The "General Games Chat" forum, which includes Retro Game Reviews, has been archived and is now read-only. You cannot post here or create a new thread or review on this forum.
Taken from PCZone:
"Moore's Law dictates that the transistor density of silicon doubles roughly once every 18-24 months, and this is also a fair indicator of performance increases over the same period of time. Currently, Intel's Pantium 4 chips use 42 million transistors while Athlons use 37 million. This is in stark contrast to the days of 10-micron, when transistors numbered 2,300 and chips operated at a lowly 108KHz.
With such an astounding rate of technological advance, the industry can expect to hit a wall in the next 10-20 years, when transistor size begins to approach the atomic level. In Intel's prediction chart, the last figure (0.032 micron) is already dangerously close. This is, of course, when the transistors themselves cannot possibly be made any smaller.
Intel's roadmap shows that chip manufacturing will hit 0.13 micron this year and 0.09micron by 2003. At this point, calculations suggest 3.6GHz processors will be possible - while 2007 will allow for scorchingly fast 20GHz units."
There's a bit more, but it goes on about needing 3D layouts for chips to get them faster than 20Ghz.
The thing is, do we need them faster than 20GHz? If we look at what they're capable of now, with a 1GHz chip, imagine what will be possible with a 20GHz chip, especially as graphics acceleration will also increase, although possibly not at the same rate. But will we notice any difference between what a 15GHz chip can do, and what a 20GHz chip can do? I personally think that we won't.
Look at graphics in games like B&W, Max Payne, games which are available now. They look stunning on a 1GHz chip. Games which are out towards the end of next year, Doom 3, Thief 3, Deus Ex 2 etc, the screenshots are unbelievable, and I can't see them needing anything more than a 2GHz processor.
Could it be, that for once CPU capabilities are growing faster than the software it runs? So every time a new game comes out it won't need the newest processor? Could it be that if you upgrade your PC it will last longer than 1 year before you need to upgrade again? I hope so.
A number of people on here have said that they don't really use their PC for gaming as they don't want to have to upgrade every 18 months or so, and it's a fair point. But would you be prepared to buy a PC for games if you were 100% sure that you could still play games in 4 years time? Thats almost the lifespan of a console. If people can be sure that when they buy or upgrade a PC, they will still be able to play the games in 4 years time, then maybe more people will buy PC's for games?
The extra numbers of people buying PC games could be an incentive for some developers to be a bit more imaginative, a bit more experimental, as there is more chance that the game will sell well enough. Perhaps we would see the end of sequels to games thnat are just a re-hash of the original, and stale genres? (Well, to an extent). Perhaps this would allow the PC to go back to its rightful place as the best platform for gaming.
Of course, there is a down side. The extra power could encourage developers to be sloppy in their programming. They may think "Oh, it's ok, we don't need to tidy the code up, machines are fast enough to handle it with it's problems." Which of course would handicap those people with slower machines.
What do you think?
"Moore's Law dictates that the transistor density of silicon doubles roughly once every 18-24 months, and this is also a fair indicator of performance increases over the same period of time. Currently, Intel's Pantium 4 chips use 42 million transistors while Athlons use 37 million. This is in stark contrast to the days of 10-micron, when transistors numbered 2,300 and chips operated at a lowly 108KHz.
With such an astounding rate of technological advance, the industry can expect to hit a wall in the next 10-20 years, when transistor size begins to approach the atomic level. In Intel's prediction chart, the last figure (0.032 micron) is already dangerously close. This is, of course, when the transistors themselves cannot possibly be made any smaller.
Intel's roadmap shows that chip manufacturing will hit 0.13 micron this year and 0.09micron by 2003. At this point, calculations suggest 3.6GHz processors will be possible - while 2007 will allow for scorchingly fast 20GHz units."
There's a bit more, but it goes on about needing 3D layouts for chips to get them faster than 20Ghz.
The thing is, do we need them faster than 20GHz? If we look at what they're capable of now, with a 1GHz chip, imagine what will be possible with a 20GHz chip, especially as graphics acceleration will also increase, although possibly not at the same rate. But will we notice any difference between what a 15GHz chip can do, and what a 20GHz chip can do? I personally think that we won't.
Look at graphics in games like B&W, Max Payne, games which are available now. They look stunning on a 1GHz chip. Games which are out towards the end of next year, Doom 3, Thief 3, Deus Ex 2 etc, the screenshots are unbelievable, and I can't see them needing anything more than a 2GHz processor.
