The Next-gen Situation discussion *spawn
- Thread starter gongo
- Start date
That is true I guess... You do have a lot of considerations and obviously the compromises have to be made even if I am not personally happy with them! There is always a possibility that the neXbox will be the best selling console in the next generation and everything MS management did/does will be considered genius, aka the Wii effect. The proof is obviously at the end of the day how the market responds to each console and if they can sell the console for $100 cheaper with little loss in visual quality then they have reached an excellent balance between the price and performance.
It would complicate the naming scheme a little if they did that. Maybe the best way would be to do a tick/tock cycle, say Xbox Next then Xbox Next + 3 years later. I would say any more than a single revision during a console generation would be far too complicated for the customers and developers to work around. They could then name the update around say major functionality improvements so for instance they could do something like:
Xbox 360 ---> Kinect ---> Xbox Next + Kinect 2.0 ---> Xbox Next + Kinect 3.0/head mounted display/eye tracking etc ---> Xbox 4
It doesn't make sense to release a major console revision unless you're doing something more useful than pushing prettier pixels. The idea would be instead of spending money and time making the same console cheaper, spend the money on making a more appealing console at the same price like what Apple does with their revisions.
Tick/tock wouldn't maintain BC. And what I propose is no different to PS2's BC, as it were. PS2 could play all PS1 games and PS2 games. Devs could target either. Consumers knew the difference. Eventually use of the older system phased out (software sales) and devs could abandon it. The future could be exactly like that only with far less time between SKUs (but more time than Apple's 6th monthly updates!), providing an upgrade path to keep the platform relevant, maintain the software library, and be managed with more flexibility. Each machine could be evaluated regard current market to see how much investment in hardware was needed, so a little more grunt to one-up the competition, or just keep a gradual update at a more welcoming price point.
A three year interval is readily serviceable IMO. It'd be enough to offer a notable upgrade for high-end gamers, maintain a relevant cheap console for entry level, and provide a suitable transition period where NextBox 2 games start with NextBox 1 games at higher IQ and then, with larger install base, start to get NextBox 2 exclusive games that NextBox 1 owners will have to upgrade to.
It would mean less utilisation of the high-end, but it would also mean better IQ and experience on the low-end, and shouldn't be any worse off than a 6 year interval anyhow. Devs are free to target the latest hardware if they want. If the business sense isn't there to, they can hardly be blamed for consumers not wanting to buy the latest, greatest AAA games!
A three year interval is readily serviceable IMO. It'd be enough to offer a notable upgrade for high-end gamers, maintain a relevant cheap console for entry level, and provide a suitable transition period where NextBox 2 games start with NextBox 1 games at higher IQ and then, with larger install base, start to get NextBox 2 exclusive games that NextBox 1 owners will have to upgrade to.
It would mean less utilisation of the high-end, but it would also mean better IQ and experience on the low-end, and shouldn't be any worse off than a 6 year interval anyhow. Devs are free to target the latest hardware if they want. If the business sense isn't there to, they can hardly be blamed for consumers not wanting to buy the latest, greatest AAA games!
There is nothing to win for the console vendors, nothing to win for the developers and very little (if any) to win for few consumers.
The console vendors never get to drive down the costs of console hardware, so never recoup the investment it is to launch a new platform.
The developers will have to choose which console spec to develop for. The weak spec with the larger install base or the fast spec with the small install base ? If they opt for the weak spec, they won't sell as many games to high spec console owners (full price for second-rate game ? NO WAY!). If they opt for the fast spec, they'll see lower sales because of the smaller install base.
The consumer will see the effect of the developers having to target two distinct install bases. Sales for each individual install base will be lower, so budgets will be lower, so the quality of the games will be lower.
Everybody loses.
Cheers
The console vendors never get to drive down the costs of console hardware, so never recoup the investment it is to launch a new platform.
The developers will have to choose which console spec to develop for. The weak spec with the larger install base or the fast spec with the small install base ? If they opt for the weak spec, they won't sell as many games to high spec console owners (full price for second-rate game ? NO WAY!). If they opt for the fast spec, they'll see lower sales because of the smaller install base.
The consumer will see the effect of the developers having to target two distinct install bases. Sales for each individual install base will be lower, so budgets will be lower, so the quality of the games will be lower.
Everybody loses.
Cheers
Last edited by a moderator:
the new platform isn't a major investment any more, with custom RnD and new branding and big reveals. It's an update, like many other products (cars, TVs, handsets). There'll be ongoing software sales, and service revenues, so the hardware can sold at cost to ensure maximum install base and fastest growth.
No different to targeting the old generation of a console or the new one. In 1999, do you target the PS1, or the upcoming PS2? And no different to PCs. The upside is you target the lower spec but the higher spec gets an improved experience.
