-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Digital Foundry - In Theory: Could Sony release PlayStation 5 in 2018?
- Thread starter Vashetti
- Start date
GinoFelino
Member
Conclusion: as much a solid no as anyone with a functioning brain was expecting
Nestunt
Member
This is a good argument. If they want to keep selling PS5 Pro, PS6 Pro, etc.. they have to show consumers how not obsolete it is.
I'm all for iterative jumps using the same tech. Has incredible benefits of making it easy for devs as well as backwards comparability.
However I hope Sony doesn't jump the gun too much. I don't have a PRo but I'm happy enough that it exists. My expectations have been that every 3 years we get jumps. So I'm hoping PS5 is not until 2019.
However I hope Sony doesn't jump the gun too much. I don't have a PRo but I'm happy enough that it exists. My expectations have been that every 3 years we get jumps. So I'm hoping PS5 is not until 2019.
potatohead
Member
lol, didnt think about it like that.
Its gonna happen like the PS3 to PS4. It might come right before the PS5. Last of Us 2 also.
RyanHardy101
Banned
Agreed. The great games are coming this year and in 2018/19, I would be happy with a PS5 only in 2021/2022, I would be fine with a PS5 in 2020 if the technological leap justifies it. 2018 just seems so rushed.
Haha no. Developers are struggling as is with development costs inflating and the time it takes to actually bring a game out ever increasing. As other's have stated there are some who haven't even released one game let alone two.
I think what is going to happen is Sony and Microsoft are just going to release updated hardware under the umbrella of their already existing brands. On top of that we'll see VR and AR added in to increase the types of experiences a player can achieve.
I think what is going to happen is Sony and Microsoft are just going to release updated hardware under the umbrella of their already existing brands. On top of that we'll see VR and AR added in to increase the types of experiences a player can achieve.
ethomaz
Banned
PS5 will be TSMC 7nm.
That means 2019/2020.
There is no beneficies in make a new consoles in 16nm FF+... Scorpio is probably close to max limitation of process.
7nm is supposed to be 60% smaller than 16nm (bigger difference than 28nm to 16nm) and 70% more power efficiency.
In theory we are looking at APU with same size of Scorpio's APU:
+ 64 CUs, 4096 SPs
+ 1800Mhz
That will give us about 14.7 TFLOPs.
With the space save in CPU and AMD advances in GPU (GCN 6?) it is probably to reach even more CUs and clocks... so 20 TFLOPs is not out of question for APUs in 7nm.
That means 2019/2020.
There is no beneficies in make a new consoles in 16nm FF+... Scorpio is probably close to max limitation of process.
7nm is supposed to be 60% smaller than 16nm (bigger difference than 28nm to 16nm) and 70% more power efficiency.
In theory we are looking at APU with same size of Scorpio's APU:
+ 64 CUs, 4096 SPs
+ 1800Mhz
That will give us about 14.7 TFLOPs.
With the space save in CPU and AMD advances in GPU (GCN 6?) it is probably to reach even more CUs and clocks... so 20 TFLOPs is not out of question for APUs in 7nm.
Sony will stretch out this generation as long as it can, it'll be Microsoft pushing for a reset where it can be on more even ground than it is now. Considering the timing of the Scorpio though, surely Microsoft will leave the Scorpio on the market for at least 24 months until they reset?
2021 looks far more likely for the next generation.
2021 looks far more likely for the next generation.
PLASTICA-MAN
Member
Exactly. And like I said in a previous post that Sony will continue this trend by releasing the PS5 with Native 4K as base resolution while releasing a midgen PS5 PRO allowing some native 8K games (remasters or non heavy PS5 games) just like what is happenong now.
jobrro
Member
I disliked the PS360 span because by the end of the gen the consoles were ancient. I feel like a ~7-8 year span with a refresh in the middle (Pro, Scorpio) is ideal. Those who want higher IQ get it, those who just want the games can ignore the upgraded systems.
