This will be the last one in the form of what we've known it imo.Only 3 more until PS9; can they make that commercial a real simulated reality?
This will be the last one in the form of what we've known it imo.Only 3 more until PS9; can they make that commercial a real simulated reality?
It doesn't matter if they kill GaaS, the unique quirky games from Sony that some of us loved are never coming back, the industry has completely changed. Instead of getting GaaS you won't be getting the next Sotc, those type of games don't sell enough anymore for what games cost to make. You will just get another lame ass Marvel game now. It's all catered to the masses and the almighty dollar now.![]()
A change in business models you say?
![]()
I keed I keed.
But theyve already abandoned the GaaS Initiative, what other "change in business model" could they be talking about outside of doing ports?
We kinda already know about the PS6 Handheld, PS6S and PS6 Proper.........is that what hes talking about as it isnt exactly common knowledge they "may" have 3 SKUs.
Maybe reducing the budgets and scope of their first party titles.......considering the PS5 generation is almost over and weve not had all their 1st parties release a new product yet, maybe they need to scale down so we get back to smaller more often releases......ask Insomniac how they manage to get so many games out within a small time frame and budget.
pushing it one more year wouldnt actually be a bad idea. But as others more credible has said it would be the same hardware2028-2030
pushing it one more year wouldnt actually be a bad idea. But as others more credible has said it would be the same hardware
I agree with you I think by then the memory and storage situation could possibly have improved. But they could launch in late 2027 or 2028 still and stick it out with lower sales tbh and lower production volumes until summer 2029 when they can drop the prices to something more affordable."Summer 2029" thats possibly the best time to sell ps6 at a much affordable price range and AI components demands is highly possible have stabilize by then.
They cant meaningfully update the hw and release before 2030. And the prices are really driven by speculation and demand for AI so they will come down. Consider that memory prices have gone up by 10x in some cases, not because there is anything technologically better but because of hoarding. So the only option is to either delay and release the same hw at the right prices or launch at low volume production until the prices come back to normal, which they will.The problem is not the hardware it uses, it's the price for the hardware.
honestly a deepseek moment for memory from CXMT and flash from YMTC would save everyone. If China can flood the global market with cheap ddr5 and SSDs then it would really make AI inference much more accessible. Right now the big players are hoarding anything they can get after Altman's antics last year.Eventually of course it will
I don't
"Too big to fail" arguments in 2026?
memory is still a commodity. there is nothing meaningfully different from ddr5 made by SkHynix or Samsung or Micron, or when CXMT starts getting it done right and in the large quantities. At some point we are going to move to DDR6 and DDR5 fabs will be plentiful still. I dont see how memory prices can remain so speculatively high even when datacenter build out cannot match demand. At a certain point these companies will be sitting on large quantitiesof hoarded memory that they have to put back into the market if they cant build out all these datacenters in time.The prices and abundance of memory early 2025 that enabled foolishness like 30GB GDDR7 on a mainstream console will never be repeated. Ultra Premium Memory is too valuable to sell at basement prices (3-5$/GB).
Memory is a commodity no longer, memory is now a premium product that commands premium margins.
SKHynix / SEC LSI Workers won't go from 1M USD Bonuses back to 60K Salaries willingly.
Yes.
![]()
The 5 hyperscalers are spending 1.6T in 2026-2027 on CAPEX. That's literally the only thing holding up the US economy. If the market for chips and semiconductors goes from that back to 2024 TAM or merely double/triple that, suddenly everyone is clogged with inventory and over leveraged.
My point is that if memory crashes to being cheap, that's a 2008 style crash. That's how far they have inflated this bubble. And thus that scenario is far worse for everyone, gamers included.
Exactly. Redesign or even tweaking to a higher spec is not going to happen. There would be no benefit to the average consumer and it would increase NRE costs and obviously cost a lot in development. It's pure hope from fanboys.And how do you do that without a new floorplan?
To some extent they can do that without AMD at all, but that would require the console shell/motherboard/cooling to be redesigned and would also increase the BOM by quite a bit. More significant changes would require new physical design and/or new node, which again is a lot of time and money.
Let's be honest, ppl likeHeisenbergFX4 and myself would totally buy that 128GB add on for 5000$.
This is just not true. and it'll increasingly become less true overtime. Just look at the HBM market to see what's on store for the client memory market eventually.there is nothing meaningfully different from ddr5 made by SkHynix or Samsung or Micron
Because the value of the memory has increased. 30GB GDDR7 could be a Game Console sure, but it also could be 50 different things that are more useful and happily able to pay the higher prices.I dont see how memory prices can remain so speculatively high even when datacenter build out cannot match demand.
Because the value of the memory has increased. 30GB GDDR7 could be a Game Console sure, but it also could be 50 different things that are more useful and happily able to pay the higher prices.
