Report: PS4 Is Selling Twice As Well As Xbox One (arstechnica.com)

The latest numbers released by analysts suggest that the Sony PlayStation 4 is selling twice as many units worldwide as the Xbox One since both systems launched in late 2013. The data comes from a new SuperData report on the Nintendo Switch, which is backed up by Niko Partners analyst Daniel Ahmad. SuperData mentions an installed base of 26 million Xbox One units and 55 million PS4 units. Ars Technica reports: Ahmad's chart suggests that Microsoft may have sold slightly more than half of the 53.4 million PS4 units that Sony recently announced it had sold through January 1. Specific numbers aside, though, it's clear Microsoft has done little to close its console sales gap with Sony over the past year -- and may have actually lost ground in that time. The last time we did our own estimate of worldwide console sales, through the end of 2015, we showed the Xbox One with about 57 percent as many systems sold as the PS4 (21.49 million vs. 37.7 million). That lines up broadly with numbers leaked by EA at the time, which suggest the Xbox One had sold about 52.9 percent as well as the PS4 (19 million vs. 35.9 million). One year later, that ratio has dipped to just above or even a bit below 50 percent, according to these reports. The relative sales performance of the Xbox One and PS4 doesn't say anything direct about the health or quality of those platforms, of course. Microsoft doesn't seem to be in any danger of abandoning the Xbox One platform any time soon and has, in fact, recently committed to upgrading it via Project Scorpio later this year. The gap between PS4 and Xbox One sales becomes important only if it becomes so big that publishers start to consider the Xbox One market as a minor afterthought that can be safely ignored for everything but niche games.

Comments

so what?

[dnorman]

Both are viable platforms with lots of games and players. Who cares if one is twice the size of the other?

Re:so what?

[Anonymous Coward]

Mostly investors, publishers and developers.

Re:so what?

[Anonymous Coward]

none of those care, they all only care about bottom line profit and it is profitable on both. PS will always win worldwide as Xbox is mostly a western audience and in the west it actually runs pretty well neck and neck with PS (sometimes above sometimes below). The Wii massively outsold both PS and Xbox last gen yet investors, publishers and developers would have ALL been pissed at the profits from it.

Re:so what?

[Anonymous Coward]

"none of those care, they all only care about bottom line profit and it is profitable on both."

_but it is *more* profitable on PS4_

So you're saying they don't care about MORE profit? Who are these philanthropists?

Re:so what?

[Anonymous Coward]

Not necessarily all games will be more profitable on the PS4.

Re:so what?

[taxman_10m]

Isn't player base critically important for multiplayer games? PS4's advantage should mean that CoD/BF1/TF2/etc online player base for the game should be sustainable longer than for the XBox One.

Re:so what?

[Anonymous Coward]

most of the profit from PS goes to pay Sony's mountains of debt, and as a percentage of Sony or MS both the playstation and Xbox brands are little more than rounding errors. Everyone always wants more money, but as long as they aren't losing money is the main point.

Re:so what?

[Anonymous Coward]

publishers don't give a shit, only dick measuring gamers care. It is total audience and profit they care about, both ecosystems have more than enough gamers to be profitable and hence both get all the 3rd party games that either company hasn't paid for an exclusive deal for. PlayStation could be 10 times the market and as long as Xbox market was still profitable they will happily service that market.

Re:so what?

[gravewax]

not really, that is only true if a game was struggling for a userbase. One having 10 million COD users and the Other 5 will not make the slightest difference. COD/BF etc are all annual or biannual releases with a constantly moving userbase. This however might be true for some small unpopular games with very low userbases but then those type of games don't matter anyway.

Re:so what?

[Anonymous Coward]

I'm a developer. I bought an Xbox One because I can write software for it without a $uper $pecial Dev Kit.

Costco had one (a 500TB "S" model) on a black friday deal for $230 with a free download of Minecraft. The cheapest I've seen for a PS4 (not a used one from gamestop) is $280. There were lots of deals this last holiday season.

Meanwhile, I haven't bothered with a PS4 because I have a seldom-used PS3, so I don't need another black box cluttering up my shelf. I'm also far too lazy to buy games from anywhere but Steam these days. (FWIW, I'm developing a UWP app for a specialty niche that would allow a "cheap" Xbox One to run it instead of requiring a PC. A PC with the same multimedia and processing capabilities that my app needs would be quite a bit more expensive than the Xbox One.)

Re:so what?

[tepples]

I'm a developer. I bought an Xbox One because I can write software for it without a $uper $pecial Dev Kit.

True, Microsoft has been generous with ID@Xbox. But if you can make more money by selling copies to PlayStation 4 owners than by selling copies to the smaller Xbox One audience, that might make up for the cost of such a devkit.

I'm developing a UWP app for a specialty niche that would allow a "cheap" Xbox One to run it instead of requiring a PC.

What keeps you from also releasing ports of the app to X11/Linux, Win32 desktop, and possibly macOS? Then you can make it available to people who already have a PC.

A PC with the same multimedia and processing capabilities that my app needs would be quite a bit more expensive than the Xbox One.

In before Hairyfeet mentions that many owners of what's now considered an entry-level desktop PC can drop in a $150 video card and get performance more than on par with Xbox One.

Re:so what?

[elrous0]

Since PS4 and Xbox One have both adopted similar x86 architectures this generation and have similar hardware specs, the issue is largely moot now. Developing a game designed for both systems is pretty easy and so a developer would just be throwing away money if it just developed for one or the other.

Now as for Nintendo, on the other hand....

Re:so what?

[Anonymous Coward]

Isn't player base critically important for multiplayer games?

Microsoft fanboys like to move the goalposts, even when they have to move them backwards to save face. Windows market share is critically important—thus making it inherently superior to all other platforms—but Xbox market share somehow doesn’t matter (as long as they’re down; otherwise, it totally matters).

Amazing

[Anonymous Coward]

Considering Destiny and Bloodborne are currently the only reasons to own a game console, they still managed to sell 75 million consoles.

Re:Amazing

[Anonymous Coward]

The main reason to own a console is that it's much cheaper than a gaming PC. Many families can't justify the cost of more than one gaming PC, and besides, life's too short to deal with driver incompatibility.

Re:Amazing

[Anonymous Coward]

Also easier to deal with and possibly better exclusives (with a PS4). However, if you need a new desktop computer in general and also want a game platform, then it may be worth investing in a PC built for gaming. Otherwise, even on the low end for a gaming PC, it will still cost more than a PS4 Pro and likely won't be that much better. You need to spend at least twice as much before you really start noticing a difference. That said, the console online services add up and over time may negate the cost savings against a lower end gaming PC.

Re:Amazing

[Ranbot]

The main reason to own a console is that it's much cheaper than a gaming PC. Many families can't justify the cost of more than one gaming PC, and besides,

I don't believe the claim that game consoles are cheaper than gaming PCs for a few reasons:

1) Most families still need or want a computer at home for reasons besides gaming (e.g. internet, word processing, tracking finances, online banking, digital storage, remote connections to workplaces). This is particularly true for families with children because many school homework assignments today require or are greatly assisted by online research. So, if a family will have a home computer anyway, but buys a separate console for gaming at nearly the same price as the PC, where's the savings?

2) The price of games should be factored into the cost of a game system and games are cheaper on PC through digital distributors like Steam, Origin, etc., which over time off-sets the initial cost of the PC.

3) If you want to play games online (which many people do) you have to add the life-time cost of an online subscription to a console.

4) You don't need an expensive PC to play games. A $500 PC (which is comparable to a new PS4+accessories) will play ~98% of the PC games. If you have to buy a new monitor a PC is a little more expensive, but it is a more versatile device (see point 1). PC tech upgrades cycles are much slower than they were years ago it shouldn't become obsolete right away either.

Yes, the initial sticker price of a console is very likely cheaper than a PC, but if you add up life-cycle costs and versatility it's much harder to justify a console over a PC based on price. Consoles used to have a cost advantage in that they double as DVD/Blueray players, but that is less of factor as more people view movies/shows in digital streaming formats.

...life's too short to deal with driver incompatibility.

10+ years ago this would be a valid argument, but it's not the norm anymore. As long as you are tech savy enough to connect your computer to online, from there most peripherals and programs search and install their own drivers in seconds.

Welcome to the Osborne Effect

[SuperKendall]

For those not up on computer history, Osborne was a computer maker that announced a great new model coming in a year... so sales started tanking while people waited... which meant there was no model in a year (or maybe there was, my memory is fuzzy on that detail).

I think MS was really dumb to try and compete with the PS4 Pro by saying they would have improved hardware next year. All they had to do was literally nothing, the PS4 Pro is not big enough of a bump that it would have effected XBox sales...

Although really it seems like XBox sales have been lagging even before the recent hardware upgrade was announced.

Re:Welcome to the Osborne Effect

[Yaztromo]

For those not up on computer history, Osborne was a computer maker that announced a great new model coming in a year... so sales started tanking while people waited... which meant there was no model in a year (or maybe there was, my memory is fuzzy on that detail).

Microsoft had a pattern of doing this throughout the 90's, and it generally worked out well for them. As soon as other PC operating systems (and OS/2 in particular) started chipping away at the badly aging Windows 3.1x line, Microsoft started promising the moon with Windows 95/PC DOS 7 -- more than two years before it shipped. They didn't deliver on most of their promises, and the end result was worse than the competition, but by that point it didn't matter -- people believed the hype and decided to skip the competition out of fear that the competition was going to be eclipsed in a years time. They did the same with Windows NT. Remember "Cairo"? Microsoft started talking about it in 1991, and continued through 1996 before dropping the release completely. WinFS probably takes the cake -- a complete redesign of how a PC OS stores information, it was first promised in 1991, and was continually touted until 2006, usually in around whenever a competing OS was being released.

