I originally titled this, "Xbox 360: Console price war in the making?" but thought it would be better to give my thoughts on this coming generation (I see it as an ugly price war), but let you guys sound off on how you think it will go. (Yes I was bored... yes, I started writing this in small pieces starting last week). No poll here--just give your thoughts on what is going to happen. My below ideas are on how I see the MS vs. Sony race. I do not know enough about what Nintendo will be doing to comment on them (although I do not believe them to only be a complete niche player).
This is meant to be fun--no flaming. Feel to disagree with me and give your own predictions.
-
In another thread I gave some thoughts on why I thought PS3 and CELL were a really killer move and how in the long run I think Sony has made a really good move. But I have also been thinking about the Xbox 360 design and have some thoughts on the issue of what I believe could be the beginnings of a very brutal pricing war. Your thoughts are always welcome.
Now that the leaks have been confirmed I have been pondering the future of the Xbox 360. It appears to me that the system really is designed to be scaled down, and quickly. They avoided the bad licensing agreements and seem to have created a design that will be cost effective down the road (something the Xbox never was).
1. 165M transister CPU and 150M transister GPU (not including eDRAM to my knowledge). A comparison: The two CELL CPUs we have seen were 234M and 250M transisters at 90nm. The NV40 (6800) is 222M transisters at 130nm. We do not know how big the PS3 GPU is, but looking at nVidias history and the fact it is being released after the Xbox 360 it would appear a safe guess that it would be a bigger chip (especially if we bear in mind the necessity of a powerful GPU to be paired with the powerful CELL).
At these sizes, it would seem plausible to consider the possibility that at 65nm (which will kick off in early 2006 and should be viable by the end of 2006/early 2007) MS could try to place the GPU and CPU on the same chip. That would save money and space in the console. This is one of the benefits over owning the IPs this generation and having more influence on the designs (something they did not have as much as when they selected off the shelf parts last gen and not owning the IPs).
2. 128bit memory. The choice of 128bit memory makes me believe MS will later replace the 128bit memory with 256bit memory, therefore saving space and cost in the future.
3. Detatchable HDD. This gives MS flexibility to offer not only different sized drives, but even different technology behind those drives. Flash prices could continue to bottom out. Who knows, but with an external drive they have a lot of flexibility in size and make.
-
It is no mystery that Sony is MS main target. And since we know almost nothing about Nintendo's system I will ignore that for now. From the little we know, it looks like the PS3 will have the following parts:
- CELL CPU
- XDR Memory
- BR Drive
- NV GPU
- HDD (or large memory sticks; either way they need something for microtransactions)
Tough Position #1: Costly parts. The first 3 items will be more expensive than the MS parts, so MS will already have an edge on price. Further, MS is going to be selling Xbox 360s for up to a year before Sony in certain regions, so it would seem reasonable to assume that a console that begins mass production sooner will begin to drop in production costs sooner.
Tough Position #2: Size. The next issue is size. The CELL and GPU will most likely be a but larger than the Xbox parts. The two CELL versions come in at about 234M and 250M transistors. That is a good 70M-85M more transistors than xCPU. So Sony has two options: fab a bigger and more expensive chip or cut back on CELL (1:4 or 1:6).
If they make a smaller CELL they deminish their technological lead.
If they make a larger CELL they put themselves in a position to fight a losing pricing battle.
Simiarly, my guess would be that the PS3 GPU will be bigger. The Xbox has eDRAM, but the PS3 GPU is rumored to not only have eDRAM but also a 2nd "chaperone" chip for lighting. That is still a rumor, but it would be hard to believe that the PS3 GPU will be smaller than the Xbox GPU as this will have a significant impact on how games look. Coming out a year later with a comparably sized GPU (and thus a simiarly performing GPU give or take) would be a mistake in my opinion. MS developers will be working on 2nd gen titles and will have already bloodied their noses on the R500. The PS3 GPU would be a newer challenge.
Tough Position #3: Memory. MS's master stroke in my opinion. MS knew two things: 1. Developers always want more memory and 2. the PS3 is set to use a high performance yet expensive memory in XDR (as you all know, Rambus has stated they expect a price premium on XDR compared to GDDR3 based on its performance advantage which is smart business sense).
Yet MS, the first out of the gate with the Xbox 360, has set the tone: Every console this generation will have 512MB of memory. Sony would lose a serious PR battle (and probably make a lot of disgruntled developers) if they decided to go with 256MB of memory. It would make porting to the PS3 difficult from the PC and Xbox 360. And if PS3 has the power advantage we expect, Xbox 360 titles having 2x as much memory could be a good way to gloss over some of the technical shortcomings.
Sony is basically stuck with 512MB of more expensive memory. I could not fathom them going with 256MB of memory--it would be a serious advantage in mindshare and would make a noticable difference in games.
Tough Position #4: Storage and transfer rates. The BR drive is more expensive than a 12X DVD drive. So that is one check against Sony in a price war. But price aside, BR brings a mixed bag to the table. Ignoring the format war (I am assuming that will be resolved with a merged format.....) BR does offer the GREAT incentive of HD movies. This is great for Sony because it gets their format into peoples homes and it is great for movie watchers with HD TVs.
A few problems: The PS3 is first and formost a game machine. Gamers are paying extra money for something that will not make any noticable difference in their games (less the few that require more space like a FF). The next problem is that this luxurious BR drive will, from what I have found online (I could be wrong... please correct me if these numbers are wrong) will increase load times
A 12x DVD can transfer about 15.85MB/s (http://www.osta.org/technology/dvdqa/dvdqa4.htm), while a 1x BR has a data transfer rate of about transfers about 6.75MB/s (http://www.blu-ray.com/faq/). (Note the BR site uses Mb which is megabits; you can test this by looking at the 36Mbps write rate that takes 93min to fill a 25GB disk, which clearly shows this is a Mb and not a MB... far too many sites use MB and Mb interchably). To load 512MB of memory we are looking at theoretical fastest transfers of 32 seconds for a 12X DVD drive and 75 seconds for a 1x BR drive. Even a 2x BR will lag behind a 12x DVD in theoretical transfer times. Load times are a big issue with games. We often neglect the little things when comparing systems, but if it takes on average 2x as long to load a game on a system it could be a turn off. This issue will be overcome to a certain degree by streaming content and allowing players to begin playing before the entire level is loaded, but with 512MB of memory this will be an issue.
Another negative for Sony is that they are also cutting into their BR sales. Unlike DVD drives that had been out for a couple years before the PS2 and were quickly dropping in price, there is a huge premium to be had with BR drives come 2006. They will be new, they will be hot and people will want them (especially with the explosive HD TV growth in the US). Why would anyone buy a $300 BR player when they can get a PS3 that has a BR for $300? The PS3 will be a unitl the lose money on, the BR player a unit they make money on. If consumers are buying expensive PS3s that they intend to use mainly for movies this could be a big negative for Sony. This would bring down their software attatched rate and lead to some hefty losses (if they are selling consoles below cost) and could affect a price war.
Another factor is a HDD. I believe Sony will have to offer some form of mass storage device because of their stated plan to offer microtransactions. They may very well go with large memory sticks. In fall 2006 in the US I would guess that 4GB memory sticks would be available with 2GB at a reasonable price. While a 2GB memory stick would work great for microtransactions, it would be less desirable for a lot of content updates, patches, new maps/levels, etc. It also would not aid in the caching of game data. While the 20GB Xbox 360 HDD will most likely be uses partially as a cache for game information to speed up load times, a PS3 without a HDD would be painfully slower than the Xbox 360 in load times.
-
I believe Sony was (is?) aiming to launch the PS3 at 65nm. This would alleviate a lot of the cost concerns. But I see a very intelligent plan from MS unfolding:
1. MS never was going to beat Sony in GFLOPs--the CELL was designed from the ground up to excell in this area. Sony, as a consumer electronics maker, has a huge advantage in this area.
2. So MS went the next best route: Release a year early with a balance machine (ala GCN) that is powerful but also cost effective, be aggressive and get developer support and maximize your tools and lead. MS has 40 titles ready for the launch window--a huge change from the Xbox 1. Obviously developers are CONVINCED MS is serious and is a real contended in the home console market. This is not grumpy Nintendo or flakey Sega. MS has really connected with some developers and 40 launch titles is proof of that.
3. When the more powerful/more expensive PS3 launches, release your killer app: Halo 3. A slew of 2nd gen titles also arrive partially glossing over some of the power difference. The Xbox 360 will have more titles, and probably more polished titles, in fall 2006.
4.Shrink your system down, get the CPU and GPU on the same chip, reduce the RAM modules by half--and then begin an aggressive price war. If the PS3 is at $300, sell for $200. MS knows Sony will be losing more per unit compared to MS, and MS also knows that they have deeper pockets to take a hit. If taking a HW hit keeps them ahead/neck-and-neck in total console sales I know MS would do it. MS then puts Sony in the position to either lose more money per unit or stick to their higher price point and possibly allow MS to make serious inroads. As has been mentioned on the forum before Sony makes about as much money as Nintendo does on games. They outsold the GCN 4:1, yet their profits were similar. Sony is not in a position to loose large portions of its marketshare with the floundering consumer electronics division.
I am not fully convinced MS's aim is to even topple Sony this generation. MS does not need to topple Sony as the market leader to adversely affect Sony. If they make Sony take larger losses on HW than expected and increase their market size MS will do enough to make a move in the following generation--all the while making a profit this generation.
I think MS has put themselves in a position to put Sony on the defensive.
-
There are a lot of reasons why this wont happen or it could backfire really bad for MS.
1. Xbox 360 is the worst kept console secret ever. MS did not launch the Xbox 360 in May 2005, but almost a year earlier with the accurate system configuration/spec leak. The patents uncovered only backed up the verasity of the leak. While Sony and Nintendo would not wholesale change their plan just because they saw what MS was up to, having info on your competitor 2 years before launch can give you meaningful insight into deciding a system configuration. Do we go with 256MB of memory or 512? Do we go with a GPU with 16 pixel shader units or 24? Do we want to go with a 1:4 CELL or a 1:8 CELL?
2. MS, with an earlier launch, may not be able to capitolize with quality software. 1st party mascot title, titles in the works for years, should be a technical tour de force. MS should be showing 3rd parties how to exploit their machine. PDZ did not do that on May 12th. Now who knows what will happen by launch (some of the new screenshoots looks leaps and bounds better), but if MS has a lot of misses at launch and Halo 3 appears rushed in 2006 MS could have put themselves in a bad position that even cutting prices wont totally alieviate.
3. The industry analysts are right for once and 2007 is when the next gen really starts. Even if MS gets 40 games at the launch window, developers could very well be waiting for the PS2 to die in 2005 to move on to next gen projects. If MS is not able to get a good stream of games to build a good gaming library and has a problem with substandard titles the early launch will be a negative than a positive.
4. Nintendo and Sony will just eat the losses and push hard on the fact their systems are more powerful. The power difference may be enough that the underpowered Xbox may be the odd man out. If MS is not able to get consumer interest with some killer titles early developer interest may turn to the more powerful consoles (if they are).
5. Sony and Nintendo launch earlier than MS expects. We expect Sony to launch in spring of 2006 in Japan. But what if Sony launches in the spring in the US as well?
6. Sony and Nintendo offer something truly different, something new, something advanced. MS already played their cards, so maybe Nintendo comes out with something so different--but good--that consumers get more interested in the more powerful PS3 and the neat Revolution?
7. The PS3 has a GPU and CPU at 65nm at launch, has a DVD player instead of a BR drive (BR player is offered in a more expensive PS3 unit). Not taking the hit at 90nm (cost, heat, power) would allow Sony to build a faster machine cheaper. This would minimize some of the price war losses and if they are able to offer a significantly more powerful system the extra cost, and wait, may be very well worth it to consumers.
-
In all I see the PS3 a really good move for the PS4. The PS4 CELL is really going to offer Sony a very well established (manufacturing and software development) platform that should have insane performance. The CELL is based on the concept of multiple units and should be better in a parallel environment than other chips.
