What a shame they spoil it by saying something like this, though:
WoW is basically capped at 64 fps with dual core systems so it is difficult to determine if the Killer NIC would provide any further frame rate increases with our test platform.
Err, try turning off the Vertical Sync checkbox in the video options and you won't be limited to the 60fps refresh rate of your screen.:)
So having a card that does all the network processing itself, without relying on the CPU, would avoid that slowdown.
Last time I checked, network packets get built in main memory by your Berkley/Winsock stack and the final packets with headers and checksums get transferred to your NIC by DMA for transmission down the wire. In a modern gaming machine we're talking a (typically) 800MHz/128-bit memory bus, versues a 33MHz/32-bit PCI bus. How does having a Killer NIC improve this situation?
All I can think of is that Killer NIC is tweaking the Priority and QoS attributes of packets to help your upstream packets reach the server more quickly, but it's not going to do anything for your downstream traffic (originating from a server that's not going to tweak Priority and QoS) which is going to be far larger in terms of bytecount. The only traffic being sent from you is your own position/action updates, but the server is sending you the state of everything in the zone you need to know about (all other players, active object states, etc).
The only thing remarkable about RFID technology is the battery-free property. Anything which needs power might as well be called a "radio transmitter" and not RFID.
RFIDs are radio transmitters, just with a bit of smarts. Battery-powered RFID is completely valid. They have drive-by applications where the tag cannot be expected to pass within a couple of meters of reader/writer devices, but because of battery life limitations need to be replaced every few years.
So-called "Passive RFID" tags are still active devices, they just get their power by rectifying the incoming RF signal from the reader/writer device. Some of the incoming signal must be snooped to see if the tag should respond, the rectification process will always be less than 100% efficient, the digital/memory circuits need power, and the RF transmission circuitry will always be less than 100% efficient, so the response transmission range of passive tags will always be fairly low - a couple of meters at best, dependant on the transmission power output of the reader/writer device and the distance between the reader/writer and the tag (power falls off at the inverse-square of the distance).
Active RFID tags have their own battery but spend most of their time in a "sleep" state. When they receive an authenticated signal they'll power their transmission using the battery and then go back to sleep. This enables them to increase their transmission range (sometimes further than the range the reader/writer device can reach itself).
I'd pay the correct 72 cents (well, 71.786 cents if you can manage it) on the invoice and then wait for someone from their accounts to phone you up about it.
It's quite tricky because, typically, the TCP/IP stack has no knowledge of which application is the originator of a particular packet.
Actually, it's completely trivial. All TCP (and UDP) activity from apps (and services) is made through the winsock and wsock32 DLL's and it's trivial to figure out the pid and tid of the caller using GetCurrentProcess() and GetCurrentThread(). I believe that this is how NetLimiter has always worked, setting itself up as a "Device Filter", and now Microsoft is "building in" that functionality to their own version of NetLimiter.
If your bank really wants to make it easy for people to rip them off, it's not really your problem is it? [blah blah, waffle waffle, etc...]
That's absolute crap. As someone who's been on the pointy end of the stick by having their Visa card abused after its details were stolen from a vendor's supposedly-secure (PCI compliance be damned) database I can tell you it is a big problem for the consumer. The bank has nothing to do with it: Visa themselves took every single one of their 45 business days to "investigate" the complaint before getting the bank to credit the charge back to my account, and the interest charges were never reimbursed.
As a result, end users will be able to more easily share files between Microsoft Office and OpenOffice.org, as documents will better maintain consistent formats, formulas and style templates across the two office productivity suites.
Maybe it's just the pessimist in me, but this sounds like a Divide and Conquer strategy to me.
With the OpenDocument format standard becoming a published ISO standard this week, who cares about Microsoft's OpenXML format? Forking OO.o just means that bugs and security problems will have to be fixed by two sources, deployed by two sources, and cause interoperability problems between users of vanilla OO.o and Novell's OO.
All to cause confusion and allow Microsoft to paint themselves in a better light than the FOSS community.
