By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Sony Discussion - Sony might be useing 3D stacked chips to be more power effecient, have better cooling.

 

Mindblown?

yes, 3D stacked chips arn... 7 50.00%
 
nah, I'm liveing in the f... 7 50.00%
 
Total:14
Pemalite said:
JRPGfan said:

Im gonna go ahead and say I'm impressed.

They are patents. They aren't actual products.
Don't be impressed when there is no demonstrable product yet.

JRPGfan said:


Sony seems to be doing alot of "new" things.
The variable clock speeds, to maximise performance without needing to waste anything on overhead.
The 3D stacked chip, and the unique cooling solution seem, like something we've heard of would be comeing in the future, but no one really expected to see anytime soon.

Variable clockrates that aren't developer controlled/influenced isn't a good thing...

JRPGfan said:

The 3D stacked chip, and the unique cooling solution seem, like something we've heard of would be comeing in the future, but no one really expected to see anytime soon.

We have been stacking NAND chips for years.

JRPGfan said:


Even the controller is new, its no longer called a dualshock, but now the dualSense, after the changes it got in tech.

Controller is just an evolution of past designs, it's different, but it's not going to be playing games any differently, it still has your normal array of buttons and two analogue sticks and a d-pad.

JRPGfan said:


This ontop of stuff like all the work they put into removeing bottlenecks, when moveing data around (from the SSD).

Please elaborate on those bottlenecks.

JRPGfan said:


Alot of this stuff just "seems" more planned out than what Microsoft did.
With the Xbox series X, there wasnt any "surprises", it was basically just more or less what everyone thought it would be.
With the Playstation 5, there seems to be alot of elegant solutions, and new stuff/thinking.

Seems? Elegant solutions? We haven't even seen the console yet!
It's all well and good to use buzzwords to hype something up, but don't use them before we have even seen the hardware.

JRPGfan said:

Sarcasm I take it?... but how many other consumer products do you know that use it currently? Do you see other CPU or GPU solutions that do so?
How many other products do you see a 3D stacked chip cooled on both side? I've never seen that before.

Chip stacking has a ton of caveats... And like you alluded to, cooling is a big one... Implementing an additional cooling layer between stacked chips increases complexity and reduces yields and is entirely unnecessary in scenarios where you have enough space (I.E. PC and Console) when you can simply go the chiplet route and interface those chips with a larger and more efficient surface area to draw heat from. - Basic thermodynamics comes into play.

However 3D chip stacking isn't new or novel... The PSP from 2004 even had a stacked Toshiba eDRAM chip.

HBM Memory uses an interposer with stacked RAM chips on top.

This isn't a new technology invented by Sony.


1) "They are patents. They aren't actual products. " - Permalite

I know that, but this appears to for the PS5, and will likely be for a real product.

 
2)  "Variable clockrates that aren't developer controlled/influenced isn't a good thing..."

Theres nothing that isnt good for something.  This is fantastic for reduceing overhead, and increaseing power effeciency.
Thats why you see it on the PC, mobil, laptop, smartphone side.  Which in turn could mean you get a small chip to run faster, and waste less power.

Is it fantastic for the developer? maybe not, but overall I'd say its a good thing (for the hardware).


3) "We have been stacking NAND chips for years."

That form of stacking is differnt than this one right? Those are layers, to reduce the size of it
(if your layout was side by side, the SSD would be thin as a sheet of paper but gigantic):
They dont really take any advantage of the fact that they are ontop of one another (afaik).

4)  "Controller is just an evolution of past designs, it's different, but it's not going to be playing games any differently,..... ."

"Wired were also given the opportunity to play a version of Gran Turismo Sport on the PS5 dev-kit, which allowed them to experience the improved haptic feedback, which they described as "Driving on the border between the track and the dirt, I could feel both surfaces."

This sensation, Wired explains, "disappeared entirely" when they replayed the same track using the DualShock 4, which emphasises exactly how much the haptic feedback has been improved in the DualSense." - Wired

Theres been other journalist and tech guys, that have tried it, and says it plays very differntly than the past controllers.


5)  "Please elaborate on those bottlenecks."

They have a custom memory controller, that allows for upto six levels of priority when reading from the SSD (when normally its ~2).

example:  "While textures are being loaded, an enemy might be shot and have to saying some dying words. That interrupts the texture loading."

On sony's SSD that npc that dies, and says some words, wont effect loading speeds of other things needed to be loaded.
While on Xbox it probably does.  Will this actually be something noticeable ingame? Who knows.

https://www.youtube.com/watch?v=ph8LyNIT9sg

Watch from 12m onwards.

Check-in (copying data from one place to another) takes roughly 1 next gen CPU core. And thats just the tip of the iceberg, if all the overheads get 100 times larger, that will cripple the player as soon as that massive stream of data comes off the ssd.

1) Custom flash controller (to handle that) and custum units inside the APU.

2) Kracken (compression)  (turns those 5,5GBS (raw) -> 8/9 GBS (typical) and can go upto 22GB/s (max theoretical))
(in terms of cpu performance (for compression), this kracken chip is like 9 cpu cores, in the PS5)

3) Dedicated DMA controller (which again is like 1 cpu core worth saved)

4) I/O Co-proccessors that direct the custom hardware parts (again offloading the cpu)
(1 of them, is for SSD IO, lets sony by-pass traditional file IO, as it bottlenecks when reading ssd)
(1 of them, is for memory mapping, mapping and remapping as part of file IO, and this too as a bottleneck)

5) Coherancy engines (to assist the co-proccessors, mainly data in the gpu caches, flushing gpu cashe when ssd is read, is a "unattractive option" and can really hurt the GPUs performance, so sony implimented a "gentle" way of doing things, were coherancy engines talk to the gpu and inform it of overhead ranges and scrubbers... bla bla bla)

honestly just go watch Mark Cerny again, he talks about these bottlenecks and how they delt with them in the video.

6) "Seems? Elegant solutions? We haven't even seen the console yet!
It's all well and good to use buzzwords to hype something up, but don't use them before we have even seen the hardware."

Fair enough, so far its just a patent that points out how the PS5 APU might be cooled, and that its useing 3D stacking.
Its not a actual product yet.

7) "Chip stacking has a ton of caveats... And like you alluded to, cooling is a big one... Implementing an additional cooling layer between stacked chips increases complexity and reduces yields and is entirely unnecessary in scenarios where you have enough space (I.E. PC and Console) when you can simply go the chiplet route and interface those chips with a larger and more efficient surface area to draw heat from. - Basic thermodynamics comes into play."

Yes you have more heat, in a smaller area, makeing cooling abit harder.
The upside is ; its more power-effecient (which means less heat overall) and has lower latency when moveing data between parts.

The idea that they just cool the 3D chip from both sides, kinda off-sets that draw back though right?
You just doubled the area were heat can disspate from.

Also this isnt a "additional cooling layer between stacked chips"
The ilusstration shows that its cooled from the front of the chip, and the back side of the chip.
Theres no cooling "inside" the layers of the chip or such as you worded it.

Last edited by JRPGfan - on 24 April 2020

Around the Network

Nm 

Last edited by sales2099 - on 24 April 2020

Xbox: Best hardware, Game Pass best value, best BC, more 1st party genres and multiplayer titles. 

 

JRPGfan said:

Im gonna go ahead and say I'm impressed.

Sony seems to be doing alot of "new" things.
The variable clock speeds, to maximise performance without needing to waste anything on overhead.
The 3D stacked chip, and the unique cooling solution seem, like something we've heard of would be comeing in the future, but no one really expected to see anytime soon.

Even the controller is new, its no longer called a dualshock, but now the dualSense, after the changes it got in tech.
This ontop of stuff like all the work they put into removeing bottlenecks, when moveing data around (from the SSD).

Alot of this stuff just "seems" more planned out than what Microsoft did.
With the Xbox series X, there wasnt any "surprises", it was basically just more or less what everyone thought it would be.
With the Playstation 5, there seems to be alot of elegant solutions, and new stuff/thinking.

I’d argue they have been behind on hardware for several years and are only now catching up. The only reason ps4 was more powerful then Xbox One was Mattrick didn’t prioritize gaming specs. Both consoles had weak specs at launch even for its time.

PS4 and even some models of PS4 Pro are extremely loud (XBX is whisper quiet) and couldn’t do 4K bluray or 4K gaming as advertised. PS5 details sound impressive, and they are, but it’s because it’s been long overdue. Same goes for the controller. I always thought it was ergonomically lacking. And the redesign now finally makes PSs controller modern and DualShock is officially obsolete. Xbox controller didn’t need to change much because it doesn’t require a change. 

People were surprised by the Series X tower design. It definitely breaks tradition. And with its design it allows for 70% better heat dissipation by simple fact that heat rises and the giant fan only helps further. If PS5 has a flat rectangle shape, they certainly have their work cut out for them to do every trick they can to make the heat dissipation work. 

I’d argue MSs Series X specs is all about consistency where as Sony’s “variable” this and “work arounds” sound a lot like an asterisk (*) should be next to all their claims. Just giving the opposite point of view. 

Last edited by sales2099 - on 24 April 2020

Xbox: Best hardware, Game Pass best value, best BC, more 1st party genres and multiplayer titles. 

 

Of course Sony also the one who invent cooling fan so all of the Sony "fan" boy are not mad at Sony . /s

On Topic it probably costly but i think it's not that very costly like every one said. Sony has made some calculation on how this will cost effectively. But i am just gald they are thinking this stuff after learned from PS4 jet cooling design.



