They're three distinct things, and Vista dropped OCA.
WER handles the reporting of the errors (formerly called Dr. Watson) OCA handles the analysis of the reports, and informs the user of the results (Vista integrated this into the WER interface) CEIP reports usability data from certain applications, such as Windows Live Messenger, and doesn't collect program crash info.
You talk about "pro-MS" propaganda, but you're the one desperately searching for things to shit on.
How exactly would they get notified of their error? The idea behind a secret ballot is that nobody but the voter sees the ballot at a point where the person voting can be identified.
In Canadian federal elections, NOBODY sees your ballot after it is first handed to you, until they break the seal on the ballot boxes for tabulation. And at that point, there's no way of tying a particular ballot to a particular person. As soon as the voter fills in their ballot, the voter folds it in a specific way before leaving the booth, and it remains so folded as they walk over to the ballot box and insert it into the slot.
Any system that allows for notifying a voter that they made an error is a flawed and dangerous democratic process. I'm shocked that this is an accepted practice in the US.
I wouldn't necessarily call the Mac Mini underpowered. The CPU is, after all, dual core, and quite fast for non-intensive uses (about what you'd expect in a $600 computer).
However, with RAM prices being what they are (last I checked 2GB cost about $50), I'd expect 2GB of RAM and a significantly larger hard drive for that price. The fact that they use notebook components to save space and power is neither here nor there.
Which would be why blu-ray discs that aren't in caddies have special coatings on their bottom surfaces that make them MORE durable than both CDs and DVDs (unless they themselves are treated).
The coatings really do work quite well. They don't make the surface completely unscratchable, but they do make them far more durable than a standard CD/DVD.
While the media itself limited the recording speed in the past
The media itself limited recording speed in the past because the discs would shatter if you spun them too fast. It'd be nice if they let us know what has changed with BluRay that means that discs won't tear themselves apart at high rotational speeds. Are they not made of the same polycarbonate materials as HD-DVD/DVD/CD are?
No, 1x was meant to mean the same speed that the audio played at; one-times real-time. 2x would mean you burn/read at twice the rate of playback. 1x never meant you could burn a CD in one hour. For example, a standard CD-R is 72 minutes, and takes.... 72 minutes to burn at 1x. Most CDs are 80 minutes these days. I'm sure you can figure out how long they take to burn at 1x.
1x happens to be 150KB/s, but that wasn't the original definition.
Not exactly. TekSavvy does pay bell a wholesale rate ($20.50 per customer), but Bell doesn't give them a complete service.
That $20.50 pays for Bell's service of maintaining the last-mile connection between the customer's modem and the DSLAM in the CO.
The ISP (TekSavvy) must pay for a connection to Bell's ATM network in order to get the traffic from the CO to the ISP's network ($1300 for a GigE). From there, the ISP is responsible for internet connectivity. TekSavvy's primary transit is through Peer 1 (premium customers) and Cogent (unlimited customers) with various other things in the mix (TorIX, Teleglobe, etc). I know they recently purchased some InterNap hardware, but I'm not sure if they're using them for transit too.
As you mentioned, the ISP is responsible for providing technical support. However, many issues require TekSavvy to open trouble tickets with Bell (who provide pretty bad service to TekSavvy) in order to get problems resolved. This is because many problems with DSL involve incorrect settings made by Bell techs at the CO, a constant annoyance for TekSavvy.
Anyhow, that base fee only provides the last-mile. The ISP themselves pay to get the traffic from Bell to themselves, and from there it's entirely through the ISP's own network.
The end result is that TekSavvy can provide far better customer service and performance at lower prices than Bell. This is why they're signing up 1500-1800 customers per month. That figure, BTW, comes directly from TekSavvy themselves, who post on DSLReports a lot of information that most ISPs keep secret.
I don't agree. The Aptera weighs 850 pounds. Let's assume it carries two 200 pound passengers, for a total weight of 1250 pounds.
The Honda Fit weighs more than twice as much, and has 109HP.