The investment is no different. You spend big on your AAA title for NextBox. NextBox 2 owners buy the game and play it at double framerate with better lighting and shadows. When NextBox 2's install base is suitably large enough, developers have the option to target it specifically. The later devices are 100% BC with the former, so there's no work to support it on top of developing the game in the first place.
The difference is that if you choose the old platform, your install base is twice as big compared to half generation upgrades. If you choose the new one, you know your game will have twice as long to sell.
You're right, it is exactly like the PC market. That's not a market to emulate. The big sellers there are all games that run on mediocre PCs: Sims, WoW, Diablo 3, etc. with few exceptions (Battle Field).
Two questions:
1. Will consumers on NB pay full price for a game without getting the full benefit ?
2. Will consumers on NB2 pay full price for a "last gen" game ?
Answer to both: No
The sentiment of the two groups will be:
NB owners: Developers spent time to improve lighting and shadowing for NB2 owners at NB owners expense. Resources could and should have been used for more content or more optimization.
NB2 owners: Developers gimped the game to make it run on old hardware (similar to how PC gamers whine about how consoles are holding PC gaming back)
You'll have an inferior product on both platforms.
Cheers
Yeah I was thinking something similar. It's puzzling the direction MS may be going with the next xbox since the people that have purchased a 360 since the launch of Kinect are the least likely to be there for a system launch.
Cater to the core at launch and as the system's price lowers, opening up new markets, you start to branch off your support in multiple directions. It worked out great for the PSone, PS2, and 360, so I don't understand why they would deviate from this.
At first it looked like they were planning to cater to the core with the studios and talent they have been building over the last few years, but bkilian's previous comment doesn't give me much faith. This is sad considering Mattrick's background in the game industry.
Last edited by a moderator:
So... besides Vita touchscreen, Sony is also exploring a touchpad controller like this:
http://www.neogaf.com/forum/showpost.php?p=46709969&postcount=1418
Pretty interesting. They propose to turn a basic touchpad into a myriad of controllers.
You have committed a serious strategic error. You have incorrectly designed your questions. In the scenario that you are discussing, while the answers to your questions are correct, neither of your questions reflect the reality described. Therefore the answer has no meaning.
That's cool. Maybe confusing for customers to know which controller config they need to play a game. Definitely interesting. My only hope is that it would be a low latency controller. That's one gripe I have with the new consoles. People complain about the touch screen in the Wii U controller, but the reason it was picked is because it is a very low-latency touch screen. Valve seems to be focussed in the same area, improving gaming responsiveness. It is definitely an interesting concept.
That makes no sense. whenever you release your game, it'll sell for as long as the platform exists. That could be 20 years (think PC download titles of old games)
That's because the PC market has rivals in the consoles. It's not the business model of the PC that ensures its top sellers are mediocre games. If consoles didn't exist, all the AAA titles would be released for PC targeting scalable hardware.
Your describing sentiments not reflected in any other parallel. In the Android space, people buy a game. When they upgrade their hardware, they upgrade the game experience with better framerate at the least. No-one's grumbling about getting an inferior experience by playing the content on a less capable machine. Nor on the PC - you don't pay less for BF3 or Crysis if your machine isn't as capable of running it. Nor even on console, where PS3 owners don't pay less for a cross-platform title if it runs with 5% framerate and 50% less grass and half the shadow resolution. Nor between console and PC, paying less for Borderlands 2 no XB360 than on PC because it is lower quality.
SteamBox is going to be exactly that upgradeable experience. Buy a game and play it. Buy new hardware and play the same game better. Play the same game 20 years later. Devs can target whatever level of hardware based on metrics (like Android also, where a develoepr can choose based on install base to target v1, 2.2, 2.3.3., 3, 4, 4.1). The only real downside is a lack of high-end focus, which effectively delays technical progress probably a couple of years with games not targeting hardware features until they've been out a bit. On the upside, it'll provide a better 'current-gen' experience as time progresses for those willing to invest. Another downside of course is the extra consolyness, such as a fancy new controller or camera with every console as opposed as a barely used peripheral, but most of the talk from core gamers seems to be against all that anyhow.
Your describing sentiments not reflected in any other parallel. In the Android space, people buy a game. When they upgrade their hardware, they upgrade the game experience with better framerate at the least. No-one's grumbling about getting an inferior experience by playing the content on a less capable machine. Nor on the PC - you don't pay less for BF3 or Crysis if your machine isn't as capable of running it. Nor even on console, where PS3 owners don't pay less for a cross-platform title if it runs with 5% framerate and 50% less grass and half the shadow resolution. Nor between console and PC, paying less for Borderlands 2 no XB360 than on PC because it is lower quality.
