I rearly don't think microsoft are going to be able to make a profit from the xbox 360, as with all consoles they will make a loss selling the unit. But the price of the GDDR alone is about £75... so there going to be making a massive loss if they sell the console for the same price as its competitors £300 ish. They may think they can make the money up selling games and through xbox live but they must be making the largest fanacial gamble of any console maker ever. I think this product wasn't intended to actuly make microsoft any money in the short term, Just to decimate the compotition so that microsoft could peddle inferior goods with a huge profit margin in the future ;) (AKA netscape)
I guess the next question is, why a PC at all anymore?
The reality is that hype is just that, and there is a chance that this thing won't work nearly like we all expect it to. If all this technology has been sitting around for the development of this, why would we not be hearing moves made towards this outside of the console market? This technology would be snapped up by PC users of many types (not just enthusiasts) with its universal architecture, assuming there was truly some benefit to it...at the very least, in a lesser format. There may be redesign of architecture that would need done, but I can't believe that NEITHER Intel nor AMD would have any interest in getting one over on each other and radically redefining things like this if it were truly the best way to be.
One will certainly be on my christmas list, but I am still going to be the skeptic and wait for the other shoe to drop...something is not right here...an advance this large has no reason to first debut on a console system.
I don't know big advances have come from unexpedcted places before and the gaming industry has become damn near more lucrative than computers, although yea I know the xbox is just a computer. but i'm sure there are going to be issues with the thing there always are and with this much new technology that's bound to add to it. i can't wait till they release it. Xbox 1 prices are goin to drop quick and i will definately snatch a few up. they can be niftly little linux boxes. I just wished they had told us if the 360 was goin to be backwards compatible with xbox 1 games. anybody out there now?
Originally Posted by Kipman725 I rearly don't think microsoft are going to be able to make a profit from the xbox 360, as with all consoles they will make a loss selling the unit. But the price of the GDDR alone is about £75... so there going to be making a massive loss if they sell the console for the same price as its competitors £300 ish. They may think they can make the money up selling games and through xbox live but they must be making the largest fanacial gamble of any console maker ever. I think this product wasn't intended to actuly make microsoft any money in the short term, Just to decimate the compotition so that microsoft could peddle inferior goods with a huge profit margin in the future ;) (AKA netscape)
"With specs this impressive, it's easy to see why a few of us wondered how they planned to make a profit, given that they were already operating at a loss as a result of hardware costs for the current platform. "This time, we own the silicon [chip designs]. We can have [the chips] manufactured at different foundries to keep costs down."
With this kind of hardware, we may well see the end of computers as a gaming platform all together; if a $300 console can fulfill all multimedia capabilities and do everything we like to do with computers, what use will people have for them other than as word processors and E-mail boxes? I can virtually garuntee there will be a distribution of Linux on these within the first week after release.
Originally Posted by Redwolf "With specs this impressive, it's easy to see why a few of us wondered how they planned to make a profit, given that they were already operating at a loss as a result of hardware costs for the current platform. "This time, we own the silicon [chip designs]. We can have [the chips] manufactured at different foundries to keep costs down."
I guess the next question is, why a PC at all anymore?
The reality is that hype is just that, and there is a chance that this thing won't work nearly like we all expect it to. If all this technology has been sitting around for the development of this, why would we not be hearing moves made towards this outside of the console market? This technology would be snapped up by PC users of many types (not just enthusiasts) with its universal architecture, assuming there was truly some benefit to it...at the very least, in a lesser format. There may be redesign of architecture that would need done, but I can't believe that NEITHER Intel nor AMD would have any interest in getting one over on each other and radically redefining things like this if it were truly the best way to be.
One will certainly be on my christmas list, but I am still going to be the skeptic and wait for the other shoe to drop...something is not right here...an advance this large has no reason to first debut on a console system.
I am inclined to agree with you, but a console is a unique opportunity to try something radically new because no backward and software compatibility issues have to be worried about. You can just start from scratch, and make your own games and peripherals to match.
In the PC market, compatibility with existing components and software is much more of an issue, but also it does not pay to jump too far ahead. Look at the CD-ROMvolution: Why jump from x8 to selling x32 CD-ROM drives if you can first make a healthy profit on x16 drives, and then do the same again a year later on x32 drives? Otherwise you get £750,-- dual core Intel CPUs which nobody buys because they are just too expensive for too little performance increase, because no software is available yet to really exploit it...
Still, it will be interesting to see how all that X-Box stuff ports to the PC a year down the line...
i can forsee the cost of the console not being to bad becuase if ATI if heavily subsidised this and if it goes well wich i'm sure it will then soon enough they will reap the benifits as they are gonna be the people with the experience of unifying the pipelines and Nvidia won't........well they got stick for putting old technology in the current xbox they really have done something about it, cvause heel you can't even get this jinda technology in desktops let alone afford it......MY GOD!