Could it be, that for once CPU capabilities are growing faster than the software it runs? So every time a new game comes out it won't need the newest processor? Could it be that if you upgrade your PC it will last longer than 1 year before you need to upgrade again? I hope so.
A number of people on here have said that they don't really use their PC for gaming as they don't want to have to upgrade every 18 months or so, and it's a fair point. But would you be prepared to buy a PC for games if you were 100% sure that you could still play games in 4 years time? Thats almost the lifespan of a console. If people can be sure that when they buy or upgrade a PC, they will still be able to play the games in 4 years time, then maybe more people will buy PC's for games?
The extra numbers of people buying PC games could be an incentive for some developers to be a bit more imaginative, a bit more experimental, as there is more chance that the game will sell well enough. Perhaps we would see the end of sequels to games thnat are just a re-hash of the original, and stale genres? (Well, to an extent). Perhaps this would allow the PC to go back to its rightful place as the best platform for gaming.
Of course, there is a down side. The extra power could encourage developers to be sloppy in their programming. They may think "Oh, it's ok, we don't need to tidy the code up, machines are fast enough to handle it with it's problems." Which of course would handicap those people with slower machines.
What do you think?
Page:
Yeah it would if you think about it!! Wonder when it will be 30ghz!!
Originaly when Gordon Moore predicted this in 1965 the number of components was to double every year, which it did until around 1975. It then slowed to around every 18 months. So I predict it will slow more maybe to 2-2.5 years before it doubles. It gets harder and harder to get them to perform higher. i had to learn all about this for my course at uni.
Moores law could be replaced by another one "Rocks law" which shows the cost of semiconductor fabrication equipment doubles every four years. This suggests that the ultimate barrier to greater chip densities might be economic, not technological. It is only due to the fact that the price of memory is falling all the time that cost of PC's change very little.
So we may see 20 GHz proccessors by 2007 or a short time after but just because they are that size dosn't mean things will run alot faster. As developers create software which require a chip like this to run.
Moores law could be replaced by another one "Rocks law" which shows the cost of semiconductor fabrication equipment doubles every four years. This suggests that the ultimate barrier to greater chip densities might be economic, not technological. It is only due to the fact that the price of memory is falling all the time that cost of PC's change very little.
So we may see 20 GHz proccessors by 2007 or a short time after but just because they are that size dosn't mean things will run alot faster. As developers create software which require a chip like this to run.
Yer! I read that post you posted in Chat anywhoo!! :-D
Yeah, it's ok. I read it too qucik the first time. After re-reading it I realised what you meant.
My fault, sorry again.
My fault, sorry again.
Your Honour wrote:
> Kapoera Gaebora wrote:
> LOL, Make some quick case copy em onto
> SR cos they are your own work
> and win a free game sounds like
> a good plan!!
"Make some quick case" I don't want to
> manufacture luggage you fool!
If you used ENGLISH and PUNCTUATION
> then I may be able to understand what you're on about.
Sorry I was touch typing...I meant Cash!!!
> Kapoera Gaebora wrote:
> LOL, Make some quick case copy em onto
> SR cos they are your own work
> and win a free game sounds like
> a good plan!!
"Make some quick case" I don't want to
> manufacture luggage you fool!
If you used ENGLISH and PUNCTUATION
> then I may be able to understand what you're on about.
Sorry I was touch typing...I meant Cash!!!
No worries, anything to relieve the boredom of work!
:-)
:-)
OK, thanks YH.
Try http://www. quietpc.com/ (SPACE!!)
They don't sell stuff, but the have reviews of products and that which you might find handy.
They don't sell stuff, but the have reviews of products and that which you might find handy.
Won't be a problem, I'll be getting a new PC soon, so I can buy the customised parts and assemble the lot by myself.
I just assumed you'd tried one as you said whisper quiet ones were still very loud.
I don't know if SR sell them.
Maybe try searching for "quiet PSU's" in google or something. Remeber, you'll probably need to get a new PSU and a new heatsink and fan for you'r CPU, although the CPU shouldn't be too loud to start with.....
I don't know if SR sell them.
Maybe try searching for "quiet PSU's" in google or something. Remeber, you'll probably need to get a new PSU and a new heatsink and fan for you'r CPU, although the CPU shouldn't be too loud to start with.....
Page: