Gabe Newell: Valve will release its own console-like PC

It would be interesting to see how many of these developers, some of which haven't been willing to add code support for things like controllers, DX11, newer msaa solutions, or basically any code upgrades at all on the pc sku presumably as they deemed it financially not worth it, will they now be willing to take on a new Linux sku and the development/financial/support/marketing responsibilities that come with it. All that to put it on a platform that charges them 30%, when windows store charges 20% and runs their existing sku. Seems like a bit of a stretch to me to just expect publishers to all hop onto yet another development platform unless the terms are exceptional and user growth can be guaranteed.

Considering the restrictions for selling a game through the Window's Store, I doubt that the Window's Store is an option for many developers that release their games through Steam.

Something like the upcoming "Don't Starve" or somewhat cute but dark "Little Inferno" wouldn't make it into the Window's Store due to rating of the content or time to load the application on a "lower power device."

That said, I mostly agree with the rest. I'm not seeing a large financial incentive for most of the games with a larger budget to be ported to linux. Porting to Mac is already fairly risky from an ROI point of view.

For smaller devs with a smaller budget/smaller scope to their game that's definitely possible. But if you are making a game that might fit in only 40-200 MB, then porting to Linux would be significantly easier/less risky...but then again at that point the Windows Store becomes a more attractive proposition.

Regards,
SB
 
Considering the restrictions for selling a game through the Window's Store, I doubt that the Window's Store is an option for many developers that release their games through Steam.

Something like the upcoming "Don't Starve" or somewhat cute but dark "Little Inferno" wouldn't make it into the Window's Store due to rating of the content or time to load the application on a "lower power device."

The rating issue I figure is temporary. They probably went very conservative with that to be able to launch the app store world wide asap without issue. They already offer movies with violence on their video store and songs with hardcore lyrics on their music store, so eventually the app store will get sorted as well with regards to more hardcore apps.

Time to load is something console coders already have to deal with, you used to get failed if your game took too long to load, unless you were a Sony 1st party of course :) I haven't looked into the specifics of time to load on windows store since my apps load right away, I'd have to check it out to see what the restrictions are. Maybe they went hardcore to start with then will relax them over time as their RT os get more mature/optimized. The last thing they probably wanted is their rev 1.0 code taking forever to load apps, they would get raped on all the review sites. So my guess is they purposely went very restrictive there to make sure 1st impressions where good.

End of the day, I can't imagine that Microsoft would not want a piece of the revenue generated by games like COD, etc...
 
Nope.
We don't know that. 18 CU/2 TFlop machines aren't amazingly powerful. We've no idea what price they are targeting, or what extras they ahve at what costs (eg. Kinect 2, or Vita controllers for PS4).
and you think that a steam box is going to fall where ? If its higher than the $300-$400 of a typical launch console then your getting into pc range pricing and if it doesn't replace either of those devices whats the point

1. Provide existing Steam users a console that plays their existing library
2. Offer a platform that can offer console gamers an upgrade path 2/4/6 years into next gen
3. Compete on features and services, which is what the next boxes will be competing on anyway, unless MS sells at a massive loss to go with the price advantage.
4. Provide current-gen games with a much better experience. eg. If you want to play Borderlands 2 now, or Battlefields, or COD, the Steambox will play them much better, whereas the XB3 and PS4 likely won't play them at all.
5. Give console gamers the option to also game on PC, providing a hardware synergy that offers value, and also a BC path that means the games they buy now will always be playable at no extra cost, unlike the consoles that like to resell old games made compatible.
Wont existing steam users have a laptop / pc that can play their existing library ?

2) aren't the current consoles the platform that offers them an upgrade path in the future ?

3) Which means they need to have a compelling set of features and a compelling price point. I don't see them having the price point

4) Wont next gen consoles do this as well as today's pc's ? Heck even the mac air will play titles like Batman AA at better resolutions and settings than the console verisons

5)Wouldn't it be cheaper for a console gamer to simply buy a cheap hardware upgrade for their pc ?Wouldn't a console owner have to rebuy all their games to play them on the steam box anyway ? So whats the difference between the two ? Esp since there will be some games the steam box doesn't have.