So while the future looks bright for Sony, the immediate console generation may be ugly if things break MS way. I certainly expect Sony to keep their lead in the market, but if MS can offer quality software to go along with the GCN2 opps: I mean Xbox 360!) I have a feeling we will see a bitter price war with MS making headway in the market.
Anyhow, that is how I see this generation shaping up.
This is meant to be fun--no flaming. Feel to disagree with me and give your own predictions.
-
In another thread I gave some thoughts on why I thought PS3 and CELL were a really killer move and how in the long run I think Sony has made a really good move. But I have also been thinking about the Xbox 360 design and have some thoughts on the issue of what I believe could be the beginnings of a very brutal pricing war. Your thoughts are always welcome.
Now that the leaks have been confirmed I have been pondering the future of the Xbox 360. It appears to me that the system really is designed to be scaled down, and quickly. They avoided the bad licensing agreements and seem to have created a design that will be cost effective down the road (something the Xbox never was).
1. 165M transister CPU and 150M transister GPU (not including eDRAM to my knowledge). A comparison: The two CELL CPUs we have seen were 234M and 250M transisters at 90nm. The NV40 (6800) is 222M transisters at 130nm. We do not know how big the PS3 GPU is, but looking at nVidias history and the fact it is being released after the Xbox 360 it would appear a safe guess that it would be a bigger chip (especially if we bear in mind the necessity of a powerful GPU to be paired with the powerful CELL).
At these sizes, it would seem plausible to consider the possibility that at 65nm (which will kick off in early 2006 and should be viable by the end of 2006/early 2007) MS could try to place the GPU and CPU on the same chip. That would save money and space in the console. This is one of the benefits over owning the IPs this generation and having more influence on the designs (something they did not have as much as when they selected off the shelf parts last gen and not owning the IPs).
2. 128bit memory. The choice of 128bit memory makes me believe MS will later replace the 128bit memory with 256bit memory, therefore saving space and cost in the future.
3. Detatchable HDD. This gives MS flexibility to offer not only different sized drives, but even different technology behind those drives. Flash prices could continue to bottom out. Who knows, but with an external drive they have a lot of flexibility in size and make.
-
It is no mystery that Sony is MS main target. And since we know almost nothing about Nintendo's system I will ignore that for now. From the little we know, it looks like the PS3 will have the following parts:
- CELL CPU
- XDR Memory
- BR Drive
- NV GPU
- HDD (or large memory sticks; either way they need something for microtransactions)
Tough Position #1: Costly parts. The first 3 items will be more expensive than the MS parts, so MS will already have an edge on price. Further, MS is going to be selling Xbox 360s for up to a year before Sony in certain regions, so it would seem reasonable to assume that a console that begins mass production sooner will begin to drop in production costs sooner.
Tough Position #2: Size. The next issue is size. The CELL and GPU will most likely be a but larger than the Xbox parts. The two CELL versions come in at about 234M and 250M transistors. That is a good 70M-85M more transistors than xCPU. So Sony has two options: fab a bigger and more expensive chip or cut back on CELL (1:4 or 1:6).
If they make a smaller CELL they deminish their technological lead.
If they make a larger CELL they put themselves in a position to fight a losing pricing battle.
Simiarly, my guess would be that the PS3 GPU will be bigger. The Xbox has eDRAM, but the PS3 GPU is rumored to not only have eDRAM but also a 2nd "chaperone" chip for lighting. That is still a rumor, but it would be hard to believe that the PS3 GPU will be smaller than the Xbox GPU as this will have a significant impact on how games look. Coming out a year later with a comparably sized GPU (and thus a simiarly performing GPU give or take) would be a mistake in my opinion. MS developers will be working on 2nd gen titles and will have already bloodied their noses on the R500. The PS3 GPU would be a newer challenge.
Tough Position #3: Memory. MS's master stroke in my opinion. MS knew two things: 1. Developers always want more memory and 2. the PS3 is set to use a high performance yet expensive memory in XDR (as you all know, Rambus has stated they expect a price premium on XDR compared to GDDR3 based on its performance advantage which is smart business sense).
