The "General Games Chat" forum, which includes Retro Game Reviews, has been archived and is now read-only.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.
We are now in the age of 128bit gaming... But where do we go next? Obviously the next step is 256bit, that's the way gaming has advanced over the years from the 8bit days of the Atari up to the modern super 128bit consoles namely the PS2, GCN and X-Box. Doubling the processing power each time has always been the "norm", but as we progress, the visual leap is getting smaller and smaller each time.
Modern processors are reaching their peak. Silicone based chips are reaching their full capability and even with each new generation of console claiming to push millions more polygons than the last we are slowly reaching a point where things will all look the same.
There was an obvious difference between the PS1/Saturn and the DC/PS2, but this was mainly due to faster processors and larger bit rate, but what about the next jump to 256bit? Remember that us mere mortals see the world around us in 32bit and hear in 16bit... So the obvious changes, again, would be the visuals and the leap to "photo realism"...
Old console architecture used to rely on the main CPU performing most of the maths behind a game. The polygon count, textures, physics, lighting etc was all handled here. With a modern console like the PS2, other devices have been created to reduce the load bearing on the main CPU. Vector Units now deal with the mathís and co-ordained, the Graphics Synthesizer deals with the Polygon counts, textures and lighting and the Emotion Engine handles the physics and the "emotions" of the game and finally you have the sound processors.
Even with all of these "added" processes, the GS is still the main CPU for putting out the final game to our screens and it's maximum capability is all based on it's bandwidth and how fast it can shove this information from one point to another.
This is where future consoles will have to adapt.
It's already known that obtaining true "photo realism" in games is possible, the problem? The bandwidth between the Ram, processors and the final CPU.
Shoving mathís, physics and polygons down a pipe to the CPU is easy, and to increase speed, you mount everything as close as possible to each other so the information has less distance to travel, these days, the BUS is mounted on the CPU omitting the distance between the to and increasing performance but bandwidth between the two is still maxing out at around 4.3GB sec and silicon chips are getting to the size where they are too small for electrons to travel around them... So even if the next 256bit consoles where to arrive today their bandwidth wouldn't be dramatically different...
The future lies in Carbon Nanotubes or "buckytubes". Carbon Nanotubes will make CPU's smaller, faster and more powerful and when compared to silicon it's like comparing your Mini to your Ferrari...
Look inside any PC, Mac or console and you'll find a silicon processor beating away at its heart. But silicone chips, made up of millions of transistors, can only get so small before electrical resistance makes their tiny channels unable to pass sufficient current.
Carbon Nanotubes are made up of carbon atoms, no more than 10 atoms across. These tiny things, 500 times smaller than silicon transistors, are able to conduct electricity, making it possible to use them to build the logic gates that form the basis of any computer.
In the future, this means that the number of transistors currently placed on a computer chip could be packaged into a much smaller space. Consequently, computers of today's size could have a much greater processing capability, making them faster at tasks such as graphics processing, for example.
This would lead to the dawn of photo-realism being a standard as the processing power would be there, on tap. Combine carbon processors with carbon based Ram and Bus's mounted directly upon the CPU and the bandwidth would almost become infinite allowing more information to be shifted around at once...
Advancement in gaming has to go beyond the amount of polys you push about and how small your silicon is for there to be a dramatic leap. 256bit maybe on its way but unless new technology like carbon Nanotubes become a reality, the impact on gaming will be insufficient.
The future of gaming is exciting, it's always been at the cutting edge of technology but now it's time for the dawn of new technology to emerge if we want to achieve that over all goal of photo realistic gaming environment, a world where the difference between what is real and what is CG becomes indistinguishable.
Modern processors are reaching their peak. Silicone based chips are reaching their full capability and even with each new generation of console claiming to push millions more polygons than the last we are slowly reaching a point where things will all look the same.
There was an obvious difference between the PS1/Saturn and the DC/PS2, but this was mainly due to faster processors and larger bit rate, but what about the next jump to 256bit? Remember that us mere mortals see the world around us in 32bit and hear in 16bit... So the obvious changes, again, would be the visuals and the leap to "photo realism"...
Old console architecture used to rely on the main CPU performing most of the maths behind a game. The polygon count, textures, physics, lighting etc was all handled here. With a modern console like the PS2, other devices have been created to reduce the load bearing on the main CPU. Vector Units now deal with the mathís and co-ordained, the Graphics Synthesizer deals with the Polygon counts, textures and lighting and the Emotion Engine handles the physics and the "emotions" of the game and finally you have the sound processors.
Even with all of these "added" processes, the GS is still the main CPU for putting out the final game to our screens and it's maximum capability is all based on it's bandwidth and how fast it can shove this information from one point to another.
This is where future consoles will have to adapt.
It's already known that obtaining true "photo realism" in games is possible, the problem? The bandwidth between the Ram, processors and the final CPU.
Shoving mathís, physics and polygons down a pipe to the CPU is easy, and to increase speed, you mount everything as close as possible to each other so the information has less distance to travel, these days, the BUS is mounted on the CPU omitting the distance between the to and increasing performance but bandwidth between the two is still maxing out at around 4.3GB sec and silicon chips are getting to the size where they are too small for electrons to travel around them... So even if the next 256bit consoles where to arrive today their bandwidth wouldn't be dramatically different...
The future lies in Carbon Nanotubes or "buckytubes". Carbon Nanotubes will make CPU's smaller, faster and more powerful and when compared to silicon it's like comparing your Mini to your Ferrari...
Look inside any PC, Mac or console and you'll find a silicon processor beating away at its heart. But silicone chips, made up of millions of transistors, can only get so small before electrical resistance makes their tiny channels unable to pass sufficient current.
Carbon Nanotubes are made up of carbon atoms, no more than 10 atoms across. These tiny things, 500 times smaller than silicon transistors, are able to conduct electricity, making it possible to use them to build the logic gates that form the basis of any computer.
In the future, this means that the number of transistors currently placed on a computer chip could be packaged into a much smaller space. Consequently, computers of today's size could have a much greater processing capability, making them faster at tasks such as graphics processing, for example.
This would lead to the dawn of photo-realism being a standard as the processing power would be there, on tap. Combine carbon processors with carbon based Ram and Bus's mounted directly upon the CPU and the bandwidth would almost become infinite allowing more information to be shifted around at once...
Advancement in gaming has to go beyond the amount of polys you push about and how small your silicon is for there to be a dramatic leap. 256bit maybe on its way but unless new technology like carbon Nanotubes become a reality, the impact on gaming will be insufficient.
The future of gaming is exciting, it's always been at the cutting edge of technology but now it's time for the dawn of new technology to emerge if we want to achieve that over all goal of photo realistic gaming environment, a world where the difference between what is real and what is CG becomes indistinguishable.
SonicRav wrote:
> Well, the mistake you make is to say that the processor speed is 128 bits. In fact the console's processor times are far from this! The PS2 has 300MHz, the GC has 400MHz and the Xbox has 700MHz (I hope I remember them all correctly).
Yeap... though if you look at the results, the graphical difference isn't too far apart. My point was regading the bandwidth and the in ability to produce Photo-realism in games, though possible, because they are unable to get the information from Bus to CPU any faster and that to obtain this sense of realism, new technology would need to be investigated like Carbon nanotubes to increase the speed that is paluging silcone. (I think that made sense?)
> After all, the Atari Jaguar was, in theory, a 64 bit console at a time that Ninty were just beginning to sell 16 bit consoles... but the games weern't as good, so the console plumetted!
The architecture in the Jaguar was near on identical to that of the N64 and capable of perfoming the same. It was down to a lack of experience and not knowing what to do with it that killed it off. Things in games we take for granted today hadn't appeared in the time of the Jaguar such as Bi-Linear filtering, Bump-mapping etc so effectively they were trying to make games using technology that, at the time, didn't exist...
I still have a Jaguar... don't play it often but it did have some craking games on it!!
Sonic
> Well, the mistake you make is to say that the processor speed is 128 bits. In fact the console's processor times are far from this! The PS2 has 300MHz, the GC has 400MHz and the Xbox has 700MHz (I hope I remember them all correctly).
Yeap... though if you look at the results, the graphical difference isn't too far apart. My point was regading the bandwidth and the in ability to produce Photo-realism in games, though possible, because they are unable to get the information from Bus to CPU any faster and that to obtain this sense of realism, new technology would need to be investigated like Carbon nanotubes to increase the speed that is paluging silcone. (I think that made sense?)
