * IPC and the amount of cores present in the system. The addition of HT is good for another 20% throughput.
What they need to do is to look at their setup, and make sure there isn't another bottleneck - have you spawned enough threads and processes to utilize the system completely? PHP may be "thread safe", but that usually means that there's a huge lock around everything that could be dangerous and one process refuses to use more than 100% cpu on 1 core, so serve it with apache-prefork + load balancer + separate static file server. Same thing for Python - fork off more copies via mod_wsgi even in threaded mode, as many as you can afford within the available RAM, or the Global Interpreter Lock will limit the CPU usage to 1 core.
If you have setup the environment well and there are no other bottlenecks, web services scale perfectly with the available CPU power. And that has increased by an insane amount for the Xeon 54xx to 55xx, it's almost doubled the performance in most server apps (OLTP, VM), but even the PHP test case which failed to scale to 16 cores in a single process was good for +39%.
It's an older chipset, but the Intel X-38 based motherboards accept ECC Dimm's. The price premium? 5 euros per 2GB, I've installed 4 of them in this system. As for something more recent, the X-58 Nehalem boards are ready for ECC, but to use it you need to install a Xeon.
This is a classic case of bad defaults. Yes, you will always have a trade off between performance and security, but going for either extreme is bad usability!
People expect that, without explicit syncing, the data is safe after a short period of time, measure in seconds. The old defaults were: 5 seconds in ext3, in NTFS metadata is always and data flushed asap with but no guarantees. In practice, people don't lose huge amount of work.
What happened is that the ext4 team thought waiting up to a *minute* to reorder writes is a good idea - choosing for the extreme end of performance.
My question is: WHY? Does it really matters to home users that KDE or Firefox starts 0.005 seconds faster? Apparently, the wait period is long enough to have real life consequences even with limited amount of testers, imaging what happens when it gets rolled out to everyone. On servers, it's redundant. Data is worth much, much more than anything you hope to gain and SSD's, battery backed write cache on controllers and SAN's have taken care of fsync's() already. If you run databases, those sync their disks anyway, so you just traded a huge chunk of reliability for "performance" on stuff like /home, /var/mail and /etc.
The "solution" of mounting the volume with the sync everything flag is just stupid. Yay, lets go for the other extreme - sync every bit moving to the disk. Isn't it already obvious that either extreme is silly?
Just set innodb^W ext4_flush_log_at_trx_commit on something less stupid already, flushing once every second shouldn't kill any disk. Copy Microsoft for config options:
* Disable flush metadata on write -> "This setting improves disk performance, but a power outage or equipment failure might result in data loss".
* Enable "advanced performance" disk write cache -> "Recommended only for disks with a battery backup power supply" etc etc.
* Enable cache stuff in RAM for 60s -> "Just don't do it okay, it's stupid."
And here's another happy user of Vista. Sure, it's not perfect, but given enough spare power, you will appreciate the new features built in.
Currently, I'm running a 64 bit setup with 8GB of ram. Windows never felt faster before as everything is prefetched and you don't have to worry about closing programs anymore (eclipse for coding, netbeans for profiling & packaging, plus WoW for distraction and all the standard background services like apache + mysql). The last combo can bring a non 64bit box to it's knees with it's 3G ram limit, not to mention the slowness of swapping in firefox again after a long idle period. Vista x64 is lightyears ahead on XP x64 on driver support and usability, this single feature is enough to convince me. With DIMM's on 140 euro for 8G, why settle for anything less? Almost all developers will appreciate the ability to run everything at the same time without touching the swap file.
There are a few other features would probably made me upgrade too: Cleartype over RDP, the new start menu (no more program hunting) and much improved administrative tools like the extended eventlog for built in features and services, kernel tracers in performance tools. Also, almost every feature was updated. Minor changes like interruptable io in explorer, ctrl-scroll for changing view mode (larger - small icon - details - small list), DWM (live preview in alt-tab + hover over taskbar is nice, too bad [win]+tab is mostly a toy / tech demo) are quite nice to have.
Overall, I don't see why people are so negative over Vista. Yes, it waste much more RAM compared to XP, don't try to run it with 512MB and some reserved for onboard VGA, but with current dimm prices that really isn't a problem. If you give it enough memory, it's not noticeable slower at all, if not much faster. I've never seen eclipse reboot in less than 4 seconds, but that's normal here after the first launch after boot, it stays file-cached after that.
And throw away the near perfect scalability of the J2EE stack so you can reinvent the wheel again? I'd love to see someone try to do something like JBoss AS + clustering in plain C/C++, there's no way that you can get a decent ROI from such a project with the current hardware prices.
Webkit nightly (r30881) with Safari 3 on Windows XP = 90%, same as OS X. Too bad the text rendering is kinda broken, the anti-aliasing looks horrible, or it could become one of the major players on this platform.