I hate to tell you this, but 35 in a 45 zone is not impeding traffic. 45 is the absolute limit under optimal conditions.
Yes, it is the absolute limit and people should slow down when the conditions inhibit it (such as fog and driving rain).
Police in New South Wales (Australia) used to routinely book people for obstructing traffic if they were driving slower than 10kmh below the signed limit. We started seeing Road Rage when they stopped. There's also a national road rule that when driving in a 90kmh or faster zone it's "keep left unless overtaking" - not nearly enough people get booked for ignoring it, though.
Long exposures (>60 seconds) will remove most moving objects (cars, trucks, people).
Or with computers, a series of short digital exposures which only keep the content "common" between the frames (moving objects will be in different parts of subsequent frames).
It does *claim* to handle.xvid files, but I haven't tried any yet. It also claims to handle.flv files: so far I get video playback of about 90% of files and audio playback of 0% (that's zero), even ones I'm playing them back directly from YouTube. Colour me unimpressed.
I haven't yet seen a notebook that didn't have integrated video. However, when I bought my Gateway notebook with a docking station (as opposed to a port replicator) way back when I was able to use a video card in one of the PCI slots to "upgrade".
Scoble responds that Microsoft's innovation can be found in the little things: 'I remember when they improved the error messages you get in Internet Explorer, or when they improved fonts in Windows with ClearType technology.
How quickly they forget that ClearType, the method as Microsoft describes it, is a direct rip-off of the font smoothing technology Apple came up with for using Apple II's on (comparatively) lo-res colour television displays in the mid-1980's.
If they can't see ads and pop-ups then they can't click on them to download and install malware. "Your computer's time is wrong, click here it fix it" works because non-technical people don't know any better.
It supports rich applications like Firefox and Thunderbird using XUL and XPCOM. Cross-platform (and unlike Microsoft's implementation of the term "cross-platform", it doesn't just mean several versions of Windows). It supports themes and plugins.
Why get rid of Web version x.x? The web is a hypermedia document service, get the applications out of it.
How can they be in total compliance with the contract if they're not running the game on a supported OS?
Compliance? Look in "World of Warcraft/Data/eula.html" and "World of Warcraft/Data/tos.html" and tell me where it places any retrictions or terms and conditions regarding the OS you run WoW on. Sure, they don't support anything but Windows, but that doesn't mean that you're not allowed to use anything other than Windows.
If this does turn out to be a Linux/Cedega ban hammer then my account will certainly be terminated before the end of the year (when I finally get rid of my last Windows machine).
The 50k/s is going to be a limit of their connection. There's not much you're going to be able to do to improve the situation for that individual aside from (1) smaller chunks and (2) compression. A lot of people here have mentioned breaking up your data into smaller chunks, which is valid and first priority. Have you also considered serving-up a compressed version of the data, say using a.gz'd version of the data file on your server with the http/1.1 header "Content-Transfer-Encoding: gzip"? There's probably going to be *lots* of redundant data in your data files so they should compress really well - smaller chunks, quicker downloads.:) I wouldn't have the server compressing the file on the fly, by the way, especially if it's likely to be getting a lot of requests (it will hammer the CPU to death).
I keep seeing a number of people bitching about FF's speed. As far as I'm concerned, FF's browsing speed is about the same as IE's and its Javascript speed leaves IE for dead when changing page content dynamically.
For example, try the Javascript code below on a page with the two multiple-select list controls also below (I'd post the whole page but the/. lameness filter doesn't like it)...
With 2,000 items in the list on my machine FF 2.0 completes a move of all items from one list box to the other in 0.631 seconds; MS-IE 6.0 takes 2.420 seconds and MS-IE 7.0 takes 4.487 seconds. With 10,000 items in the list the times become 17.395, 81.579 and 123.828 seconds respectively.