*faints and wakes up again to finsh post*
Originally Posted by Nexxo I am inclined to agree with you, but a console is a unique opportunity to try something radically new because no backward and software compatibility issues have to be worried about. You can just start from scratch, and make your own games and peripherals to match.
To a point...I know I'm not the only one who still goes back and plays my ps1 games on my ps2. It's a feature that has spoiled us on PC for years, and Sony put it to the console market. There have already been 4 posts between the 2 threads asking "Will this support my old XB1 games?" Console has more forgiveness in this field, but it is no longer a non-issue, particularly after one pays $300+ for a next-gen unit.
And if all of this technology works as planned, Intel and AMD lose the enthusiast game market. That is not something they will do lying down...Bye bye p4EE, bye bye FX57. There'd be no real need. And ATi's and NVidia's v-card business? Gone. Why would we spend thousands upgrading systems that could never beat the $300 one in our living room, that displays on a big screen?
All in all, I'm just saying caveat emptor...if this works like it's supposed to, then all of the pc companies sat around and fiddled while Rome burned. And I could see one or two miss the boat, but all of them? Tells me they might know something we don't.
If you look the 1meg L2 cache and whatever L1 I/D caches has been split 3 ways, so each core has it's own independent bit to access. It looks far more complex than Intel's "join 2 together and have done" method, although im not sure what the bit to the left is but there's clearly 2 sets of the same thing.
Originally Posted by Da Dego And if all of this technology works as planned, Intel and AMD lose the enthusiast game market. That is not something they will do lying down...Bye bye p4EE, bye bye FX57. There'd be no real need. And ATi's and NVidia's v-card business? Gone. Why would we spend thousands upgrading systems that could never beat the $300 one in our living room, that displays on a big screen?
I would guess that it will take a long time to use the full capabilities of both the CPU and GPU in the Xbox 360, and we're not going to see early games with much better graphics than what can be seen on the PC right now. By the time the Xbox 360 games start to use the consoles capabilities, we're likely to see quad cores from AMD and Intel - yes they're due in 2006 AFAIK, and unified shader architectures from both ATI and NVIDIA on the desktop are likely to appear as Longhorn rolls out.
G70 and R520 are the stop gap until Longhorn. Also, don't forget that the pipelines are unified, if the new GPU's roll up with 32 pixel pipelines and 8 vertex shaders, the difference in 'total' number of pipelines is not that massive IMHO. I'm not sure that this round of consoles will put an end to PC gaming as we know it, as there are plenty of things happening in the PC world that will keep things ticking over nicely.
Think of Xbox as a teaser for what is coming to the PC over the next couple of years. That's the way that I see consoles. I'm excited about this generation consoles, but I still believe that there will be a PC market running along side it for a good while yet.
To further add to this with some more technical overviews of R500...
I believe that the memory is on a 128-bit memory interface, running at 700MHz DDR (1400MHz effective), as that's more efficient than running at 350MHz DDR (700MHz effective) on a wider bus. Also, there'd be no point in using GDDR3 if it was running at only 350MHz.
Also, comparing pipelines in Xbox 360 to what we currently have in PC-based video cards is like comparing apples to oranges. I know we do that in our video card reviews, but this is slightly different in that retrospect. We are talking about unified shaders and specialised shaders here.
Some interesting tidbits here:
R500 has roughly twice the shader operation power per cycle compared to R420
and R500 also has about half the aliased (0xFSAA) fill rate power per cycle of R420.
The R420 has 96 ALU's in total. Each pixel shader has 5 ALUs: two vector, two scalar and one texture ALU per pipeline. Each vertex shader has two ALU's: one vector and one scalar.
The R500 has 48 unified ALU's that are each capable of 1 vector and 1 scalar op per cycle. In that sense the R500 could work similarly to a dual core - it's capable of multiple operations per cycle. That's where the 48 billion shader ops per second come from.
I was starting to wonder, as 48 x 500MHz does not equal 48 billion shader ops per sec. However, it's almost certain that the other 24 billion shader ops will be instruction duplicates of each other, so you can't actually achieve 48 billion shader ops without doing both vector and scalar ops simultaneously. This isn't the case in PC graphics, which is why comparing R500 to a current GPU is not quite as easy as it sounds. :)
Comments 1 to 25 of 82
Tch.
:)
Any idea on the price of this thing?
*saves for xmas*
BBC said ~$300 this morning I think, which means it'll probably be around the £300 mark ;)
...looks stunning, shame it really needs an expensive TV, wonder how much the kit to hook it up to a monitor will cost.
I guess the next question is, why a PC at all anymore?
The reality is that hype is just that, and there is a chance that this thing won't work nearly like we all expect it to. If all this technology has been sitting around for the development of this, why would we not be hearing moves made towards this outside of the console market? This technology would be snapped up by PC users of many types (not just enthusiasts) with its universal architecture, assuming there was truly some benefit to it...at the very least, in a lesser format. There may be redesign of architecture that would need done, but I can't believe that NEITHER Intel nor AMD would have any interest in getting one over on each other and radically redefining things like this if it were truly the best way to be.
