A PC based console, a thought experiment

Is the rest of your environment protected?
The rest of the environment is irrelevant ...
Because if it isn't I could e.g. make an image of the HD, load it in VirtualBox, find a way to shoehorn SoftIce/Ollydbg into it (it's a Windows-based OS, right) and patch your dongle protection.
Patching does not quite cover the gravitas of your task ... this isn't about reversing jumps, this is about replicating the input/out behaviour of a blackbox. The dongle isn't just returning a "legit/not legit" signal, it is returning substantial amounts of computed data created by unknown parts of the original program (enough to make LUT attacks impossible).

It's important to note that the encrypted code running on the dongle only runs on that specific dongle, the dongle is tied to an account and the encrypted code for it is supplied during activation. You can compromise the rest of the system and look at it, but it won't do you any good without the secret key (which the dongle knows, but for which you need those probes on a <90 nm circuit to find out).
Do you expect to make a profit on hardware? Because I could just open one of your boxes, see what's inside and clone it as everything's off the shelf, then sell it for $50 less.
With the dongle tech it would be trivial to only allow the special consolized versions of the games to be activated on the official Valve boxes (the consolized versions are almost the same as the normal version BTW, the executables are the same they just come with profiles and install/configuration scripts to integrate them into the consolized environment). Personally though I'd say that if you can make it for 50$ less more power to you. I think the official stamp of approval and knowledge that the exact hardware spec. will be used during QA justifies the extra 50$ ... those which find it doesn't justify it (or who want the convenience of a consolized UI with more powerful hardware, at the cost of a larger potential for incompatibilities) can buy your solution. They will want Steam games either way.

Hell, if you make a really stylish box I might even allow you to add it to the official line-up for a small fee. This is the PC world, where we can make money without being anal retentive control freaks.

(I'm being hopelessly optimistic about Valve here, but that's just what I would do if I were them.)
Note that I have nothing against your idea, I just don't think it's financially sound. Steam as a service makes a lot of sense to me, but tying it to hardware you can get elsewhere for less money does not.
The console doesn't tie Steam to the hardware, it just provides a way to play Steam games in a rock solid environment with all windows configuration, gamepad support issues etc etc well hidden away ...
 
Last edited by a moderator:
Ok, so we're talking about replicating the functionality of the Cell's isolated SPU in the form of a dongle that is connected to your system. That's a different kettle of fish and could certainly work. I have no idea of what the cost of such a solution would be, though.

So the question that remains is: Does Valve have enough cash to pull this off? You need to throw gazillions of dollars at it to afford marketing and manufacturing and, of course, offer something that neither Sony nor MS offer already. I'm not sure better looking games alone would cut it (console owners already have the OS stability part). Valve is a well known developer in the PC world but I don't think Joe Sixpack out there has a clue of what Valve is.

What's the intended target, existing PC owners or would you want to attract console owners as well?
 
A 32 bits 66 MHz smart-card chip costs 2 bucks (SIM type carrier). Small FPGAs are pretty cheap as well. Not lightning fast, but it's running over USB ... so you couldn't run a whole lot on the dongle in the first place, it just has to be "enough".

As for what to offer. Blizzard games are getting a ridiculous amount of hype, so get together with them and see if they will let you market the versions for the Valve box as "console exclusives" :) (It's disingenuous, but in marketing everything is.) Get together with Acer to market the access to it's Chrome OS app store (I fail to see the big deal about app stores, but idiots seem to like them). Also market the hell out of it's storage capacity, the only console to come with enough storage to be a HTPC out of the box, license the Powered by Boxee trademark perhaps and put that on there.

The intended target is the console and HTPC crowd.
 
Seems easier just to have a unified gaming os pc gamers can use than the other way around. I'm surprised we don't have an os built for gaming that get rids of the bloat of windows in general.
 
That doesn't really help with hardware variations, and as long as people can just install random shit/drivers/etc they will always manage to fuck things up.
 
I don't believe MS wants PC gaming to die, otherwise they would allow/encourage mouse/keyboard support for 360 games. They just aren't solely focused on it at the moment. Probably due in large part to it being a platform where it's difficult to protect your IP.
 