With only 3 buttons, no way I'll be using this mouse in WoW. I ran out of buttons here on a Logitech G5: left and right for the usual select and action, thumb is an instant cast I use often so you can hold your hand on WSAD, tilt left is autorun, tilt right is the "oh shit" button, desperate prayer on my priest, ice block on my mage.
It may seem an overkill at first, but it really is much easier to keep moving / casting with your left hand and not have to worry about the actions mapped on the mouse.
Small batch problem? I think OpenGL solved that one with display lists - basically, you create a list with commands once, then you can execute the whole batch with a single call, instead of calling glVertex3f() for every single vertex.
Fixed pipeline isn't really an api problem either, the gpu's added a function to allow a programmer to change pipeline type, from vertex to pixel and the other way around. It doesn't look like it's hard to implement in OpenGL either, it's just a setPipelines(int, int, int) call.
From the article: To put this in real terms this new shader can take triangles and treat them like objects controlling them at their vertices. These primitives are then passed on to the pixel shader.
The source of all incompatibility - the geometry shader. But it looks like only another pass of vertex manipulation after the vertex shaders...
Conclusion: DX10 is just another step in the evolution of 3D graphics, nothing spectacular, and definately not fixing the problems with PC games. It will take a lot more than new api's to fix the gameplay (shooter #2674) and qa = release and patch afterwards.
This isn't AJAX - It's java, embedded in a web browser. They are creating 'light' versions with AJAX, but what you saw is the full version, created with swing.
Trying to create anything as complex as an office suite with a clientside interpreted language, and html + browser as graphical toolkit is just plain stupid, imo. ThinkFree got the message, and used the right tools to do the job.
For example, there aren't much variation in ways to code a doubly linked list. If a project in java needs one, you need to write it yourself, because it isn't in java.util.* yet. With a standard coding style in that language, I've seen quite a few near identical looking implementations for an assignment.
It's about time to stop suing over one snippet of code in a project - there are only so many ways to do the basic tasks. It's how you use the individual lego blocks to build something that counts - if you copy the whole design and claim it as your own, then you deserve to be sued, not for using five white ones to build a wall, as everyone does that.
After the little crash, most of us went away for food, shopping etc, be back at 21:00 and hope that someone noticed that the server needed a little restart. It's a good thing you have voice comm instead of only the ingame chat.
I think that anyone that plays this game thinks the same at the moment... With the world servers overloaded all the times (it's already fragmented in 50+ small ones in europe), and recently even crashing every other day (EU-Neptulon, thu 2100, today at 1700 till now), Blizzard really need to start fixing stuff instead of dreaming about expansions and new features. I wonder if they are paying any attention, because today, it's one of the two continents that have crashed on my realm. About everyone and his dog has paged a gamemaster and the forum is full of it, but there was no reaction from Blizzard till 3 hours later - [SERVER] Restart in 15:00. I guess it's too much to give an ingame reply, or post it in the status page or forum...
I don't know how it is elsewhere, but here (in the netherlands), we are taught to use and understand all kinds of programming languages, the usual c/c++/java, scripting languages and functional/logical ones, haskell and prolog, and even a 100 hours assignment about powerpc asm.
Comparing ArrayList in java to List in C# is kinda nonsense, as you're always open to write your own classes implementing the java.util.List interface.
Depending where you're using the huge list of ints for, you can write your own implementation, if it ever becomes the bottleneck. While I haven't tested c#, in java (1.4.2b5), it spent 0.4 sec (-Xprof, so that's including some vm startup/gc) adding a million Integers(i) using the plain ArrayList, so it's not gonna be the bottleneck anytime soon.
But lcd's don't work that way. lcd's sync on the signal, and redirects the current line/pixel to the correct location on the screen, most of them without any sort of buffering. The no ghosting part means that the response time is not the problem, in fact, this monitor has been nominated the best gamer's lcd a while ago by anandtech.
Try another (few) pc's, video cards (you got an spare GF2 mx?), laptops etc. For sure, you'll see soon enough for yourself that it isn't the display that's faulty.
BTW, I'm working on one right now, I use it for games, console games (via composite in, en picture-in-picture while gamefaq-ing), and I've never seen anything like that special effect in your video. Oh, and try disable the secondary display, it might help (here, the secondary gone blurry, guess club3d (rad 9600xt) don't have the quality to use both)
Yesterday I was helping my friend get set up with a Java application she needs. She's a fairly experienced Windows user. She needed to install the Sun JRE 1.4 to get this thing to work. She needed help from me, and a bit of exploring on java.sun.com, to find out which is the right file for her to download to get this thing working. If Sun wants MS Windows users to upgrade to be able to run Java apps easily and painlessly, there needs to be a big link right on java.sun.com saying "Windows users, click here to download Java for your system." Such a link does not exist and the user has to figure out "do I want NetBeans (no), do I need a SDK (no), do I need J2ME (no), do I need the JRE (yes)". You realized that java.sun.com is the developers site? There's a huge button on sun.com saying "FREE JAVA DOWNLOAD" pointing to java.com, follow another "FREE DOWNLOAD" link and you're at the JRE.