> After all, the Atari Jaguar was, in theory, a 64 bit console at a time that Ninty were just beginning to sell 16 bit consoles... but the games weern't as good, so the console plumetted!
The architecture in the Jaguar was near on identical to that of the N64 and capable of perfoming the same. It was down to a lack of experience and not knowing what to do with it that killed it off. Things in games we take for granted today hadn't appeared in the time of the Jaguar such as Bi-Linear filtering, Bump-mapping etc so effectively they were trying to make games using technology that, at the time, didn't exist...
I still have a Jaguar... don't play it often but it did have some craking games on it!!
Sonic
Thats a very good point that Sonics just made. The consoles we are all waiting for are in fact inferior to our PCs (well some of ours) in terms of speed.
300mhz for the Ps2? 2hgz for the latest Pentium 4 based PC.
However in terms of graphics PC still haven't reached the same standards in my own opinion...now I haven't seen what the Geforce 3 is capable of....or in fact what the Geforce 4 will be capable of, but at the moment PC games can't compete graphics wise with our modern consoles....however fast they may be.
300mhz for the Ps2? 2hgz for the latest Pentium 4 based PC.
However in terms of graphics PC still haven't reached the same standards in my own opinion...now I haven't seen what the Geforce 3 is capable of....or in fact what the Geforce 4 will be capable of, but at the moment PC games can't compete graphics wise with our modern consoles....however fast they may be.
Strafex wrote:
Console designer 2:
> Lets put more pins in our processing chip! Instead of 128, lets put in 256!
> Nothing will ever be better than that!
Believe it or not, this is what Sega did with the Saturn... here's the story:
It's the mid 90's at Sega HQ, and all is looking good. With their console, the Saturn, finally being completely designed, Sega had nothing to do but give dev kits to developers and go on their merry way...
And then disaster struck- Sony released the PSX specs. Not only were they far supperior to the Saturn, but Sony had expected something that Sega didn't- games in the next generation would become 3d!
Sega, struck down by this news, went back to the drawing board... and managed to come up with a cunning idea. Stick another processor in the console. Rush the American launch, therefore getting into the market before Sony. Launch at £400.
So, not only was the console impossible to program for (due to its odd processor layout), but it was coming out 6 months before developers expected, and at a ridiculous price....
Good job SEGA! To be that good took AGES!
Sonic
Console designer 2:
> Lets put more pins in our processing chip! Instead of 128, lets put in 256!
> Nothing will ever be better than that!
Believe it or not, this is what Sega did with the Saturn... here's the story:
It's the mid 90's at Sega HQ, and all is looking good. With their console, the Saturn, finally being completely designed, Sega had nothing to do but give dev kits to developers and go on their merry way...
And then disaster struck- Sony released the PSX specs. Not only were they far supperior to the Saturn, but Sony had expected something that Sega didn't- games in the next generation would become 3d!
Sega, struck down by this news, went back to the drawing board... and managed to come up with a cunning idea. Stick another processor in the console. Rush the American launch, therefore getting into the market before Sony. Launch at £400.
So, not only was the console impossible to program for (due to its odd processor layout), but it was coming out 6 months before developers expected, and at a ridiculous price....
Good job SEGA! To be that good took AGES!
Sonic
Well, the mistake you make is to say that the processor speed is 128 bits. In fact the console's processor times are far from this! The PS2 has 300MHz, the GC has 400MHz and the Xbox has 700MHz (I hope I remember them all correctly).
The "bits" rating of a console is now totally irrelavant- the architecture (the way the console processes games) and the clock speed are now what is used to judge a console's "power".
Now, before I go on, you may ask why anyone would buy a 300MHz PS2 when a PC can go at over 1GHz- over 10 times faster! The answer to that is that console games can be designed to utilise every last inch of power from the machine... making them supperbly fast. PC games have to be made so that they can run on any machine in a general spec range, with any number of graphics cards being supported. For that reason, consoles don't need such a high clock speed to reach the same graphical standard of a PC.
All that said though, all that matters is the quality of the games that appear on a console, and not the specs of the console itself. After all, the Atari Jaguar was, in theory, a 64 bit console at a time that Ninty were just beginning to sell 16 bit consoles... but the games weern't as good, so the console plumetted!