There's a whole list of active threads at the moment where that discussion is more relevant than in this one NRP :) A console like environment for PC games can be accomplished regardless of what Microsoft wants, I don't think they are in the best position to do it in the first place (apart from getting the OS for free). Valve is.
 
Well the Phantom never materialized.........

A Valve centric box would be interesting, with the right price, marketing, and appeal to consumers. Issue is that the box would need to be a closed system to maintain some kind spec that could be powerful enough to last the amount of time it needed. You can't really treat it like a computer, upgrading would just confuse and frustrate the normal console like user. Who would be the market? You might get a few console gamers who want some of Steam's PC exclusive games and of course games that run better on PC than the consoles like Source games. PC gamers probably would want to stick with their PCs since they can do everything else on them as well as game. Basically you have a truly stripped down PC that unless you can aggressively price and market, as well as push it's benefits, has no real value outside gaming, but the idea is intriguing. Valve could tailor PC specific games to run on the hardware much better since it's a specific hardware set, provide a possibly better social experience, though Steam provides a decent one for those who actually care. As far as the OS goes, you'd need Windows of some kind, probably a stripped down version of Vista or 7 so something better than DX9 support is there for future releases, but MS wouldn't create such a version for Valve since it would be in direct competition with their 360. That doesn't stop them from taking a ready made OS, installing in each machine and tweaking on their own to lower resource usage as much as possible. They could install the box with an OEM copy of Windows, then run a program that retweaks Vista or whatnot to there exact vision, deleting programs and background tasks unneeded for a purely gaming machine. MS would still be highly upset regardless...........

But for the fun of it, plausible specs I would expect in such a box that would make it cheap to produce, yet still powerful for current games:

- High end 45 nm Dual Core processor like an Athlon II X2, or maybe a Core i5.
- 1 GB Radeon 4670, preferably 1 GB 5650 for DX11 capability w/ HDMI out
- 2 GB DDR3-1066 SRAM
- 1 TB 7200 RPM HDD
- Built in Wireless LAN
- Stripped down Win7 based OS
- Steam built in with messaging/friend services a la Xbox Live and media functions with at least music support
- Internet browser
- Aesthetically pleasing form factor no bigger than the 360.
- Get the power consumption down below 200, preferably 150W. Quite doable thanks to current mobile computing tech, but cost is the big issue here.
- Packaged with wireless mouse and keyboard
- USB controller support
Get down to $300 and you'd have an attractively priced product that would certainly get someone a good yet cheap PC experience.
 
Last edited by a moderator:
I think its a pity that MS haven't been pushing the PC console.

There were attempts to get the PC into the living room or more accessible for high quality gaming, but they have failed (if not coming close to release in some cases).

Xbox should be an IP with special software and hardware designs consumer electronics companies can use. It should also be fully compatible with PC and the other way round too.

I think we're in a time where multiple consoles are making less sense every generation yet the PC is theoretically the ideal platform in terms of openess and flexibility.
 
It wouldn't be marketed on the basis of exclusives, it would be marketed on the basis that it makes the multiplatform games look better. That said, timing it with the release of Diablo 3 would not be an entirely bad idea.

Aside from all the other obstacles, Diablo 3 will never ever come anywhere near Steam, Blizzard will push their games through retail, their own blizzard store and possibly through the new Battle.net if there will be some sort of separate client coming for it.

And honestly, even with all the success Steam has, it has GfW games too, which use Live!, which makes it already one step more complicated, you'll have 2 chat interfaces, 2 places to log to etc etc etc.

And then "stripped down windows" - do you see MS providing it as plausible option, possibly eating both XBox and normal Windows sales to some extent. And the ChromeOS? Dead idea, it's separate OS, not something you run on top of Windows, + it doesn't support DX and most games on Steam don't support OpenGL.
And what was that talk about "emulated graphics"?
 
It doesn't really need to be on Steam ... the system works with any PC game for which Valve writes a profile (the beauty of using a locked down windows).

Microsoft doesn't do the stripping down, all Microsoft has to do is give you the same volume discount it gives the other big boys (ie. Windows Tax prices). There are laws to convince them to do that.
 
Performance numbers I have seen for CXGames have not been good enough to really make it an option IMO.
 
What ever happened to that Phantom console, is that's similar what you have on your mind ?