The top selling games are typically cross-platform. Steam has of plenty exclusives too, even if indie titles. But the Steambox software will be a hell of a lot cheaper in the long run. See my thread on the costs of switching to PC - any hardcore console gamer buying lots of multiplat titles will save a bundle with a Steambox and buying from Steam.
Steam indie games play on really old hardware. I can play the majority of the indie games on a 6 year old laptop .

I don't see how steambox will be cheaper for software in the long run , perhaps if you want to wait months or years for price drops it could be slightly cheaper but looking at steam

Black ops 2 is $60
Assassin's creed 3 is $50

I don't see a huge price difference. I know that we get holiday sales and some games can be had at a huge discount but then again I got black ops 2 for $40 bucks on my 360 and I doubt its going to drop more than that on steam for quite a long time
 
The thing is they could just release profiles for popular configurations from oems (dell, Asus, acer etc). Steam certified pc or whatever, no need to get into the hardware business if you're not going to make an effort to actually compete with consoles there is no point in taking the financial risk.

How many of those "popular configurations" are? I am sure more than a hundred, at least.
Add Shifty Geezer #77 post, Valve desire to be free from Microsoft clutches, the possibility to install Libreoffice, GIimp, etc. and you have it all.
 
They could do a dozen or a hundred, it doesn't matter. If they build hardware they will need to spend hundreds of millions to sell enough units to get developers to even think about optimizing for it. MS spent half a billion marketing kinect.
 
Steam indie games play on really old hardware. I can play the majority of the indie games on a 6 year old laptop .

I don't see how steambox will be cheaper for software in the long run , perhaps if you want to wait months or years for price drops it could be slightly cheaper but looking at steam

Black ops 2 is $60
Assassin's creed 3 is $50

I don't see a huge price difference. I know that we get holiday sales and some games can be had at a huge discount but then again I got black ops 2 for $40 bucks on my 360 and I doubt its going to drop more than that on steam for quite a long time

There's the assumption that Microsoft and Sony are going to stick to the same digital distribution pricing they have for current gen because they still need retail, whereas Steam's digital only approach makes pricing more flexible.

Then again for Steambox to be anything but a niche, it's going to require retail presence closer to consoles. Retailers stand to make more money from the console model so this will be a challenge.
 
They could do a dozen or a hundred, it doesn't matter. If they build hardware they will need to spend hundreds of millions to sell enough units to get developers to even think about optimizing for it. MS spent half a billion marketing kinect.

The thing with Valve doing a console, is I don't think they have to be nearly as aggressive marketing it and I doubt they'd be concerned about "winning" the console wars. Convincing devs to port their DirectX based Windows games might be difficult if they don't sell a good number.
 
While they might not be concerned with winning I expect they will be concerned with making a profit. It's going to take more than a closed box with a valve brand to accomplish that. They will need to convince Activision/Blizzard and EA to get on board or it's dead before it starts. To convince them they are going to need to invest significantly, either in the hardware to get a big consumer interest or in development of exclusive software and or money hats to pass around for 3rd parties. Without real investment I fail to see how this project will accomplish anything.
 
The thing with Valve doing a console, is I don't think they have to be nearly as aggressive marketing it and I doubt they'd be concerned about "winning" the console wars. Convincing devs to port their DirectX based Windows games might be difficult if they don't sell a good number.

I tend to agree, Steam seems like the best place to advertise a steam "console", and I suspect they can get a pretty good advertising rate there.
I'm sure Valves business model would include making a significant profit on the HW.
I think betting on Linux ports is a bit of a stretch in the short term, I just don't see publishers paying for the ports without some monetary incentive.
 
I tend to agree, Steam seems like the best place to advertise a steam "console", and I suspect they can get a pretty good advertising rate there.
I'm sure Valves business model would include making a significant profit on the HW.
I think betting on Linux ports is a bit of a stretch in the short term, I just don't see publishers paying for the ports without some monetary incentive.

Couldn't they just cut them a deal where a certain title gets released with a lowered "Valve tax" on revenues? Seems like the easiest way to do it. The publisher basically has zero distribution costs, good advertising on Steam, and they collect pretty much all of the money from the consumer.

