Support NeoGAF

[mckmas8808]

The gap in what games can do between generations is shrinking. Partly due to hardware but also partly due to excess development time. Sony can do what they can in making PS6 games, but 3rd parties can't be forced into it. What with AAA games taking 6 years, we will be lucky if 3rd parties even make PS6 games at launch. Money is also tight these days. People are willing to wait longer to upgrade. This is just how the world is now.

Making a PS5 handheld is actually a valid way to offer something new to the public who don't want a PS6 or don't feel the need to buy one yet.

Gotcha! 100% agree with you here. It's why the crossgen period from PS5 to PS6 will be a minimum of 3 years for 90% of games for Sony. Probably a minimum of 4 years for 90% of 3rd party AAA games. And a minimum of 5 years for 90% of 3rd party AA games.

If this PS handheld (in handheld mode) can be a bit stronger than the Switch 2, then it'll be a great device until 2032 for the extreme amount of games that come out for it (assuming it releases in 2027) over a 5 year period. A 5 TF PS handheld that does FSR4, has a 8-inch 120 hz screen, with ray-tracing built-in would be an amazing device. The only question is........does it cost less than $600?

 

[ResilientBanana]

I mean I don't think anyone should expect thing to do PS6 exclusive games...or least not many. I honestly don't think it even has to anyway.

To me this device will be worth it on whether Sony manages to get general/universal PS5 compatibility implemented. Then you have a very sizable library and will ensure you do keep getting you new (crossgen) games up until ~2030 if PS4 longevity is anything to go by.

Then why buy it? Doesn't seem quite as enticing to just play PS5 and below.

Based on what? Portable is a different market and they aren't obligated to make run a ps6 games on portable.

Based on mobile handhelds not being able to reach 250W like the base consoles can. They need to sip power not chug it.

Historically, devkits have had twice the amount of RAM than production consoles.

 

[Audiophile]

I'm actually not as concerned as I initially was regarding getting native PS6 games on a 'PS6 Portable' and I genuinely think it'll be the case and it'll be mandated. Power draw doesn't scale down linearly for starters, so at ~1/15th power draw you should be able to get 1/6th to 1/5th the overall performance.

The GPU side can scale quite easily with resolution, frames & small fx tweaks. I was concerned about bandwidth, but if PS6 is making very efficient use of 640GB/s for 18-23TF RDNA5 (36-46 Dual-Issue), then I expect PS6P to make efficient use of the ~205GB/s for ~4TF RDNA5 (8 Dual-Issue); so there's a relative B/W per flop of 1.4-1.9x on the PS6P vs PS6...

Plus, there's the rumoured 16MB MALL/Infinity Cache for the PS6 Portable which hasn't been alluded to for the main PS6 console. So this offers not just an additional, relative bandwidth bump that with expectedly good cache hit rates puts you closer to an effective ~300GB/s overall and probably ~512GB/s for actual data in cache, it offers much-improved latency (about 1/8th I believe vs GDDR7 and 1/5th vs LPDDR5X) which in tandem with lower latency LP5X can allow the CPU a little more efficiency, not to mention vastly reduced power consumption for cache hits vs LP5X and especially G7.

Also, if the PS6 Portable has 4MB L2 on the GPU while the PS6 is in the 10-12MB range, then relative to the overall compute, that could give it another effective bandwidth, latency and power use/distribution advantage.

My one concern that remains somewhat is the CPU side. World simulation, complexity and the fundamental functionality of your game world which will be just as important in making lively next-gen games, may face some limitation.

But, let's say PS6 is using [for games] 8c/16t zen6c @ ~4.35GHz and PS6P is 4c/8t Zen6c @ ~3GHz, then you need to work with ~35% of the CPU throughput. The LP5X + MALL Cache latency advantage might give you a few percent bump in efficiency from less stalls, so let's say 38%. Significant downgrades in visual settings/complexity may get you down to around the 70% region from the initial load but if you're already batting up against the CPU limits on PS6 and pushing a complex/dense world simulation it may bottom out around there and not scale down anymore with visuals.

Cut from 60fps to 40fps gets you down to around 46%; then from there you can maybe par back some tick rates, complexity, density etc. on distant or objects less essential to gameplay without parring back the core experience too much. Perhaps Sony could allow for a small CPU<>GPU trade-off by devs as needed to clock up a bit. And then you should be in the ballpark. Or you could of course just take the hit and smash it right down to 30fps; and they've done that on Switch for many games, but I'd hope that for games where better response and reasonable clarity are required they'd try and aim for a nice middleground like 40fps. Of course a smaller display tends not to accentuate motion issues as much as big displays, but I still think 30fps is too low for specific games, 40 really does make a surprising difference.