Memory is memory. There is no magic most memory technologies use the same or similar process nodes. GDDR7 supply can be trivially converted to LPDDR5X for Rubin NVL72. That needs 74TB LPDDR5X Memory per rack, after all.There really aren't that many products that use GDDR other than the consoles and graphics cards. Some crypto cards used it in the past and there's also Rubin CPX.
I mean delaying isn't really going to improve the cost...
The only way around this issue is a lower spec'd model...
I feel like it would make sense. And then you can charge whatever for the standard edition, which would have that 30GB of RAM.
The latter the bette. There is no reason to release a new console now specially when pro exists
Yes there is when Sony has deemed it needs to sell it for $900. They need a newer, more efficient design with a lower BOM. PS6 seems to be designed with that mindset.The latter the bette. There is no reason to release a new console now specially when pro exists
They never did and, in fact, Totoki's quote doesn't exactly counter any of their claims (internal release target, SoC manufacturing for Q2 next year). It's the opposite, actually, he pretty much confirmed that they are internally aiming for a FY 2027 release:You guys expecting Kepler or any other insider to just come over and post release date, plans and official specs 1+ year before launch are absolutely delusional.
About the upcoming generation, future generation gaming consoles. We have not yet decided on at what timing we will launch the new console at what prices. So we would like to really observe and follow the situation. The memory prices, looking at the current circumstances, the memory prices is expected to be very high also in FY27, because there will still be a shortage in supply. So under that assumption, what can, we will like to think about, think carefully what we can do. The, how can we reduce the other costs of the hardware other than the semiconductor? And also, we might think of new ways of selling the product. So in, we would like to think about various, we will do various simulations, including changing business models to come up with the best solution and strategy.
Ps5 was kinda underwhelming, dunno how many hardcore players are excited to upgrade. That and the difference wil be even less than ps4->ps5. There is no need to rush this.
Sony speaking about out it now definitely feels like 2027 is still the plan, unless they see some drastic problems in what ever modelling they are doing.I never said "locked", I said it was the plan, which it was (and still is).
lol, lmao
Oh no the AI bubble pop will be far worse than 2008
we can still get that from indies with pared down graphics and over-all production, but amplified by AI to compensateIt doesn't matter if they kill GaaS, the unique quirky games from Sony that some of us loved are never coming back, the industry has completely changed. Instead of getting GaaS you won't be getting the next Sotc, those type of games don't sell enough anymore for what games cost to make. You will just get another lame ass Marvel game now. It's all catered to the masses and the almighty dollar now.
All you need is great exclusive games for a console priced for mass market?
Possibly ps6 related too. I really dont see a 2027 launch. Might be earliest is Christmas 2028 for ps6.
You guys still dreaming about 499 prices are completely delusional .... the dogshit switch2 is 499.I still think they should release it in Q4 2027 as planned. Pare down the RAM to 20GB and offer a 500GB SSD sku.
PS6 500GB - $499
PS6 1TB - $599
PS handheld - $499
Release PS6 Pro in 2030 with 30GB RAM and a more power APU for $899.
I still think they should release it in Q4 2027 as planned. Pare down the RAM to 20GB and offer a 500GB SSD sku.
PS6 500GB - $499
PS6 1TB - $599
PS handheld - $499
Release PS6 Pro in 2030 with 30GB RAM and a more power APU for $899.
Those prices
![]()
Even with those cut down specs are completely unrealistic (ancient PS5 is 600$). 500GB when one game can be 200GB+?
Ancient PS5 is 1TB and PS5 Pro is 2TB.Those prices
![]()
Even with those cut down specs are completely unrealistic (ancient PS5 is 600$). 500GB when one game can be 200GB+?
Ancient PS5 is 1TB and PS5 Pro is 2TB.
![]()
According to this estimate to which Kepler agreed, 30GB RAM alone is $300 and 1TB SSD is $142.
PS6 APU is only $110.
I propose to cut down on memory because it's the most expensive parts. They could probably save $100 for a 20GB RAM and $70 for a 500GB SSD. The current estimate cost for PS6 is $743. 743 minus 170 is 573.
At $567, Sony could sell a PS6 for $499 subsidized. With the 1TB version, the estimate would be $643, Sony can sell it for $599 subsidized.
They would lose less money with 1TB, so that would be the abundant sku and the 500GB only sold for marketing purposes. Just so they can say PS6 is available for $499.
Okay probably the subsidy is a bit steep.
PS6 500GB - $549
PS6 1TB - $ 649
At that price, Sony would only subsidize $24 for the 500GB and sell the 1TB at no profit.
The main point is to cut down on the most expensive prices which is memory to reduce the price. When memory pricess start to normalize, PS6 can probably sell for $399 in the future. But if memory prices is here to stay, at least the PS6 will not be very expensive that will keep it from selling in critical mass.