Sonic
The "bits" rating of a console is now totally irrelavant- the architecture (the way the console processes games) and the clock speed are now what is used to judge a console's "power".
Now, before I go on, you may ask why anyone would buy a 300MHz PS2 when a PC can go at over 1GHz- over 10 times faster! The answer to that is that console games can be designed to utilise every last inch of power from the machine... making them supperbly fast. PC games have to be made so that they can run on any machine in a general spec range, with any number of graphics cards being supported. For that reason, consoles don't need such a high clock speed to reach the same graphical standard of a PC.
All that said though, all that matters is the quality of the games that appear on a console, and not the specs of the console itself. After all, the Atari Jaguar was, in theory, a 64 bit console at a time that Ninty were just beginning to sell 16 bit consoles... but the games weern't as good, so the console plumetted!
Sonic
Console designer 1: So how do we make consoles better?
Console designer 2: Lets put more pins in our processing chip! Instead of 128, lets put in 256! Nothing will ever be better than that!
Seriously though, this new material will be good for speeding up consoles in general and boosting preformance as well as reducing over heating, but improvements are alround.
This new material will boost all parts of new systems from the RAM to the graphics processor.
256 processor is all very fine and dandy, but what's a brain with out it's organs and muscles?
Console designer 2: Lets put more pins in our processing chip! Instead of 128, lets put in 256! Nothing will ever be better than that!
Seriously though, this new material will be good for speeding up consoles in general and boosting preformance as well as reducing over heating, but improvements are alround.
This new material will boost all parts of new systems from the RAM to the graphics processor.
256 processor is all very fine and dandy, but what's a brain with out it's organs and muscles?
We are now in the age of 128bit gaming... But where do we go next? Obviously the next step is 256bit, that's the way gaming has advanced over the years from the 8bit days of the Atari up to the modern super 128bit consoles namely the PS2, GCN and X-Box. Doubling the processing power each time has always been the "norm", but as we progress, the visual leap is getting smaller and smaller each time.
Modern processors are reaching their peak. Silicone based chips are reaching their full capability and even with each new generation of console claiming to push millions more polygons than the last we are slowly reaching a point where things will all look the same.
There was an obvious difference between the PS1/Saturn and the DC/PS2, but this was mainly due to faster processors and larger bit rate, but what about the next jump to 256bit? Remember that us mere mortals see the world around us in 32bit and hear in 16bit... So the obvious changes, again, would be the visuals and the leap to "photo realism"...
Old console architecture used to rely on the main CPU performing most of the maths behind a game. The polygon count, textures, physics, lighting etc was all handled here. With a modern console like the PS2, other devices have been created to reduce the load bearing on the main CPU. Vector Units now deal with the mathís and co-ordained, the Graphics Synthesizer deals with the Polygon counts, textures and lighting and the Emotion Engine handles the physics and the "emotions" of the game and finally you have the sound processors.
Even with all of these "added" processes, the GS is still the main CPU for putting out the final game to our screens and it's maximum capability is all based on it's bandwidth and how fast it can shove this information from one point to another.
This is where future consoles will have to adapt.
It's already known that obtaining true "photo realism" in games is possible, the problem? The bandwidth between the Ram, processors and the final CPU.
Shoving mathís, physics and polygons down a pipe to the CPU is easy, and to increase speed, you mount everything as close as possible to each other so the information has less distance to travel, these days, the BUS is mounted on the CPU omitting the distance between the to and increasing performance but bandwidth between the two is still maxing out at around 4.3GB sec and silicon chips are getting to the size where they are too small for electrons to travel around them... So even if the next 256bit consoles where to arrive today their bandwidth wouldn't be dramatically different...
The future lies in Carbon Nanotubes or "buckytubes". Carbon Nanotubes will make CPU's smaller, faster and more powerful and when compared to silicon it's like comparing your Mini to your Ferrari...
Look inside any PC, Mac or console and you'll find a silicon processor beating away at its heart. But silicone chips, made up of millions of transistors, can only get so small before electrical resistance makes their tiny channels unable to pass sufficient current.