One will certainly be on my christmas list, but I am still going to be the skeptic and wait for the other shoe to drop...something is not right here...an advance this large has no reason to first debut on a console system.
i really cant say more apart from triple core hyperthreading proccessor. and 48 GFX pipelines. jus holey crap
"With specs this impressive, it's easy to see why a few of us wondered how they planned to make a profit, given that they were already operating at a loss as a result of hardware costs for the current platform. "This time, we own the silicon [chip designs]. We can have [the chips] manufactured at different foundries to keep costs down."
http://www.tomshardware.com/game/200505121/xbox_360-05.html
lol you trust toms hardware
In the PC market, compatibility with existing components and software is much more of an issue, but also it does not pay to jump too far ahead. Look at the CD-ROMvolution: Why jump from x8 to selling x32 CD-ROM drives if you can first make a healthy profit on x16 drives, and then do the same again a year later on x32 drives? Otherwise you get £750,-- dual core Intel CPUs which nobody buys because they are just too expensive for too little performance increase, because no software is available yet to really exploit it...
Still, it will be interesting to see how all that X-Box stuff ports to the PC a year down the line...
*faints and wakes up again to finsh post*
To a point...I know I'm not the only one who still goes back and plays my ps1 games on my ps2. It's a feature that has spoiled us on PC for years, and Sony put it to the console market. There have already been 4 posts between the 2 threads asking "Will this support my old XB1 games?" Console has more forgiveness in this field, but it is no longer a non-issue, particularly after one pays $300+ for a next-gen unit.
And if all of this technology works as planned, Intel and AMD lose the enthusiast game market. That is not something they will do lying down...Bye bye p4EE, bye bye FX57. There'd be no real need. And ATi's and NVidia's v-card business? Gone. Why would we spend thousands upgrading systems that could never beat the $300 one in our living room, that displays on a big screen?
All in all, I'm just saying caveat emptor...if this works like it's supposed to, then all of the pc companies sat around and fiddled while Rome burned. And I could see one or two miss the boat, but all of them? Tells me they might know something we don't.
http://image.com.com/gamespot/images/2005/features/xbox360/xbox360_screen006.jpg
If you look the 1meg L2 cache and whatever L1 I/D caches has been split 3 ways, so each core has it's own independent bit to access. It looks far more complex than Intel's "join 2 together and have done" method, although im not sure what the bit to the left is but there's clearly 2 sets of the same thing.
I would guess that it will take a long time to use the full capabilities of both the CPU and GPU in the Xbox 360, and we're not going to see early games with much better graphics than what can be seen on the PC right now. By the time the Xbox 360 games start to use the consoles capabilities, we're likely to see quad cores from AMD and Intel - yes they're due in 2006 AFAIK, and unified shader architectures from both ATI and NVIDIA on the desktop are likely to appear as Longhorn rolls out.
G70 and R520 are the stop gap until Longhorn. Also, don't forget that the pipelines are unified, if the new GPU's roll up with 32 pixel pipelines and 8 vertex shaders, the difference in 'total' number of pipelines is not that massive IMHO. I'm not sure that this round of consoles will put an end to PC gaming as we know it, as there are plenty of things happening in the PC world that will keep things ticking over nicely.
Think of Xbox as a teaser for what is coming to the PC over the next couple of years. That's the way that I see consoles. I'm excited about this generation consoles, but I still believe that there will be a PC market running along side it for a good while yet.
I believe that the memory is on a 128-bit memory interface, running at 700MHz DDR (1400MHz effective), as that's more efficient than running at 350MHz DDR (700MHz effective) on a wider bus. Also, there'd be no point in using GDDR3 if it was running at only 350MHz.
Also, comparing pipelines in Xbox 360 to what we currently have in PC-based video cards is like comparing apples to oranges. I know we do that in our video card reviews, but this is slightly different in that retrospect. We are talking about unified shaders and specialised shaders here.
Some interesting tidbits here:
The R420 has 96 ALU's in total. Each pixel shader has 5 ALUs: two vector, two scalar and one texture ALU per pipeline. Each vertex shader has two ALU's: one vector and one scalar.
The R500 has 48 unified ALU's that are each capable of 1 vector and 1 scalar op per cycle. In that sense the R500 could work similarly to a dual core - it's capable of multiple operations per cycle. That's where the 48 billion shader ops per second come from.
I was starting to wonder, as 48 x 500MHz does not equal 48 billion shader ops per sec. However, it's almost certain that the other 24 billion shader ops will be instruction duplicates of each other, so you can't actually achieve 48 billion shader ops without doing both vector and scalar ops simultaneously. This isn't the case in PC graphics, which is why comparing R500 to a current GPU is not quite as easy as it sounds. :)