The advantage of a handheld (or at least handheld modes) is you have a guaranteed specification on the display, so if they spec it with VRR+LFC then they don't have to adhere perfectly, allowing a little more slippage and can have modes in the region of ~40fps (or anywhere really) without having to worry whether you have a display to support it.

The one issue remains if you have a game only doing 30fps and maybe 40fps on PS6 with everything cranked and you're CPU-limited (think a GTA7 at the end of the gen [a little optimistic probably, lol]). Then what do you do on the portable? In some cases you might at best be able to trim back to the 50-60% region in terms of required throughput relative to the PS6 and you'd have a gulf to make up that can only be done by fundamentally limiting the core game functionality. You might just argue that devs have to make that decision and make those trade-offs, but it'd be a shame for this to significantly limit the scope of landmark titles that really try to push the boundaries. So for that reason I really hope that the final hardware is very favourable to the CPU side and every plausible optimisation is made and every dev-side option is entertained.


Some separate, rough examples for how PS6 Games could scale..

Eg. 1 (Advanced, more latency sensitive gameplay)
- PS6: ~1080p @ 60fps > 4x PSSR3/FSR5 > 2160p w/ Path Tracing
- PS6P: ~540p @ 40fps > 4x PSSR3/FSR5 > 1080p w/ RT GI + RT Reflections + reduced tick rate/complexity for distant view animations/physics & altered distant tri : pixel ratio/bias in virtual geometry.

Eg. 2 (Advanced, less latency sensitive gameplay)
- PS6: ~1080p @ 30/40fps > 4x PSSR3/FSR5 > 2160p w/ Comprehensive Path Tracing + Ultra-High Visuals
- PS6P: ~540p @ 30fps > 2x PSSR3/FSR5 > 720p > 2x HQ Spatial Upscale > 1080p w/ RT GI + RT Reflections + Med/High Visuals + reduced tick rate/complexity for distant view animations/physics, reduced density for non-essential distant simulated objects & altered distant tri : pixel ratio/bias in virtual geometry [CPU Priority Mode w/ CPU Clocks increased & GPU/Mem Clocks reduced].

Eg. 3 (Middleground)
- PS6: ~1440p @ 60fps > 2x PSSR3/FSR5 > 2160p w/ Higher Precision RT GI + RT Reflections/Shadows/AO
- PS6P: ~720p @ 40fps > 2x PSSR3/FSR5 > 1080p w/ Lower Precision RT GI + RT Reflections/AO + reduced tick rate/complexity for distant view animations/physics & reduced density for non-essential distant simulated objects.

Eg. 4 (Traditional)
- PS6: ~1800-2160p @ 30/40fps > PSSR3/FSR5 AA w/ Baked Lighting + RT Reflections/AO/Shadows
- PS6P: ~900-1080p @ 30fps/+ > PSSR3/FSR5 AA w/ Baked Lighting + RT Reflections/AO + reduced asset LoD distance.

 

[paolo11]

I want to ask, will the ps6 handheld be as powerful as the handheld with AMD 395 with 8060 mobile GPU?

 

[KeplerL2]

RDNA2 is about as efficient as RDNA3, RDNA4 is about 15% better than these two.
For RDNA5 to cut down power from ~100W to a handheld level it should be 4X as efficient per watt as RDNA4.
This is very obviously not going to happen, and a "far smaller node" won't help with this either as all these newer nodes past N7 have very limited gains in power/clocks/density.
Even a "docked" mode doesn't seem likely to happen with such wattage at these settings as you'd probably need about 50-60W and that's 2X down from RDNA2/3 and some -40% from RDNA4.

So the concern which @Loxus brought up is valid and it will be interesting to see how far this "eco mode" will end up being from the actual handheld h/w.

? RDNA4 is ~75% more efficient than RDNA2 (220W 9070 is >50% faster than 250W 6800)

Also an ultra low power design with custom physdes using low VT/low leakage xtors and LPDDR will have massively better efficiency than a standard high power/high clocked design even on the same architecture and node.

 

[viveks86]

I'm actually not as concerned as I initially was regarding getting native PS6 games on a 'PS6 Portable' and I genuinely think it'll be the case and it'll be mandated. Power draw doesn't scale down linearly for starters, so at ~1/15th power draw you should be able to get 1/6th to 1/5th the overall performance.

The GPU side can scale quite easily with resolution, frames & small fx tweaks. I was concerned about bandwidth, but if PS6 is making very efficient use of 640GB/s for 18-23TF RDNA5 (36-46 Dual-Issue), then I expect PS6P to make efficient use of the ~205GB/s for ~4TF RDNA5 (8 Dual-Issue); so there's a relative B/W per flop of 1.4-1.9x on the PS6P vs PS6...