2018 feels too soon for a PS5, especially after a 2016 Pro launch. Generally a gen upgrade is approximately 10x the GPU power and 16x memory. Maybe the PS5 won't need 128GB RAM even if it is launched in 2020s, maybe 32GB will suffice, but for 4K displayed I feel the GPU power need to be there. We are getting PS4 games at 4K-ish on the Pro, should be PS4 quality games at native 4K on Scorpio but that about the thermal limit for a console for 2017. Unless there is a big unforeseen jump I don't see something being about to hit even close to 18TF in a console thermal envelope in the next couple of years.
The PS4 was approximately 10x the GPU of the PS3, some of that went into a generally 1080p frame buffer over 720p of the previous gen, but I remember people who couldn't tell the difference early on in the gen. Hell I even remember people calling the Xbox 360 an Xbox 1.5, but the Xbox 360 was leagues ahead of OG Xbox especially once developers took true advantage of it.
I don't think Sony or MS want to put out a new generation console where the masses can't tell the difference in the games. Would be bad PR. Better to iterate on the current platform as a premium SKU for the hardcore and make a significant jump when they can.
2018 feels too soon for a PS5, especially after a 2016 Pro launch. Generally a gen upgrade is approximately 10x the GPU power and 16x memory. Maybe the PS5 won't need 128GB RAM even if it is launched in 2020s, maybe 32GB will suffice, but for 4K displayed I feel the GPU power need to be there. We are getting PS4 games at 4K-ish on the Pro, should be PS4 quality games at native 4K on Scorpio but that about the thermal limit for a console for 2017. Unless there is a big unforeseen jump I don't see something being about to hit even close to 18TF in a console thermal envelope in the next couple of years.
The PS4 was approximately 10x the GPU of the PS3, some of that went into a generally 1080p frame buffer over 720p of the previous gen, but I remember people who couldn't tell the difference early on in the gen. Hell I even remember people calling the Xbox 360 an Xbox 1.5, but the Xbox 360 was leagues ahead of OG Xbox especially once developers took true advantage of it.
I don't think Sony or MS want to put out a new generation console where the masses can't tell the difference in the games. Would be bad PR. Better to iterate on the current platform as a premium SKU for the hardcore and make a significant jump when they can.
Not true at all, transitioning from Bulldozer CPUs will make a jump even in 16nm a highly relevant one, plus the RAM limitations of consoles have always been pretty significant, more memory would alleviate significant bottlenecks. But I don't disagree console makers will want a full cycle to monetise their current install bases properly, after not making much money for the past 3 years while their install base has ramped up.
Panajev2001a
GAF's Pleasant Genius
We do not have consumer GPU's based on Volta. NVIDIA showed an early sample of an HPC card.
Marmaladefire
Member
I really want to play Kingdom Hearts III one day, so let's not. Plus the jump probably wouldn't be remotely worth the price.
it's funny to see the "5 year" cycle being brought up since that has been mostly for Nintendo Consoles.
the only PS console that lasted a similar timeframe was the PS1. PS2 had more than 6 years between the time it released and when the PS3 came out. in japan, the difference between the PS2-PS3 timeframe and PS3-PS4 timeframe is only 3 months (yeah, the "absurdly long" last gen lasted almost as long as the PS2 gen in japan. in the rest of the world, the PS2 had just one less year than the PS3 since launch until the sucessor was out.)
i don't think they'll release anything before late 2019. the only thing i could see pushing them ahead is the Switch eating the PS4's lunch overnight and outpacing it by a big margin everywhere.
the only PS console that lasted a similar timeframe was the PS1. PS2 had more than 6 years between the time it released and when the PS3 came out. in japan, the difference between the PS2-PS3 timeframe and PS3-PS4 timeframe is only 3 months (yeah, the "absurdly long" last gen lasted almost as long as the PS2 gen in japan. in the rest of the world, the PS2 had just one less year than the PS3 since launch until the sucessor was out.)
i don't think they'll release anything before late 2019. the only thing i could see pushing them ahead is the Switch eating the PS4's lunch overnight and outpacing it by a big margin everywhere.