I'm not so sure Valve will really care about profits on the hardware, as long as they break even after marketing and distribution. The real money is just in getting people to buy the thing and have the pc market grow in general, whether they buy a Valve console or any PC that's equivalent or better. I'm not so sure the aim is to get everyone to buy one of these consoles so much as advertise PCs as Steam as a great place to buy and enjoy games. I can see them making profits on the gamepad, which they'll hope to sell to pretty much anyone that owns a PC.
 
The rating issue I figure is temporary. They probably went very conservative with that to be able to launch the app store world wide asap without issue. They already offer movies with violence on their video store and songs with hardcore lyrics on their music store, so eventually the app store will get sorted as well with regards to more hardcore apps.

Time to load is something console coders already have to deal with, you used to get failed if your game took too long to load, unless you were a Sony 1st party of course :) I haven't looked into the specifics of time to load on windows store since my apps load right away, I'd have to check it out to see what the restrictions are. Maybe they went hardcore to start with then will relax them over time as their RT os get more mature/optimized. The last thing they probably wanted is their rev 1.0 code taking forever to load apps, they would get raped on all the review sites. So my guess is they purposely went very restrictive there to make sure 1st impressions where good.

End of the day, I can't imagine that Microsoft would not want a piece of the revenue generated by games like COD, etc...

The load time requirements are currently 5 seconds to start and 2 seconds to resume from a standby state on "low power" devices. I'd imagine that if developers started to abuse level load times that the requirements will be adjusted to include that as well.

As to the ratings, I get a different vibe from it. It seems that Microsoft is really wanting to push the Windows Store as being family friendly. Everything about it is geared towards making sure everything viewed or experienced in there is geared towards appropriateness of 12 year olds. Although I do question how the Adult Friend Finder app managed to make it in. :p

For "adult" or "mature" games, Microsoft has the Xbox console which has parental controls.

Perhaps MS will eventually include parental controls in Windows which would then open up the possibility of more mature games in the Windows store.

They could do a dozen or a hundred, it doesn't matter. If they build hardware they will need to spend hundreds of millions to sell enough units to get developers to even think about optimizing for it. MS spent half a billion marketing kinect.

That all depends on whether they go with a Wintel machine or a Linux/other based machine. If it's Wintel based, then the investment is relatively small and there isn't a need to have a smash hit immediately. They can seed the market and gradually grow their share (if successful) or drop it with relatively small losses if it doesn't.

Granted it's a strategy predicated on patience, but I view it as the most likely to succeed if Valve does it right. IE - locked down machine, profiles for games (guaranteed performance), multiplayer profiles that allow play only among locked machines (unlocked machines running Steam is also an option) to provide a "cheat free" online experience as well as limited possibility for hardware/software conflicts and/or malware, etc.

I tend to agree, Steam seems like the best place to advertise a steam "console", and I suspect they can get a pretty good advertising rate there.
I'm sure Valves business model would include making a significant profit on the HW.
I think betting on Linux ports is a bit of a stretch in the short term, I just don't see publishers paying for the ports without some monetary incentive.

I agree with this. I really don't see Linux as being a real option until/unless Valve's Linux initiative manages to succeed. Until then, I'm almost positive that whatever machine they release will be Wintel (Arm need not apply obviously) based. The only question in my mind is whether it will be completely open (traditional PC) or locked down to some degree in hardware and/or software.

Regards,
SB
 
Linux is a chicken-and-egg situation. If there are few Linux gamers, devs won't target them. Perhaps a Linux Steambox will be enough to draw attention to development for it.

Anyone know the porting costs from Windows to Linux?
 
If they do go Linux, I don't really see them as having any other choice but to go with Google's Android. Otherwise they are basically not going to have a draw for any significant number of people to buy it.

That's been the problem of pretty much every single linux console that has been proposed or launched in the past 15+ years.

Valve has a chance to change this, but not with a console. By making it possible for developers to sell Linux based games through a well established and trusted storefront, they open up the possibility that the Linux game scape may expand faster than it is currently.

Once there is a decent catalog of games including a steady supply of new releases, at that point I can see a possibility for a successful (even if minor) console.

As to the costs, I'd guess it would scale with the scope of a title. Small indy titles should be relatively easy and inexpensive as Zed pointed out earlier.

