Feasibility of an upgradeable or forwards compatible console *spawn*

[expletive]

I refuse to upgrade my console in less than 5 years. Just as I won't buy new TVs, cell phones, PCs, tablets and such every second year either. I also don't want to bother with selling them online or such.

I've got enough things to spend on, gadgets are quite behind on the list. I suppose a lot of other people feel the same way. If MS tries to adapt to such a strategy, they'll probably fail.

Its not so much about you getting your money's worth but more about people entering the buying market every two years having to compare the console offerings with every other set top box. As an early adopter the tech will be advanced so you will base your purchase on that, but what about the person who enters the market in 2-3 years? Now he is looking at games on an Apple TV with an A7x chip that look close enough to your Durango.

With a 'thick' API is it really that hard for developers to tick a few extra boxes on AAA titles so they run better on newer console hardware? Developers are already targeting an infinite number of platforms on PC, is having 3 possible SKUs of Xbox really that big a deal?

Yeah, and beyond screwing with customers and developers, there are other costs that would hurt MS - developing new hardware every 2 years, and inability to shrink manufacturing costs as time goes on.

So who exactly would benefit from this business model, apart from people who have too much money to throw away on what are basically expensive toys?

Good, Better, Best - $99, $199, $299. They will still shrink the current platform and that will slide to the "Good" tier and the newer one with the more powerful chip will take its place in the "best" slot. The benefit is always being able to be competitive with every tier of product fighting for the space under your TV. I see this as the only way to compete with both a Sony and APple simultaneously.

 

[TheChefO]

It's my belief they are going to maintain the multi sku model they have except instead of the differentiator being the HDD it will be extended to the hardware.

They already have patents showing multiple skus based off different number of gpus and CPUs. AMD already has plans that allows upgrading PC apus in such fashion.

They are probably going to release an arcade at a base config with a premium or elite with a higher CU/GPU count that's accomplished with an addon card.

It allows you to address the whole market from casual to enthusiast without catering to one specific sub market or hoping a significant amount of people buy into refreshes every two years. All at the release. It's pretty much no difference than what's the PC market offers us already except instead of performance based on bigger and faster CPUs and GPUs it will be based on CPU and GPU count.

It makes sense that they would drop the high performing hot and high cost hardware for energy efficient parts that are scalable. Jaguar cores aren't even meant for desktop PC, they are made for form factors even more TPD restrictive than consoles. Consoles have more space to work with and you don't have to contend with battery life.

Energy efficient CPU, GPU and RAM will allow for upgradeability from the start and cater to all markets without being to force to contend with hardware cost for parts meant to satisfy the core market on lower priced skus for the casual market who will still have the option of upgrading performance as they transition to a core gamer.

As we move down to 20 nm and below it will allow for more CPUs and GPUs and help creat hardware that scales smoothly over time without disrupting the market.

MS and or Sony changes the dynamic of the console market every gen. We got Live and a HDD with xbox 1 and we got multiple skus and digital distribution with the PS360.

That's a fine strategy to have, but it must be accompanied by a *clear* reveal of this strategy from day one.

Otherwise, customers will choose based on the options available on day one, which at the moment, technically favor Orbis. Unlike phones, friends lists aren't transferable like a contacts list are ... Once a direction is chosen for a group, the likelihood of switching is minimal.

Technically, I quite like the concept of a multi-tier level console, but I don't think it's necessary to delay the release of the higher sku.

Arcade sku = 720p standard
Pro sku = 1080p standard
Elite = 1080p 3D
Elite pro = 1080p 3D 60FPS

Etc.

None of which would require devs to labor over new versions of the same game. Just target those tiers and viola, everyone is happy.

MS isn't losing money on hardware, gamers aren't pissed about weak hardware, and xblg is still making a boat load of cash.

Everyone wins.

 

[anexanhume]

Care to elaborate?

The ideas that MS is going to performance tier SKUs or rapidly iterate hardware.

They're completely antithetical to how the console market works. And not just tradition. They shred up everything the business model is.

 

[Phil]

Perhaps it's worth expanding on this a little. We are already seeing a few rival systems appearing who will offer just that. Ouya will offer a box where your games get better over time, to some degree, as you upgrade the hardware. SteamBox will provide the same. If Apple create a set-top box, that'll provide the same. And MS already have a forwards compatible platform that allows your old library to be played in better and better quality in Windows. That could be a significant advantage next-gen against a rival who's library is a one-stop deal. If MS communicate effectively that it's Xbox Forever, than my options will be: PS4 that may be notably better than XB3, but who's library will be outdated and a dead-end in 4ish years; or XB3 and a family of improving boxes I can buy into when I feel its time for an upgrade, including PC even, taking all my library with me. That's a compelling argument.