MS-IE's main problem appears to be it's "Instant Redraw", causing it to redraw the list controls immediately after every change. Gecko seems to postpone its redraws so it only redraws the page once at the very end.
function MoveItems(oSourceList, oTargetList, bAllItems) { var oStartDate = new Date; var dtStartTime = oStartDate.getTime(); var iSourceIndex=0; while (iSourceIndex<oSourceList.options.length) { var oOption = oSourceList.options[iSourceIndex]; if (oOption.selected || bAllItems) { oTargetList.options.add( new Option(oOption.text, oOption.value) ); oSourceList.options[iSourceIndex] = null; } else { iSourceIndex++; } } var oFinishDate = new Date; var dtFinishTime = oFinishDate.getTime(); var iMilliseconds = dtFinishTime - dtStartTime; if (bAllItems) { alert("Completed in " + (iMilliseconds/1000) + " seconds."); } return 0; }
Hey Dean, in your response to Question 4 you said this:
I think there's a clear difference between the protection offered in IE7 and other places. I suggest readers look here and here and decide for themselves.
Unfortunately there were no links on "here" and "here"... any chance you could provide them for us? Thanks!
Slashdot Mirror
Last time I checked, network packets get built in main memory by your Berkley/Winsock stack and the final packets with headers and checksums get transferred to your NIC by DMA for transmission down the wire. In a modern gaming machine we're talking a (typically) 800MHz/128-bit memory bus, versues a 33MHz/32-bit PCI bus. How does having a Killer NIC improve this situation?
All I can think of is that Killer NIC is tweaking the Priority and QoS attributes of packets to help your upstream packets reach the server more quickly, but it's not going to do anything for your downstream traffic (originating from a server that's not going to tweak Priority and QoS) which is going to be far larger in terms of bytecount. The only traffic being sent from you is your own position/action updates, but the server is sending you the state of everything in the zone you need to know about (all other players, active object states, etc).
RFIDs are radio transmitters, just with a bit of smarts. Battery-powered RFID is completely valid. They have drive-by applications where the tag cannot be expected to pass within a couple of meters of reader/writer devices, but because of battery life limitations need to be replaced every few years.
So-called "Passive RFID" tags are still active devices, they just get their power by rectifying the incoming RF signal from the reader/writer device. Some of the incoming signal must be snooped to see if the tag should respond, the rectification process will always be less than 100% efficient, the digital/memory circuits need power, and the RF transmission circuitry will always be less than 100% efficient, so the response transmission range of passive tags will always be fairly low - a couple of meters at best, dependant on the transmission power output of the reader/writer device and the distance between the reader/writer and the tag (power falls off at the inverse-square of the distance).
Active RFID tags have their own battery but spend most of their time in a "sleep" state. When they receive an authenticated signal they'll power their transmission using the battery and then go back to sleep. This enables them to increase their transmission range (sometimes further than the range the reader/writer device can reach itself).
I'd pay the correct 72 cents (well, 71.786 cents if you can manage it) on the invoice and then wait for someone from their accounts to phone you up about it.
Actually, it's completely trivial. All TCP (and UDP) activity from apps (and services) is made through the winsock and wsock32 DLL's and it's trivial to figure out the pid and tid of the caller using GetCurrentProcess() and GetCurrentThread(). I believe that this is how NetLimiter has always worked, setting itself up as a "Device Filter", and now Microsoft is "building in" that functionality to their own version of NetLimiter.
How about 36.62 uLOC/sec?
/.)
(Sure would be nice if ampersand-mu or ampersand-#956 worked on
That's absolute crap. As someone who's been on the pointy end of the stick by having their Visa card abused after its details were stolen from a vendor's supposedly-secure (PCI compliance be damned) database I can tell you it is a big problem for the consumer. The bank has nothing to do with it: Visa themselves took every single one of their 45 business days to "investigate" the complaint before getting the bank to credit the charge back to my account, and the interest charges were never reimbursed.
Maybe it's just the pessimist in me, but this sounds like a Divide and Conquer strategy to me.