The Aptera should, therefore, do quite nicely with 50HP. Equivalent to a 5-door Honda Fit.
At 50HP, an 8-hour charge (144KWh) gets you ~3.9 hours of driving at full load.
For most people, considering city driving (far from full load all the time), that is far more than enough.
However, when it comes to vehicles like the Tesla Roadster, I'd tend to agree with you. On the other hand, we're rapidly approaching the time where 150A isn't enough for a household. Modern high-performance gaming computers can use 8 amps just for themselves, for example. Throw in air conditioners or heaters, refrigerators, appliances, garage door openers, hair dryers, and all the other myriad devices that might be running at a time, and 150A is starting to look limiting.
4x80GB drives for $40 each? That's a waste of money.
Your cost is $240 and you get 320GB of storage.
Consider that 500GB drives cost about $100. So raise your cost by 25% (3x500GB for RAID-5) and you get 467% of the storage.
The time where 80GB drives are cost effective is long past; they can't compete with larger drives for cost-per-gig.
In fact, you suggest (for some reason) spending $200 on a RAID card, which may be a bit overkill for a bunch of tiny 80GB drives. I'd suggest that you're better off shifting $60 of that to drives. The 500GB drives will be MUCH faster (areal density), and because you've got only 3 drives (so 1/3 parity), the array will be more reliable.
You might even try to save more money by ditching the RAID card altogether and using ZFS/RAIDZ on OpenSolaris or whichever BSDs supports it these days. You may lose in performance in certain performance cases, but you'll gain in reliability over hardware RAID, and save money by ditching the extra controller.
That's only until you start considering the cost of multiple discs. We're constantly told that HD-DVD discs are cheaper to press. So, what is the cost of two dual-layer HD-DVD discs (60 GiB) versus a single dual-layer Blu-Ray disc (50 GiB)?
If HD-DVD is cheaper even with two discs, the extra capacity is meaningless. Also worth considering is that most HD movies on either format are still in the ballpark of 20 GiB. While that may change how much room is available for extra content, the core feature itself (and let's not kid ourselves, that's what really counts) isn't significantly different in bitrate/size regardless of how much actual storage space is available.
Personally, I'd then turn around and guess that most HD-DVD discs are shipping as dual-layer and most Blu-Ray discs are shipping as single-layer, but it'd be just that; a guess. Either way, we have yet to see a conclusive advantage as far as capacity goes.
Another angle to consider is that one advantage the studios gave to Blu-Ray was the extra protection (BD+) on top of AACS. That has since been cracked, so the DRM on BluRay is currently no more secure than HD-DVD.
8GiB (64gbit) capacity, 1GiB/s write speed, 8 seconds to write every bit on the chip.
31,536,000 / 8 = 3,942,000
So, you would completely rewrite the chip almost 4 million times per year. Scale it back to 100MB/s writing constantly and you'd generate almost 400 thousand writes per year.
But they're not marketing it as a first-time expo. They're marketing it as the successor to E3, they've chosen a similar name "E4 All", and have people like Tommy Tallarico pimping it as "E4".
This year, those 18000 attendees were probably mostly drawn in by the attempt to tie it to E4. Now, the cat is out of the bag. The media has more or less panned the event, and playing the E3 card isn't going to work again.
Next year, they'll be going up against PAX, which was likely double E4A's size this year. PAX tends to double their attendance every year, so it's no stretch to think that next year's PAX could easily reach 60k attendees.
Can E4A really compete with that? I wouldn't be surprised if E4A actually shrunk next year, or grew only slightly.
On the other hand, it's not too late for IDG to save their event. All they need to do is reschedule at least a month (or three) in either direction, and drop the major arrogance. For example, performing guerrilla marketing outside PAX 2007 wasn't a very smart move.
TekSavvy is currently rolling out service to all of Telus' operating area, which I'm pretty sure includes Vancouver. The CRTC's regulations don't apply just to Bell, they apply to all the incumbents, which includes Telus.