Now wait just a moment. That... that looks like a tower. Omg, they actually did it, they copied Microsoft. Unbelievable.

Edit: Sarcasm.

Last edited by Zoombael - on 25 April 2020

Hunting Season is done...

Around the Network
JRPGfan said:

1) "They are patents. They aren't actual products. " - Permalite

I know that, but this appears to for the PS5, and will likely be for a real product.

Sony has a patent for cartridges. Didn't mean the Playstation 5 will be using carts.

And how do you know it will be for the Playstation 5? Gut feeling? Need more than that to assert that something is true I am afraid.

JRPGfan said:

2)  "Variable clockrates that aren't developer controlled/influenced isn't a good thing..."

Theres nothing that isnt good for something.  This is fantastic for reduceing overhead, and increaseing power effeciency.
Thats why you see it on the PC, mobil, laptop, smartphone side.  Which in turn could mean you get a small chip to run faster, and waste less power.

Is it fantastic for the developer? maybe not, but overall I'd say its a good thing (for the hardware).

It doesn't reduce overhead. (Whatever that is supposed to mean?)

PC and mobile uses it because not all components are taxed at 100%, thus there is headroom to drive up clockrates or reduce power consumption and thus heat.

Consoles are innately different in that regard where we can assume that the developer will be using every component to it's maximum potential in order to drive overall simulation quality.

JRPGfan said:



3) "We have been stacking NAND chips for years."

That form of stacking is differnt than this one right? Those are layers, to reduce the size of it
(if your layout was side by side, the SSD would be thin as a sheet of paper but gigantic):
They dont really take any advantage of the fact that they are ontop of one another (afaik).

It's still stacking, it's just a variant of it.
I also mentioned HBM which stacks memory on top of a Silicon Interposer and logic.

https://www.anandtech.com/show/15469/jedec-updates-hbm2-memory-standard-to-32-gbps-samsungs-flashbolt-memory-nears-production



It's nothing new, Sony didn't go out and invent the wheel again.

And you are correct, that if an SSD had all it's NAND laid out in a planar form factor the size would be massive, hence the need for stacking, there isn't the same need for stacked processors, otherwise AMD would have done the same with it's chiplets.

JRPGfan said:


4)  "Controller is just an evolution of past designs, it's different, but it's not going to be playing games any differently,..... ."

"Wired were also given the opportunity to play a version of Gran Turismo Sport on the PS5 dev-kit, which allowed them to experience the improved haptic feedback, which they described as "Driving on the border between the track and the dirt, I could feel both surfaces."

This sensation, Wired explains, "disappeared entirely" when they replayed the same track using the DualShock 4, which emphasises exactly how much the haptic feedback has been improved in the DualSense." - Wired

Theres been other journalist and tech guys, that have tried it, and says it plays very differntly than the past controllers.

It's still a controller that is based on past controller designs, it's nothing new, novel or different.

It still uses the same button layout, it still uses two analogue sticks, still uses a D-Pad, it even copied the ergonomics from Microsoft. - It has a few extra features bolted-on in a new shell, it's NOT a new or revolutionary change in how you play games with a controller, it's an evolution of past designs... Let's not try and frame it as anything else other than that... Otherwise that is just being disingenuous.

JRPGfan said:

5)  "Please elaborate on those bottlenecks."

They have a custom memory controller, that allows for upto six levels of priority when reading from the SSD (when normally its ~2).

example:  "While textures are being loaded, an enemy might be shot and have to saying some dying words. That interrupts the texture loading."

On sony's SSD that npc that dies, and says some words, wont effect loading speeds of other things needed to be loaded.
While on Xbox it probably does.  Will this actually be something noticeable ingame? Who knows.

https://www.youtube.com/watch?v=ph8LyNIT9sg

Watch from 12m onwards.

Check-in (copying data from one place to another) takes roughly 1 next gen CPU core. And thats just the tip of the iceberg, if all the overheads get 100 times larger, that will cripple the player as soon as that massive stream of data comes off the ssd.

It is all well and good to parrot information you heard elsewhere... But you didn't really elaborate on what those actual bottlenecks are.

Because the Xbox Series X and PC also removes those "bottlenecks" by moving to an SSD from a mechanical disk and it doesn't use that priority level system.

Is the only bottleneck the storage? What about the rest of the system?
I am not asking these questions because I don't know, I am asking these questions to see if you know and actually understand them.

JRPGfan said:


1) Custom flash controller (to handle that) and custum units inside the APU.