Over-promising way in advance and under-delivering was MS's modus operandi through much of the 1990's and early 2000's. A lot of people fell for it, and a lot of people continue to fall for it (I follow some PlayStation forums now and then, and have seen more than one person claim they're waiting until Scorpio ships because it's going to blow everything that ever came before it out of the water...sound familiar?).

Yaz

Re:Welcome to the Osborne Effect

[gman003]

Good theory, but I think you're wrong. That's not what's hurting Microsoft - Sony did the same thing, with rumors of the "Neo" appearing shortly after the console itself launched. And yet the PS4 still sold quite well from day one.

What hurt the Xb1 is that it's demonstrably weaker than the PS4, but cost significantly more at launch ($500 compared to $400). Most games are available on both, so the natural inclination is to go with the cheaper and more powerful console. With a wide library of shared games, there's lots of direct comparisons to make, and even before they launched, it was easy to tell the PS4 would be more powerful. That gave the PS4 a very strong advantage during the first year or two.

Even now, they only have price-parity, with both having an entry price around $250-$300. But more people already have a PS4, making that the more attractive option both for multiplayer gaming (if all your friends are on PS4, you'd want one too) and for the larger percentage of third-party exclusive titles (it's nowhere near as big a deal as it once was, since porting is so easy, but there's still some studios that are deciding to skip the Xb1 because the audience is smaller). And it seems to me (as a non-Xb1, non-PS4 gamer) that Sony's shoveling the money from their console sales into more first-party games, giving it a still stronger library, which is ultimately what every gamer cares about.

Microsoft doesn't have a lot of options for coming back from this, just as the PS3 struggled to come back from the Xb360's early lead and the XbC never came close to the PS2. They could make the Scorpio be *substantially* more powerful than the PS4 Pro, making it more future-proof and maybe able to handle 4K/VR better. They could slash the price, and hope to catch up that way, but that's a risky move. They could pin it on VR or AR, but that's riskier still. They could double-down on their cross-play with PC bets - make every single Xb1 game PC-compatible and bundle a PC version, which would widen their library (although it would cannibalize Scorpio somewhat). Or they could go on a spending spree and buy up every developer they can, and kill off the PS4's third-party support - Sony is no Nintendo, they can't survive on first-party games alone (even Nintendo might not do so much longer).

Re:Welcome to the Osborne Effect

[zerocommazero]

Microsoft's XBox division and OS division are cannibalizing each other as I see it. And since Windows is paramount to Microsoft's overall sustainability, OS will always get the focus. They can try to sell both together as a strength but really, they are both competing for a gamer's attention. You can only play games on one at a time (for a single gamer obviously, I'm not including a family in this example). There's no defining reason to own an XBox if you are also an avid PC gamer. Cross-play makes it even more noticeable. In fact, if you compare all of the gaming outlets (PC, PS4, XBOne, Nintendo), XBOne is the weakest out of all of them based on strengths. PC has everything and if you are willing to pay for high grade hardware, the sky is the limit. Sony has a variety of exclusives and majority support, Nintendo has the greatest (in number and in quality for some titles) exclusives but not much else. XBOne is kinda... there. Even its exclusives are mostly in a single genre, primarily FPSes. If you go with a primary/secondary console to give yourself the a bigger overall selection to choose from, pairing any with the XBOne is not a great option for diversity.

Re:Welcome to the Osborne Effect

[Anonymous Coward]

XBox is great if you're into US games, particularly first person shooters. Also, didn't they come out slightly ahead last gen (360 vs PS3)? I personally enjoy a lot of Japanese games and am not into first person shooters though, so getting an XBox would be a waste.

Re:Welcome to the Osborne Effect

[Anonymous Coward]

That could be due to microsoft cancelling or pushing back a lot of their exclusive titles.

Re: Welcome to the Osborne Effect

[Shepanator]

By the end of the generation Ps3 actually managed to sell pretty much the exact same amount as the 360, and by some counts even slightly more. The 360 performed quite poorly towards the end of the gen, while on ps3 there was a constant stream of great exclusives. Besides that, the ps4 has more first person shooters than the Xbox one. So unless you're a diehard halo fan there really isn't a reason to get one

Games though?

[TheOuterLinux]

They sell twice as many consoles but have about 1/5 the amount of games as Xbox. This is what happens when people get too dependent on DirectX when OpenGL is soooo much better and available to everyone.

Re:Games though?

[Mashiki]

OGL is effectively dead for gaming, and has been superseded by mantle/vulkan. Better FPS then either OGL or DX, less overhead cost for the API. Supports all platforms and the FPS difference on team AMD or Nvidia cards is so small that it makes no difference. Where you can see +10FPS swings/drops with OGL or DX between either card. And OGL created their own death all on their own with the massive 3.0 fuckup by stripping and ripping out features that developers were waiting for, that pushed a lot of companies to use DX.

The current console generation was just a shitshow from the start, especially when you had developers pushing the "but 30FPS is cinematic." Gotta thank both MS and Sony for helping PC gaming in that regard.

Re:Games though?

[exomondo]

They sell twice as many consoles but have about 1/5 the amount of games as Xbox. This is what happens when people get too dependent on DirectX when OpenGL is soooo much better and available to everyone.

But the Playstation doesn't use OpenGL anyway, one advantage of those high level APIs is they allow abstraction of a vast range of hardware with a cost of overhead for doing so. The Playstation does not have a range of graphics hardware so suffering that overhead would be pointless, they use their own low-level graphics API.

Re:Games though?

[Dorianny]

They sell twice as many consoles but have about 1/5 the amount of games as Xbox. This is what happens when people get too dependent on DirectX when OpenGL is soooo much better and available to everyone.

But the Playstation doesn't use OpenGL anyway, one advantage of those high level APIs is they allow abstraction of a vast range of hardware with a cost of overhead for doing so. The Playstation does not have a range of graphics hardware so suffering that overhead would be pointless, they use their own low-level graphics API.

But the Playstation does have a range of hardware with the release of the PS Pro. MS will have the advantage here as the DirectX API originating from the PC was designed from the start to easily develop for a range of hardware. It will be significantly easier for developers to take advantage of the extra power in Project Scorpio then it is for the PS Pro

Re: Games though?

[Anonymous Coward]

Bullcrap. Directx or whatever indo-chimps are "developing" is windows code - you have to use visual studio, and it sucks. No sane linux, android, mac developer will ever be touching windows. One exception I can think of - when Monkeyshit corp goes bust and sells out - whoever their new owner that brings back OpenGL will win.

Re: Games though?

[Anonymous Coward]

>Bullcrap. Directx or whatever indo-chimps are "developing" is windows code - you have to use visual studio, and it sucks.

Why do you think you have to use Visual Studio? You realize Visual Studio is an IDE and you don't need to use that to compile software for Windows right? Not only do you not have to use Visual Studio but you dont even have to use the Microsoft compilers, you can use GCC if you want and just convert the DX libs to .a files.

>No sane linux, android, mac developer will ever be touching windows.

By definition they would have no reason to.

>One exception I can think of - when Monkeyshit corp goes bust and sells out - whoever their new owner that brings back OpenGL will win.

Like it or not OpenGL is on its deathbed, it will still be supported going forward but Khronos has made it quite clear that Vulkan is the future.

Re:Games though?

[exomondo]

But the Playstation does have a range of hardware with the release of the PS Pro.

Yes, a range of 2, just like the XBox One and Project Scorpio. And it is forward compatible meaning that you can develop for the PS4 and it works just as good on the PS4 Pro, you only need to put in additional effort if you want create a unique experience on the PS4 Pro.

MS will have the advantage here as the DirectX API originating from the PC was designed from the start to easily develop for a range of hardware. It will be significantly easier for developers to take advantage of the extra power in Project Scorpio then it is for the PS Pro

Well given the assumption that Project Scorpio will be backwards compatible with the XBox One what is a specific example of something that will be so much easier to do with DirectX on Project Scorpio to create a unique experience that will be difficult to do on the PS4 Pro with PSGL?

Re: Games though?

[91degrees]

Like it or not OpenGL is on its deathbed, it will still be supported going forward but Khronos has made it quite clear that Vulkan is the future.

No, it really isn't. Vulkan is a hardware abstraction layer more than an API. Really good for getting down and dirty with the hardware, but way too low level for actual practical use.

Re:Games though?

[ilguido]

They [...] have about 1/5 the amount of games as Xbox

It's the other way around, a lot of indie and not so indie titles (mainly Japanese stuff, but also games like Tropico 5) are out on PS4, but not on Xbone.

Re: Games though?

[Anonymous Coward]

Not really. An API is an interface with which application-level programs communicate, which is what Vulkan is. The hardware abstraction layer is between Vulkan and the driver.

I think you have Vulkan mixed up with another recently talked-about project called Gallium3D.

Re: Games though?

[91degrees]

No. I realise I'm using the jargon inaccurately, but the point is that the applications programmers will rarely use the Vulkan API as the programming interface. They'll use higher level libraries. The benefit of Vulkan is that the high level libraries can access the hardware without needing any specific details about the hardware. In other words, it abstracts the hardware.

Re: Games though?

[Shepanator]

The ps4 pro has the exact same Cpu as the ps4 just clocked considerably higher, and is has the exact same gpu too, just with twice as many compute units and again clocked higher. The hardware is very closely related, optimisations on one can directly carry over to the other. Also ps4 supports directx if devs choose to go that way.

Re: Games though?

[Anonymous Coward]

Vulkan is a hardware abstraction layer more than an API.

No it is not. Perhaps step through this most basic of tutorials and understand it before making statements like this. What part of what you learned from that tutorial makes you think it is so much more low level than what you do in OpenGL?

Really good for getting down and dirty with the hardware, but way too low level for actual practical use.

While you *can* get more control over a lot of the hardware aspects it is not mandatory to do so. There is some boilerplate code for doing things like device enumeration and initialization with respect to capabilities, you do similar things in OpenGL too but it has become reasonably standard to just use GLEW for that boilerplate stuff.

Re:Games though?