PS6 Pro can be 30GB with more powerful APU at $999 and it wouldn't matter. The Pro does not need to sell 100M.
PS5 is at its end cycle and hardware profitability is priority. Sony said so themselves. They are no longer vying to increase adoption. 90 million + have already bought in to the system.Consoles with old ass memory systems and slow SSD modules are 500$ (S2), 600$ (PS5 DE) and 900$ (PS5 Pro).
I wouldn't give it credit if Kepler didn't agree with the estimate.You think this MLID BOM is worth anything?
Anything spent additionally will be part of the loss that Sony will gain and recoup through game sells (that 30% cut) and peripherals. You know how this works right?What about billions spend on R&D and packaging, transport, storage, retailers cut etc.?
That's why the estimate is $300 for the RAM alone. That's why propose a cut down.PS6 will use (expensive) GDDR7
Probably, or they can keep it as is. It's already very fast.and more likely than not SSD speed will be higher than on PS5 - that means more expensive NAND chips.
No. Only memory is ridiculously priced. Pare that down and you can get PS6 to a critical mass price point.If you believe that they can offer that hardware for 500$ (or even 600$) you live in a world of make believe.
lol. Sony was subsidizing almost $300 per unit during PS3 era. That's not what I'm proposing here mr. straw man.Unless Sony wants to repeat loses/per unit they had with PS3...
Nerfing the specs so much kinda defeats the point of having a next-gen system. I think the only reasonable cutbacks are keeping SSD at 1TB and (if they are really desperate) cutting down mem bus to 128-bit and VRAM to 24GB.I still think they should release it in Q4 2027 as planned. Pare down the RAM to 20GB and offer a 500GB SSD sku.
PS6 500GB - $499
PS6 1TB - $599
PS handheld - $499
Release PS6 Pro in 2030 with 30GB RAM and a more power APU for $899.
PS5 is at its end cycle and hardware profitability is priority. Sony said so themselves. They are no longer vying to increase adoption. 90 million + have already bought in to the system.
That's not the same with PS6. Sony will sell it at a loss or at bom price to increase adoption early on. The PS6 has to gain traction right off the bat.
I wouldn't give it credit if Kepler didn't agree with the estimate.
Anything spent additionally will be part of the loss that Sony will gain and recoup through game sells (that 30% cut) and peripherals. You know how this works right?
That's why the estimate is $300 for the RAM alone. That's why propose a cut down.
Probably, or they can keep it as is. It's already very fast.
No. Only memory is ridiculously priced. Pare that down and you can get PS6 to a critical mass price point.
lol. Sony was subsidizing almost $300 per unit during PS3 era. That's not what I'm proposing here mr. straw man.
Nerfing the specs so much kinda defeats the point of having a next-gen system. I think the only reasonable cutbacks are keeping SSD at 1TB and (if they are really desperate) cutting down mem bus to 128-bit and VRAM to 24GB.
Delaying for prices to come down would be the most stupid thing Sony, or anyone, can do. And they surely aren't that stupid, they have a front row seat into this current gen, and we have, for the first time in gaming history see prices go up over the course of a gen and not go down. And yet you somehow think that people in a conference room at Sony are considering how waiting for prices to go down on future and even more expensive tech is the right play?They cant meaningfully update the hw and release before 2030. And the prices are really driven by speculation and demand for AI so they will come down. Consider that memory prices have gone up by 10x in some cases, not because there is anything technologically better but because of hoarding. So the only option is to either delay and release the same hw at the right prices or launch at low volume production until the prices come back to normal, which they will.
if fully adopting a cross-gen strategy, while Sony would still subsidize costs, they likely wouldn't need to do so too aggressively. They can comfortably release a PS6 and not plan on selling more than 10-12M units in its first year. And sell those to the people more likely to pay the most. And when prices start coming down, they can start dropping the price and increasing supply. There is no need to "gain traction right off the bat" when all your games on the PS6 will also be on the PS5.That's not the same with PS6. Sony will sell it at a loss or at bom price to increase adoption early on. The PS6 has to gain traction right off the bat.
You don't just cut RAM down to 20GB and call it a day. Those are not the kinda specs you want to fuck with when launching a new console that is supposed to be supported for the next 10-14 years. That kinda decision would hamstring your entire generation before it even starts.That's why the estimate is $300 for the RAM alone. That's why propose a cut down.
No. Only memory is ridiculously priced. Pare that down and you can get PS6 to a critical mass price point.
Cutting the bus to 128-bit but still doing clamshell wouldn't save enough to be worth it.
PS6 having less raw memory BW than PS5 for sure would be something lol.
It'd be more at 32 gbps speed. At 28 gbps it'd be 448, same as PS5.
Ah, I don't know why I had false memory in my head that PS6 will have 256bit memory bus