SteamBox is going to be exactly that upgradeable experience. Buy a game and play it. Buy new hardware and play the same game better. Play the same game 20 years later. Devs can target whatever level of hardware based on metrics (like Android also, where a develoepr can choose based on install base to target v1, 2.2, 2.3.3., 3, 4, 4.1). The only real downside is a lack of high-end focus, which effectively delays technical progress probably a couple of years with games not targeting hardware features until they've been out a bit. On the upside, it'll provide a better 'current-gen' experience as time progresses for those willing to invest. Another downside of course is the extra consolyness, such as a fancy new controller or camera with every console as opposed as a barely used peripheral, but most of the talk from core gamers seems to be against all that anyhow.
LOl, you certainly answered that a lot nicer than me.
I don't see why it will be any different to 256MB cards are the start of the current generation. GPU's like the X1900 Pro and 8800GT came with 256MB configurations that could handle the full spec console versions of ports for at least a couple of years into the generation and possibly a lot more. And those GPU's are relatively computationally weaker compared with RSX and Xenos than a 680 is over Orbis and Durgango.
Don;t get me wrong, I'm not saying a 2GB GPU will be enough to match either console in every way and certainly not for their full life. But there are plenty of ways to easily scale down memory use (texture resolution, draw distance, asset quality etc...) which can and will be built into PC versions of ports to support the many configurations out there which do have less then 4GB of RAM. So it's entirely possible that some games may require you to lower draw distance or texture resolution on a 2GB 680 compared with the console version of the game but at the same time that 680 may be able to push out 60fps to the consoles 30fps or 1080p to the consoles 720p, or better shadows/more particles/higher details shaders etc.... A 2GB 680 will have trade offs with the current generation in games that use more than 2GB of the memory in Orbis for rendering. But that doesn't mean they won;t be able to run those games at all. And in other ways those games will run better on the 680. Of course, if you want the no compromise experience guaranteed every time you'll have to grab a 4GB card. I expect the 7xx and 8xxx series GPU's wuill be coming with a lot more memory than we're used to seeing as standard though.
Actually in any comparison like this I generally assume a 2X average performance advantage from a closed box environment. Not because I think that's the actual number as a few of the really reputable guys have already said is not true but simply stem the responses of "but what about the console advantage".
The 680 generally keeps good pace with the 7970 Ghz Edition which is a 4.2 GFLOP GPU. So I'd say that puts it at a little more than twice the speed of the Orbis GPU - assuming the rest of the GPU components scale linearly with flops in comparison to Tahiti obviously. So giving the console a very generous 2x closed box advantage means that later in their lives when they are actually achieving that advantage, a 680 should still be able to keep up or even slightly exceed them computationally. Don't forget that when your comparing 192GB/s in the 680 to the consoles you're comparing how NV manage memory bandwidth to ATI. The 7970 Ghz addition needs 288GB/s to stay roughly on par with the 680 for example.
Or lost
, remember there is also included in E&D the highly profitable MS hardware & mac software divisions.We can only go by the facts, and the facts are the E&D division has lost between 0.5-1.5billion since the time of the 360
(youre forgetting office? which IIRC has become MS's largest money earner i.e. overtaken windows
Latest quarter numbers (in profits)
Windows & Windows Live: $2.95 billion
Business: $3.77 billion )
But in that case hes disillusioned to suggest that MS is doing far better than the 2 japanese companies is completely nuts
IIRC (I can dig out exact figures if ppl want)
~>30 billion profit nintendo lots of consoles,handhelds
~4 billion profit ps1,2,3,psp,vita
~7 billion loss XB,360
even idi amin would be hardpressed coming up with that conclusionLike I said before I dont think theres any reason for MS/sony to release a console this year its not like the wii U is romping away, I'll wait for the current craze of phone/tablets dust to settle first
I've not read the whole thread yet so apologies if this has already been pointed out but you're flipping the target of your argument to suite whatever agenda your current post has.
The thread of this discussion is about how the new generation of consoles compare to PC's at their launch in relation to how the last generation did.
You can either do that by comparing the high end now (actually at the end of this year) with the high end in 2005 or you can compare the "PCs with the latest technologically advanced standardized single components available in most markets to consumers" whatever the hell that means.
If you compare the former, then that means 680/7970 level components with large memory configurations. If its the latter then we need to compare with the same thing from 2005. Which I'll get to in response to your next point.
And for the record, one of those links was to a Geforce 670. I'd hardly describe that as ""the most powerful PC money can buy". Would you have described a single 7800GT in such a way in late 2005? Or if I really want to get pedantic, a 6800GT since that's what the 670 will be equivalent to by the end of this year.