GhostTrick
Banned
It already does. Hence why you get concessions on framerate, sometimes IQ and shadow. Sure, you can make some improvements, it's a different matter.
I don't think we'll reach 20tflops in the console space in 2019/2020 from AMD. VEGA, which is for this year, is supposedly reaching 12Tflops. That'd be a 60% increase from their previous line, in 2 years.
Fury X was 8Tflops in 2015 and was their most expensive offering at +600€ (and it was watercooled) and 28nm.
VEGA is 12Tflops in 2017 and is supposed to be their most expensive offering at the moment and 14nm. It's a 50% increase.
If we follow that 50% increase in 2019 with a lower process node (7nm), that'd be 18Tflops for their most expensive GPU at the time.
family_guy
Member
Rockstar would release GTA 6 by 2019 on PS4 assuming its in development. They wouldn't just skip it for PS5. Cross gen development will be happening well through 2020.
Rumblebones
Banned
After he last generation I was adamant this one would have to be in 4-5 years range, especially when I still gamed primarily on an XB1, but I since moved to PC gaming and got a PS4 Pro and now I can't imagine why Sony would want to kickstart a new gen when they're on their way to a 100m+ seller and released the Pro less than a year ago. Unless maybe they think the revenue/profit boost a PS5 would generate would be worth it. I just don't see it.
wesleyshark
Banned
Simply based on how well the ps4 is selling, it would make no sense to release a new generation console next year.
BlitzerRadic
Member
To test consumer response to iterative hardware. Sony has a history of doing stuff like that.
Anyways, I don't think 2018 is unrealistic. Remember PS2? Sony waited 6 years and MS launched 360 a year earlier... that didn't turn out so well for Sony.
Agent Gibbs
Member
2018 is about 2 years too soon! The PSVR is barely out, the PRO is barely out at all, its stupid, i think most of the millions of people who currently own a PS4 would not be making an upgrade any time soon either.
On this Digital Foundry are reaching or i hope they are
On this Digital Foundry are reaching or i hope they are
There is no need for this in 2018. Late 2019 is the earliest that PS5 should arrive, giving PS4 a 6 years on the market. Late 2020 is even more likely.
As for the ram, Hynix has announced arrival of 2GB GDDR6 chips in 2018. Sixteen of those [same ammount that PS4 had on its launch mobo] would make more than respectable 32GB of ram, and the speed should be around half a terabyte per second.
As for the ram, Hynix has announced arrival of 2GB GDDR6 chips in 2018. Sixteen of those [same ammount that PS4 had on its launch mobo] would make more than respectable 32GB of ram, and the speed should be around half a terabyte per second.
SolidSnakex
Member
The PS3 launching later wasn't the issue. The PS3 launching later and struggling with multiplatform games, general delays, and a higher price were the issues.
miller.skippins
Banned
i feel that the OG PS4 hasn't even peaked yet........just look at Horizon.
I can easily see another 3 years before Sony unveil PS5.
Also, there's no reason to get into a TFLOP arms-race with MS at this point as Sony is selling at a profit per box in addition to the digital software sales and PS+ sub....
I can easily see another 3 years before Sony unveil PS5.
Also, there's no reason to get into a TFLOP arms-race with MS at this point as Sony is selling at a profit per box in addition to the digital software sales and PS+ sub....
Jack The Nipper
Banned
Two to three times power is what they would have to shoot for tomorrow make noticeable strides in visuals right now and id feel a little closer to four times would be better.
15 tf plus a significant upgrade on cpu should definitely do it.
I could definitely see more than 16gb of ram but not sure what the system may actually need.
We would probably be needing to be near 10tb of storage but 5tb would probably be more standard.
Games are likely to hit 200gb soon and games at higher resolutions at much higher quality settings on new hardware will likely be much higher than that.
I wouldn't be surprised to see 1tb games in ten years of not five. Id assume they will have to come up with special hardware compression techniques until hdd space can go far beyond what it is now. Hdd space has never had such a low size in comparison to content I believe.