Plus, there's the rumoured 16MB MALL/Infinity Cache for the PS6 Portable which hasn't been alluded to for the main PS6 console. So this offers not just an additional, relative bandwidth bump that with expectedly good cache hit rates puts you closer to an effective ~300GB/s overall and probably ~512GB/s for actual data in cache, it offers much-improved latency (about 1/8th I believe vs GDDR7 and 1/5th vs LPDDR5X) which in tandem with lower latency LP5X can allow the CPU a little more efficiency, not to mention vastly reduced power consumption for cache hits vs LP5X and especially G7.

Also, if the PS6 Portable has 4MB L2 on the GPU while the PS6 is in the 10-12MB range, then relative to the overall compute, that could give it another effective bandwidth, latency and power use/distribution advantage.

My one concern that remains somewhat is the CPU side. World simulation, complexity and the fundamental functionality of your game world which will be just as important in making lively next-gen games, may face some limitation.

But, let's say PS6 is using [for games] 8c/16t zen6c @ ~4.35GHz and PS6P is 4c/8t Zen6c @ ~3GHz, then you need to work with ~35% of the CPU throughput. The LP5X + MALL Cache latency advantage might give you a few percent bump in efficiency from less stalls, so let's say 38%. Significant downgrades in visual settings/complexity may get you down to around the 70% region from the initial load but if you're already batting up against the CPU limits on PS6 and pushing a complex/dense world simulation it may bottom out around there and not scale down anymore with visuals.

Cut from 60fps to 40fps gets you down to around 46%; then from there you can maybe par back some tick rates, complexity, density etc. on distant or objects less essential to gameplay without parring back the core experience too much. Perhaps Sony could allow for a small CPU<>GPU trade-off by devs as needed to clock up a bit. And then you should be in the ballpark. Or you could of course just take the hit and smash it right down to 30fps; and they've done that on Switch for many games, but I'd hope that for games where better response and reasonable clarity are required they'd try and aim for a nice middleground like 40fps. Of course a smaller display tends not to accentuate motion issues as much as big displays, but I still think 30fps is too low for specific games, 40 really does make a surprising difference.

The advantage of a handheld (or at least handheld modes) is you have a guaranteed specification on the display, so if they spec it with VRR+LFC then they don't have to adhere perfectly, allowing a little more slippage and can have modes in the region of ~40fps (or anywhere really) without having to worry whether you have a display to support it.

The one issue remains if you have a game only doing 30fps and maybe 40fps on PS6 with everything cranked and you're CPU-limited (think a GTA7 at the end of the gen [a little optimistic probably, lol]). Then what do you do on the portable? In some cases you might at best be able to trim back to the 50-60% region in terms of required throughput relative to the PS6 and you'd have a gulf to make up that can only be done by fundamentally limiting the core game functionality. You might just argue that devs have to make that decision and make those trade-offs, but it'd be a shame for this to significantly limit the scope of landmark titles that really try to push the boundaries. So for that reason I really hope that the final hardware is very favourable to the CPU side and every plausible optimisation is made and every dev-side option is entertained.


Some separate, rough examples for how PS6 Games could scale..

Eg. 1 (Advanced, more latency sensitive gameplay)
- PS6: ~1080p @ 60fps > 4x PSSR3/FSR5 > 2160p w/ Path Tracing
- PS6P: ~540p @ 40fps > 4x PSSR3/FSR5 > 1080p w/ RT GI + RT Reflections + reduced tick rate/complexity for distant view animations/physics & altered distant tri : pixel ratio/bias in virtual geometry.

Eg. 2 (Advanced, less latency sensitive gameplay)
- PS6: ~1080p @ 30/40fps > 4x PSSR3/FSR5 > 2160p w/ Comprehensive Path Tracing + Ultra-High Visuals
- PS6P: ~540p @ 30fps > 2x PSSR3/FSR5 > 720p > 2x HQ Spatial Upscale > 1080p w/ RT GI + RT Reflections + Med/High Visuals + reduced tick rate/complexity for distant view animations/physics, reduced density for non-essential distant simulated objects & altered distant tri : pixel ratio/bias in virtual geometry [CPU Priority Mode w/ CPU Clocks increased & GPU/Mem Clocks reduced].

Eg. 3 (Middleground)
- PS6: ~1440p @ 60fps > 2x PSSR3/FSR5 > 2160p w/ Higher Precision RT GI + RT Reflections/Shadows/AO
- PS6P: ~720p @ 40fps > 2x PSSR3/FSR5 > 1080p w/ Lower Precision RT GI + RT Reflections/AO + reduced tick rate/complexity for distant view animations/physics & reduced density for non-essential distant simulated objects.