With the OpenDocument format standard becoming a published ISO standard this week, who cares about Microsoft's OpenXML format? Forking OO.o just means that bugs and security problems will have to be fixed by two sources, deployed by two sources, and cause interoperability problems between users of vanilla OO.o and Novell's OO.
All to cause confusion and allow Microsoft to paint themselves in a better light than the FOSS community.
Yes, it is the absolute limit and people should slow down when the conditions inhibit it (such as fog and driving rain).
Police in New South Wales (Australia) used to routinely book people for obstructing traffic if they were driving slower than 10kmh below the signed limit. We started seeing Road Rage when they stopped. There's also a national road rule that when driving in a 90kmh or faster zone it's "keep left unless overtaking" - not nearly enough people get booked for ignoring it, though.
Nobody said the vehicle is taking pictures while on the move.
If you read the The Road to Knowhere: Microsoft Virtual Earth with "Street Side Views" you'll note that it's stills, not video, being presented.
Long exposures (>60 seconds) will remove most moving objects (cars, trucks, people).
Or with computers, a series of short digital exposures which only keep the content "common" between the frames (moving objects will be in different parts of subsequent frames).
It does *claim* to handle .xvid files, but I haven't tried any yet. It also claims to handle .flv files: so far I get video playback of about 90% of files and audio playback of 0% (that's zero), even ones I'm playing them back directly from YouTube. Colour me unimpressed.
I haven't yet seen a notebook that didn't have integrated video. However, when I bought my Gateway notebook with a docking station (as opposed to a port replicator) way back when I was able to use a video card in one of the PCI slots to "upgrade".
Don't bother. None of the editors around here have even seen a dictionary, let alone realize that "procede" doesn't exist in any of them.
If they can't see ads and pop-ups then they can't click on them to download and install malware. "Your computer's time is wrong, click here it fix it" works because non-technical people don't know any better.
XULRunner
It supports rich applications like Firefox and Thunderbird using XUL and XPCOM. Cross-platform (and unlike Microsoft's implementation of the term "cross-platform", it doesn't just mean several versions of Windows). It supports themes and plugins.
Why get rid of Web version x.x? The web is a hypermedia document service, get the applications out of it.
How does "spider.txt" get an Insightful when it's "robots.txt"? Sheesh, bump the Mods Roster.
The 50k/s is going to be a limit of their connection. There's not much you're going to be able to do to improve the situation for that individual aside from (1) smaller chunks and (2) compression. .gz'd version of the data file on your server with the http/1.1 header "Content-Transfer-Encoding: gzip"? There's probably going to be *lots* of redundant data in your data files so they should compress really well - smaller chunks, quicker downloads. :)
A lot of people here have mentioned breaking up your data into smaller chunks, which is valid and first priority.
Have you also considered serving-up a compressed version of the data, say using a
I wouldn't have the server compressing the file on the fly, by the way, especially if it's likely to be getting a lot of requests (it will hammer the CPU to death).
I keep seeing a number of people bitching about FF's speed. As far as I'm concerned, FF's browsing speed is about the same as IE's and its Javascript speed leaves IE for dead when changing page content dynamically.
For example, try the Javascript code below on a page with the two multiple-select list controls also below (I'd post the whole page but the /. lameness filter doesn't like it)...
With 2,000 items in the list on my machine FF 2.0 completes a move of all items from one list box to the other in 0.631 seconds; MS-IE 6.0 takes 2.420 seconds and MS-IE 7.0 takes 4.487 seconds. With 10,000 items in the list the times become 17.395, 81.579 and 123.828 seconds respectively.
MS-IE's main problem appears to be it's "Instant Redraw", causing it to redraw the list controls immediately after every change. Gecko seems to postpone its redraws so it only redraws the page once at the very end.
andIt's always nice to see that Technical Writers for IT magazines are savy enough to know the difference between a Domain Name and a Host Name.
Erk, has anyone taken a look at Dr. Narang's website, The fusion of two sciences, Web Vastu?
It was only five months ago that we were being told: Ozone Layer Improving Faster Than Expected.