There's a whole other two or three threads discussing this (Upgradeable console etc.) for those who want to argue for or against, but the key point is this is clearly a valid business strategy for MS and so shouldn't be looked on with surprise or consternation. We might well see a disparity between Orbis and Durango that goes beyond hardware differences, but that may also result in Durango (DirectX, Windows) being the preferred platform for gaming in the coming years.

Interesting stuff, this.

I wonder though, how feasable this is from a software point of view?

Take an example: Microsoft launched with Project Gotham Racing 2013 this fall on Xbox2013. Then, 2 years down the road, Microsoft bringts out Xbox2015 and by then, Blizzard (or whoever develops PGR) bringts out PGR2015 with it. Now, the interesting thing would be that both games both run on both Xbox versions - hence the thick API.

Is this really possible?

I know, this is more or less doable on PC gaming if you look at it. Though at the same time, the PC gamer audience is very different. With the million of different hardware, people there expect a bit of fiddling, framerate hickups, bugs to get things to work properly. Those are some of the reasons to why many PC gamers converted to a fixed hardware platform for gaming back in the day...

If Microsoft (or anyone) can pull this off - the games and hardware that are forward compatible would need to run well and most importantly remain playable. I would expect framerate to remain constant, meaning that in my hypothetical example, PGR2015 would need to run as well as PGR2013 on the Xbox2013 - but both would run (especially PGR2015) with nicer graphics on Xbox2015.

Now, I guess if newer games are limited to graphical upgrades, this is probably quite easy to achieve.... but my thoughts are, if the next iteration of hardware is that much better than the old one it's technically replacing, how do you deal with games pushing the envelope (well pushing the API) with better AI, physics, better gameplay mechanics?

If you want to remain compatible with the first SKU you've launched with, you are also limiting developers to the possibilities that the 2013 hardware offered - meaning, probably less innovation on that front. So 2015 hardware might not be all that attractive afterall, with the exception for better graphics. I don't think it's possible to design a game with better 2015 hardware in mind (think of bigger worlds or gameplay mechanics that aren't possible on 2013 hardware) and then simple turn off features to make it run on 2013 hardware without completely changing the gaming experience or altering the game itself. Well, maybe you could, but then you would be designing the game with those limitations of the weaker 2013 hardware in mind. I'm not sure a API can do that for you, which IMO sounds like a lot of headaches.

And if you simply ship games at some point that don't run on 2013 hardware... well... then you're not forwards compatible anymore.... which, to some degree defeats the thick API in the first place. Well, I guess you could say, new games are at least compatible with 4 year old hardware before you phase it out which would smooth things considerably... Would it work?


Dealing with this kind of problems from a software centric point of view... This would make one interesting topic, that's for sure.

 

[Xenio]

The ideas that MS is going to performance tier SKUs or rapidly iterate hardware.

They're completely antithetical to how the console market works. And not just tradition. They shred up everything the business model is.

who cares about tradition?

just give me the best option.

 

[Xbat]

A good game takes anywhere from 3 to 4 years to make, especially in the begining of the new cycle. So which tier would you target? Surely you would have to target the base console which means the so called better machines could only really offer better resolution and frame rate not really worth the 300 or 400 dollars they will cost.

 

[dobwal]

The ideas that MS is going to performance tier SKUs or rapidly iterate hardware.

They're completely antithetical to how the console market works. And not just tradition. They shred up everything the business model is.

Prior to two generation ago, the console market did not operate with multiple skus nor hdds, nor online ecosystems nor the level performance that rivaled PCs even when gpus weren't standard hardware. The presence of MS has tremendously changed the landscape of gaming. And the most beneficial aspect of MS's presence to gamers has been performance. The Wii doesn't warrant the level of hardware in the PS3 nor the level of hardware losses incurred by Sony. That level and generational leap in performance is not sustainable in the current model even for MS who was the most profitable at it last gen.

You can't have 2/3rds of the manufacturers selling 70-80 millions of console a piece and still loses billions of dollars. Either you fall back on performance like Nintendo or you operate under a different paradigm. One sku - one market with high performance consoles doesn't work. If you divide the market into segments, the highend gamers have been very profitable. The more you move toward the casual gamer segment the less profitable the PS360 becomes. Its has not been profitable to drive the current hardware into hands of more mainstream and casual gamers. Their software purchases don't support it nor do the hardware costs support the price level needed for mass market appeal.