Eg. 4 (Traditional)
- PS6: ~1800-2160p @ 30/40fps > PSSR3/FSR5 AA w/ Baked Lighting + RT Reflections/AO/Shadows
- PS6P: ~900-1080p @ 30fps/+ > PSSR3/FSR5 AA w/ Baked Lighting + RT Reflections/AO + reduced asset LoD distance.

Thoughtful post! But I Ctrl + F'd "Unreal Engine 6" and found no results in these projections

 

[mckmas8808]

? RDNA4 is ~75% more efficient than RDNA2 (220W 9070 is >50% faster than 250W 6800)

Also an ultra low power design with custom physdes using low VT/low leakage xtors and LPDDR will have massively better efficiency than a standard high power/high clocked design even on the same architecture and node.

Thanks for responding to that. Logic told me that what he said couldn't possibly be true. Do you think it's possible or even likely that RDNA5 will be more than 100% more efficient than RDNA2?

 

[KeplerL2]

Thanks for responding to that. Logic told me that what he said couldn't possibly be true. Do you think it's possible or even likely that RDNA5 will be more than 100% more efficient than RDNA2?

Easily

 

[Justin9mm]

Same reason I have a PC and a Steam Deck? Play on the go anywhere without streaming hiccups. I play my SD outside my house often (in the summer) which doesn't get good rception and I don't have unlimited hotspot with my phone.

I also stream to my PC to the Steam Deck for games that it can't run too.

Personally this isn't for me since I barely have a library with PS, but I can totally see why if someone does have a good sized library.

Well for me, I don't need something to play out of the house, but I do have a Portal and love it, because I can game anywhere in my house, mainly in my bed or sitting outside the back etc. The Portal fits the bill. Buying an actual handheld would be a waste for me. The Portal was only a streaming device and it is/was expensive, I can't see an actual handheld being less than the price of base PS5 now. I dunno, I get where your coming from, but most Sony gamers who are going to buy a PS6, probably don't need this unless you are an avid gamer on the go. I feel like it's going to confuse consumers who are casual gamers.

 

[nial]

I feel like it's going to confuse consumers who are casual gamers.

PS6, home console.
PS6 Portable / Go / Lite / Whatever, handheld console.

 

[Justin9mm]

It's not for you, it's for markets that don't want or can't afford the full fat PS6 or for people who want a secondary system. Two SKUs was perhaps, a good idea, but the Series S was in the wrong form factor. This will widen Sony total addressable market considerably.

Yep, it's definitely not for me, secondary system would be overklill when I don't game out of my house. Portal I have and great for bedtime, sitting out the back in summer, using it instead of the TV whilst the wife watches said TV and can still be together.. So this is not necessary for me. Interesting though.

 

[ThatOneGrunt]

I'm getting tired of being a consumer. I think I'm just going to play with my old games. Download some ROMs. Stop contributing to landfills. Start growing vegetable maybe.

No interest in buying things anymore.

 

[Justin9mm]

PS6, home console.
PS6 Portable / Go / Lite / Whatever, handheld console.

I get that much, but when you dive down into the specs/differences, the mums and dads and kiddies will probably have a hard time weighing up whats what. I think people are going to be shocked how much this handheld will be, I'm not expecting it to be that much cheaper than the PS6 console itself.. It's Sony.. Look at the price of the Portal that does nothing but streaming.

 

[nial]

I get that much, but when you dive down into the specs/differences, the mums and dads and kiddies will probably have a hard time weighing up whats what.

It won't make much difference. Can it play Fortnite, Genshin Impact, Roblox, Call of Duty? Yes, it can.

I think people are going to be shocked how much this handheld will be, I'm not expecting it to be that much cheaper than the PS6 console itself..

Nah, there's no way that it's more expensive. Nothing points out to that being the case.

It's Sony.. Look at the price of the Portal that does nothing but streaming.

Console add-ons are meant to be hugely profitable, that's why.

 

[mckmas8808]

Well for me, I don't need something to play out of the house, but I do have a Portal and love it, because I can game anywhere in my house, mainly in my bed or sitting outside the back etc. The Portal fits the bill. Buying an actual handheld would be a waste for me. The Portal was only a streaming device and it is/was expensive, I can't see an actual handheld being less than the price of base PS5 now. I dunno, I get where your coming from, but most Sony gamers who are going to buy a PS6, probably don't need this unless you are an avid gamer on the go. I feel like it's going to confuse consumers who are casual gamers.

I think a full portable PS handheld can sell 10% of the total PS6 market. If priced right it may do 15% AT MOST!!! So if the PS6 ever hit 100 million, I can see this device doing 10-15 million in sales. That's not setting the world on fire, but it is a decent market for Sony to focus on to help keep them in their ecosystem.