There's nothing exotic about a regular off the shelf GPU with a high memory configuration. And if there were, then you can forget about using the 7800GTX 512MB as a comparison point to Xenos back in 2005. Now we're down to comparing it to the old 430Mhz Geforce 7800GTX. This harms your argument rather then helps it.
At what point did anyone mention those?
At what point did anyone mention those?
When did anyone mention any of this?
I have not tried to compare to anything "non-standard" whatever that means in PC terms at any point. And as far as developers recommendations go, by late 2013, early 2014 there will certainly be nothing at all unusual about developers recommending 3/4GB GPU's and 8GB system RAM for the best performance.
You're just making this stuff up as you go along aren't you? At no point did I even hint at such a ridiculously configured system. Why? Because it's not needed. All I've ever compared to is a single stock clock GPU with the higher of two optional memory configurations in a PC sporting standard DDR3 1600Mhz. Please stop the massive exaggerations of my argument in an attempt to make your own look more reasonable.
We can settle this very simply. I'll tell you what I consider a high end configuration from October 2005 and what the equivalent of that totay would be. Then we can put all the ridiculous-nous of the above few paragraphs to bed. Sound good?
October 2005
AthlonX2 4800+
2GB DDR 400
1x GeForce 7800 GTX 512MB (no they did not have 1GB single GPU's in 2005).
January 2013
i7 3770K
16GB DDR3 1600
1x Radeon 7970Ghz Edition 6GB
If you would care to argue with either of those definitions, go right ahead, but please drop this "the most powerful PC money can buy" foolishness, it was never part of the equation.
Sigh...
There are standard, readily available, off the shelf PC's GPU's with enough memory available today. By the end of 2013 there will be lots more. The option would be included in the PC version, of course it would. What is the harm in including it for people with the capability to use it as long as there is an option to scale down texture resolution for those that need it? This has always been the way it works.
Non-answer? It's a fact and thus I understand why you chose not to comment. You can't dedicate 100% of any systems memory to graphical assets. You've got to allocate something to actually running the game.
That simple eh? Or perhaps, a little more simple, the PC version has the option to reduce texture resolution / draw distance / asset details for those systems unable to achieve the console levels while systems that are able to match or exceed those levels can use higher settings. And if a GPU is more powerful but has less memory then it's a simple matter for it to run with lower texture resolution but higher framerate/resolution/shader effects etc... as an example. Where have I seen this done before... ah yes, normal PC games. In 2005/6 there were lots of 256MB GPU's but that never stopped developers making console ports that would run just fine on them.
if the rumored specifications of PS4 are right, than PC gamers are in real trouble, they wont see the same graphical quality of core gaming assets as in PS4, or there is a solution, if they want to, they could buy one.
But again if you can answer my question, I will seriously reconsider my position.
Yes, they are in real trouble. 3x the computation power and at lest the top 6 single GPU configurations offering more memory than the PS4 is a very worrying situation to be in. Obviously PC's were in a far better position in late 2005 with the um one GPU with equal or slightly greater computation power to Xenos and the two GPU's that offered the same amount of memory if you ignore the edram.
EDIT: Huh, I just read the rest of the thread and realised I've just repeated what everyone else already said. That'll teach me for going away for 2 days. Sorry!
Last edited by a moderator:
I take it from this you haven't played a games in proper 3D? Not that your point is invalid, it's certainly possible for core graphics to make up for all of the above improvements but it would have to be a fairly big increase in core graphics to overcome the visual effect of true 3d on a big screen at high resolution. That's just my opinion of course.
Halo ran great on PC's back in the day. I remember running it on my Ti 4200 at 1024x768 with great frame rates. That GPU wasn't hugely beyond the XGPU. MGS2 also ran perfectly on that GPU but that's obviously a fair bit higher spec than the PS2. I guess you'd have to consider how well it runs on a GF2 and of that I have no idea.
GTA4 was badly CPU limited but if you had a quad core I don't think it needed a crazy amount of GPU power to run at console settings (the in game graphics options in GTA4 scaled to crazy levels - way beyond the capabilities of the best PC's or consoles at it's launch)
i keep always remember a long time ago it was said ea had battlefield 3 running at pc high settings 1080p, 30 fps on durango. this was in context of durango not necessarily being as powerful as wanted.
i look at crysis 2 which runs well on my 4890 rig, at high or highest settings mostly 30 fps, it should look fantastic on durango. ditto bf3, which my 4890 can also surprisingly actually run at mostly ultra settings 30 fps campaign.
just those facts actually make me pretty damn excited for durango. considering how shit optimized pc games are. throw in a 8 core cpu and 5gb ram, the rest of the specs are there too.
my 4890 is 800 shaders factory overclocked to 925 mhz i think, but the roughly 7770 in durango should easily outpower it due to gcn. and have tesselation, etc.