Falling back on performance will upset the most important segment of the console gaming market. It would practically destroy the differentiation between them and Nintendo and make Nintendo more relevant to the market segment dominated by the PS360. The adoption of a model thats already standard across the tech market as a whole makes sense. You provide console skus at prices that drive adoption of the bigger but less software hungry casual mainstream market with hardware that can justify those prices and you provide enthusiast and their tremendous software consumption with high performance consoles at price points that they are more willing to accept.

That type of model wasn't as feasible in previous generation because highly parallel multi core processors and the software know how to take advantage of it weren't as prevalent. We don't live in a world mostly driven by single thread single core performance anymore. Scalability is more of a reality than its ever been on both the hardware and software side, there is no reason that to operate under the old model anymore especially using PC based hardware when PCs and the tech market in general abandoned such a model decades ago.

One sku one market is an archaic model that does not do a good job of servicing a market thats become as big, diverse and demanding as the console market today.

Either console market adopt multiple skus based on performance or cheaper hardware. Tell me how cheaper hardware is going to compete against other gaming markets that don't have problem with investing in cutting edge performance with multicore hardware.

 

[Xenio]

seems like Sony drop the competition with kinect, the dual camera misses infrared z-axis camera http://www.vgleaks.com/orbis-dual-camera-whats-this/

 

[expletive]

The ideas that MS is going to performance tier SKUs or rapidly iterate hardware.

They're completely antithetical to how the console market works. And not just tradition. They shred up everything the business model is.

Obviously i'm not saying with certainty this is where MS is headed but i do think that the console market is not insulated from every other type of device aiming to take its place connected to your TV. The market and dynamics are changing, the competition and expectations are different. You're implying that everything else revolves around the 'game console' business model and i think its the opposite. MS and Sony will need to adapt, how and by how much isn't clear but i don't think its debatable that the functionality and price pressures are much different now than they were in 2005.

 

[dobwal]

A good game takes anywhere from 3 to 4 years to make, especially in the begining of the new cycle. So which tier would you target? Surely you would have to target the base console which means the so called better machines could only really offer better resolution and frame rate not really worth the 300 or 400 dollars they will cost.

A console thats heavily tessellation based will allow for targetting multiple skus at different performance levels even more efficiently than what we have now. The PC market has supported multiple skus since the advent of the gpu all with a smaller userbase.

Microsoft isn't Sony or Nintendo whose gaming business has been totally shaped by consoles. MS for the most part is a PC business and their gaming business started with the PC. They have driven the adoption of multiple skus, hdd, online service, software updates and digital purchases which were practically new in terms of acceptance on consoles but pretty standard fare on PCs. They are basically taking whats been widely accepted in the PC space and applying it to the console space.

 

[Laa-Yosh]

Dealing with this kind of problems from a software centric point of view... This would make one interesting topic, that's for sure.

Quality Assurance alone would be a terrifying undertaking...

 

[patsu]

I believe both Sony and MS will explore quick console refresh and forward game compatibility. They have to do so because of iOS style gaming. MS seems to take the API route. Sony may take the API + JIT compile route based on their dual GPU patent.

When people buy new consoles, they will have to download purchased games from the respective online stores anyway. For disc games, they can be patched.

In parallel, they are also going ahead with various business ideas, like PS Mobile, cross-play and PS+. These will provide valuable data on the actual spending and usage patterns.

Essentially, they cannot afford to ignore the iOS/Android economy.

 

[Xenio]

Both Sony and MS will explore quick console refresh and forward game compatibility. They have to do so because of iOS style gaming. MS seems to take the API route. Sony may take the API + JIT compile route based on their dual GPU patent.

Sony is pushing developers to go as close to metal as they can

 

[Xbat]

Yes pc has supported multi sku but for the mainstream console gamer, like some of my friends are, dont have a clue about how to adjust settings and whatnot. Just the other day I told a buddy of mine to drop his resolution in far cry to get better performance and he asked me what resolution was. So you can imagine how hard it was for me to tell him about ambient occlusion and all that other jazz. A big percentage of console gamers just want to put a disk in and play after a hard days work. Lots of people are going on about features and services because of the iphone success but a major part of that is because it's become a fashion accessory, lots of the users dont even use half of the features.

 

[Laa-Yosh]

I don't give a damn about iOS style gaming! If I did I'd play games on my phone or tablet, but I want to play console games, with depth, story, complex gameplay instead. Don't f*** my consoles up and don't urge the vendors to do so either

 

[expletive]

Quality Assurance alone would be a terrifying undertaking...

If it isn't properly automated, then yes it could add quite a bit to costs.

 

[patsu]

Sony is pushing developers to go as close to metal as they can

Yes, they have a patent on JIT compile for 2 architecturely different GPUs. Those guys have implemented assorted emulators, h/w compat, recompile, JIT compile, up/down porting, and other techniques between all the Playstations. Once the business trend is set, they will do whatever they can to move their business forward.