Yet MS, the first out of the gate with the Xbox 360, has set the tone: Every console this generation will have 512MB of memory. Sony would lose a serious PR battle (and probably make a lot of disgruntled developers) if they decided to go with 256MB of memory. It would make porting to the PS3 difficult from the PC and Xbox 360. And if PS3 has the power advantage we expect, Xbox 360 titles having 2x as much memory could be a good way to gloss over some of the technical shortcomings.
Sony is basically stuck with 512MB of more expensive memory. I could not fathom them going with 256MB of memory--it would be a serious advantage in mindshare and would make a noticable difference in games.
Tough Position #4: Storage and transfer rates. The BR drive is more expensive than a 12X DVD drive. So that is one check against Sony in a price war. But price aside, BR brings a mixed bag to the table. Ignoring the format war (I am assuming that will be resolved with a merged format.....) BR does offer the GREAT incentive of HD movies. This is great for Sony because it gets their format into peoples homes and it is great for movie watchers with HD TVs.
A few problems: The PS3 is first and formost a game machine. Gamers are paying extra money for something that will not make any noticable difference in their games (less the few that require more space like a FF). The next problem is that this luxurious BR drive will, from what I have found online (I could be wrong... please correct me if these numbers are wrong) will increase load times
A 12x DVD can transfer about 15.85MB/s (http://www.osta.org/technology/dvdqa/dvdqa4.htm), while a 1x BR has a data transfer rate of about transfers about 6.75MB/s (http://www.blu-ray.com/faq/). (Note the BR site uses Mb which is megabits; you can test this by looking at the 36Mbps write rate that takes 93min to fill a 25GB disk, which clearly shows this is a Mb and not a MB... far too many sites use MB and Mb interchably). To load 512MB of memory we are looking at theoretical fastest transfers of 32 seconds for a 12X DVD drive and 75 seconds for a 1x BR drive. Even a 2x BR will lag behind a 12x DVD in theoretical transfer times. Load times are a big issue with games. We often neglect the little things when comparing systems, but if it takes on average 2x as long to load a game on a system it could be a turn off. This issue will be overcome to a certain degree by streaming content and allowing players to begin playing before the entire level is loaded, but with 512MB of memory this will be an issue.
Another negative for Sony is that they are also cutting into their BR sales. Unlike DVD drives that had been out for a couple years before the PS2 and were quickly dropping in price, there is a huge premium to be had with BR drives come 2006. They will be new, they will be hot and people will want them (especially with the explosive HD TV growth in the US). Why would anyone buy a $300 BR player when they can get a PS3 that has a BR for $300? The PS3 will be a unitl the lose money on, the BR player a unit they make money on. If consumers are buying expensive PS3s that they intend to use mainly for movies this could be a big negative for Sony. This would bring down their software attatched rate and lead to some hefty losses (if they are selling consoles below cost) and could affect a price war.
Another factor is a HDD. I believe Sony will have to offer some form of mass storage device because of their stated plan to offer microtransactions. They may very well go with large memory sticks. In fall 2006 in the US I would guess that 4GB memory sticks would be available with 2GB at a reasonable price. While a 2GB memory stick would work great for microtransactions, it would be less desirable for a lot of content updates, patches, new maps/levels, etc. It also would not aid in the caching of game data. While the 20GB Xbox 360 HDD will most likely be uses partially as a cache for game information to speed up load times, a PS3 without a HDD would be painfully slower than the Xbox 360 in load times.
-
I believe Sony was (is?) aiming to launch the PS3 at 65nm. This would alleviate a lot of the cost concerns. But I see a very intelligent plan from MS unfolding:
1. MS never was going to beat Sony in GFLOPs--the CELL was designed from the ground up to excell in this area. Sony, as a consumer electronics maker, has a huge advantage in this area.
2. So MS went the next best route: Release a year early with a balance machine (ala GCN) that is powerful but also cost effective, be aggressive and get developer support and maximize your tools and lead. MS has 40 titles ready for the launch window--a huge change from the Xbox 1. Obviously developers are CONVINCED MS is serious and is a real contended in the home console market. This is not grumpy Nintendo or flakey Sega. MS has really connected with some developers and 40 launch titles is proof of that.