[pezezin]

We all know Vulkan is the future of computer graphics, but OpenGL is far from dead. It's right now the standard API for Linux, Android and the web. Also, the new Doom uses OpenGL, and is probably the best performing game of 2016 (although the Vulkan version performs even better).

Almost identical architectures

[Actually, I do RTFA]

The danger of not supporting a platform is increased as the platforms differences increase. The Xbox One and PS4 both have similar underlying hardware. Therefore, there is not as much danger as a publisher will ignore one of those two compared to the Switch.

Re:Almost identical architectures

[guises]

The Switch is an ARM SoC. Everyone but everyone is developing for ARM right now - if anything I'd be more worried about the other two.

Re:Almost identical architectures

[gravewax]

Switch is competing for game developers with the Xbox and PS, Nintendo has struggled massively with 3rd party support previously and it doesn't look like that will change much. There architecture is massively different and that makes them much more expensive to develop for. Xbox and PS are similar with a combined userbase of more than 80 million. Switch isn't destined for disaster but it has a lot of work ahead to be successful as apart from the gimmicky Wii Nintendo have been in a userbase slide for a long time now.

Re:Almost identical architectures

[guises]

Switch is competing with the Xbox and PS, among other platforms. At this point there are basically two architectures for gaming: x86 on PC, Xbox, and PS, and ARM on Switch and mobile. Both are very big, very well established architectures with extensive development tools available - neither pose any significant obstacles for developers. Nintendo may have a small edge in this respect, since ARM is the big thing right now.

Nintendo has struggled with third parties in some cases in the past because of some difficulty with supporting their architecture and interface choices. Lack of third party support is certainly not a given with Nintendo, however. Let me remind you that they were masters of third party support in the 8 and 16 bit generations, and continue to completely dominate portable console gaming with no shortage of third party support.

Neither of these issues apply to the switch: completely standard architecture, and almost completely standard interface. There is no technical reason why third parties wouldn't flock to this thing.

Re:Almost identical architectures

[gravewax]

if Nintendo plans to just compete with mobile then they are fucked before they even start. They need to compete with Xbox and PS and get ports of the games targeting those platforms not the shitty mobile games which when given a choice the majority will just buy them on their mobile

Re:Almost identical architectures

[Actually, I do RTFA]

Everyone but everyone is developing for ARM right now

AAA titles are not developed for ARM. I cannot think of any game with more than $10MM in dev. costs that was for an ARM platform.

Re:Almost identical architectures

[guises]

Where did you get the idea that they were just competing with mobile? That doesn't make any sense.

I'm a little excited for the Switch for exactly the reason you hint at: I see this as maybe a chance to raise the bar on mobile gaming. Mobile games are, as you say, shit. If devs start porting over Switch games though, real games, it could perhaps help to pull mobile gaming out of that hole. At least to some degree.

Probably not. This may just be wishful thinking on my part, the interface is different between the Switch and mobile and that's really more important than architecture, but... Maybe? I hope.

Re:Almost identical architectures

[guises]

I can think of a few, but that doesn't matter: (almost) every studio makes ARM games now, regardless of their budget, and that means that (almost) every studio has ARM experience. Thus, ARM does not pose a barrier to entry.

Re:Almost identical architectures

[Actually, I do RTFA]

The issue isn't "studio has ARM experience", its "how many hours does it take to port from the initial configuration to ARM" Because a AAA product will be designed first for Xbox/PS4/Computers.

Re:Almost identical architectures

[guises]

This is circular logic: "AAA products won't be designed for ARM because AAA products are not designed for ARM." The fact that it hasn't been done so far means squat, this is the first time that a major gaming console is... Actually, no it isn't the first time. There have been a ton of major gaming products designed for ARM - the DS, 3DS, and PS Vita are all ARM based.

Regardless, even if all that you care about is ports of other console games the fact that everyone is currently developing for ARM means that the tools for porting from x86 are already in place. Most major gaming engines support both: Unreal, Gamebryo, Blitztech, CryEngine, etc. Really, out of every generation of consoles to date, this seems like the one for which porting will be the easiest - support for porting between ARM and x86 is certainly at a much more mature state than porting between PowerPC (Xbox 360) and Cell (PS3), for example.

Re:Almost identical architectures

[gravewax]

well they are, how the fuck doesn't it make sense, they have a mobile type console. they have put themselves in no mans land, they are a poor underpowered console and an overpriced mobile device. I like Nintendo and hope they succeed but I just can't see it happening with the switch, everything is against them from lack of games, underpowered console, competing to get mobile market share and they are introducing paid online to top it off.

Re:Almost identical architectures

[Actually, I do RTFA]

It's not circular logic - ports of AAA products won't be designed for ARM because the current AAA products aren't designed for ARM. New AAA products won't be designed for ARM, because ARM doesn't have a history of AAA products that you can point to to demonstrate to the money people that AAA titles are worth it on ARM

a ton of major gaming products designed for ARM - the DS, 3DS, and PS Vita are all ARM based.

Handhelds. Not AAA domain.

Most major gaming engines support both: Unreal, Gamebryo, Blitztech, CryEngine, etc

You really overestimate the "write once, run everywhere" nature of game engines.

ut of every generation of consoles to date, this seems like the one for which porting will be the easiest - support for porting between ARM and x86 is certainly at a much more mature state than porting between PowerPC (Xbox 360) and Cell (PS3), for example.

It will be the easiest... because all AAA titles will be ported from XBox (x86) to PS4 (x86) to PC (x86). All indie games will be published across all platforms that they can push to.

Re:Almost identical architectures

[iampiti]

I was under the impression that nowadays there's very little to none low level code in games. Also, most bigger engines support a lot of platforms. I don't think using an ARM CPU makes porting games too difficult.

Re:Almost identical architectures

[Anonymous Coward]

Experienced game developer here. C/C++ games and engines have no problem supporting any CPU architecture. We've been using PowerPC for many years and usually the codebase also worked on PC. And, to be honest, it's the x86 architecture that gets in the way with it's un-sexyness making it a very boring job.

After having to eat x86 for a few years, trust me, AAA developers are VERY excited about being able to "switch" to 64 bit ARM assembler :-) It's another thing if it'll make sense from a economic perspective or not, that's on gamer's hands right now. If Nintendo sells 30 million Switch consoles you bet it's going to have a similar treatment as the other systems. My theory? They can sell 100 million. So maybe it's Sony and Microsoft that should be worried. The hardware is not that underpowered.

Re:Almost identical architectures

[Anonymous Coward]

The Switch can work as a handheld. But it's a TV system.

I think kids will love being able to play FIFA multiplayer local games at the school. And it's going to look the same as the PS4 version.

Re:Almost identical architectures

[Anonymous Coward]

I port linux software to ARM(Tegra K1) and it is 95% of the time as simple as flipping a few compiler flags. The endian mapping for ARM is both big and little.
People should be less concerned about the CPU and more concerned about the graphics API since that is likely going to be vastly different.
As long as Unreal 4 and Unity get ported though, all of that lower level abstraction should be pulled away from the developer anyways.

Re:Almost identical architectures

[tlhIngan]

Switch is competing with the Xbox and PS, among other platforms. At this point there are basically two architectures for gaming: x86 on PC, Xbox, and PS, and ARM on Switch and mobile. Both are very big, very well established architectures with extensive development tools available - neither pose any significant obstacles for developers. Nintendo may have a small edge in this respect, since ARM is the big thing right now.

Nintendo has struggled with third parties in some cases in the past because of some difficulty with supporting their architecture and interface choices. Lack of third party support is certainly not a given with Nintendo, however. Let me remind you that they were masters of third party support in the 8 and 16 bit generations, and continue to completely dominate portable console gaming with no shortage of third party support.

Neither of these issues apply to the switch: completely standard architecture, and almost completely standard interface. There is no technical reason why third parties wouldn't flock to this thing.

Yes, there are many.

3rd parties flocked to 3DS mostly because there's no other option around. Except in Japan, the only other portable system around is mobile. Sony abandoned the Vita so quickly that I don't think they have any chance of recovery with a 3rd portable console anytime soon - developers and consumers would be worried about Sony dropping support pretty much a couple of years later.

But even the Wii U, which Nintendo promised would have large 3rd party support (and had a wide variety of 3rd party games at launch) pretty much fizzled. Which basically came from Nintendo being, well, NIntendo.

And we're seeing Sony becoming same - with the PS4 doing so well, they're getting like what Microsoft was last gen with the 360 - fat, arrogant and lazy.

Try to buy a Nintendo 3DS - it's quite difficult these days as Nintendo is short-shipping them now. This is rather stupid, since retailers are still moving a lot of units - if they'd get any in. It's no surprise the NES Classic is doing the same - this is standard Nintendo behavior. For some reason or other, they always short ship - which resulted in Tengen and the NES10 clones when Nintendo short shipped developers cartridges for the NES. Well, we're seeing it again - maybe retailers can't sell 100+ 3DS units per week per store like the used to, but they can still move 50 units a week per store, but they only get 10 per store, every month. That's really leaving money on the table. You can bet Switch will be same. Nintendo doesn't have to compete with Sony on the portable space, When they did, they didn't short ship - they kept retailers well stocked (it may appear the units sit on the shelf, but they're regularly clearing out and the retailer puts more on the shelf because product that moves makes profit).

We see Sony doing same now - they're doing so well, they are getting fat and lazy. PS+ used to have lots of good games, now it's pretty much a wasteland of maybe a couple of recognizable indie games and crap elsewhere. Same goes with the PS4 Pro - it's a better game machine, but it's lacking a UHD Blu-Ray drive to play UHD Blu-Rays. Something Microsoft added much to many consumer electronics manufacturers dismay. (Prior to the Xbox One S, UHD players sold for $400. After, they're all going for $200 because they have to match Xbox pricing).