 

[nial]

I think a full portable PS handheld can sell 10% of the total PS6 market. If priced right it may do 15% AT MOST!!! So if the PS6 ever hit 100 million, I can see this device doing 10-15 million in sales. That's not setting the world on fire, but it is a decent market for Sony to focus on to help keep them in their ecosystem.

You're greatly underestimating its potential.

 

[mckmas8808]

You're greatly underestimating its potential.

Maybe I am. Trust and believe, I'm all in on it's possiblities. But it's sales potential will be limited by how much it cost. And I'm sorry, but it's not the Switch or Switch 2. It'll mainly be a companion machine for people already with a PS5 or PS6.

Some will buy it that have only a Switch 2, maybe........but that won't be most people. At $500 or $600, no way it can sell over 50 million units. So chance in hell.

 

[nial]

Maybe I am. Trust and believe, I'm all in on it's possiblities. But it's sales potential will be limited by how much it cost. And I'm sorry, but it's not the Switch or Switch 2. It'll mainly be a companion machine for people already with a PS5 or PS6.

Some will buy it that have only a Switch 2, maybe........but that won't be most people. At $500 or $600, no way it can sell over 50 million units. So chance in hell.

If it's $500 like most of us assume, it's going to do well. It doesn't make much sense for it be marketed as a companion machine when it has a dedicated library. People will either choose this or the main PS6.

 

[GoldenEye98]

I would absolutely buy this if it has full PS5 compatibility. That's a longgg ass library between PS4 and PS5.

I mean on a certain level having just having a portable PS4 beats Switch 2 for me because that BC is so much more vast and technically superior to Switch 1 BC.

 

[LiquidMetal14]

I want to ask, will the ps6 handheld be as powerful as the handheld with AMD 395 with 8060 mobile GPU?

Based on speculation, yes. In terms of certain features and aspects, it may excel while in others be on par or close. One thing that will elevate it is closed HW and optimizations that Sony is known for given the last 3 generations.

I have a lot more faith in this that I did with PS5. They just need to get their house in order and I have strong suspicion that they will have a rebound generation.

 

[Crispy Gamer]

How strong is this unit supposed to be when docked compared to PS5?

 

[th4tguy]

Don't devkits have 2x the RAM as the consumer console? I would guess 24GB.


This would be way worse than XSS

They often have more but not always double. It's not a rule.

 

[Justin9mm]

If it's $500 like most of us assume, it's going to do well. It doesn't make much sense for it be marketed as a companion machine when it has a dedicated library. People will either choose this or the main PS6.

If I know Sony, it's going to be noticeably underpowered compared to PS6 and cost not much less. Look at Portal and PSVR2 prices at launch. Sony is going to price gouge the same as they aways do.. So it's either going to be a secondary system to the PS6 or a small market of causal gamers will choose this as a cheaper option. The only way its going to do really well is if Sony gets the price right to make it a valuable option.. It won't be!

 

[Crispy Gamer]

Maybe I am. Trust and believe, I'm all in on it's possiblities. But it's sales potential will be limited by how much it cost. And I'm sorry, but it's not the Switch or Switch 2. It'll mainly be a companion machine for people already with a PS5 or PS6.

Some will buy it that have only a Switch 2, maybe........but that won't be most people. At $500 or $600, no way it can sell over 50 million units. So chance in hell.

Sony seems to be serious about converting those active PS4 users playing GAAS titles I think it'll sell very well at an attractive price I just hope they're prepared for the demand.

 

[digoutyoursoul]

Don't let this parasite bastard infect the PlayStation side of things as he has with Xbox

 

[Jesb]

I passed on consoles this gen but I'm looking forward to jumping back in with the PS6 handheld. I recently got a steamdeck and for me I can't go back to playing on a tv. Way more immersive with a screen in front of your face.

 

[nial]

If I know Sony, it's going to be noticeably underpowered compared to PS6 and cost not much less. Look at Portal and PSVR2 prices at launch. Sony is going to price gouge the same as they aways do.. So it's either going to be a secondary system to the PS6 or a small market of causal gamers will choose this as a cheaper option. The only way its going to do really well is if Sony gets the price right to make it a valuable option.. It won't be!

You keep saying that because you keep comparing it to console accessories, which makes zero sense.

 

[viveks86]

Don't let this parasite bastard infect the PlayStation side of things as he has with Xbox

Wait… is he testing the waters to see if he can jump ship?

 

[Tobimacoss]

Don't let this parasite bastard infect the PlayStation side of things as he has with Xbox

Too late, he's all yours. NO takesies backsies.

 

[Fafalada]

I want to ask, will the ps6 handheld be as powerful as the handheld with AMD 395 with 8060 mobile GPU?