What would Valve gained from taking a loss with every machine ? They're already 3rd party to 360 and PS3 and PC base already have one.

Making machine that cost $500 and competitive to 360 and PS3 isn't that hard but taking that machine to sub $200 is going to be a challenged that's difficult to overcome for a PC based console.
 
What ever happened to that Phantom console, is that's similar what you have on your mind ?
The big difference is that I'm suggesting a windows machine, it doesn't need developer/publisher support.
What would Valve gained from taking a loss with every machine ? They're already 3rd party to 360 and PS3 and PC base already have one.
Where did I suggest they would take a loss? This is both a PC and a console ... they don't need to win at the console wars, they already have market share. Every single one of these sold would increase their market share though.
Making machine that cost $500 and competitive to 360 and PS3 isn't that hard but taking that machine to sub $200 is going to be a challenged that's difficult to overcome for a PC based console.
As I said in the beginning, some people are saying the next gen consoles would take another 5 years. Making a 200$ machine might be pie in the sky ... but give it a year and they can launch a 350$ PC and with very good justification call it next gen compared to the current gen of consoles.
 
In a sense Sega and Nintendo were software vendors and they were following a model first thought up by companies like Atari.

One of the problems with a software house like Valve making a console and distributing via Steam is the lack of other developers wanting to get on board. The other problem is being dictated to buy very few companies regarding the design of the console unless Valve were not bothered about developing software tools for the hardware for another ISA like ARM or Power.
 
One of the problems with a software house like Valve making a console and distributing via Steam is the lack of other developers wanting to get on board.
A game doesn't need to be distributed on Steam to run on this box ... it's the ideal case, but for popular PC games not available on Steam they could simply write install scripts for the boxed versions.
The other problem is being dictated to buy very few companies regarding the design of the console unless Valve were not bothered about developing software tools for the hardware for another ISA like ARM or Power.
Windows, PC, x86 ... you know, the entire point of the thought experiment :)

They don't need the cooperation of Microsoft, they don't need the cooperation of developers, they don't need a large install base to start profiting (because it expands their existing market, PCs, which run the exact same games).
 
The big difference is that I'm suggesting a windows machine, it doesn't need developer/publisher support.

How would Valve make any royalties if all third party goes the Windows route ?

Where did I suggest they would take a loss? This is both a PC and a console ... they don't need to win at the console wars, they already have market share. Every single one of these sold would increase their market share though.

Why don't they just let people like Dell or Alienware sell their gaming PC and Valve can help by including their software in the package. A PC is a PC, it's going to increase their market share too without getting their hands dirty.

I still don't understand why it needs to be a console at all if it is a PC. A console is a subset of PC, if the machine is already a PC why does it needs to be a console too ?

As I said in the beginning, some people are saying the next gen consoles would take another 5 years. Making a 200$ machine might be pie in the sky ... but give it a year and they can launch a 350$ PC and with very good justification call it next gen compared to the current gen of consoles.

$350 ? They won't survived at $350 without fat royalties from other publisher. If they want to do this they got to see if there is a market for $1000 console to differentiate enough from PS3 and 360 in their later year. 3D0 tried that they failed, but who knows Valve and Blizzard do have the fan base. Though I just don't see them making anymore money than what they already are with just PC.
 
How would Valve make any royalties if all third party goes the Windows route ?
They would all leave Steam? Even though even with the royalties they get more per game sold than for a boxed version? I don't think so.
Why don't they just let people like Dell or Alienware sell their gaming PC and Valve can help by including their software in the package.
Because a normal PC doesn't act like a console, the fixed hardware spec is not an option and neither are the cut throat margins ... if Dell or Alienware were willing to build to spec they could be allowed to make their versions, but building boxes to specs with only cosmetic differences and small margins isn't going to be very attractive to them. If you want something done right you have to do it yourself first, you have to set the standard.
They won't survived at $350 without fat royalties from other publisher.
If we assume 50$ for Windows (Windows Tax pricing) and 100$ for Fusion it should be doable without selling below cost.
 
I think if they made this it would be easilly hacked and people would just buy a box and it and install Windows 7 or whatever on it for a cheapo gaming PC.
 
Back
Top