Microsoft's games with gold are still somewhat decent - there's a bit of shovelware there, but at least Microsoft seems to be trying. They got fat and lazy and arrogant with the Xbox360, and consumers smacked them down. Just like they did to Sony on the PS3 for being fat and lazy with the PS2. And now, Sony's getting fat and lazy - while the Pro brings some benefits, it's really just a hardware upgrade - like a PS4.5.

Competition is good.

Re:Almost identical architectures

[guises]

Yes, there are many what? You have not given any technical reasons why third parties wouldn't flock to this thing. Sure there are plenty of other potential pitfalls, I would never claim otherwise, but as you say: "3rd parties flocked to 3DS mostly because there's no other option around. Except in Japan, the only other portable system around is mobile." This remains true for the Switch.

Re:Almost identical architectures

[guises]

That is basically what I was saying, yes. Though the PS4/Switch use OpenGL, while the Xbox only uses DirectX... That's a quibble though - the barrier which that poses is nothing compared to previous generations of consoles.

Re:Almost identical architectures

[gravewax]

you want reasons that developers may not flock to this. Will give you lots

Zero userbase and a past set of consoles that have failed for 3rd party
Cost of porting, switch is a very different architecture to the 2 consoles it needs to get ports from which means games cost a lot more for a much smaller userbase
Nintendo userbase has previously shunned 3rd party games, even when they had a large userbase 3rd party ports sold poorly
The console is significantly underpowered compared to the other 2 consoles it needs to get ports from, this will make 3rd parties very cautious
The console needs to support both a mobile display and a scaled up display both running with different levels of power creating additional complexity
lastly Nintendo has had a declining userbase so the uncertainty as to whether there will be enough buyers to justify the development costs.

Re:Almost identical architectures

[guises]

I do not want reasons why developers might shun the Switch. I never asked for any such reasons. I have made no indication at any point that I was seeking these reasons.

This thread started with someone claiming that system architecture would be a barrier to porting between home consoles and the Switch, i pointed out that that really wasn't true and the rest of this conversation has just been fanboi garbage.

Also: your reasons are terrible. I mean, come on - you're citing zero userbase on a product which hasn't been released. There are some reasons why the Switch might not do well, but they're mostly about marketing. There are no technical barriers to its success.

Re:Almost identical architectures

[gravewax]

sorry but that is bullshit. you just whined in the previous post that no technical reasons had been given for developers not flocking to this. secondly the architecture IS MASSIVELY different so it most definitely is true that it will be a significant barrier. the switch compared to the XB1 or Ps4 is about as significantly different as you can get right down to completely different graphics chipset architectures.

Re:Almost identical architectures

[exomondo]

That is basically what I was saying, yes. Though the PS4/Switch use OpenGL, while the Xbox only uses DirectX...

No, neither of them use OpenGL. The PS4 uses a custom PSGL graphics library by Sony and the Switch uses the NVN API, a low-level, low-overhead graphics API that was designed for it by nVidia as it uses the Tegra processor.

Re:Almost identical architectures

[guises]

All right, I should have known consoles would be fully proprietary. Thanks for filling me in.

Well of course.

[Anonymous Coward]

People in Asia won't touch an American product with a 10 foot pole. They've got too much national pride.

Meanwhile people in America are too busy making fun of America to even think about their future.

Re:Well of course.

[Anonymous Coward]

iPhone has 66% market share in Japan.

Re:Well of course.

[Anonymous Coward]

So they like Chinese smartphones. Imagine that.

Re:Well of course.

[Anonymous Coward]

Japanese aren't as into first person shooters, that's the XBox's main appeal. Most of the exclusive Playstation and Nintendo games are developed by Japanese companies while for the XBox, it's US based companies. Also, Windows is even more widely used over there than it is in the US.

Re: Well of course.

[Anonymous Coward]

Great. Tons of chinese apps to run on a chinese made phone.

Switch is coming!

[Anonymous Coward]

a minor afterthought that can be safely ignored for everything but niche games.

Not going to happen. When it comes to home consoles, Nintendo already has the "minor afterthought that can be safely ignored for everything but niche games" market sewn up.

PS4 is for LUDDITES.

[Anonymous Coward]

Appbox One only has appy app apps on Appdows 10! LUDDITE PS4 only has LUDDITE games!

Apps!

Pro

[malditaenvidia]

How is the PS4 Pro doing, though? From what I've heard it hasn't been selling all that well.

Scorpio

[sexconker]

Scorpio isn't an upgraded Xbox One. The Xbox One S is. Just like the PS4 Pro is the upgraded PS4.

Scorpio is a new generation, though it'll likely be backward compatible, feature a similar UI/OS (Windows 10 everywhere...), tie into the same backend services, etc.

Outside of Nintendo, the days of console generations being completely new shit are likely dead. Development costs are too high, and established libraries (especially digital) are a huge consideration. Both Sony and MS are using AMD's shit for CPU and GPU. Staying on x86 (and AMD) makes perfect sense because it lowers cost, lets them maintain compatibility with older titles so people can keep their existing games as they upgrade to later versions of the console, and makes development easier. It also helps make maintaining the back end services (PSN and Xbox Live) simpler if you don't radically change things every few years.

Scorpio will almost certainly have games that ONLY play on Scorpio, but the PS4 Pro games are (for now, at least) all guaranteed to run on the PS4. This is what demarcates a generational divide. MS is trying hard to get marketshare back, and they had a great opportunity when Sony fucked up with the PS4 Pro (no 4K BluRay support), but they can't seem to get any traction. This is why they're making the first move with Scorpio.

Nintendo, as always, does its own thing on its own time.

Re:Scorpio

[HalAtWork]

Scorpio software will be hamstrung by mandatory compatibility with Xbox One hardware. http://www.gamespot.com/articl...

Scorpio's higher price will slow adoption, developers will still focus on the much larger install base of PS4 and Xbox One onwers, meaning developer resources won't be put into significant enhancements for Scorpio compatible games. http://www.gamespot.com/articl...

Scorpio's hardware will be outclassed 2 years later when the PS5 is released. http://www.gamezone.com/news/a...

It will be an expensive flop, not a good way to start a new generation.

Re:Scorpio

[Anonymous Coward]

Scorpio software will be hamstrung by mandatory compatibility with Xbox One hardware. http://www.gamespot.com/articl...

Except for VR experiences, the PS4 Pro has the same issue except it's worse because the VR titles have to work on the PS4 as well which is much less powerful. Since the architectures are effectively PC ones writing separate shaders and scaling back polygon counts is not a big deal, it's done in PC gaming to target millions of different configurations, developers can easily handle targeting just 2 configurations.

Scorpio's higher price will slow adoption, developers will still focus on the much larger install base of PS4 and Xbox One onwers, meaning developer resources won't be put into significant enhancements for Scorpio compatible games. http://www.gamespot.com/articl...

The same as the PS4 Pro, except again that VR is exempt so the VR experiences on Scorpio will be far better than PSVR. Of course we have PSVR right now as opposed to XBox not having it at all at the moment so that is advantageous.

Scorpio's hardware will be outclassed 2 years later when the PS5 is released. http://www.gamezone.com/news/a...

Can you give us the specifications of both? No of course you cant so it is exactly as likely that some Scorpio Mk2 will be released at the same time as the PS5 and outclass it, or maybe that wont happen, there is no evidence either way so if you are making any kind of blanket statement like yours then you are an idiot.

It will be an expensive flop, not a good way to start a new generation.

A slashdot prediction of the future success of technology? On that alone I would be betting on Scorpio.

Re:Scorpio

[sexconker]

I seriously doubt MS is going to hamstring Scorpio and treat it as the same generation as the Xbox One. We haven't even had an official reveal of it yet. All we've had is the announcement that it's coming. We'll get the full scoop this E3.

Re:Scorpio

[Anonymous Coward]

I seriously doubt MS is going to hamstring Scorpio and treat it as the same generation as the Xbox One. We haven't even had an official reveal of it yet. All we've had is the announcement that it's coming. We'll get the full scoop this E3.

And yet somehow he knows that the also unannounced PS5 will outclass it...go figure.

Re:Scorpio

[HalAtWork]

VR experiences don't matter, wonder why you never see reviews of VR games? VR has been out for about a year but there are no amazing games, and it's not due to lack of hardware power, games are no better on PC (which already has 10 tflops w/Nvidia 1080 vs Scorpio's 6)

Specs for Scorpio? As I said, 6 tflops, going to be AMD GPU due to compatibility with Xbox One, the rest doesn't really matter because it's already outclassed on PC and certainly will be with PS5, why wouldn't a system 2 years later have better tech? http://gamingbolt.com/xbox-one...

Sure get a Scorpio, your money

Re:Scorpio

[HalAtWork]

Did you look at the link? This is confirmed. http://www.gamespot.com/articl...

Why would MS cut off their legs? Scorpio will have an install base of 0 until end of 2017, PS4 and Xbox One will have at least 100m combined by then (about 80m now combined). Why would devs give a system that will have such a low ownership for such a short time before the next real generation the money and resources required to significantly improve games?

Re:Scorpio

[HalAtWork]

Please go figure.
- Scorpio is 6 tflops
- Nvidia 1080 already outclasses it at 10 tflops (yet Scorpio is touting VR as a pro? It will be stripped down VR like we have now)
- 2 years is a long time in tech
- even low end GPU in 2 years will easily outclass high end options now
- PS5 expected 2 years after Scorpio, will break hardware/software lineage where Scorpio needs to maintain hardware/software compatibility

Why would it not outclass Scorpio is a better question

Re:Scorpio

[Anonymous Coward]

VR experiences don't matter, wonder why you never see reviews of VR games?

Because game reviewers are morons? I dont know and frankly I dont care what game reviewers say, they can say whatever they like and it doesnt change the fact that Doom on the HTC Vive is amazing for example. You can find plenty of people talking up their experiences of it.

VR has been out for about a year but there are no amazing games, and it's not due to lack of hardware power

Games are not about how much you can push through the GPU, nobody cares who has got the biggest GPU (maybe you are just insecure about your small "GPU" :P...it's ok, it just a joke, dont get upset), just that it is enough to produce the vision of the designer.