Last I heard, they were hoping to have service go into testing in late October, but it's still a bit up in the air. You should check out TekSavvy's official forum on DSLReports for the latest.
Except if you run netstat, hey look, that IP is now stored in memory until you clear the console buffer. Wups, can't keep the IP around, better not run netstat anymore.
1) The factual accuracy here is questionable. This is 5 years old. Right after, somebody claimed that HL2 was "100%" coming to Mac because he "knew a guy" who worked at Valve as a mapper.
2) According to the forum posts, Sierra cancelled the project, not Valve
3) It would have run in software mode, most likely, since I don't recall Macs having dedicated 3D hardware back then. Half-Life limited to software mode isn't worth having at all. It made sense when HL1 was initially released to include a software renderer, but by the time the mac port would have come out (1999), software rendered FPS games were pretty much dead.
Mac gamers were given a different choice two years later, anyhow, when Half-Life was (officially) released on the PS2 and (unofficially) on the Dreamcast in 2001. The cost of getting that purchased and running is less than a decent graphics card, so other than the two year delay, would have been a better option for Mac gamers.
A common error people make is to compare the sales of console games for EVERY console platform to PC game sales.
When you compare PC game sales to individual consoles, the PC sells more games than the 360, PS3, or Wii. Halo 3 might be an exception, but that's a blip, a temporary boost that doesn't happen every week.
So of the four platforms, the PC is on top. How, then, is PC gaming dying if it's the leading platform?
I disagree. For an 11" widescreen display, if you compare that to a 4:3 display, you're looking at roughly 1024x768 at a similar DPI. That's perfectly reasonable for a 12" notebook, considering how tiny that screen would be.
With desktop LCDs still shipping 19" at 1280x1024, 12" at 1024x768 isn't half bad.
"The students in the dorms playing the new game were going through the dorm's T1"
I think that's more the problem. A single T1 isn't even enough for one residential user anymore, let alone dozens or hundreds. To have a dorm on a T1 in this day and age is just incompetence. Most DSL connections (5-7mbit where I'm from) are far faster than a T1 for downstream. And yes, I realize the difference between a DSL line and a T1, but when you're using that T1 for residential use, all the non-oversubscribed and SLA business-oriented stuff doesn't really matter.
Internet games don't use multicast packets. Playing Halo 3 over the internet therefore would only produce one stream of low-bandwidth packets to the XBL server. Even LAN games only do multicast for locating hosts. Once you're in-game, they tend to only send packets to the host.
You speak of having 1000 hosts sending multiple packets per second. Having 1000 hosts on a T1 line in a dorm is almost criminal. While a game might use, let's say, 20KB/s of bandwidth in any given direction (very little, many multiplayer games use significantly less), if you're going to cram hundreds or thousands of users onto a connection that can only do about 200KB/s, then you're going to have problems, and it's not the game's fault.
The dorms should either provide their tenants with a reasonably fast connection, or allow them to purchase their own DSL/cable connections.
It is, of course, possible to compensate by simply adjusting the colour balance to compensate. The colour balance would become corrected at the expense of overall display brightness.
I think, though, that this is an important first step; for years we've been waiting for large commercial OLEDs to become available. Now, they finally are, with this small first step. Now we'll start to see larger and cheaper screens slowly develop, until we can finally get "big screen TVs" that use OLEDs.
In addition, at 11 inches, this is pretty close to laptop territory. I expect to see 12 inch OLED laptops as soon as the price comes down a bit (say, to $500 for just the panel).
Except game servers are by and large not disk IO limited. They're usually either CPU limited or transit limited. You might be memory limited, but that's far cheaper to fix than the other problems, so it isn't really a concern.
Game servers only really touch the disk during map loads.