Depends what TDP they target.
There is literally a handheld now that runs 395 @ 80Watts (and that's not even max TDP the chip is rated for). Even if PS6 chipset is 100% more efficient - that'd be TDP of 40W. While cooling assemblies clearly exist to handle this handheld (and Sony would only do better than the PC companies in this space) - I have doubts they'd consider ever going that high - much as I'd love to see it.

I'm actually not as concerned as I initially was regarding getting native PS6 games on a 'PS6 Portable' and I genuinely think it'll be the case and it'll be mandated.

I'm personally very mixed on it.
Your analysis on scaling CPU/simulation side already shows how things get very complex (in game-dev terms, that complexity really just means lowest common denominator will be used 95% of the time - we've got 25 years of industry proving this right in practice), but in the supposed 'ML is the new coolness' world, the GPU scaling will be impacted by the same problem.
It's not just about putting pixels on screen anymore.

Now sure - people will say 'cross gen' will be the limiter anyway for years to come - but this just gives another excuse to devs not to bother.

 

[Justin9mm]

You keep saying that because you keep comparing it to console accessories, which makes zero sense.

I'm comparing Sony's history with prices, it doesn't matter if its a console, controller, portal etc. I'm in my 40s and have had every PS console since day 1, I'm a PS fan boy and I know Sony's behaviour inside and out. You make zero sense because Sony selling it's console and console accessories are one and the same. Why is this handheld any different? The price compared to the PS6 is going to be the be all and end all on it's success.

 

[kevboard]

I'm comparing Sony's history with prices, it doesn't matter if its a console, controller, portal etc. I'm in my 40s and have had every PS console since day 1, I'm a PS fan boy and I know Sony's behaviour inside and out. You make zero sense because Sony selling it's console and console accessories are one and the same. Why is this handheld any different? The price compared to the PS6 is going to be the be all and end all on it's success.

the difference between consoles and accessories is that consoles are subsidised while accessories are literally part of the revenue stream that is used to subsidise the console.

the Dualsense is an insanely overpriced controller, because they make up lost revenue from the consoles.
that's why 8bitdo is able to sell superior controllers for less than half the price of the Dualsense... they aren't using them to subsidise and make up for something else they sell.

 

[nial]

I'm comparing Sony's history with prices, it doesn't matter if its a console, controller, portal etc. I'm in my 40s and have had every PS console since day 1, I'm a PS fan boy and I know Sony's behaviour inside and out. You make zero sense because Sony selling it's console and console accessories are one and the same. Why is this handheld any different? The price compared to the PS6 is going to be the be all and end all on it's success.

Nope, if that were the case, we would see the PS5 price close to 1K USD. Console accessories have a WAY higher profit margin, because that's the point of them, they're supposed to make money that helps to make up for the relatively low profit margins that keep their consoles on an affordable price point.

 

[Katatonic]

Wait… is he testing the waters to see if he can jump ship?

More like a concerted effort to propagate and control the PS6 narrative in order to stall PS5 sales and set the stage for the next xbox power king astroturf fiesta. He gets his marching orders from Greenberg.

 

[The Fartist]

 

[VolticArchangel]

Don't let this parasite bastard infect the PlayStation side of things as he has with Xbox

 

[Justin9mm]

Nope, if that were the case, we would see the PS5 price close to 1K USD. Console accessories have a WAY higher profit margin, because that's the point of them, they're supposed to make money that helps to make up for the relatively low profit margins that keep their consoles on an affordable price point.

Yes, because the PS5 Pro was a great example of that.. Is that also an accessory to the base PS5? lol

If we were talking about previous generations and going into PS5 generation, I would agree, but right now in 2025, Sony are going to price gouge because they are back to greedy Sony, price of the acessories for PS5 showed that. PS6 handheld is rumoured to be $600 USD lol.. Is it even going to be faster hardware than base PS5? And you think that's Sony keeping it affordable!?

 

[nial]

Yes, because the PS5 Pro was a great example of that.. Is that also an accessory to the base PS5? lol

It's premium hardware for their hardcore fanbase. Why isn't the PS5 much more expensive if what you're saying is true?

If we were talking about previous generations and going into PS5 generation, I would agree, but right now in 2025, Sony are going to price gouge because they are back to greedy Sony, price of the acessories for PS5 showed that.

PSVR in 2016 was literally more expensive than a PS4.

PS6 handheld is rumoured to be $600 USD lol..

No, it isn't. Not even MLiD or KeplerL2 are expecting it to be that expensive, lol.

Is it even going to be faster hardware than base PS5? And you think that's Sony keeping it affordable!?

Faster in what sense? It's definitely not more powerful.

 

[dgrdsv]

There are only 4 cores, not 8.

Yeah, and the 100W result is on 4 cores, not 8. To get from that down to 25W _including_ the GPU you'd need 8X perf/watt improvement on CPU cores at the very least.