Specs for Scorpio? As I said, 6 tflops, going to be AMD GPU due to compatibility with Xbox One, the rest doesn't really matter because it's already outclassed on PC and certainly will be with PS5, why wouldn't a system 2 years later have better tech?

...and if history is any guide a year after that the Project Scorpio S will be released and eclipse the PS5 so gives a crap? If you are waiting 2 years then why does it even matter, just buy whatever is best at the time.

Being outclassed in raw performance by the PC is a completely pointless metric because games need generic implementations to support millions of PC configurations which includes the capabilities of the CPU and GPU, the speeds of the CPU and GPU, the amount of system RAM, the speed of the system RAM, the amount of GPU memory, the speed of the GPU memory, the amount of CPU cache, the speed of the CPU cache, the speed of the system interconnects, etc, etc, etc. Which means the abstraction results is FAR FAR less efficiency of PC games vs consoles, obviously because the system is has so many unknowns that you can't possibly optimize for all those configurations.

Fanboys dont understand any of this of course.

Re:Scorpio

[Anonymous Coward]

VR is fun for a couple of hours and then there's not much reason right now to keep coming back and titles are few and far between. It hasn't found its Mario 64 moment yet.

Anyway no matter what it's guaranteed PS5 will definitely exceed the by then 2 year old Scorpio.

Re:Scorpio

[Anonymous Coward]

Anyway no matter what it's guaranteed PS5 will definitely exceed the by then 2 year old Scorpio.

...and then a year later that will be supersceded by some Project Scorpio S or whatever and then 2 years after that it will be supersceded by the PS5 Pro and on and on. Chasing specs on the basis that at an arbitrary point in time in the future X will be the most powerful is pointless, why would that even matter?

Re:Scorpio

[Anonymous Coward]

- 2 years is a long time in tech

So by then project scorpio will be old news and its successor will be on the horizon if not already available. No point buying the PS4 Pro because less than 2 years after its release the more powerful project scorpio will arrive, but no point buying that because 2 years later the more powerful PS5 will be available, but no point buying that because shortly after that the more powerful XBox Two will be available, so buy that? no don't be stupid because in just 2 long years the more powerful PS5 Pro will be available, but of course you wouldnt want to buy that when the more powerful next XBox is right around the corner...

Re:Scorpio

[Anonymous Coward]

^ This. VR is really hard to adapt to for more than 15 min at a time for many people. The experience is great initially, but it's a lot more complicated and expensive. It's really only useful for first person type games, but not everyone wants to play games like that or enough to justify the extra cost of buying VR gear.

Re:Scorpio

[Billly Gates]

First off the specs leaked show it is a AMD Ryzen with an RX 480 underclocked to match performance with an RX 470. No for the parent this is not bad hardware anymore.

So no developers do not have to target PS 4 pro or Scorpio? They just simply make the same games for the xbox One which are binary compatible and up the graphic details and resolution when a higher end console is available. This is quite smart actually.

Also MS is making with game mode DirectX xbox compability (which oddly is very different from PC directX) which means a larger marketshare to port the console to the PC market. The gaming PC market is exploding! Yes, last decade gaming companies obsessed with piracy tried to kill the PC, but the demand is growing if you look at marketshare for the PC again due to the large millennials and lower costs of decent GPUs and CPUs now on the market.

Re:Scorpio

[Anonymous Coward]

What matters is the cutoff point where compatibility with a generation's software ends, and then how well new hardware is suited to handling a 6 year stint on a new generation of software. The Scorpio is either too late for this gen to accrue a proper library given the target deveopment spec of the PS4 due to its alice of the pie, or too early for the next gen and won't keep up with the PS5 just like X1 vs PS4, and will suffer because of it.

Re:Scorpio

[Anonymous Coward]

Exactly, no point in buying these half step consoles. Look at PS4 Pro improvements. Despite double the processing power of the PS4 there are only slight enhancements to the games because developers are focused on making the PS4 and Xbox One version run great, and then enhancements are an afterthought if they have it in the budget at all.

Scorpio will go through the same thing and by the time the PS5 comes out and cuts off PS4 compatibility on the SDK side, Scorpio will be underpowered for what's coming.

The point of buying the PS5 will be to have compatibility with new software that won't work on the PS4. PS4 Pro and Scorpio I will happily skip over though.

Re:Scorpio

[Anonymous Coward]

Even then, first person games where you teleport around and do not actually walk around.

Re:Scorpio

[ausekilis]

This got me thinking about a conversation I had with a colleague last week.

We were talking about emulation and how the underlying design of systems, which chips are used, makes that emulation difficult. It seems consoles at first had off-the-shelf chips that were put to a new and interesting use (NES/Master System era). As the generations went on, the chips got more specialized (XB360/PS3 era), but did not veer too far from the established general-purpose PC architectures. After all, the xbox series used custom x86 chips with a special instruction set and odd behaviors, and the PS3 used a powerPC derived chip. Now both the PS4 and XB1 are using x86, a more general-purpose chip.

Which direction are the going to go next? Are we going to see a media/gaming optimized x86 series of chips? Or are we going the way of Sony/MS branded Steam machines with custom OS's?

I think the former would be a more fun experiment. X86 has that whole backwards compatibility and bolted-on instructions baggage it's been carrying around for years. Maybe between Sony, MS and (maybe) Nintendo, they can prune out some of the unnecessary crap and make the chips feel faster.

Re:Scorpio

[HalAtWork]

Exactly, upping the resolution, fps, and anti aliasing are minor upgrades on basically the same game. Scorpio games won't really poush the hardware with more detailed models, physics, AI, bigger worlds with more going on, etc... and they won't be able to either since Scorpio gamers will be matched up with Xbox One gamers online, so they can't even have different physics and AI models going on if they wanted to.

Haha DirectX and Compatible are funny to have in the same sentence!! Right now on PC they can't even get games to work well across Nvidia and AMD GPUs, games that release perfectly playable on console launch in a broken state on PC, including MS's own games using UWP like Forza Horizon 3 and Quantum Break! Nothing is going to change here.

Re:Scorpio

[HalAtWork]

It's an interesting idea but emulation of GPUs is a significant undertaking, and games on PS4 and Xbox One go low level enough to the hardware that it won't be possible to just trap high level API calls and execute those on newer hardware. The Xbox One especially has a unique architecture with eSRAM management that will need to be handled somehow, as games use many tricks to acheive performance on that platform.

Direct hardware lineage makes more sense, but then console manufacturers are tying themselves to one vendor, and one basic architecture that can't vary as much as might make sense in the future. The beauty of a new generation of consoles is that everything is out the window and we get to make the best decisions for hardware from scratch with no compromises.

Otherwise we might just end up with what we have on PC with a lot of cruft in for the sake of compatibility and lower return on hardware upgrades.

Re:Scorpio

[SuiteSisterMary]

The Xbox used virtually bog-standard Celeron CPU and NVidia GPU; the only real special thing was a unifed memory architecture. The 360 was, IIRC, three PowerPC chips and a fairly standard AMD CPU.

The PS2, on the other hand, was custom silicon; the Emotion Engine and what not, while the PS3 was Cell architecture.

The PS1 and Dreamcast were both also fairly standard components, while the N64 was SGI custom.

Consoles are absolutely migrating towards being PCs with purpose-built cases, and backwards compatibility is becoming an expectation rather than a bonus, especially in this digital era, when you can get old titles from digital marketplaces.

Re:Scorpio

[Anonymous Coward]

Compatibility will hapoen under UWP, but UWP will bring with it some restrictions, it's MS's push for a trusted computing platform with full accountability for DRM. As a result it will be a walled garden that sits on top of Windows. Win32 will stagnate as up to date APIs become inaccessible behind UWP, with the reasoning that access could fracture trust. Not looking good for PCs as an open platform

Re:Scorpio

[exomondo]

Who cares about any of that? The only people interested in spec wars are braindead fanboys. The logical thing is to buy the console that has the games you want to play at the time regardless of its specifications. And quite frankly with consoles only costing a few hundred dollars anybody seriously into gaming that wants to play exclusives on both consoles will just buy both consoles since you can have both of them and some change for the price of a single highend PC gaming GPU.

Re:Scorpio

[Anonymous Coward]

Exactly why I am sticking with the base PS4 and just waiting until the next generation to get any new console

Avoiding the PC though because there seems to be as more troubleshooting than actual playing on launch games, it's like playing russian roulette with games instead of a hassle free plug and play experience, I have a budget on my relaxing time and don't want to spend it fixing anything if I can help it and so far have not been let down on consoles (to that end I avoid Ubisoft and Bethesda games until GotY editions come out)

Re:Scorpio

[Anonymous Coward]

What matters is the cutoff point where compatibility with a generation's software ends, and then how well new hardware is suited to handling a 6 year stint on a new generation of software. The Scorpio is either too late for this gen to accrue a proper library given the target deveopment spec of the PS4 due to its alice of the pie, or too early for the next gen and won't keep up with the PS5 just like X1 vs PS4, and will suffer because of it.

When you talk about "current gen" and "next gen" what exactly is defining that? It seems you just define that to mean "whatever Sony is doing".

If all you're doing is adding more processing power then you just end up with the same situation we had with the PS3/360 and PS4/One where the latter pair were not effectively utilized broadly for a long time because developers wanted to capitalize on the existing installbase and just having better graphics doesn't make a better game. What killer feature do you think the PS5 is going to have that developers won't hamstring in order to capitalize on existing PS4, PS4 Pro, One and Scorpio install base?

It's probably a smart thing to wait a year and release a more powerful console while developers are still supporting the broadly installed consoles.

Re:Scorpio

[Anonymous Coward]

Why would devs give a system that will have such a low ownership for such a short time before the next real generation the money and resources required to significantly improve games?