If you're building a really beefy box to use as a game server, say a quad core quad processor box with 16GB of RAM, and you're running, say, 32 game servers on there, then yes, the chance that two servers will try to load a map at the same time increases, but that's probably better solved with a cheap RAID array. For perspective, said server could easily saturate a 100mbit connection, so transit required could be somewhere in the area of 48TB/mth. I know that Cogent charges my ISP about 4 cents per gig for transit, so we're talking about $1500 for transit and $1000 for the physical connection. So internet connectivity would cost $2400 per month, and that'd be a far more pressing concern than disk IO.
On the other hand, if you start talking about MMO-type games, that's a different story. I have no idea what the bottlenecks tend to be on that sort of server.
Don't be so sure. The lead time might have dropped emulating the N64 generation, but for the generation after that, where are we?
No fully working Dreamcast, PS2, GameCube, or XBox emulators. Some of those platforms have working emulators, but none come anywhere close to running most games. PCSX2 is probably the furthest along, and last I checked, it was only at about 1/3 compatibility. And it didn't really run on commodity hardware.
The Dreamcast came out 9 years ago, and the likes of NullDC are still not "done" yet. The XBox came out 6 years ago and there aren't even any emulators (or virtualizers) under development anymore (cxbx sort of runs Turok, and nothing else).
I'd expect THAT generation to be fully working by 2012, not the PS3/360 generation. The Wii might be an exception, since the hardware is so similar to the GameCube, and the controllers work with regular PCs.
Re:And yet, it is being held back for North Americ
on
Orange Box Turns Gold
·
· Score: 1
I do, but apparently you don't. There's a difference between being able to activate a game and being able to purchase/download a game. While the retail discs are little more than copies of the GCF cache files, it's not exactly difficult for them to only activate retail copies (which I assume have some sort of unique identifier).
That said, yes, my information (taken from Wikipedia) was incorrect. Wikipedia has also updated their own information, October 10th for Steam/North America and October 12th for the rest of the world. That's what I get for trusting the factual accuracy of Wikipedia.
Slashdot Mirror
They've examined ONE SINGLE game and used this to (try to) invalidate the testing method for EVERY game. Sorry, doesn't work like that.
All they've proven is that there is something wrong with the timedemo system in Crysis.
They're three distinct things, and Vista dropped OCA.
WER handles the reporting of the errors (formerly called Dr. Watson)
OCA handles the analysis of the reports, and informs the user of the results (Vista integrated this into the WER interface)
CEIP reports usability data from certain applications, such as Windows Live Messenger, and doesn't collect program crash info.
You talk about "pro-MS" propaganda, but you're the one desperately searching for things to shit on.
How exactly would they get notified of their error? The idea behind a secret ballot is that nobody but the voter sees the ballot at a point where the person voting can be identified.
In Canadian federal elections, NOBODY sees your ballot after it is first handed to you, until they break the seal on the ballot boxes for tabulation. And at that point, there's no way of tying a particular ballot to a particular person. As soon as the voter fills in their ballot, the voter folds it in a specific way before leaving the booth, and it remains so folded as they walk over to the ballot box and insert it into the slot.
Any system that allows for notifying a voter that they made an error is a flawed and dangerous democratic process. I'm shocked that this is an accepted practice in the US.
But then again on the plus side, it can't be accidentally used to destroy airplanes.
Well, not from the outside, no.
I wouldn't necessarily call the Mac Mini underpowered. The CPU is, after all, dual core, and quite fast for non-intensive uses (about what you'd expect in a $600 computer).
However, with RAM prices being what they are (last I checked 2GB cost about $50), I'd expect 2GB of RAM and a significantly larger hard drive for that price. The fact that they use notebook components to save space and power is neither here nor there.
Which would be why blu-ray discs that aren't in caddies have special coatings on their bottom surfaces that make them MORE durable than both CDs and DVDs (unless they themselves are treated).
The coatings really do work quite well. They don't make the surface completely unscratchable, but they do make them far more durable than a standard CD/DVD.
While the media itself limited the recording speed in the past
The media itself limited recording speed in the past because the discs would shatter if you spun them too fast. It'd be nice if they let us know what has changed with BluRay that means that discs won't tear themselves apart at high rotational speeds. Are they not made of the same polycarbonate materials as HD-DVD/DVD/CD are?
