You have to think about what those processes are doing. Is google's search time sensitive to the ms range? Does the NYSE have to be more accurate than the nearest second? You do understand that google doesn't actually search the live data of the web when you make a search at google.com, right? You response indicates a grave misunderstanding of the difference between the way a company like google serves a precomputed search response vs WOW computing and serving a complex simulation in ms time sensitive real time.
This is much like how things work on slash: you have a massive number of people involved, and if even 1 in 10 people complains about something, it will seem like a huge number of complaints. There's no reason to believe that the contrary complaints are coming from the same people.
Sorry, that was a habit typo. Hopefully it is still clear that the numbers we're comparing are 180 (wireless Mbps) vs standard ethernet at 100 (wire Mbps) but that because wireless won't deliver the full 180, which is actually faster is going to be unclear, and will depend upon the installation.
No, I meant 1080/60p, and yes, that is heavy duty, but it's clearly the direction everyone is headed. TV's are trickling out already that will handle that, and I have little doubt that in 5 years it'll be every tv. The video sources are lagging behind, but I doubt it will be long before the HDDVD people have upconverting players. It certainly looks like it is likely to be the display format of choice, with most sources being upconverted for some time to come.
If you are utterly confident your provider will give you the update. Many of the low-end providers (dlink,linksys,airgo,etc) provide updates rarely/slowly, or that are very buggy. Be sure to put down good money for cisco or the like if you want to go down this path. Otherwise, wait for a stable product that won't require updates.
Unfortunately, as you only have 0x9 reasons so far, and I think many can be undermined, I don't know if you'll reach 0xA legitimately.
1) 180MBps is indeed the limit, but achieving 100MBps to beat standard home networking won't be that hard. And frankly, I think people will take 80 MBps if they can lose the wires. 2) I can achieve real 20MBps on my g network. They only need to quadruple that to come pretty close to what 100MBps networks can do. 3) True. But the difference is so small compared to the internet at large, will anyone care? And if they do, won't they just plug their gaming PC into one of the network ports that comes with every wireless router/switch? 4) You must have bought the wrong wireless setup. Mine took me all of 5 minutes to get up and running. 10 more minutes later on to filter out the ports that were frequently attacked so that my network wouldn't carry that data. 5) True, wireless can't ever be fully secure. But unless you live in a high density apartment complex, this isn't likely to be a problem. 6) Well... my g network hasn't had to change in 1.5 years now. I wouldn't be an n network till the spec is out of draft for 6 months, then I think you'll be buying a pretty stable platform. 7) Not much more expensive than wired, and pretty much built into every computer you can buy now. My $50 wireless router does a great job with my g network, and was the only piece of equipment I had to buy, everything else came with a wireless network card in it. Even if all you have is $300 boxes, I wouldn't consider that egregiously expensive. 8) That's true. 9) BellSouth will. They explicitly advertise compatibility with wireless networks for their DSL service, and will help you set up common routers/switches. I imagine they're not the only one.
It'll be close. You might get enough bandwidth out of this for hidef video... 1080p/60 prefers 135MBps to look good. Given a max of 180MBps, the likelihood that you'll deliver that kind of bit rate over any distance is not good. Lower res formats will probably be fine, but the so called 'true hi def' won't. I guess we'll all still wire up gigabit networking or wait for the next generation wireless networks for our ultra cool hi def wireless entertainment.
I think the ESRB's (or really the people pressuring the ESRB's) argument is that if the game comes with the data, it should be rated for that data, even if the default codepath will never show that data. Personally, I think developers should just go along with this, and maybe start offering rated and unrated versions of their games like the movies do. The smart thing to do in general, though, is to be very very strict and compliant when you've made the decision to be compliant. Otherwise you're seen as trying to undermine a system that to many people has legitimate value.
Yeah, I realized after submit you meant more on the hardware side. But I think the answer is that the hardware side has become even more regularized than the software side: the software APIs are playing catchup with hardware that has become very regular (already, hardware is mostly handling the oddball old codepaths using work done in the driver to convert to the more standard/modern codepaths). The next generation will continue on that path, leaving, I think, very little room for improvement by the OpenHardware people. They won't even have to support the old features in their drivers thanks to the dx9/10 split.
You missed my point entirely. The parent poster suggested that game devs won't target dx10 because the market will be small due to the tie to vista. My claim is that much like the dx1 tie to win95 (which was a huge success with game developers), the feature advantages of dx10 will draw in plenty of developers regardless of the tie to vista. I'll guarantee the first B titles for dx10 within 6 months of release, and A titles to hit the first christmas more than 6 months away.