Because they are now on a PC-based architecture, there is no massive re-investment in training and learning a new system like there was with the horrible mess of the Cell processor, more to the point Microsoft has the advantage of leveraging the skills of PC game developers' experience with DirectX whereas Sony comes out with a new version of their custom API that nobody else uses for each generation of their hardware. So what money and resources are required here? Improving shader quality, running at a higher resolution, and generating higher polygon meshes from the original artist models isn't a significant investment now that we are using a PC architecture so there are no problems with system code optimization because there is no longer a change in architecture like the PowerPC and Cell systems that came before the current ones.

Scorpio makes no sense

[HalAtWork]

Scorpio, an upgraded Xbox One, is said to have about 3x the power of the PS4, 1.5x the power of the PS4 Pro. And 5x the power of the Xbox One, which it has to be compatible with. Scorpio software must run adequately on the Xbox One despite the huge power gap.
This is a premium system with a rumored high price tag, launching about 2 years before the expected next generation of consoles debuts.

Having to compete with the installed base of PS4 and Xbox One so late in the game, developer focus is expected to reflect this, leading to only minor updates to games such as higher resolution, framerate, and anti aliasing. Not expected are more complex models, more detailed worlds, significant shader and texture differences, increased complexity in physics models, or AI differences (especially since Scorpio gamers are expected to match with Xbox One gamers online, and the games are supposed to remain basically the same, only superficially different).

Bottom line is they expect knowledgable gamers (casuals won't care about this) to pay a high price for minor cosmetic differences not long before newer powerful machines come out that aren't resterained by compatibility with older less powerful systems. It doesn't seem like this will do much to bolster the Xbox brand, and may even make adopters upset. Probably only Microsoft's own titles will bother to put the resources behind making any significant improvements to games, especially considering how much games already cost, and how much testing will have to be done specifically for a significantly different version, and that's only 2-3 games a year?

It does not seem like Scorpio will help MS....

Re:Scorpio makes no sense

[Anonymous Coward]

Pachter believes consoles games won't require a traditional console, but instead a simple CPU/GPU in your house that's connected to your television (think of an iPhone or a Fire TV box from Amazon).

Is so obviously not going to happen I'm surprised you even bring up the article.

And 5x more powerful? I dunnow, if the price is just a little more, why not? Plenty people buy/upgrade to 4K TVs, even though it makes no real difference, just because the price is basically the same as a 1080p TV.

Re:Scorpio makes no sense

[CrashNBrn]

Unless it's an AMD Steam Box... with upgradeable parts.

Re:Scorpio makes no sense

[DrXym]

Scorpio, an upgraded Xbox One, is said to have about 3x the power of the PS4, 1.5x the power of the PS4 Pro. And 5x the power of the Xbox One, which it has to be compatible with. Scorpio software must run adequately on the Xbox One despite the huge power gap.

It wouldn't be the first time Microsoft have gimped specs between announcement and release. When Kinect was called Project Natal it had an onboard CPU/DSP that could do motion tracking of 4 people independently, track fingers & hands, even facial expressions. Then they decided to do all the processing on the 360 instead and the thing could barely recognize a person flailing their arms in an exaggerated bowling motion.

Microsoft might take a look at the PS4 Pro and decide there is no reason to exceed it in performance in any substantial way. If games have to remain backwards compatible with the XB1, then perhaps there are limits on what they could even do with the added compute power even if it were available to them.

Re:Scorpio makes no sense

[Anonymous Coward]

Because even at 5x more power, game upgrades will be minimal. Even on PC where gamers are willing to spend extra to get SLI 1080s, there is little benefit other than higher fps and better anti aliasing, because devs focus on the lowest common denominator so the majority of people can buy and enjoy a game. Hardware options like HDR were supported in high end cards since 2014 but PC games supporting it are only trickling out now as it becomes a feature that the majority of gamers have, not just the elite few that buy expensive hardware.

Devs will focus on where the money is, PS4 and Xbox One, tailoring the experience to those with a giant installed base compared to what the Scorpio will be able to scrap up. Like PS4 Pro there will only be small rewards.

Largest yet deployment of FreeBSD?

[Brian Feldman]

I am quite curious if Sony is now the leading manufacturer of consumer electronics powered by FreeBSD. The only other manufacturer that I think may be in the running would be Panasonic, with FreeBSD as the basis for their televisions.

Re:Largest yet deployment of FreeBSD?

[ilguido]

Sony's TV sets, bluray players, smartphones (Android) etc. are all based on Linux. Their PS3/4 OS is said to be based on FreeBSD and NetBSD, but it is highly modified and integrated with proprietary components, so it may be a FreeBSD system as MacOSX is.

Microsoft Hurt Themselves Early

[Thyamine]

I have generally purchased an Xbox first, then later on purchased the same gen PlayStation. But this last go around, MS really shot themselves in the foot when they announced all their 'features' that were going to limit owners and limit how/where games could be used. Then either because they were never going to do that, or they just seized on the moment, Sony said 'we aren't doing that' and basically many of us rushed to buy a PS4 instead of an Xbox One. I still haven't really seriously looked at getting one.

Re:Microsoft Hurt Themselves Early

[Anonymous Coward]

Sony definitely "seized on the moment" with their plans of "We're not doing what Microsoft is doing". It's clear that they were initially going to do the same thing until Microsoft was booed off the stage that year - 3 months after launch Sony admitted to changing tactics on the spur of the moment.

But regardless, the fact that they were willing to change gears that quickly let people decide to take up the PS4 for this gen.

I still remember the xbox fanboys complaining about microsoft eventually scrapping their plans over the loss of the ability to "lend their games" to their friends. They didn't seem to realize that lending was nothing more than allowing their friend access to a timed demo once. (Couldn't re-lend the same game for instance, and likely the person couldn't receive the same game from different friends.)

For me, I wasn't planning on picking up the Xbone regardless since I never really took to the original or 360 that much. Got some games for both but most exclusives of the library isn't my cup of tea. But when microsoft attempted to pull that crap they did at launch, I decided to boycott the line entirely - to the point if someone gave me xbone, I'd pawn it off and buy some ps4 games with it.

Re:Microsoft Hurt Themselves Early

[DrXym]

It depends on what you mean by "limit". Having to insert a disc to play a game is a pain in the ass, so if the XB1 did uniquely register a game to a console to avoid that hassle then it has obvious advantages.

Where they fucked up is they didn't provide a way for people to "de-register" the disc so they could sell or loan it. The obvious way to do it would be to say that whoever owns the disk can play it and other images are invalid. If you try to play from the image the console will check online to see if the disk has been used elsewhere and ask you to insert the disk if it had. It could also occasionally challenge the user to insert the disc, more frequently based on usage if it thought the disc was a rental copy.

But instead they junked the feature. So now the console has a cache of the game but you always need to insert the disc to play it.

VGChartz

[Anonymous Coward]

VGChartz is a useful site if you're interested in console and game sales.

Re:VGChartz

[antek9]

If only their numbers were up to date...

MS was caught off guard

[HalAtWork]

They did not know the PS4 Pro was coming, so they had nothing ready in time... but had to counter with something. They couldn't say they weren't doing anything with sales already languishing.

Re:MS was caught off guard

[Anonymous Coward]

They did not know the PS4 Pro was coming

Of course they did. The Project Scorpio official announcement came just shortly after the Project Neo rumors first surfaced. The PS4 Pro is pretty pointless anyway: it is hamstrung by the need to be compatible with PS4.

Kinect should have died with the Xbox 360

[Anonymous Coward]

Between paying for Xbox Live to use online multiplayer and having that worthless Kinect my Xbox One is mostly a dust collector these days. I'd trade it for a PS4 in a heartbeat.

It was a bad start...

[Grim Beefer]

Microsoft unrolled a plan that put major roadblocks up to trading or selling used games. Consumers told them to go fuck themselves. ...fast forward...

Sales are 2 to 1, in favor of the company that didn't try to pull this blatantly anti-consumerist bullshit, and rightfully so. Apparently, plenty of idiots were swayed by their last minute reversals.

No surprise. The XBone launch was a disaster.

[Qbertino]

The XBone launch was a disaster. They had to backpedal on just about any announcement made, having sold countless lock-ins as "features", type A Microsoft style. It's only for about a year now that people can trust the XBone to be reasonably fair to the consumer in most areas. And this is the stage of a console lifetime were those interested will go and ask around which console was better marketshare and is likely to have more people playing on- and offline. Hence even potential XBone buyers are craning their necks for the PS4s offerings.

I own the last iteration of the Xbox 360 and a stack of games, most of which would run on the XBone, and even I am reluctant of the XBone, due to the lock-in and lack of convenience in this generations consoles.

Consoles are too much of an online service extension and not really that convenient anymore these days. Pop in a disk, run a game used to be. Now it's download the update of Mafia 3 for 4 days flat until you can actually play. People who have no problem with that get a PC. XBones+Kinect "allways-on" non-sense and similar stuff was just the straw that broke the camels back, vis-a-vis the (slightly) less invasive and pretentious Sony and their PS4.

Not surprising really

[DrXym]

Microsoft fucked up the launch of the XBox One. It was overpriced, bundled with a peripheral that nobody wanted, was less powerful and looked uglier. That gave the PS4 the lead and it's been widening since.

Whether that continues when "Project Scorpio" turns up in some form remains to be seen. The PS4 Pro and PSVR didn't exactly take the world by storm so perhaps there is an opportunity for Microsoft to seize or maybe the same pit to fall into.

TV TV TV TV

[Anonymous Coward]

TV TV... tv TV..tv Tv MTV

when your console sucks and your presentation is about its television capabilities to the point you mention "tv" like 50 times, people that already have a television might not buy it

lololol

https://www.youtube.com/watch?v=KbWgUO-Rqcw

All I know is...

[neo-mkrey]

Forza Horizon 3 and Gears of War 4 look stunning on my 4K 70" Samsung when played on my XBOX 1S.

