Predict: The Next Generation Console Tech

[(((interference)))]

The problem with such expansions is they don't get properly used. Devs have to design for the minimum spec, and then spend extra to support a higher spec console. I suppose for texture fidelity a RAM pack is an easy-to-accommodate upgrade, but that's the only expansion you could comfortably fit, and you be better off putting such an expansion on a proper internal bus designed for the purpose. No need for cross-device portable connectivity such as Lightpeak for that! We haven't had a RAM expansion or anything close for 2 generations, and I doubt we'll see a return. The closest we got was Kutaragi's hopes of networked boxes over Gigabit, that went nowhere.

Well, they got properly used on N64 and SNES (Perfect Dark, Star Fox) - of course not everyone is going to take advantage of them and will just tack on some enhancements or ignore it completely but there will always be the technically inclined studios who want to squeeze every last drop from the system and they will see it being put to good use.

I agree that any add on would have to slot into the motherboard much like the cartridge slots on old systems.

But your right we're probably not going to see anything like this next gen.

 

[phenix]

Sony May Launch PlayStation 4 in 2012 – Rumours

http://www.xbitlabs.com/news/multim...May_Launch_PlayStation_4_in_2012_Rumours.html

 

[pc999]

Given that we get each time less gfx oriented users and that to such a expansion to make a immediate and perceptible difference (much more than "high" DX9 crysis to "ultra" DX11 crysis) you would need a major, but really major HW upgrade (something a xenos --> 1Gb 5670 jump minimum , for the crysis example). I have a real hard time believing anyone would bother with such a thing.

 

[Rangers]

Given that we get each time less gfx oriented users

Huh?

Didn't the wii prove the opposite? Nintendo eventually had to abandon it due to lack of graphics...

It still remains about segments, some segments (some games, some users, even I dont care about graphics if playing Tetris) dont care about graphics (even there it must not look "old", 16 bit Wii wouldn't fly), but in the hardcore space they remain of utmost importance.

 

[tongue_of_colicab]

Stop trolling.

1. Nintendo isn't abandoning anything. Wiiu is still 1.5 years away and it's very unlikely that they will just pull wii off the market. They probably lower the price to something ps360 can't match (atleast not without losing money) and probably still continue to sell plenty (just like they are doing now btw, its not like wii sales are dead) to people who want a cheap console to have fun with friends/family or enjoy a decent enough sized library of nintendo games.

2. By the time wiiu comes out, the wii will be on the market for more than 6 years. Given how generations normally lasted about 5~6 years there isn't anything wrong with that. It's just that ms/sony are trying to make things last as long as possible to try and make up for their investment.

3. If ''hardcore'' would really care that much about gfx why stick with 6+ year old hardware in the ps360? They would all buy a pc. And if they did, we'd get a lot more games utilizing high end pc hardware. But they don't. The only conclusion can be that gfx just need to be ''good enough''.

Anyway I agree that it wont take long before the leap in gfx of every new generation becomes getting smaller and smaller. We are already at a point were gfx are pretty good and people will start noticing differences less and less. Also, looking at how many devs went belly up this gen there must be a point at which devs just can't afford it anymore to produce such high detail games. Sure, some will be able to but the majority won't. Unless you don't mind games getting even shorter.

 

[RudeCurve]

3. If ''hardcore'' would really care that much about gfx why stick with 6+ year old hardware in the ps360? They would all buy a pc. And if they did, we'd get a lot more games utilizing high end pc hardware. But they don't. The only conclusion can be that gfx just need to be ''good enough''.

The only reason why PS360 graphics is good enough is because both MS and SONY havn't yet released a more powerful successor. If they did PS360 wouldn't be good enough anymore because everyone's expectations would suddenly be higher, they'd want the better than current generation graphics after seeing the superior graphics improvement.

 

[pc999]

Huh?

Didn't the wii prove the opposite? Nintendo eventually had to abandon it due to lack of graphics...

It still remains about segments, some segments (some games, some users, even I dont care about graphics if playing Tetris) dont care about graphics (even there it must not look "old", 16 bit Wii wouldn't fly), but in the hardcore space they remain of utmost importance.

No, it didnt.

The many comments in this tech oriented forum about gfx should prove this.

Anyway there is plenty of others evidences:

1) Zelda for Wii is one of the most anticipated games;
2) many of the latest hotter games dont sport the bigger and latest gfx, eg: Fear3, Call Of Duty Series, Dead Space 2... In fact only a few of them care to push them very far (but IMO all of them have good gfx/art)
3) good sales of XBL/PSN game sales

Anyway we all like better specs, but beyond some point this accelerated evolution in gfx starts to be meaningless.

 

[Rangers]

but beyond some point this accelerated evolution in gfx starts to be meaningless.

There's no evidence of this. Their just seems to be a cadre of Nintendo fans who have adapted this claim since the announcement of the Wii U, but I cant agree at all. I think it's more like wishful thinking.

If next gen comes out, has 8X better specs, and the games honestly dont look that much better I'll agree. However, I am almost certain that will not happen. There's no way a console with 4GB of RAM isn't going to produce vastly better graphics than one with 512mb, to narrow it down to one specific for the sake of illustration.

I suppose I begrudgingly accept the diminishing returns argument eventually, but I've never seen any evidence that we're near it yet. Each console gen looks stunningly better than the last to date. And I see enough evidence (UE3 next gen demo, Crysis 2 DX11 pack, BF3 PC) already that next gen consoles will provide a vast improvement. Just look at all the complaining over Crysis 2 console's "reduced LOD, 25 FPS, no AA etc etc whatever". Now imagine a Crysis+Crysis 2 dual pack, 1080P 60 FPS all ultra high settings for just 39.99 (since it's a remake) for next gen consoles and drool Hell, you can probably argue whether even next gen consoles could pull off the Crysis games with AA, 1080P, 60 FPS etc (think they might have to settle for 30), which just shows even more how far there is to go.

 

[RudeCurve]

There's no evidence of this. Their just seems to be a cadre of Nintendo fans who have adapted this claim since the announcement of the Wii U, but I cant agree at all. I think it's more like wishful thinking.

If next gen comes out, has 8X better specs, and the games honestly dont look that much better I'll agree. However, I am almost certain that will not happen. There's no way a console with 4GB of RAM isn't going to produce vastly better graphics than one with 512mb, to narrow it down to one specific for the sake of illustration.

I suppose I begrudgingly accept the diminishing returns argument eventually, but I've never seen any evidence that we're near it yet. Each console gen looks stunningly better than the last to date. And I see enough evidence (UE3 next gen demo, Crysis 2 DX11 pack, BF3 PC) already that next gen consoles will provide a vast improvement. Just look at all the complaining over Crysis 2 console's "reduced LOD, 25 FPS, no AA etc etc whatever". Now imagine a Crysis+Crysis 2 dual pack, 1080P 60 FPS all ultra high settings for just 39.99 (since it's a remake) for next gen consoles and drool Hell, you can probably argue whether even next gen consoles could pull off the Crysis games with AA, 1080P, 60 FPS etc (think they might have to settle for 30), which just shows even more how far there is to go.

I totally agree with this, even 1080P@60fps ALONE would be a nice improvement let alone more RAM and upgraded graphics processing power.

 

[pc999]

There's no evidence of this. Their just seems to be a cadre of Nintendo fans who have adapted this claim since the announcement of the Wii U, but I cant agree at all. I think it's more like wishful thinking.

If next gen comes out, has 8X better specs, and the games honestly dont look that much better I'll agree. However, I am almost certain that will not happen. There's no way a console with 4GB of RAM isn't going to produce vastly better graphics than one with 512mb, to narrow it down to one specific for the sake of illustration.

Things seems much more dependent on good art and big budget, so I dont have that much hope, even if they can look that much better how many games can really make it even better?

Personally they should put as much as good tech as possible in next gen, as long as they keep the console affordable, and keep investing in innovation.

Tech dont need to be the focus anymore IMO, like a mp3 player, they keep getting better and having better sound, just not like in the first years, on the other side they really are cheap, varied and popular.

I expect the same to happen with console, they keep getting better with better visuals/physics/... just not the focus anymore.

I suppose I begrudgingly accept the diminishing returns argument eventually, but I've never seen any evidence that we're near it yet. Each console gen looks stunningly better than the last to date. And I see enough evidence (UE3 next gen demo, Crysis 2 DX11 pack, BF3 PC) already that next gen consoles will provide a vast improvement. Just look at all the complaining over Crysis 2 console's "reduced LOD, 25 FPS, no AA etc etc whatever". Now imagine a Crysis+Crysis 2 dual pack, 1080P 60 FPS all ultra high settings for just 39.99 (since it's a remake) for next gen consoles and drool Hell, you can probably argue whether even next gen consoles could pull off the Crysis games with AA, 1080P, 60 FPS etc (think they might have to settle for 30), which just shows even more how far there is to go.