No, 1x was meant to mean the same speed that the audio played at; one-times real-time. 2x would mean you burn/read at twice the rate of playback. 1x never meant you could burn a CD in one hour. For example, a standard CD-R is 72 minutes, and takes.... 72 minutes to burn at 1x. Most CDs are 80 minutes these days. I'm sure you can figure out how long they take to burn at 1x.
1x happens to be 150KB/s, but that wasn't the original definition.
Not exactly. TekSavvy does pay bell a wholesale rate ($20.50 per customer), but Bell doesn't give them a complete service.
That $20.50 pays for Bell's service of maintaining the last-mile connection between the customer's modem and the DSLAM in the CO.
The ISP (TekSavvy) must pay for a connection to Bell's ATM network in order to get the traffic from the CO to the ISP's network ($1300 for a GigE). From there, the ISP is responsible for internet connectivity. TekSavvy's primary transit is through Peer 1 (premium customers) and Cogent (unlimited customers) with various other things in the mix (TorIX, Teleglobe, etc). I know they recently purchased some InterNap hardware, but I'm not sure if they're using them for transit too.
As you mentioned, the ISP is responsible for providing technical support. However, many issues require TekSavvy to open trouble tickets with Bell (who provide pretty bad service to TekSavvy) in order to get problems resolved. This is because many problems with DSL involve incorrect settings made by Bell techs at the CO, a constant annoyance for TekSavvy.
Anyhow, that base fee only provides the last-mile. The ISP themselves pay to get the traffic from Bell to themselves, and from there it's entirely through the ISP's own network.
The end result is that TekSavvy can provide far better customer service and performance at lower prices than Bell. This is why they're signing up 1500-1800 customers per month. That figure, BTW, comes directly from TekSavvy themselves, who post on DSLReports a lot of information that most ISPs keep secret.
I hear there's terrorists on the, uh, Internets...
I don't agree. The Aptera weighs 850 pounds. Let's assume it carries two 200 pound passengers, for a total weight of 1250 pounds.
The Honda Fit weighs more than twice as much, and has 109HP.
The Aptera should, therefore, do quite nicely with 50HP. Equivalent to a 5-door Honda Fit.
At 50HP, an 8-hour charge (144KWh) gets you ~3.9 hours of driving at full load.
For most people, considering city driving (far from full load all the time), that is far more than enough.
However, when it comes to vehicles like the Tesla Roadster, I'd tend to agree with you. On the other hand, we're rapidly approaching the time where 150A isn't enough for a household. Modern high-performance gaming computers can use 8 amps just for themselves, for example. Throw in air conditioners or heaters, refrigerators, appliances, garage door openers, hair dryers, and all the other myriad devices that might be running at a time, and 150A is starting to look limiting.
4x80GB drives for $40 each? That's a waste of money.
Your cost is $240 and you get 320GB of storage.
Consider that 500GB drives cost about $100. So raise your cost by 25% (3x500GB for RAID-5) and you get 467% of the storage.
The time where 80GB drives are cost effective is long past; they can't compete with larger drives for cost-per-gig.
In fact, you suggest (for some reason) spending $200 on a RAID card, which may be a bit overkill for a bunch of tiny 80GB drives. I'd suggest that you're better off shifting $60 of that to drives. The 500GB drives will be MUCH faster (areal density), and because you've got only 3 drives (so 1/3 parity), the array will be more reliable.
You might even try to save more money by ditching the RAID card altogether and using ZFS/RAIDZ on OpenSolaris or whichever BSDs supports it these days. You may lose in performance in certain performance cases, but you'll gain in reliability over hardware RAID, and save money by ditching the extra controller.
That's only until you start considering the cost of multiple discs. We're constantly told that HD-DVD discs are cheaper to press. So, what is the cost of two dual-layer HD-DVD discs (60 GiB) versus a single dual-layer Blu-Ray disc (50 GiB)?