Scorpio still won't help MS

[HalAtWork]

As you pointed out the PS4 has a giant installed base, with the X1 costing more and selling less, despite promises of enhanced games for free using the cloud to power some physics and graphical elements in games.

Again the Scorpio will be more expensive, and despite promising more power, it will not be used as developers target the much larger installed PS4 and Xbox One user base. Why should developers spend money and resources enhancing games for a tiny market segment? They don't really do that now with the PS4 Pro.

We'll have to wait until the next generation and a break in compatibility to get games that are significantly different from what we have now. No matter how much punch the Scorpio packs, it will only get minor enhancements like resolution and frame rate. This may well anger gamers who spent a lot of money on a new box where MS is marketing the sheer hardware performance that doesn't manifest in actual practice.

My guess is only MS's own games will take any real advantage of Scorpio's hardware, while third parties continue to target the base PS4 and Xbox One. Just wait until the next generation, these half step consoles are expensive and offer little benefit in practice. Their advantages are paper-only.

DirectX doesn't seem to help here anyway

[HalAtWork]

Curiously though DirectX does not seem to help with compatibility. Games on Windows still have major issues with different hardware, and interestingly on the GPU side we have Nvidia and AMD modifying their drivers to aid with game compatibility!! It seems like DirectX barely does anything to help with the situation... if at all.

Re:DirectX doesn't seem to help here anyway

[exomondo]

Curiously though DirectX does not seem to help with compatibility. Games on Windows still have major issues with different hardware

I'm not quite sure what you mean specifically on how this relates to DirectX (or what you are implying is fundamentally different between it and say OpenGL), it in fact does help with compatibility in that you aren't coding for a specific GPU but for an abstraction where the implementation is provided by the driver. This is the same concept with both OpenGL and DirectX.

and interestingly on the GPU side we have Nvidia and AMD modifying their drivers to aid with game compatibility!!

Of course you do, that's where the implementation of the interface is. The driver provides the interface between the application and the hardware, now in the case of OpenGL and DirectX<=11 there is a lot of work done in the driver that the application developer has no control over (important things like memory management) so the driver writers then often include application-specific optimizations for popular programs. This is where DirectX12 and Vulkan come in, they reduce the amount of resource management that the driver has to do and the application becomes responsible for it, and of course there are frameworks starting to emerge that provide generic implementations for applications that do not need that level of optimization.

It seems like DirectX barely does anything to help with the situation... if at all.

Actually it seems like you fundamentally misunderstand what DirectX (and OpenGL for that matter) is and what they are designed to do. You see back before we had these APIs the issue was that you needed to write your rendering code specifically for every GPU (Glide for 3Dfx, S3d for the Virge, MSI for Matrox, etc) but DirectX and OpenGL meant that the code could be written for those APIs and the manufacturers just needed to provide an implementation in their driver so then that same code could run on any GPU without having to explicitly support it.

Re:DirectX doesn't seem to help here anyway

[HalAtWork]

"I'm not quite sure what you mean specifically on how this relates to DirectX (or what you are implying is fundamentally different between it and say OpenGL), it in fact does help with compatibility in that you aren't coding for a specific GPU"

Then why are there often fixes on the game dev's end to fix issues with a specific graphics card? Shouldn't the patch notes say "fixed improper DirectX call" instead of say "fixed issue with lighting on Radeon 480" or "added support for Intel HD"... If the game is just asking DirectX to interpret calls, why should it have to worry about specific chipsets?

"Of course you do, that's where the implementation of the interface is. The driver provides the interface between the application and the hardware, "

But DX/OGL are between the driver and the user application, the software asks DX to throw something up on the screen, and DX sends those calls to the driver. All the driver should have to worry about is properly handling DirectX requests, not about specific functions one game is trying to perform. Why do driver patch notes then say "improved compatibility with game X" instead of "improved compatibility wit DirectX"... why would only one game benefit instead of all DirectX compatible software? Why is the driver tested only against specific games and not some sort of reference DirectX implementation?

"DirectX and OpenGL meant that the code could be written for those APIs and the manufacturers just needed to provide an implementation in their driver"

This is basically my point as well! Yet as you figure, "of course" drivers are updated for specific games and also somehow games are implementing discrete fixes for specific GPUs instead of just for the broad implementation of DirectX... This is where I start getting confused and my understanding falls apart...

Thanks for addressing my post btw and any info would be helpful, there are things here that just don't seem to align properly in my understanding...

Re:DirectX doesn't seem to help here anyway

[exomondo]

Then why are there often fixes on the game dev's end to fix issues with a specific graphics card? Shouldn't the patch notes say "fixed improper DirectX call" instead of say "fixed issue with lighting on Radeon 480" or "added support for Intel HD"... If the game is just asking DirectX to interpret calls, why should it have to worry about specific chipsets?

It doesn't, the implementation is in the driver, just like OpenGL. What "improper DirectX call" are you talking about? I think you need to be more specific about what the issue is you're referring to to be able to explain but if there is a fix for a specific GPU then it is in the driver implementation of OpenGL/DirectX for that architecture. Sometimes on the application side there are specific extensions that might be used to target specific GPUs more efficiently or an effect might be re-written with in a less efficient way to support GPUs that don't have the featureset to do it efficiently but are still powerful enough to at least support it. For example there are some great features in nVidia's Pascal architecture that I could use for effects but they wouldn't work on earlier GPUs and DirectX, Vulkan, OpenGL, etc do not abstract this away.

But DX/OGL are between the driver and the user application, the software asks DX to throw something up on the screen, and DX sends those calls to the driver. All the driver should have to worry about is properly handling DirectX requests, not about specific functions one game is trying to perform.

No, that's wrong. The DX/OGL libs that application developers link to are the functions that call into the implementation that resides in the driver. This is why getting support for newer versions of DirectX or OpenGL is a matter of upgrading your driver, not just upgrading the version of the application libs.

Why do driver patch notes then say "improved compatibility with game X" instead of "improved compatibility wit DirectX"...

I don't think they do, it's usually specific things like the way shader code is compiled to address compiler bugs or things specific to the architecture. The way the patch notes are written doesn't really matter, what matters is the message that X game will now work better or a bug has been resolved.

why would only one game benefit instead of all DirectX compatible software?

Because improvements for a specific game can come from resolving a shader compiler bug, shader optimizations or things like memory management optimizations for particular games dictated by the specific executable.

Why is the driver tested only against specific games and not some sort of reference DirectX implementation?

It is tested against a reference, but that doesn't mean it is completely optimized and bugfree in every possible combination of use case. Just the same as any programming language, API and even CPUs often have errata documentation.

This is basically my point as well! Yet as you figure, "of course" drivers are updated for specific games and also somehow games are implementing discrete fixes for specific GPUs instead of just for the broad implementation of DirectX...

Well it is like any language, you don't just write a compiler and be done with it, it is evolutionary. As developers use the features of the API and language (HLSL or GLSL) in different ways opportunities for optimizations (and of course bugs) present themselves and can then be addressed in updates.

Re:DirectX doesn't seem to help here anyway

[HalAtWork]

Again thanks for your time....

I see, so the driver itself mplements the actual API function's behavior based on what *should* happen in a reference DirectX implementation?

I guess I previously imagined that the driver opened a very basic interface to the hardware, and that somehow a DirectX library was responsible for translating an API call into some sort of low level code for the driver to run on bare hardware (sort of how binaries are x86 CISC bytecode, and those x86 CISC instructions are translated into an even lower level instruction set internally and then finally executed in whatever way the silicon actually works inside)

But most of the heavy lifting of translating the API call into machine instructions is actually done in the driver directly? So DirectX is just a spec... I guess I was confused because there is software called "DirectX" which installs libraries to the system, now I'm wondering what that huge install is, just basic SDK framework stuff like headers?

Thanks, hopefully I am catching on

Re:DirectX doesn't seem to help here anyway

[exomondo]

I see, so the driver itself mplements the actual API function's behavior based on what *should* happen in a reference DirectX implementation?

Basically yes.

I guess I previously imagined that the driver opened a very basic interface to the hardware, and that somehow a DirectX library was responsible for translating an API call into some sort of low level code for the driver to run on bare hardware

The reason this isn't the case is that the low level code is GPU-specific (or at least specific to each GPU architecture) so that gets implemented by the driver vendor instead.

But most of the heavy lifting of translating the API call into machine instructions is actually done in the driver directly? So DirectX is just a spec... I guess I was confused because there is software called "DirectX" which installs libraries to the system, now I'm wondering what that huge install is, just basic SDK framework stuff like headers?

Well DirectX is more than just Direct3D, but the Direct3D component includes the libraries with the entry points that are bound to the driver implementation, helper libraries and functions for doing things like specifying state, building command buffers, math libraries, shader classes, etc. But the SDK itself also includes things like the debug libraries, debugging tools, documentation, samples, etc for 32 and 64bit as well and possibly the C# libraries for interfacing with it as well (the equivalent of something like OpenTK for C# + OpenGL). The Vulkan SDK is fairly similar, in the order of about 130MB but of course this is the first version of Vulkan, the DX SDK also includes a lot of stuff for backwards compatibility.

Re:DirectX doesn't seem to help here anyway

[HalAtWork]

Thank you very much for your time in explaining all this :)

Re:DirectX doesn't seem to help here anyway

[exomondo]

No problem ;)

Multiplayer snowball

[DarthVain]

The biggest factor I think is the whole multiplayer snowball effect. As once a particular system gets a bit of an advantage for one reason or another, and people start migrating to it, the effect becomes magnified the longer it goes on. As you say probably one of the largest decisions to buy a particular system outside of exclusive games is about what system all your friends are on. Once they move, you move, the more, etc... To the point where one system starts crushing the other.

Case in point most of my friends used to be Xbox, now many of them have moved on to the PS4.

Devil's advocacy

[tepples]

I tend toward PC in principle, but sometimes I argue the other side to help keep both sides honest and help bring out both sides' strongest arguments.

Most families still need or want a computer at home for reasons besides gaming (e.g. internet, word processing, tracking finances, online banking, digital storage, remote connections to workplaces).