Funny I use the UE samaritan demo and Crysis DX11 (runing on HW many times more powerfull than PS360) to prove the oposite point:smile: They really have a hard time impressing me and I probably would hardly notice any difference if they didnt point side by side them.

Anyway a gfx/visual oriented user will probably notice them a lot easier than me (like I do in Audio), the big question is if there is many users noticing and caring enough to not buy the worst gfx games.

 

[bomlat]

There's no evidence of this. Their just seems to be a cadre of Nintendo fans who have adapted this claim since the announcement of the Wii U, but I cant agree at all. I think it's more like wishful thinking.

If next gen comes out, has 8X better specs, and the games honestly dont look that much better I'll agree. However, I am almost certain that will not happen. There's no way a console with 4GB of RAM isn't going to produce vastly better graphics than one with 512mb, to narrow it down to one specific for the sake of illustration.

I suppose I begrudgingly accept the diminishing returns argument eventually, but I've never seen any evidence that we're near it yet. Each console gen looks stunningly better than the last to date. And I see enough evidence (UE3 next gen demo, Crysis 2 DX11 pack, BF3 PC) already that next gen consoles will provide a vast improvement. Just look at all the complaining over Crysis 2 console's "reduced LOD, 25 FPS, no AA etc etc whatever". Now imagine a Crysis+Crysis 2 dual pack, 1080P 60 FPS all ultra high settings for just 39.99 (since it's a remake) for next gen consoles and drool Hell, you can probably argue whether even next gen consoles could pull off the Crysis games with AA, 1080P, 60 FPS etc (think they might have to settle for 30), which just shows even more how far there is to go.

I just played on the COD:BO with split screen with my son,and I have to say that the main reason of the graphics is not connected to the power of the machines,but to the money that they spend for the graphics.
The transformers (that they like very much,because there is many on every level) are not detailed enough,and even the visible details are wrong ( they showing plates that interconnecting the high voltage side bushing ),the radiators are not detailed,the Bucholz relay completely missing with the step changer.
On the level where there is an R-7 rocket the whole structure unrealistic,the support pillions and the service level of the R-7 completely missing,the main part of the pipework missing,and the remaining part is not make any sense.
The same for the level with the radar,the mechanism that supposed to move it is way away from any realistic rotating machine.
There was a level with a few lathes,and all of them badly designed, with design which not make sense,or with missing parts.

However at the end of the day, how much would it cost to draw all this details correctly, and who could be interesting in them?
It is not simply just the hardware limitations: obliviously the guy who designed all of this details doesn't have the slightest idea how should look like a lathe,or a transformer.
So even if you give more computational power,the trouble will be that to fill it up with details.The cost of the CODBO is around 30 million $.The required detail level could push it up to the 100 million level.Will it worth it?

 

[mikiex]

I just played on the COD:BO with split screen with my son,and I have to say that the main reason of the graphics is not connected to the power of the machines,but to the money that they spend for the graphics.
The transformers (that they like very much,because there is many on every level) are not detailed enough,and even the visible details are wrong ( they showing plates that interconnecting the high voltage side bushing ),the radiators are not detailed,the Bucholz relay completely missing with the step changer.
On the level where there is an R-7 rocket the whole structure unrealistic,the support pillions and the service level of the R-7 completely missing,the main part of the pipework missing,and the remaining part is not make any sense.
The same for the level with the radar,the mechanism that supposed to move it is way away from any realistic rotating machine.
There was a level with a few lathes,and all of them badly designed, with design which not make sense,or with missing parts.

However at the end of the day, how much would it cost to draw all this details correctly, and who could be interesting in them?
It is not simply just the hardware limitations: obliviously the guy who designed all of this details doesn't have the slightest idea how should look like a lathe,or a transformer.
So even if you give more computational power,the trouble will be that to fill it up with details.The cost of the CODBO is around 30 million $.The required detail level could push it up to the 100 million level.Will it worth it?