Seems some new hdd or storage tech needs to be crafted that is not only huge but affordable.
TechJunk
Member
The thing is, back in the 90's/early 00's, technology was advancing so quickly that 5 years meant so much more computing power... look at the difference between NES and SNES games for example.
And now look at how little Intel has advanced in terms of actual computational power... people still rocking I5 2500K's for a gaming rig, a CPU that was released what... 6 years ago?
I agree it's a long time to wait, but I have a PS4 Pro and I'm not at the point yet where I see frame stutter and am disappointed in the visual quality of the games.
I think 2020 would be a logical time for a next generation.
ethomaz
Banned
14/16nm to 7nm is a bigger shrink than 28nm to 14/16nm.
There is a 60-70% decrease in size... no 50%.
15 TFs is the minimum in 7nm with similar die size of actual consoles... it could ends better than that.
Said that 7nm can delay but I'm confident in late 2019 or 2020 release.
CJY
Banned
If Sony want to get back into being in the realm of the top-dogs in tech, alongside Apple, Google, MSFT, they need to get into the r&d of quantum computing.
It may be a little early on the timeline, but the earlier they get in, the more potential there will be for them to have the technology to drive development of PlayStation iterations until the mythical "9".
That's only 5 generations away, and it we believe what Mark Cerny is saying, and the belief in the "generational" model still exists on their roadmap, that's only 20-25 years out. Can you imagine the progress that will be made with d-waves quantum computers in that time?
With current processor technology development slowing to a crawl, it's seems unreasonable to expect an 8X increase in power from the previous gen going forward.
At the very least, Sony should purchase a D-Wave and throw some engineers at it to see what they can come up with. With that foundation, maybe enter into an arrangement with Samsung to develop quantum computers themselves, or with funding from the Japanese government.
If anyone has any doubt about quantum computing, it's viability, or its potential, see this video explaining a lot:
https://youtu.be/PqN_2jDVbOU
Take a look at Apple; while other companies such as Samsung, HTC and Google are using chipsets (SOC) from Qualcomm, Apple bought an Isreaili company called P.A Semi back in 2008 and have since been designing and using their own chips in their iPhone and IPad products, since the A4 chip introduced with the original iPad. They're on the cusp of introducing their A11 chip with the iPhone 8. Apple are on a mission of owning all the relevant technologies in their products and that is one major factor of what makes them as successful and their products more powerful and efficient than their competitors.
I miss the days of Sony developing their own chips, the last of which I believe was the emotion engine of the PS3. Granted, this was a chip that almost bankrupted the company, but the top brass at some point must have approved it and saw it as the correct way to progress.
Quantum Computing is something completely different and it's applications are currently unimagined. Video games and Sony's industry-leading position offers them the opportunity to invest in quantum computing, using Videogames as a justification, while continually developing the technology that will likely bring untold changes to our lives in the very near future.
Tl;dr Sony needs to do R&D on Quantim Computing STAT!
Edit: saw another video on Computing power: https://youtu.be/JhHMJCUmq28. Obviously, everyone knows we are hitting the physical limits of of current technologies. It's with this in mind that makes it difficult to imagine how we are going to get to a PS9 in our lifetimes
It may be a little early on the timeline, but the earlier they get in, the more potential there will be for them to have the technology to drive development of PlayStation iterations until the mythical "9".
That's only 5 generations away, and it we believe what Mark Cerny is saying, and the belief in the "generational" model still exists on their roadmap, that's only 20-25 years out. Can you imagine the progress that will be made with d-waves quantum computers in that time?
With current processor technology development slowing to a crawl, it's seems unreasonable to expect an 8X increase in power from the previous gen going forward.
At the very least, Sony should purchase a D-Wave and throw some engineers at it to see what they can come up with. With that foundation, maybe enter into an arrangement with Samsung to develop quantum computers themselves, or with funding from the Japanese government.