First, these non-gaming applications can be done with a cheap eight-year-old PC with a Core 2 Duo and Intel integrated graphics. I'm told just dropping a video card into a PC with a CPU that old isn't enough to run AAA games from the present generation (2014 and later), which would quickly become CPU-bound. Second, these can be done with a laptop, and I've seen no evidence that people routinely upgrade a laptop with a separately purchased MXM video card. Third, a console can be used while someone else is using the family PC.

The price of games should be factored into the cost of a game system and games are cheaper on PC through digital distributors like Steam, Origin, etc., which over time off-sets the initial cost of the PC.

First, though Steam has sales. PlayStation Store also has sales. Second, console games have historically been more likely than PC games to support same-screen multiplayer with two to four gamepads, and if you have more than one gamer in the house, one copy of a $60 game that supports multiple gamepads is cheaper than three copies of a $30 game that requires a separate copy per player. Third, if everybody were to wait for the sale instead of buying in release month at full price, publishers would have no money to continue to fund development of high-production-value games.

If you want to play games online (which many people do) you have to add the life-time cost of an online subscription to a console.

PlayStation Plus and Xbox Live Gold cost $60 per year. But in addition to online matchmaking, this includes rental of a rotating selection of games (PlayStation Plus Free Games and Games with Gold respectively). What's the analogous way to try PC games?

You don't need an expensive PC to play games. A $500 PC (which is comparable to a new PS4+accessories) will play ~98% of the PC games.

Or you could go for a pre-owned PlayStation 4 console with a 500 GB HDD, which costs $280 (source). Which accessories were you including in the price?

Re:Devil's advocacy

[elrous0]

First, though Steam has sales. PlayStation Store also has sales. Second, console games have historically been more likely than PC games to support same-screen multiplayer with two to four gamepads, and if you have more than one gamer in the house, one copy of a $60 game that supports multiple gamepads is cheaper than three copies of a $30 game that requires a separate copy per player. Third, if everybody were to wait for the sale instead of buying in release month at full price, publishers would have no money to continue to fund development of high-production-value games.

Not to mention the fact that, unlike PC games, you can often find dirt-cheap used physical copies of console games on Amazon and other sites. A while back I bought a used copy of Battlefront for PS4 for $10 on Amazon. Good luck getting it for sale on PC for that price.

And even if you could, big Steam sales only come a few times a year and many devlopers don't even support Steam (including big names like EA and Ubisoft). By contrast, cheap used copies of console games can be found anytime.

Re:Devil's advocacy

[Ranbot]

I tend toward PC in principle, but sometimes I argue the other side to help keep both sides honest and help bring out both sides' strongest arguments.

OK

First, these non-gaming applications can be done with a cheap eight-year-old PC with a Core 2 Duo and Intel integrated graphics. I'm told just dropping a video card into a PC with a CPU that old isn't enough to run AAA games from the present generation (2014 and later), which would quickly become CPU-bound. Second, these can be done with a laptop, and I've seen no evidence that people routinely upgrade a laptop with a separately purchased MXM video card. Third, a console can be used while someone else is using the family PC.

Coincidentally, I recently upgraded my PC from a 7-8 year old Core 2 Duo, and even before that upgrade I was still playing AAA games, like Dying Light, just not at the highest settings. If you think about the age of the hardware in the current generation of consoles it's roughly equivalent to an 6-8 year old PC, although they are optimized and coded better for gaming than PCs, but AAA games with cross-platform support to PC still work quite well on older machines. As I said before the tech upgrade cycle for PCs has slowed considerably. I thought my old PC was around $800 when I bought it and the only non-original part was some really cheap RAM I added once, and I used it for more than just gaming.

The main reason I upgraded my PC was not because I felt lacking for games, but because I was running into compatibility issues with Win7. That may reinforce some people's belief that PC are inherently buggy and consoles are not, but mind you most people have a PC around for other uses anyway, and these can happen to any older PC regardless of its gaming use. For example my wife's parent's upgraded their laptop twice in the period of time I had my old PC, they are not gamers [or very tech savy] but they still spent more money than me on their PCs, irrespective of gaming usage.

Third, a console can be used while someone else is using the family PC.

Of course, but the original assertion was only that a console (singular) is much cheaper than a gaming PC (singular). If a family desires two devices over one despite the PC's versatility to do both, that's their issue not the PC.

First, though Steam has sales. PlayStation Store also has sales. Second, console games have historically been more likely than PC games to support same-screen multiplayer with two to four gamepads, and if you have more than one gamer in the house, one copy of a $60 game that supports multiple gamepads is cheaper than three copies of a $30 game that requires a separate copy per player. Third, if everybody were to wait for the sale instead of buying in release month at full price, publishers would have no money to continue to fund development of high-production-value games.

Sure, one $60 game and sitting on the couch together is cheaper than two or three PC games. but similar to my last response I think that's changing the parameters of the original assertion that consoles are always cheaper than gaming PCs. I will give this, consoles can be cheaper depending on the specific use.

PlayStation Plus and Xbox Live Gold cost $60 per year. But in addition to online matchmaking, this includes rental of a rotating selection of games (PlayStation Plus Free Games and Games with Gold respectively). What's the analogous way to try PC games?

Consoles make you pay that yearly subscription if you want to get full use out of your games and PCs do not. If you want to compare apples to apples then you have add those costs in. If you own a console with a $60/year subscription for 6 years you'd have an extra $360 to off-set the additional PC sticker price, subsequent PC upgrades (if needed), or pocket it.

If you want to rent ga

Re:Devil's advocacy

[tepples]

I recently upgraded my PC from a 7-8 year old Core 2 Duo, and even before that upgrade I was still playing AAA games, like Dying Light, just not at the highest settings.

Thanks for the data point. I'll mark ability to scale down to a C2D as "works for some".

most people have a PC around for other uses anyway

Except for people like one of my former co-workers, who owned only a smartphone. Mobile-only users adapt with Bluetooth keyboards and occasionally blocking out time for errands during library hours. Likewise, people who have only a laptop with Intel graphics and no standard MXM slot can't cheaply upgrade to a gaming GPU.

If you own a console with a $60/year subscription for 6 years you'd have an extra $360 to off-set the additional PC sticker price, subsequent PC upgrades (if needed), or pocket it.

Or theoretically spend it on some corresponding PC game rental service, if only one existed. Though "you can't play online on a console without also buying game rental" appears to be an argument in the PC's favor, console fans could reply "you can't buy game rental at all on PC". What do PC gamers typically use as a substitute for PlayStation Plus Free Games or Games with Gold? Or do they just rely on reviews? Or buying into a Humble Bundle every few months?

If you want to go used or older generations then the costs for consoles go down.

Likewise, I ought to try to remember to recommend a more recent used desktop PC with a new GPU to others making price comparisons.

Re:Devil's advocacy

[tepples]

Sorry, I overlooked this part the first time I read your reply:

If you want to rent games on PC you can sign up with Gamefly, but it would cost more.

From the front page of GameFly:

To enjoy GameFly, you'll need to enable JavaScript in your web browser. Please click here to learn how.

Once I got past that barrier, all games listed on the All Games page appear to be either for consoles or for handhelds made by console makers, not for PC. This is because PC game rental infringes U.S. copyright, unlike console game rental.

Steam allows users to trial games and gives full refunds within a certain time, which is a good substitute for "renting."

Until Valve decides that you've abused the refund policy and takes away access to all games that use Steam authentication.

Cairo vs. Copland

[tepples]

Remember "Cairo"? [...] WinFS probably takes the cake

I agree that Microsoft has talked a good vapor game. But each component of the Cairo project appears to have seen eventual release in some form.

• Windows NT 3.1 included DCE/RPC.
• Windows 95 and Windows NT 4 included Windows Explorer.
• Windows 2000 included Windows Search as part of MSN Toolbar. It became a core operating system component in Windows Vista.
• Windows XP included Windows NT Home. At this point, the majority of Cairo technologies had been released, and incidentally the Greek letters Chi-Rho look like the Latin letters X-P.
• Object File System, later called Windows Future Storage (WinFS), eventually became features of Microsoft SQL Server.

Was Apple any worse with its "Pink" and "Copland" projects?

Re:Cairo vs. Copland

[Yaztromo]

Was Apple any worse with its "Pink" and "Copland" projects?

I think the difference here was that Apple wasn't announcing their plans from a monopoly position in order to keep people away form the competition. Indeed, when Pink became Taligent, one of the idea of the AIM Alliance was to use a microkernel architecture that would permit various OS "flavours" to run on top of it, including Mac OS "Pink", OS/2, and Windows NT, all running on PowerPC CHRP.

My feeling was always that the problem with Apple surrounding Copland and Pink was more incompetence rather than malice, whereas Microsoft knew they were promising things they would never be able to deliver purely as a way to keep people from leaving the Windows ecosystem. Of course, it helped them quite a bit that their biggest PS OC competitor in the 90's, IBM, had a policy not to announce any product releases until 60 or 90 days before shipping (as I understand things, this was a legacy of the IBM antitrust case in the 70's). Microsoft took advantage, announcing things years in advance that they would never ship while a major competitor would basically not give anyone any information on what they were planning until it was pretty much in beta.

Maybe I'm jaded by experience, but Project Scorpio feels much the same. Sony has made no announcement about a PlayStation 5, the PS 4 and PS4 Pro are now known quantities, so now MS promises "the most powerful console ever built" before even showing anyone a prototype. Sony at least had a PS4 Pro at the PS4 Pro announcement (sure, the rumour mill expected the announcement for months, but Sony didn't officially announce anything until they were nearly ready to ship, so it wasn't a vapour announcement). This pattern feels all too familiar.

Yaz

x86-only libraries

[tepples]

I don't think using an ARM CPU makes porting games too difficult.

Unless one of the binary-only middleware libraries you're using is available only for x86 and x86-64 architectures.