It is a big mix of things more complex than just money, but the power of the machine is very important. Remember playing split screen it may be totally different stripped down as you are drawing twice as much. More power takes pressure off optimizing leaving artists more free.

Often you strip things out because you don't have time to optimize, time = money, but less optimizing = more time.

The more interesting tech you add can create more problems, so if WiiU is DX10 class people may not be able to create interesting tech, but will be able to run with similar engines to the xbox quickly but hopefully with more power.

 

[Entropy]

I just played on the COD:BO with split screen with my son,and I have to say that the main reason of the graphics is not connected to the power of the machines,but to the money that they spend for the graphics.
The transformers (that they like very much,because there is many on every level) are not detailed enough,and even the visible details are wrong ( they showing plates that interconnecting the high voltage side bushing ),the radiators are not detailed,the Bucholz relay completely missing with the step changer.
On the level where there is an R-7 rocket the whole structure unrealistic,the support pillions and the service level of the R-7 completely missing,the main part of the pipework missing,and the remaining part is not make any sense.
The same for the level with the radar,the mechanism that supposed to move it is way away from any realistic rotating machine.
There was a level with a few lathes,and all of them badly designed, with design which not make sense,or with missing parts.

However at the end of the day, how much would it cost to draw all this details correctly, and who could be interesting in them?
It is not simply just the hardware limitations: obliviously the guy who designed all of this details doesn't have the slightest idea how should look like a lathe,or a transformer.
So even if you give more computational power,the trouble will be that to fill it up with details.The cost of the CODBO is around 30 million $.The required detail level could push it up to the 100 million level.Will it worth it?

I don't have time to be on these forums now, so can't follow up, but - in some cases designers are putting in too much where their knowledge is limited, simply because the eye likes details. So armor, for instance is outlandishly gnarled, and not at all designed to let an actual attack smoothly slide off.

Designers spend effort on things that (they think) look cool or (they think) their customers like. It affects modelling and rendering both. That's why you see bizarre levels of detail and work on armor and weapons, and boobie physics, but less on interactive environments. That's why you have DOF and motion blur, even though neither makes sense for games, but lousy frame rates - control and smoothness doesn't translate as well to the promotional material.

Think in the context of marketability - to ten year olds - and many things about the games industry becomes clear.

 

[-tkf-]

More rumours

http://www.hardocp.com/article/2011/07/07/e3_rumors_on_next_generation_console_hardware/

 

[liolio]

However at the end of the day, how much would it cost to draw all this details correctly, and who could be interesting in them?
It is not simply just the hardware limitations: obliviously the guy who designed all of this details doesn't have the slightest idea how should look like a lathe,or a transformer.
So even if you give more computational power,the trouble will be that to fill it up with details.The cost of the CODBO is around 30 million $.The required detail level could push it up to the 100 million level.Will it worth it?

Well you have also to take in account the saving allowed by new engines in regard to content creation. Things like real time global illuminations will save time for artists as they don't have to tweaks I don't know how many lights to achieve the aspect they want. I don't expect artists to get everything right but clearly there is a lot of room to do more out of existing assets.
This whole content development costs discussion is getting a bit excessive imho.

 

[liolio]

More rumours

http://www.hardocp.com/article/2011/07/07/e3_rumors_on_next_generation_console_hardware/

AMD getting all the deal for all three manufacturers sounds pretty believable but I'm pretty surprised by their talk about CPUs

 

[Rangers]

It is highly likely that Microsoft's 3rd generation Xbox will be sporting a new IBM cell processor as well although it is slightly, ever so slightly possible, that this could change

LOL

 

[holsty101]

LOL

Yeah, wasn't there some talk of "cell exclusive to sony for console use" or something like that back around launch time?

 

[Prophecy2k]

The Next MS Xbox having a CELL processor?

Well, they already had the PPUs anyway. All they were missing was the SPUs ;-)

 

[Tea2]

To people who follow this close: based on the rumours we have so far, what are the chances of X360 getting full backwards compatibility? Still not enough info/too early to tell? Is there anything in particular about the 360's design that could be a potential issue when it comes to maintaing b.c. in the future? I've heard a lot about xbox developers not being allowed to "code to the metal" and all that (directx api). Does this guarantee a proper emulation of the hardware next gen or is there still something than can "go wrong" in this regard? Not being able to play the games I already own (preferably with some visual upgrades:smile would be a huge deal-breaker to me.