I'll stop there (I could go on with the others components) because you are really in bad faith. That seems to really bother you that porting those PS5 games to the handheld could be so easy.

It bothers me when people can't do basic math.

Power curves don't share a linear relationship with performance. The Steam Deck's APU draws about 15W under full load, less than 10% of what the PS5 draws, yet it's about ~20% of its performance. If you increase the power by 10x, you won't suddenly get a 150W APU that's twice the performance of the PS5.

It's all about finding the sweet spot between the two.

While they don't share a linear relation with performance we're still looking at the need to have 1/4th of power with the same performance which is still a 4X perf/watt improvement.
Note that the perf/watt is including both perf and watt in itself.

? RDNA4 is ~75% more efficient than RDNA2 (220W 9070 is >50% faster than 250W 6800)

It's not "more than 50% faster" but barely at +50% (unless we go into RT territory) and 250->220 is just a 12% improvement. This isn't "~75%".
And the example itself is questionable as these both are salvage parts meaning that you're getting all sort of funkiness in binning and on-chip voltage.

Also an ultra low power design with custom physdes using low VT/low leakage xtors and LPDDR will have massively better efficiency than a standard high power/high clocked design even on the same architecture and node.

This remains to be seen as so far we haven't seen such "massively better" difference between similar APU and GPU parts at all. Not saying that it's impossible just improbable from what we have now.

 

[SABRE220]

Looking at the results, honestly, this thing looks like a headscratcher. If it's barely running performance mode settings of the PS5 at 30fps and still dropping frames when pushed, how the hell are they going to support this thing going forward? Let's assume most games in the future run at 60fps on the ps6 with a reduced settings mode at 30fps mode on the ps5. What the hell happens to the handheld version?

Also, how well will it handle 30fps titles on the PS5? From all the discussion that was being had with the leaks, it was pretty much painting a picture of it running the PS5 version of the game at a lower resolution a streamlined and stable profile, but with it targeting performance mode settings at half the framerate and a lower res on top this becomes a much messier scenario.

If they put in 8 CPU cores, even clocked lower like in the current handhelds, this thing could have scaled much better, considering GPU settings are much easier to scale down, but with 4 CPU cores....honestly, this isn't looking good, and if Sony makes support mandatory for PS6 games, this could be a real anchor on the PS6. Hopefully, the actual device performs better because for me this looks like a headache in the making.

 

[nial]

Looking at the results, honestly, this thing looks like a headscratcher. If it's barely running performance mode settings of the PS5 at 30fps and still dropping frames when pushed, how the hell are they going to support this thing going forward? Let's assume most games in the future run at 60fps on the ps6 with a reduced settings mode at 30fps mode on the ps5. What the hell happens to the handheld version?

Also, how well will it handle 30fps titles on the PS5? From all the discussion that was being had with the leaks, it was pretty much painting a picture of it running the PS5 version of the game at a lower resolution a streamlined and stable profile, but with it targeting performance mode settings at half the framerate and a lower res on top this becomes a much messier scenario.

If they put in 8 CPU cores, even clocked lower like in the current handhelds, this thing could have scaled much better, considering GPU settings are much easier to scale down, but with 4 CPU cores....honestly, this isn't looking good, and if Sony makes support mandatory for PS6 games, this could be a real anchor on the PS6. Hopefully, the actual device performs better because for me this looks like a headache in the making.

Can't we wait until we actually get stuff natively developed for it before concerning over anything?

 

[ChiefDada]

Can't we wait until we actually get stuff natively developed for it before concerning over anything?

This happens in every lead up to a new generation, without fail. As if forum dwellers could possibly come up with novel questions that the engineers who are actually working on the hardware haven't thought of.

 

[BennyBlanco]

This happens in every lead up to a new generation, without fail. As if forum dwellers could possibly come up with novel questions that the engineers who are actually working on the hardware haven't thought of.

Yeah, it's much more sensible to wait for marketing YouTube videos from Sony to draw grand conclusions from

Analysis  Thread 'PS5 Pro/PSSR Appears to Provide Better Image Reconstruction than DLSS Running on a 4090 GPU'

Sep 11, 2024

I was curious and went looking back on DF comparisons of PS5 resolution/fidelity modes with a 4090 DLAA max settings, and preliminary comparison suggests to me that PSSR is potentially a vastly superior solution. I gathered the limited samples from today's Cerny tech talk where base PS5 fidelity mode was compared to PS5 Pro that doubled frame rate in addition to improving effective resolution. For the DF comparisons, I have Spider-Man Miles Morales and Horizon Forbidden West where, again, Alex compares 4K DLAA max settings image quality to PS5 fidelity modes (unfortunately, DF didn't use...

