Next Generation Hardware Speculation with a Technical Spin [2018]

[iroboto]

Wait. Sorry it’s hard to keep up. Now we’re back to a release prediction of 2019? That’s like the day after tomorrow. Too early! Get a Pro, enjoy the stupidly amazing games that are still coming out. Why so soon?

A lot of information is being released quickly from all areas of the industry, which is why the rumours of 2019 are sparking back up again.
If MS is making noise, Google is making noise, Sony is making noise, people are expecting them to move forward now than say later.
It's not definitive, but it's right now the strongest indicator that a 2019 launch is in the cards. I personally think a 2019 announcement makes sense. Still not sure on the launch.

 

[London Geezer]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

 

[iroboto]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

It probably does, but it becomes a question of whether they want it to be a couple more years of PS4/XBO delivering these games solo, and then spend the next 2 years transitioning to next gen? Or transition sooner and be able to still be relevant during a longer transition period.

 

[chris1515]

https://segmentnext.com/2018/10/09/call-of-duty-2019-cross-gen/

The next Infinty Ward Call of duty seems to be a cross gen game.

 

[Metal_Spirit]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

I understand what you are saying, but that's tech. If you sit still, others will move, and you must move too. Besides, production costs do not keep decreasing, and at a point they will in fact increase!
28 nm and 14 nm are now obsolete products that are holding production lines. Have you checked Sonys latest numbers? The operating profits chart from FY 2017 refers "Hardware units sales decline and cost increase" as one of the main factors for profits drops from FY 2016.
Although this is a Portuguese website, you can check the charts here.
But its also a fact that FY 2017 was a year of record sales of the PS4 (check here), so this reference is in fact only refering to cost increase of hardware units, and not any real sales decrease.
This shows that the time to launch new consoles is near. 2019 is a possibility! In fact, speaking of new tech, there will be nothing worthy of note that may be ground breaking launching in 2020, and that could be used on consoles keeping prices atractive, so 2019 may be a special strategic window of oportunity.

 

[Jay]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

There's always a couple years of cross gen that goes on.
they don't stop selling them, and games are still made for them, but the base consoles are already struggling especially the xo.
xo will limp over the generation cross over, looking like it would be more humain to put it down.

 

[McHuj]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

I think PS4 will enjoy something close to a PS2 like life post PS5 release, a long tail of sales with some big games still to be announced and released for it.

Whether it's 2019 or 2020, I'm not convinced that PS5 will sell as well as PS4 did out of the gate. I kind of expect it to be a slower burn so I think it's ok that Sony release in 2019 and slowly build the user base.

 

[Tkumpathenurpahl]

Financial Times has also mentioned a rumor of a Playstation tablet:
"There have been industry rumours that Sony might be planning a tablet that would connect to multiple devices as online streaming of games becomes more widespread and the company seeks to derive ever greater synergies between its games and movie divisions."

So, miniaturized PS4 in a portable form factor [7nm APU, full PS4 speed in docked mode, reduced PS4 speed in handheld mode] or regular android tablet with support for Sony's new multimedia projects.

Advanced remote play [flawless like that from WiiU, this time with dedicated wireless hardware [PS4 [badly] shares its single WiFi chip between RP and the internet connection]], or native game rendering?

I think the best thing Sony can do is release their own type of Fire tablet: a cheap device with HDMI out via USB C, giving people an easy way of accessing the store, Remote Play, and PSNow.

Some low powered hardware that's capable of some apps and games would be nice, especially if they're compatible with the PS4.

 

[Deleted member 11852]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

Sony boarded the PS5 R&D train some years back, they can't just slow it down and add some extra carriages (features, capabilities) because PS4 is doing well.

Complex CE devices like a console which have 24/7 production runs and bespoke chips need fab and production capacity booked way in advance. If Sony do plan on PS5 for 2019 (which I believe) the chip designs are pretty much taped out at this point so delaying it 12 months is just going to make that specification machine launch a year later when it will be less impressive than it launching on time. Adding that extra 4Gb to PS4 was a relatively easy change to the board thanks to GDD5 clamshell mode but the idea of Sony/AMD lowering the heat and adding features to a chip so it's equally competitive when it does launch isn't a thing that happens in the real world.

 

[phoenix_chipset]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

Oh god yes, it's so stupid that we get the ps5 in 2019 esp. since how slow tech is advancing. To me the whole point of the pro is to stretch the gen out! Graphically there's still gas in the tank. Maybe devs are getting sick of cpu limitations? On the gpu and ram side, a new gen right now seems like a waste.

On pro, i'm still amazed by games like ratchet, shadow of the colossus, UC4 etc.

 

[Shifty Geezer]

I think PS4 will enjoy something close to a PS2 like life post PS5 release, a long tail of sales with some big games still to be announced and released for it.

PS2 had such a long life because it got so cheap, I think, coupled with specialist titles like SingStar that brought in particularly demographics. I can't see PS4 doing that well. The price likely won't get that low, and casual gamers are served by mobile these days.

 

[DieH@rd]

I just find it hard to believe that the current hardware doesn’t have another couple years at least in it, seeing the quality and tech achievements of the software being released.