I don't give a damn about iOS style gaming! If I did I'd play games on my phone or tablet, but I want to play console games, with depth, story, complex gameplay instead. Don't f*** my consoles up and don't urge the vendors to do so either

I meant the economy side, not the gameplay or visual side.

 

[Sonic]

Obviously i'm not saying with certainty this is where MS is headed but i do think that the console market is not insulated from every other type of device aiming to take its place connected to your TV. The market and dynamics are changing, the competition and expectations are different. You're implying that everything else revolves around the 'game console' business model and i think its the opposite. MS and Sony will need to adapt, how and by how much isn't clear but i don't think its debatable that the functionality and price pressures are much different now than they were in 2005.

Nuts. Obviously everything doesn't revolve around the gaming console market, but big profits and dollars sure seem to make a lot of sense. When the next gen consoles are able to make a decent profit on the hardware, and sell millions of games that will mean there's little to no reason to release a new or upgraded system. If another company does decide to jump in then the onus is on them to establish themselves while spending billions of dollars doing so. If both Sony or MS see that these new competitors aren't making headway into the console market then there's little reason to release a newer console. You act as if any old company can come in and release something and it will change the entire dynamic of the industry. The costs of reality say otherwise.

 

[dobwal]

Interesting stuff, this.

I wonder though, how feasable this is from a software point of view?

Take an example: Microsoft launched with Project Gotham Racing 2013 this fall on Xbox2013. Then, 2 years down the road, Microsoft bringts out Xbox2015 and by then, Blizzard (or whoever develops PGR) bringts out PGR2015 with it. Now, the interesting thing would be that both games both run on both Xbox versions - hence the thick API.

Is this really possible?

I know, this is more or less doable on PC gaming if you look at it. Though at the same time, the PC gamer audience is very different. With the million of different hardware, people there expect a bit of fiddling, framerate hickups, bugs to get things to work properly. Those are some of the reasons to why many PC gamers converted to a fixed hardware platform for gaming back in the day...

If Microsoft (or anyone) can pull this off - the games and hardware that are forward compatible would need to run well and most importantly remain playable. I would expect framerate to remain constant, meaning that in my hypothetical example, PGR2015 would need to run as well as PGR2013 on the Xbox2013 - but both would run (especially PGR2015) with nicer graphics on Xbox2015.

Now, I guess if newer games are limited to graphical upgrades, this is probably quite easy to achieve.... but my thoughts are, if the next iteration of hardware is that much better than the old one it's technically replacing, how do you deal with games pushing the envelope (well pushing the API) with better AI, physics, better gameplay mechanics?

If you want to remain compatible with the first SKU you've launched with, you are also limiting developers to the possibilities that the 2013 hardware offered - meaning, probably less innovation on that front. So 2015 hardware might not be all that attractive afterall, with the exception for better graphics. I don't think it's possible to design a game with better 2015 hardware in mind (think of bigger worlds or gameplay mechanics that aren't possible on 2013 hardware) and then simple turn off features to make it run on 2013 hardware without completely changing the gaming experience or altering the game itself. Well, maybe you could, but then you would be designing the game with those limitations of the weaker 2013 hardware in mind. I'm not sure a API can do that for you, which IMO sounds like a lot of headaches.

And if you simply ship games at some point that don't run on 2013 hardware... well... then you're not forwards compatible anymore.... which, to some degree defeats the thick API in the first place. Well, I guess you could say, new games are at least compatible with 4 year old hardware before you phase it out which would smooth things considerably... Would it work?


Dealing with this kind of problems from a software centric point of view... This would make one interesting topic, that's for sure.

Tessellation is one area that makes forward compatibility possible. You keep low level meshes in RAM and then use the hardware and software to drive up geometry and lighting based on the performanc capabilities of the hardware. If its all jaguar and AMD CU based, a good tessellating hardware makes forward compatibility as practical as ever.

 

[TheChefO]

The ideas that MS is going to performance tier SKUs or rapidly iterate hardware.

They're completely antithetical to how the console market works. And not just tradition. They shred up everything the business model is.

Actually the business model would remain quite the same actually.

HW sold for small loss or breakeven, and make cash on sw.

For the devs, it would be pretty transparent if the hardware is designed correctly. Same content, same gameplay, same "target". Just let the api handle the game behind the scenes and render as best as possible to the machine's ability.

For the bottom tier that means 720p and perhaps 30fps (frame rate might be a bit too much)

For mid tier, 1080p.

Upper tier, 1080p + 3D.

This happens on a daily basis for pc gamers, but with more variables.