If HD-DVD is cheaper even with two discs, the extra capacity is meaningless. Also worth considering is that most HD movies on either format are still in the ballpark of 20 GiB. While that may change how much room is available for extra content, the core feature itself (and let's not kid ourselves, that's what really counts) isn't significantly different in bitrate/size regardless of how much actual storage space is available.
Personally, I'd then turn around and guess that most HD-DVD discs are shipping as dual-layer and most Blu-Ray discs are shipping as single-layer, but it'd be just that; a guess. Either way, we have yet to see a conclusive advantage as far as capacity goes.
Another angle to consider is that one advantage the studios gave to Blu-Ray was the extra protection (BD+) on top of AACS. That has since been cracked, so the DRM on BluRay is currently no more secure than HD-DVD.
I don't follow your math.
8GiB (64gbit) capacity, 1GiB/s write speed, 8 seconds to write every bit on the chip.
31,536,000 / 8 = 3,942,000
So, you would completely rewrite the chip almost 4 million times per year. Scale it back to 100MB/s writing constantly and you'd generate almost 400 thousand writes per year.
But they're not marketing it as a first-time expo. They're marketing it as the successor to E3, they've chosen a similar name "E4 All", and have people like Tommy Tallarico pimping it as "E4".
This year, those 18000 attendees were probably mostly drawn in by the attempt to tie it to E4. Now, the cat is out of the bag. The media has more or less panned the event, and playing the E3 card isn't going to work again.
Next year, they'll be going up against PAX, which was likely double E4A's size this year. PAX tends to double their attendance every year, so it's no stretch to think that next year's PAX could easily reach 60k attendees.
Can E4A really compete with that? I wouldn't be surprised if E4A actually shrunk next year, or grew only slightly.
On the other hand, it's not too late for IDG to save their event. All they need to do is reschedule at least a month (or three) in either direction, and drop the major arrogance. For example, performing guerrilla marketing outside PAX 2007 wasn't a very smart move.
TekSavvy is currently rolling out service to all of Telus' operating area, which I'm pretty sure includes Vancouver. The CRTC's regulations don't apply just to Bell, they apply to all the incumbents, which includes Telus.
Last I heard, they were hoping to have service go into testing in late October, but it's still a bit up in the air. You should check out TekSavvy's official forum on DSLReports for the latest.
Except if you run netstat, hey look, that IP is now stored in memory until you clear the console buffer. Wups, can't keep the IP around, better not run netstat anymore.
A few points:
1) The factual accuracy here is questionable. This is 5 years old. Right after, somebody claimed that HL2 was "100%" coming to Mac because he "knew a guy" who worked at Valve as a mapper.
2) According to the forum posts, Sierra cancelled the project, not Valve
3) It would have run in software mode, most likely, since I don't recall Macs having dedicated 3D hardware back then. Half-Life limited to software mode isn't worth having at all. It made sense when HL1 was initially released to include a software renderer, but by the time the mac port would have come out (1999), software rendered FPS games were pretty much dead.
Mac gamers were given a different choice two years later, anyhow, when Half-Life was (officially) released on the PS2 and (unofficially) on the Dreamcast in 2001. The cost of getting that purchased and running is less than a decent graphics card, so other than the two year delay, would have been a better option for Mac gamers.
Except it isn't.
A common error people make is to compare the sales of console games for EVERY console platform to PC game sales.
When you compare PC game sales to individual consoles, the PC sells more games than the 360, PS3, or Wii. Halo 3 might be an exception, but that's a blip, a temporary boost that doesn't happen every week.
So of the four platforms, the PC is on top. How, then, is PC gaming dying if it's the leading platform?
I disagree. For an 11" widescreen display, if you compare that to a 4:3 display, you're looking at roughly 1024x768 at a similar DPI. That's perfectly reasonable for a 12" notebook, considering how tiny that screen would be.