• ChiefDada
• Replies: 892
• Forum: Gaming Discussion

• 
• 
• 
• 
• 

 

[Ashamam]

If it's barely running performance mode settings of the PS5 at 30fps and still dropping frames when pushed, how the hell are they going to support this thing going forward? Let's assume most games in the future run at 60fps on the ps6 with a reduced settings mode at 30fps mode on the ps5. What the hell happens to the handheld version?

I'm assuming there is a legacy cost to running PS5 games on the handheld. There is no ML upscaling for a start, let alone texture models etc that in theory could allow huge bandwidth gains. No doubt there are other architectural changes that will need to be explicitly coded in and wont be available for legacy PS5 titles.

It depends a lot on whether the device is being targetted as a true PS6 handheld or a cross gen PS5 enabling device just piggybacking off the new hardware to enable decent handheld performance. I'm in favour of option 1, I think they will primarily balance it around running cut down PS6 games and the PS5 back compat is gravy. Its just we are seeing the gravy be tested first.

 

[BlackTron]

Does the handheld really have its own exclusive games? Seems crazy, I'd think it would be forced 2 profiles like Switch has. Well, hopefully that handheld has a pretty nice CPU or its gonna Series S the PS6

 

[hussar16]

Sony could be up to a hybrid considering switch success and if I did have to pick it's probably whoever is the traditional console, there's no circumstance I ever do portable gaming

I dont understand the hybrid console thing for sony. Why sell a handheld hybrid when your main console is doing great. Only reason nintendo did a hybrid was because of their handheld market going down. Nintendo was forced to put 2 together to make sales. Unless there are 2 ps6 versions this just does not make sense bussiness wise as they could easily sell 2 versions of ps6 and make more profit. Im guessing ps6 comes out with 2 versions one handheld and one main console but even this is very weird move

 

[nial]

This happens in every lead up to a new generation, without fail. As if forum dwellers could possibly come up with novel questions that the engineers who are actually working on the hardware haven't thought of.

My main issue comes from the fact that PS5 titles (in Power Saver mode or otherwise) will obviously not have newer upscaling features from RDNA5/Zen6 architecture and it's dumb to see them as the most of what the device will be capable of.

I dont understand the hybrid console thing for sony. Why sell a handheld hybrid when your main console is doing great. Only reason nintendo did a hybrid was because of their handheld market going down. Nintendo was forced to put 2 together to make sales. Unless there are 2 ps6 versions this just does not make sense bussiness wise as they could easily sell 2 versions of ps6 and make more profit. Im guessing ps6 comes out with 2 versions one handheld and one main console but even this is very weird move

It is two PS6 consoles.

 

[SuperC]

Could be a great handheld if Sony released it now with native PS4 support plus whatever they could
port from the PS5 library but by 2027 it would be too late

 

[SABRE220]

I'm assuming there is a legacy cost to running PS5 games on the handheld. There is no ML upscaling for a start, let alone texture models etc that in theory could allow huge bandwidth gains. No doubt there are other architectural changes that will need to be explicitly coded in and wont be available for legacy PS5 titles.

It depends a lot on whether the device is being targetted as a true PS6 handheld or a cross gen PS5 enabling device just piggybacking off the new hardware to enable decent handheld performance. I'm in favour of option 1, I think they will primarily balance it around running cut down PS6 games and the PS5 back compat is gravy. Its just we are seeing the gravy be tested first.

I mean, even taking out the PS6 from the equation, the question becomes, how is the thing going to support the demanding PS5 games at acceptable settings? Assuming this a cross gen ps5 enabling device consumers would look to play their ps5 library scaled down to the 1080p screen res at good performance levels at minimum especially with it being launched alongside the ps6 but with it hitting half the performance levels of the ps5 at performance settings and still dropping frames when the engine is pushed what happens when games like dragons dogma, baldursgate, and ue5 titles that run at 40ish fps in performance mode at 720-900p on the ps5? How will the 4 cores handle cpu workloads designed around the ps5 when workloads cant be scaled down even with the average patches? Devs sure as hell arent making dedicated ports like with the switch1.

 

[Tiago Rodrigues]

Well for me, I don't need something to play out of the house, but I do have a Portal and love it, because I can game anywhere in my house, mainly in my bed or sitting outside the back etc. The Portal fits the bill. Buying an actual handheld would be a waste for me. The Portal was only a streaming device and it is/was expensive, I can't see an actual handheld being less than the price of base PS5 now. I dunno, I get where your coming from, but most Sony gamers who are going to buy a PS6, probably don't need this unless you are an avid gamer on the go. I feel like it's going to confuse consumers who are casual gamers.

Playstation 6 - home console
Playstation 6 Portable - handheld

There. Everything explained in a few words.