It's hard to imagine that a rearchitecting will gain you very much, particularly vs dx10. DX9 was already a big rearchitecting, it fixed most of the api inconsistencies, and unified all functionality under the 3d code path (up to dx8 functionality was split between direct2d and direct3d). With DX10 they're leaving the old 2d features completely behind (no backwards compatibility required: 2d features supported only under separate directx 9 drivers), and further enhancing the consistency of the API (with even less support than previous versions for obscure hardware features).
With a from scratch redesign you might be able to manage slight improvements, but nothing that would be really amazing, nor any huge boon to the software devs working with the APIs.
Yep, it'll no doubt be a dismal failure the likes of which have not been seen since the time that technology was developed that required everyone to have win95 and excluded all the dos & win3.x users... what was it called... oh yeah, DirectX 1.0.
I owned a tv, I just didn't have cable service in an area with no OTA service. Humorously enough, I first saw that article during the time in question. In my defense, I don't talk about it much, and I did it out of poorness rather than some sense of moral superiority.:-)
I recently went 3 years with no tv / movies. It was fine. Now I watch some tv, but mostly just in the background while I read or play games. It's pretty easy to live without TV. The wierdest part was not knowing any of the common cultural references. People would talk about shows or commercials and I'd not seen any of them.
Well, that's definitely the core problem of cross platform vs mono platform. If you write an MFC app it looks nice on windows, and runs nowhere else. If you write a java app it looks crummy on windows, and exactly the same level of crummy on solaris, linux, and mac.
Doesn't take much brilliance to put that right. Maybe some enterprising developer will read this discussion and get it right in some future game that will look a lot better than this one.
The question is really whether or not there is anyone dumb enough to think so, not whether there is anyone else smart enough not to think so. And IMO there is an overabundance of morons out there, so it seems like a plausible explanation for why you'd deliberately make your game ugly like that.
Slashdot Mirror
You have to think about what those processes are doing. Is google's search time sensitive to the ms range? Does the NYSE have to be more accurate than the nearest second? You do understand that google doesn't actually search the live data of the web when you make a search at google.com, right? You response indicates a grave misunderstanding of the difference between the way a company like google serves a precomputed search response vs WOW computing and serving a complex simulation in ms time sensitive real time.
This is much like how things work on slash: you have a massive number of people involved, and if even 1 in 10 people complains about something, it will seem like a huge number of complaints. There's no reason to believe that the contrary complaints are coming from the same people.
I posted a similar comment to a previous article, no one listens.
And actually, not even google has experience running an application as scaled as WOW, their frontend is so trivial there really is no comparison.
mbits/sec, sorry, habitual typo.
Sorry, that was a habit typo. Hopefully it is still clear that the numbers we're comparing are 180 (wireless Mbps) vs standard ethernet at 100 (wire Mbps) but that because wireless won't deliver the full 180, which is actually faster is going to be unclear, and will depend upon the installation.
No, I meant 1080/60p, and yes, that is heavy duty, but it's clearly the direction everyone is headed. TV's are trickling out already that will handle that, and I have little doubt that in 5 years it'll be every tv. The video sources are lagging behind, but I doubt it will be long before the HDDVD people have upconverting players. It certainly looks like it is likely to be the display format of choice, with most sources being upconverted for some time to come.
If you are utterly confident your provider will give you the update. Many of the low-end providers (dlink,linksys,airgo,etc) provide updates rarely/slowly, or that are very buggy. Be sure to put down good money for cisco or the like if you want to go down this path. Otherwise, wait for a stable product that won't require updates.
Unfortunately, as you only have 0x9 reasons so far, and I think many can be undermined, I don't know if you'll reach 0xA legitimately.
... my g network hasn't had to change in 1.5 years now. I wouldn't be an n network till the spec is out of draft for 6 months, then I think you'll be buying a pretty stable platform.
1) 180MBps is indeed the limit, but achieving 100MBps to beat standard home networking won't be that hard. And frankly, I think people will take 80 MBps if they can lose the wires.
2) I can achieve real 20MBps on my g network. They only need to quadruple that to come pretty close to what 100MBps networks can do.
3) True. But the difference is so small compared to the internet at large, will anyone care? And if they do, won't they just plug their gaming PC into one of the network ports that comes with every wireless router/switch?
4) You must have bought the wrong wireless setup. Mine took me all of 5 minutes to get up and running. 10 more minutes later on to filter out the ports that were frequently attacked so that my network wouldn't carry that data.