With desktop LCDs still shipping 19" at 1280x1024, 12" at 1024x768 isn't half bad.
"The students in the dorms playing the new game were going through the dorm's T1"
I think that's more the problem. A single T1 isn't even enough for one residential user anymore, let alone dozens or hundreds. To have a dorm on a T1 in this day and age is just incompetence. Most DSL connections (5-7mbit where I'm from) are far faster than a T1 for downstream. And yes, I realize the difference between a DSL line and a T1, but when you're using that T1 for residential use, all the non-oversubscribed and SLA business-oriented stuff doesn't really matter.
Internet games don't use multicast packets. Playing Halo 3 over the internet therefore would only produce one stream of low-bandwidth packets to the XBL server. Even LAN games only do multicast for locating hosts. Once you're in-game, they tend to only send packets to the host.
You speak of having 1000 hosts sending multiple packets per second. Having 1000 hosts on a T1 line in a dorm is almost criminal. While a game might use, let's say, 20KB/s of bandwidth in any given direction (very little, many multiplayer games use significantly less), if you're going to cram hundreds or thousands of users onto a connection that can only do about 200KB/s, then you're going to have problems, and it's not the game's fault.
The dorms should either provide their tenants with a reasonably fast connection, or allow them to purchase their own DSL/cable connections.
It is, of course, possible to compensate by simply adjusting the colour balance to compensate. The colour balance would become corrected at the expense of overall display brightness.
I think, though, that this is an important first step; for years we've been waiting for large commercial OLEDs to become available. Now, they finally are, with this small first step. Now we'll start to see larger and cheaper screens slowly develop, until we can finally get "big screen TVs" that use OLEDs.
In addition, at 11 inches, this is pretty close to laptop territory. I expect to see 12 inch OLED laptops as soon as the price comes down a bit (say, to $500 for just the panel).
Except game servers are by and large not disk IO limited. They're usually either CPU limited or transit limited. You might be memory limited, but that's far cheaper to fix than the other problems, so it isn't really a concern.
Game servers only really touch the disk during map loads.
If you're building a really beefy box to use as a game server, say a quad core quad processor box with 16GB of RAM, and you're running, say, 32 game servers on there, then yes, the chance that two servers will try to load a map at the same time increases, but that's probably better solved with a cheap RAID array. For perspective, said server could easily saturate a 100mbit connection, so transit required could be somewhere in the area of 48TB/mth. I know that Cogent charges my ISP about 4 cents per gig for transit, so we're talking about $1500 for transit and $1000 for the physical connection. So internet connectivity would cost $2400 per month, and that'd be a far more pressing concern than disk IO.
On the other hand, if you start talking about MMO-type games, that's a different story. I have no idea what the bottlenecks tend to be on that sort of server.
Don't be so sure. The lead time might have dropped emulating the N64 generation, but for the generation after that, where are we?
No fully working Dreamcast, PS2, GameCube, or XBox emulators. Some of those platforms have working emulators, but none come anywhere close to running most games. PCSX2 is probably the furthest along, and last I checked, it was only at about 1/3 compatibility. And it didn't really run on commodity hardware.
The Dreamcast came out 9 years ago, and the likes of NullDC are still not "done" yet. The XBox came out 6 years ago and there aren't even any emulators (or virtualizers) under development anymore (cxbx sort of runs Turok, and nothing else).
I'd expect THAT generation to be fully working by 2012, not the PS3/360 generation. The Wii might be an exception, since the hardware is so similar to the GameCube, and the controllers work with regular PCs.
I do, but apparently you don't. There's a difference between being able to activate a game and being able to purchase/download a game. While the retail discs are little more than copies of the GCF cache files, it's not exactly difficult for them to only activate retail copies (which I assume have some sort of unique identifier).
That said, yes, my information (taken from Wikipedia) was incorrect. Wikipedia has also updated their own information, October 10th for Steam/North America and October 12th for the rest of the world. That's what I get for trusting the factual accuracy of Wikipedia.