3. When the more powerful/more expensive PS3 launches, release your killer app: Halo 3. A slew of 2nd gen titles also arrive partially glossing over some of the power difference. The Xbox 360 will have more titles, and probably more polished titles, in fall 2006.
4.Shrink your system down, get the CPU and GPU on the same chip, reduce the RAM modules by half--and then begin an aggressive price war. If the PS3 is at $300, sell for $200. MS knows Sony will be losing more per unit compared to MS, and MS also knows that they have deeper pockets to take a hit. If taking a HW hit keeps them ahead/neck-and-neck in total console sales I know MS would do it. MS then puts Sony in the position to either lose more money per unit or stick to their higher price point and possibly allow MS to make serious inroads. As has been mentioned on the forum before Sony makes about as much money as Nintendo does on games. They outsold the GCN 4:1, yet their profits were similar. Sony is not in a position to loose large portions of its marketshare with the floundering consumer electronics division.
I am not fully convinced MS's aim is to even topple Sony this generation. MS does not need to topple Sony as the market leader to adversely affect Sony. If they make Sony take larger losses on HW than expected and increase their market size MS will do enough to make a move in the following generation--all the while making a profit this generation.
I think MS has put themselves in a position to put Sony on the defensive.
-
There are a lot of reasons why this wont happen or it could backfire really bad for MS.
1. Xbox 360 is the worst kept console secret ever. MS did not launch the Xbox 360 in May 2005, but almost a year earlier with the accurate system configuration/spec leak. The patents uncovered only backed up the verasity of the leak. While Sony and Nintendo would not wholesale change their plan just because they saw what MS was up to, having info on your competitor 2 years before launch can give you meaningful insight into deciding a system configuration. Do we go with 256MB of memory or 512? Do we go with a GPU with 16 pixel shader units or 24? Do we want to go with a 1:4 CELL or a 1:8 CELL?
2. MS, with an earlier launch, may not be able to capitolize with quality software. 1st party mascot title, titles in the works for years, should be a technical tour de force. MS should be showing 3rd parties how to exploit their machine. PDZ did not do that on May 12th. Now who knows what will happen by launch (some of the new screenshoots looks leaps and bounds better), but if MS has a lot of misses at launch and Halo 3 appears rushed in 2006 MS could have put themselves in a bad position that even cutting prices wont totally alieviate.
3. The industry analysts are right for once and 2007 is when the next gen really starts. Even if MS gets 40 games at the launch window, developers could very well be waiting for the PS2 to die in 2005 to move on to next gen projects. If MS is not able to get a good stream of games to build a good gaming library and has a problem with substandard titles the early launch will be a negative than a positive.
4. Nintendo and Sony will just eat the losses and push hard on the fact their systems are more powerful. The power difference may be enough that the underpowered Xbox may be the odd man out. If MS is not able to get consumer interest with some killer titles early developer interest may turn to the more powerful consoles (if they are).
5. Sony and Nintendo launch earlier than MS expects. We expect Sony to launch in spring of 2006 in Japan. But what if Sony launches in the spring in the US as well?
6. Sony and Nintendo offer something truly different, something new, something advanced. MS already played their cards, so maybe Nintendo comes out with something so different--but good--that consumers get more interested in the more powerful PS3 and the neat Revolution?
7. The PS3 has a GPU and CPU at 65nm at launch, has a DVD player instead of a BR drive (BR player is offered in a more expensive PS3 unit). Not taking the hit at 90nm (cost, heat, power) would allow Sony to build a faster machine cheaper. This would minimize some of the price war losses and if they are able to offer a significantly more powerful system the extra cost, and wait, may be very well worth it to consumers.
-
In all I see the PS3 a really good move for the PS4. The PS4 CELL is really going to offer Sony a very well established (manufacturing and software development) platform that should have insane performance. The CELL is based on the concept of multiple units and should be better in a parallel environment than other chips.
So while the future looks bright for Sony, the immediate console generation may be ugly if things break MS way. I certainly expect Sony to keep their lead in the market, but if MS can offer quality software to go along with the GCN2 opps: I mean Xbox 360!) I have a feeling we will see a bitter price war with MS making headway in the market.
Anyhow, that is how I see this generation shaping up.