If anyone has any doubt about quantum computing, it's viability, or its potential, see this video explaining a lot:
https://youtu.be/PqN_2jDVbOU
Take a look at Apple; while other companies such as Samsung, HTC and Google are using chipsets (SOC) from Qualcomm, Apple bought an Isreaili company called P.A Semi back in 2008 and have since been designing and using their own chips in their iPhone and IPad products, since the A4 chip introduced with the original iPad. They're on the cusp of introducing their A11 chip with the iPhone 8. Apple are on a mission of owning all the relevant technologies in their products and that is one major factor of what makes them as successful and their products more powerful and efficient than their competitors.
I miss the days of Sony developing their own chips, the last of which I believe was the emotion engine of the PS3. Granted, this was a chip that almost bankrupted the company, but the top brass at some point must have approved it and saw it as the correct way to progress.
Quantum Computing is something completely different and it's applications are currently unimagined. Video games and Sony's industry-leading position offers them the opportunity to invest in quantum computing, using Videogames as a justification, while continually developing the technology that will likely bring untold changes to our lives in the very near future.
Tl;dr Sony needs to do R&D on Quantim Computing STAT!
Edit: saw another video on Computing power: https://youtu.be/JhHMJCUmq28. Obviously, everyone knows we are hitting the physical limits of of current technologies. It's with this in mind that makes it difficult to imagine how we are going to get to a PS9 in our lifetimes
oldergamer
Member
It makes no sense for sony to release another console 2018. Not after launching the pro last year. It makes more sense to wait until they can make a much bigger leap and avoid fracturing the PlayStation market. they also need to offer a reason for current PS owners to upgrade.
ThoseDeafMutes
Member
2019 is the right year in terms of the generation being "the right length", not as long as the horrendously drawn out previous generation, but not so short that games didn't have time to set themselves up.
2019 won't be where they want it, technically. The mid-gen refreshes only make sense if they're planning to draw it out longer than 2018, and probably more like 2020-2021. Even if we assume sub-native 4k is the target, as it is on PS4 Pro, that still means that you need essentially 16TF to get a 4x increase over PS4 Pro. If you say "the next gen will do 4k native" the comparisons get even worse since a portion of that 4x figure will be eaten by the increase in rendering resolution, and the actual power left over for other graphical things will be highly underwhelming for a generational transition.
In 2017, we have Nvidia chips doing 14tf. In 2018, we'll probably end out the year with the "Titan" level Nvidia chip pulling 20tf if it over-clocks well. But in terms of when that level of power will be available on a $399 console, the answer is 2020 and beyond.
2019 won't be where they want it, technically. The mid-gen refreshes only make sense if they're planning to draw it out longer than 2018, and probably more like 2020-2021. Even if we assume sub-native 4k is the target, as it is on PS4 Pro, that still means that you need essentially 16TF to get a 4x increase over PS4 Pro. If you say "the next gen will do 4k native" the comparisons get even worse since a portion of that 4x figure will be eaten by the increase in rendering resolution, and the actual power left over for other graphical things will be highly underwhelming for a generational transition.
In 2017, we have Nvidia chips doing 14tf. In 2018, we'll probably end out the year with the "Titan" level Nvidia chip pulling 20tf if it over-clocks well. But in terms of when that level of power will be available on a $399 console, the answer is 2020 and beyond.
I don't understand why people think this generation hasn't suffered from off-the-shelf hardware. I have a PS4 Pro and it's pretty weak compared to what I imagined it to be. I agree with Cerny, for hardware to be considered a generational leap it needs to be bespoke (although Microsoft threw power at it to try and get it to work).
In that case I welcome a whole new bespoke architecture, maybe even as mad as Ken's designs, to really push boundaries and for that reason I don't expect a PS5 until 2020 at the earliest.
Come on Sony...take your time and really push the boat out.
In that case I welcome a whole new bespoke architecture, maybe even as mad as Ken's designs, to really push boundaries and for that reason I don't expect a PS5 until 2020 at the earliest.
Come on Sony...take your time and really push the boat out.
Similar threads
- Locked