Well, same could be said for PS3 in 2013.

It got Ascension, TLOU2, Beyond Two Souls and GTA5 [and GT6 in early 2014], all among the best looking titles of the generation.

To me the whole point of the pro is to stretch the gen out!

IMO, the point of the Pro was to over visual bump for 4K TV owners and nothing more.

 

[Deleted member 13524]

Recent rumors come from those resetera posters who claim Sony had a PS5 ready for 2019, but they looked at the market and at some point entertained the idea of postponing the launch to 2020. Right now with ongoing pressure from Microsoft they're thinking of going back to 2019, but it hasn't been decided yet.

They’ll have a hard time with memory bandwidth for a minituarized PS4. I think they’d have to convert to WideIO to try and maintain the GDDR5 bandwidth.

LPDDR5 is planned for 2019 and will clock at 6.4MT/s. Original PS4's GDDR5 was 5.5MT/s. They could pick up low-binned LPDDR5 to operate at the same 5.5MT/s.
Of course a 256bit bus is heavy for a mobile system, but it's not like people are expecting 8 hours of battery life in a mobile console.
Another alternative would be 96bit GDDR6. I don't know which would be more efficient.

 

[McHuj]

PS2 had such a long life because it got so cheap, I think, coupled with specialist titles like SingStar that brought in particularly demographics. I can't see PS4 doing that well. The price likely won't get that low, and casual gamers are served by mobile these days.

Yes, it won't sell 150+ million like PS2, but I think it will surpass PS1 which was around ~102M. I think even with PS5 coming out potentially in 2019, PS4 still has room to sell another 30M units in the next 2-3 years which would put it in the 120M ballpark. If they want more, they'll have to come up with some cost cutting, maybe a really ugly top loading PS4.

 

[phoenix_chipset]

Well, same could be said for PS3 in 2013.

It got Ascension, TLOU2, Beyond Two Souls and GTA5 [and GT6 in early 2014], all among the best looking titles of the generation.

Ehh all those games were really showing their age even at that point. Except maybe beyond, but it's hardly a game and has very limited interactivity. After 2011 those consoles were pretty much tapped, even 360.

Gta V dips into the 20fps region on ps3, with plenty of pop in and low res textures. Still impressive all things considered, but the consoles were kicking and screaming trying to run games post 2011.

But look at the games coming out on ps4, and the high profile ones do a really great job of keeping solid framerate, and have great image quality on the pro. X just got 4H4. There's a limitation in game design happening on current gen, but not graphically. Well much less than those late 7th gen games had anyway.

 

[Gemüsepizza]

But look at the games coming out on ps4, and the high profile ones do a really great job of keeping solid framerate, and have great image quality on the pro.

Huh? There are already big performance issues on base consoles. A recent example, Assassin's Creed Odyssey:

The framerate dips to 26fps on PS4, and 23fps on Xbox One. And this is with dynamic resolution which goes down as low as 1344x756 on Xbox One and 1600x900 on PS4! Even the mid-gen consoles are struggling a bit, according to DF, the PS4 Pro can get down to 2227x1242 in demanding scenes, which is only 33% of 4K, and barely above 1080p.

Current gen consoles are simply showing their age, and with all those new tech improvements, Sony/MS could deliver a big jump. I don't mind new consoles every 3 years, I want to run my games on fast, new hardware.

 

[anexanhume]

LPDDR5 is planned for 2019 and will clock at 6.4MT/s. Original PS4's GDDR5 was 5.5MT/s. They could pick up low-binned LPDDR5 to operate at the same 5.5MT/s.
Of course a 256bit bus is heavy for a mobile system, but it's not like people are expecting 8 hours of battery life in a mobile console.
Another alternative would be 96bit GDDR6. I don't know which would be more efficient.

I question whether 8 discrete chips is practical for a tablet form factor. They certainly can’t stack them PoP.

 

[phoenix_chipset]

I don't think of AC as impressive though. Impressive for a yearly factory line game maybe.

Uc4, gears 4, shadow of the Colossus, 4h4, spiderman and ratchet and clank all run like butter 99 percent of the time. The heavy hitting console exclusives run great.

 

[Deleted member 13524]

I question whether 8 discrete chips is practical for a tablet form factor. They certainly can’t stack them PoP.

I don't know about the others, but Samsung sells 64bit (2*32bit) stacks of LPDDR4X.

A PS4 Go solution could use 4x64bit LPDDR5 stacks around the SoC in a single MCM, like they did with the RSX + GDDR3 on the PS3 Super Slim:

Again: I don't know if a 3-chip 96bit GDDR6 solution would be more efficient. For connected standby it probably wouldn't, though. There's also the issue with memory density, as there isn't any GDDR6 chip with more than 2GBytes at the moment, AFAIK.

Bottom line is 176GB/s wouldn't be that hard to achieve with a low-power-ish budget.

 

[MrFox]

GDDR6 parts support a slightly reduced clock at 1.25v instead of 1.35v, so 11gbps with four chips would be sort of low power. Lpddr5 would be better.

But there's no way the ps4 SoC would have a usable wattage on 7nm, not for a portable console. It would need to be about 4 times less power than the 14nm slim?