Carbon Nanotubes are made up of carbon atoms, no more than 10 atoms across. These tiny things, 500 times smaller than silicon transistors, are able to conduct electricity, making it possible to use them to build the logic gates that form the basis of any computer.
In the future, this means that the number of transistors currently placed on a computer chip could be packaged into a much smaller space. Consequently, computers of today's size could have a much greater processing capability, making them faster at tasks such as graphics processing, for example.
This would lead to the dawn of photo-realism being a standard as the processing power would be there, on tap. Combine carbon processors with carbon based Ram and Bus's mounted directly upon the CPU and the bandwidth would almost become infinite allowing more information to be shifted around at once...
Advancement in gaming has to go beyond the amount of polys you push about and how small your silicon is for there to be a dramatic leap. 256bit maybe on its way but unless new technology like carbon Nanotubes become a reality, the impact on gaming will be insufficient.
The future of gaming is exciting, it's always been at the cutting edge of technology but now it's time for the dawn of new technology to emerge if we want to achieve that over all goal of photo realistic gaming environment, a world where the difference between what is real and what is CG becomes indistinguishable.
Modern processors are reaching their peak. Silicone based chips are reaching their full capability and even with each new generation of console claiming to push millions more polygons than the last we are slowly reaching a point where things will all look the same.
There was an obvious difference between the PS1/Saturn and the DC/PS2, but this was mainly due to faster processors and larger bit rate, but what about the next jump to 256bit? Remember that us mere mortals see the world around us in 32bit and hear in 16bit... So the obvious changes, again, would be the visuals and the leap to "photo realism"...
Old console architecture used to rely on the main CPU performing most of the maths behind a game. The polygon count, textures, physics, lighting etc was all handled here. With a modern console like the PS2, other devices have been created to reduce the load bearing on the main CPU. Vector Units now deal with the mathís and co-ordained, the Graphics Synthesizer deals with the Polygon counts, textures and lighting and the Emotion Engine handles the physics and the "emotions" of the game and finally you have the sound processors.
Even with all of these "added" processes, the GS is still the main CPU for putting out the final game to our screens and it's maximum capability is all based on it's bandwidth and how fast it can shove this information from one point to another.
This is where future consoles will have to adapt.
It's already known that obtaining true "photo realism" in games is possible, the problem? The bandwidth between the Ram, processors and the final CPU.
Shoving mathís, physics and polygons down a pipe to the CPU is easy, and to increase speed, you mount everything as close as possible to each other so the information has less distance to travel, these days, the BUS is mounted on the CPU omitting the distance between the to and increasing performance but bandwidth between the two is still maxing out at around 4.3GB sec and silicon chips are getting to the size where they are too small for electrons to travel around them... So even if the next 256bit consoles where to arrive today their bandwidth wouldn't be dramatically different...
The future lies in Carbon Nanotubes or "buckytubes". Carbon Nanotubes will make CPU's smaller, faster and more powerful and when compared to silicon it's like comparing your Mini to your Ferrari...
Look inside any PC, Mac or console and you'll find a silicon processor beating away at its heart. But silicone chips, made up of millions of transistors, can only get so small before electrical resistance makes their tiny channels unable to pass sufficient current.
Carbon Nanotubes are made up of carbon atoms, no more than 10 atoms across. These tiny things, 500 times smaller than silicon transistors, are able to conduct electricity, making it possible to use them to build the logic gates that form the basis of any computer.
In the future, this means that the number of transistors currently placed on a computer chip could be packaged into a much smaller space. Consequently, computers of today's size could have a much greater processing capability, making them faster at tasks such as graphics processing, for example.
This would lead to the dawn of photo-realism being a standard as the processing power would be there, on tap. Combine carbon processors with carbon based Ram and Bus's mounted directly upon the CPU and the bandwidth would almost become infinite allowing more information to be shifted around at once...
Advancement in gaming has to go beyond the amount of polys you push about and how small your silicon is for there to be a dramatic leap. 256bit maybe on its way but unless new technology like carbon Nanotubes become a reality, the impact on gaming will be insufficient.
The future of gaming is exciting, it's always been at the cutting edge of technology but now it's time for the dawn of new technology to emerge if we want to achieve that over all goal of photo realistic gaming environment, a world where the difference between what is real and what is CG becomes indistinguishable.