Larger titles with a large scope are going to be a fair bit more costly meaning delaying the Linux port and/or bringing on a second developer. iD Software back in the early Doom/Quake days used to port their titles themselves I believe. But as titles got more complex, I believe they ended up having to hand off linux ports to other parties or delay them significantly. I can't quite remember which it was. I think at some point they stopped doing Linux ports as well as it just wasn't selling enough to recoup the cost/effort to port them to Linux.

Regards,
SB
 
Anyone know the porting costs from Windows to Linux?

It's going to depend on the game, how it was written and who does the port.
D3D to OpenGL is not a trivial port, then you have to deal with broken drivers. I have no idea what the state of Linux GL drivers is.
EA used to pay external companies for the Mac ports of the PC stuff, I don't know what the cost was, but it's going to be low to mid 6 figures probably.
You'd have to add QA costs onto that, which are not trivial.

If the original team did the work, it would be cheaper, if you targeted OpenGL on both platforms it would be cheaper. But then you still have opportunity cost, plus the distraction of trying to maintain/ship two SKU's simultaneously.
 
EA used to pay external companies for the Mac ports of the PC stuff, I don't know what the cost was, but it's going to be low to mid 6 figures probably.

Maxis has done the Mac ports internally so far. Beside of FIFA 12 and Dragon Age all current Mac offers from EA are from Mais. In the past there was a deal with TransGaming to bring games to the Mac.But these were not ports. It was done by using Cider. Cider is the same as WineLib for Linux but as a commercial product.

If the original team did the work, it would be cheaper, if you targeted OpenGL on both platforms it would be cheaper.

Rule of the Thumb: Don't use OenGL for a A+ games on Windows if you don't have at least 3 email addresses from driver developers from each company that builds GPUs.
 
Maxis has done the Mac ports internally so far. Beside of FIFA 12 and Dragon Age all current Mac offers from EA are from Mais. In the past there was a deal with TransGaming to bring games to the Mac.But these were not ports. It was done by using Cider. Cider is the same as WineLib for Linux but as a commercial product.

This is not entirely true.
I actually worked at Maxis, or what was left of it after most of it was relocated to Redwood Shores.
Spore Mac was done externally, by I believe TransGaming, call it not a port if you want, but the final exe's are different.
I believe SimCity is being co developed on both platforms, but that has as much to do with Andrew wanting to develop on a Mac as much as anything else.
 
This is not entirely true.
I actually worked at Maxis, or what was left of it after most of it was relocated to Redwood Shores.
Spore Mac was done externally, by I believe TransGaming, call it not a port if you want, but the final exe's are different.

For some reason I always forget about Spore. But we have not that much contact with Maxis anyway as we do different stuff.

Sure you need to recompile it. Because of this I have compared it with WineLib and not Wine.
 
I can't see this console running anything other than Linux. They aren't going to make their own custom OS. They surely won't be putting Windows on it. They're doing this because they don't like Windows 8, and I doubt MS will make any deal with them to put out a Windows game console when they're going to have their own game console coming out in the next year or two. Deal with Apple is incredibly unlikely. Even a deal with Google seems unlikely. I don't know if Android is really suitable for performance gaming. I think it'd be fairly easy for them to make a very stripped-down custom Linux OS that basically includes drivers, APIs, X Window System and Steam client. Steam has a webkit browser built in, and enough management for your account settings and basic content file management. The thing is going to be a fixed hardware platform, so whoever supplies their GPU just needs to make one stable device driver for that gpu in linux. Would it even need to be updated? The OS itself would probably change very little internally year by year.

The main problem, as everyone has stated, is most games are directx(windows) based.

I suppose there's a chance I'm totally wrong and this thing will just be a small form-factor PC that you can install whatever OS (Linux, Windows) on, like buying a Dell or HP, and maybe it will come with Windows boot straight into Steam. I just don't see why Valve would bother doing that. What would be the reason? What PC issues would it help avoid, and how would that make the experience easier for gamers?
 
How near are OGL "on PC" to the use of OGL/ES on "PS3"?
It could help in porting to Linux multiplatform titles?

AlphaWolf It all depends on the easiness of porting. For the hardware, ask Asrock to make a balanced 400$ PC, and sell for cost.
 
Back
Top