5) True, wireless can't ever be fully secure. But unless you live in a high density apartment complex, this isn't likely to be a problem.
6) Well
7) Not much more expensive than wired, and pretty much built into every computer you can buy now. My $50 wireless router does a great job with my g network, and was the only piece of equipment I had to buy, everything else came with a wireless network card in it. Even if all you have is $300 boxes, I wouldn't consider that egregiously expensive.
8) That's true.
9) BellSouth will. They explicitly advertise compatibility with wireless networks for their DSL service, and will help you set up common routers/switches. I imagine they're not the only one.
It'll be close. You might get enough bandwidth out of this for hidef video ... 1080p/60 prefers 135MBps to look good. Given a max of 180MBps, the likelihood that you'll deliver that kind of bit rate over any distance is not good. Lower res formats will probably be fine, but the so called 'true hi def' won't. I guess we'll all still wire up gigabit networking or wait for the next generation wireless networks for our ultra cool hi def wireless entertainment.
I think the ESRB's (or really the people pressuring the ESRB's) argument is that if the game comes with the data, it should be rated for that data, even if the default codepath will never show that data. Personally, I think developers should just go along with this, and maybe start offering rated and unrated versions of their games like the movies do. The smart thing to do in general, though, is to be very very strict and compliant when you've made the decision to be compliant. Otherwise you're seen as trying to undermine a system that to many people has legitimate value.
Yeah, I realized after submit you meant more on the hardware side. But I think the answer is that the hardware side has become even more regularized than the software side: the software APIs are playing catchup with hardware that has become very regular (already, hardware is mostly handling the oddball old codepaths using work done in the driver to convert to the more standard/modern codepaths). The next generation will continue on that path, leaving, I think, very little room for improvement by the OpenHardware people. They won't even have to support the old features in their drivers thanks to the dx9/10 split.
True. I think it made the point though. :-)
You missed my point entirely. The parent poster suggested that game devs won't target dx10 because the market will be small due to the tie to vista. My claim is that much like the dx1 tie to win95 (which was a huge success with game developers), the feature advantages of dx10 will draw in plenty of developers regardless of the tie to vista. I'll guarantee the first B titles for dx10 within 6 months of release, and A titles to hit the first christmas more than 6 months away.
It's hard to imagine that a rearchitecting will gain you very much, particularly vs dx10. DX9 was already a big rearchitecting, it fixed most of the api inconsistencies, and unified all functionality under the 3d code path (up to dx8 functionality was split between direct2d and direct3d). With DX10 they're leaving the old 2d features completely behind (no backwards compatibility required: 2d features supported only under separate directx 9 drivers), and further enhancing the consistency of the API (with even less support than previous versions for obscure hardware features).
With a from scratch redesign you might be able to manage slight improvements, but nothing that would be really amazing, nor any huge boon to the software devs working with the APIs.
Yep, it'll no doubt be a dismal failure the likes of which have not been seen since the time that technology was developed that required everyone to have win95 and excluded all the dos & win3.x users ... what was it called ... oh yeah, DirectX 1.0.
I owned a tv, I just didn't have cable service in an area with no OTA service. :-)
Humorously enough, I first saw that article during the time in question. In my defense, I don't talk about it much, and I did it out of poorness rather than some sense of moral superiority.
I recently went 3 years with no tv / movies. It was fine. Now I watch some tv, but mostly just in the background while I read or play games. It's pretty easy to live without TV. The wierdest part was not knowing any of the common cultural references. People would talk about shows or commercials and I'd not seen any of them.
How I desperately wish the FFs had gone further with that design. Should have required 3/4 agreement to pass anything, and 7/8ths to override veto.
Source code is provided, but will anyone bother to check that the compiled result matches the binary provided?
Well, that's definitely the core problem of cross platform vs mono platform. If you write an MFC app it looks nice on windows, and runs nowhere else. If you write a java app it looks crummy on windows, and exactly the same level of crummy on solaris, linux, and mac.
I have to say, I always felt bad for that Hoare girl who lived near me. She never got a break.
Doesn't take much brilliance to put that right. Maybe some enterprising developer will read this discussion and get it right in some future game that will look a lot better than this one.
Yes, this is benchmarking based on real world usage, but discounting jvm start time, which actually gives java an unfair advantage.
The problem is that I'm better in java than c++.
The question is really whether or not there is anyone dumb enough to think so, not whether there is anyone else smart enough not to think so. And IMO there is an overabundance of morons out there, so it seems like a plausible explanation for why you'd deliberately make your game ugly like that.