2) Kracken (compression)  (turns those 5,5GBS (raw) -> 8/9 GBS (typical) and can go upto 22GB/s (max theoretical))
(in terms of cpu performance (for compression), this kracken chip is like 9 cpu cores, in the PS5)

3) Dedicated DMA controller (which again is like 1 cpu core worth saved)

4) I/O Co-proccessors that direct the custom hardware parts (again offloading the cpu)
(1 of them, is for SSD IO, lets sony by-pass traditional file IO, as it bottlenecks when reading ssd)
(1 of them, is for memory mapping, mapping and remapping as part of file IO, and this too as a bottleneck)

5) Coherancy engines (to assist the co-proccessors, mainly data in the gpu caches, flushing gpu cashe when ssd is read, is a "unattractive option" and can really hurt the GPUs performance, so sony implimented a "gentle" way of doing things, were coherancy engines talk to the gpu and inform it of overhead ranges and scrubbers... bla bla bla)

honestly just go watch Mark Cerny again, he talks about these bottlenecks and how they delt with them in the video.

Points, 1, 2 and 3 all relate to the same thing.

What are those custom units you speak of?

What Kraken chip? You aren't confusing things are you? Because the Kraken decompressor isn't actually a chip, it's a fixed function unit on the APU.

DMA controller isn't anything new.

I/O Co-Processors isn't anything new.

Coherancy engines? Again, nothing new. Playstation 4 gamers talked up coherency last generation... Still waiting to see the results.

And I don't want to re-watch Cerny's video, I know exactly what he is talking about, I am trying to see what your perception of his technical explanations are.

JRPGfan said:

6) "Seems? Elegant solutions? We haven't even seen the console yet!
It's all well and good to use buzzwords to hype something up, but don't use them before we have even seen the hardware."

Fair enough, so far its just a patent that points out how the PS5 APU might be cooled, and that its useing 3D stacking.
Its not a actual product yet.

"might" being the key operative word.

JRPGfan said:

7) "Chip stacking has a ton of caveats... And like you alluded to, cooling is a big one... Implementing an additional cooling layer between stacked chips increases complexity and reduces yields and is entirely unnecessary in scenarios where you have enough space (I.E. PC and Console) when you can simply go the chiplet route and interface those chips with a larger and more efficient surface area to draw heat from. - Basic thermodynamics comes into play."

Yes you have more heat, in a smaller area, makeing cooling abit harder.
The upside is ; its more power-effecient (which means less heat overall) and has lower latency when moveing data between parts.

The idea that they just cool the 3D chip from both sides, kinda off-sets that draw back though right?
You just doubled the area were heat can disspate from.

Also this isnt a "additional cooling layer between stacked chips"
The ilusstration shows that its cooled from the front of the chip, and the back side of the chip.
Theres no cooling "inside" the layers of the chip or such as you worded it.

How is it more power efficient? And how would it effect latency? And how does any of that even matter?







--::{PC Gaming Master Race}::--

"How is it more power efficient? And how would it effect latency? And how does any of that even matter?"

You know that exsample with discribeing the a wormhole, as a piece of paper you fold, and you punch through it? to shorten the distance?
Its like that.

The futher apart these things are spaced out, the higher the power draw (and travel time).
By makeing usage of the 3rd dimension, you can have a chip ontop of a chip, with a bridge (thats a very short distance) between the two.
This means that parts of the chip, that would sometimes be spaced out far apart in a 2D flat chip, and thus have to move longer distances, can now, have parts that dont need to do that, because the stacked chip ontop is bridged to the one underneath/ontop of it.

You sound like you dont understand the differnce from stacking something ontop of one another, with no comuncation between chips, vs a chip designed around a 3D plane.

Theres massive gains for doing so.


From neogaf: mckmas8808

"Compared to this traditional 2D architecture, 3D ICs provide several significant advantages:

1. Footprint
Obviously, stacking multiple dies atop one another produces a chip that takes up less space than if those dies were side by side. If the layers are aggressively thinned, a multi-layer 3D-IC is actually no thicker than a traditional 2D chip. The tiny size of 3D-ICs is extremely valuable in miniaturized devices such as cell phones and IoT applications.

2. Speed
Dies stacked in a 3D chip are much closer together than chips on a circuit board. The shorter distances allow electronic signals to travel more quickly from one component to another. 3D stacked devices have shown as much as 5x speed improvement over comparable 2D solutions.

3. Power
Shorter connections automatically require less power, but 3D ICs have another power-saving trick. When an electronic signal travels from one chip to another, it passes through special circuitry that screens out any accidental electrostatic discharge (ESD). These ESD filters consume energy. Signals that travel from one layer to another within a 3D-IC do not require ESD checks. Tests have seen as much as 90% reduction in power consumption.

4. Heterogeneous Integration
Because the layers in a 3D IC are manufactured separately, they can be built differently. This is more important than it might seem! The process in which a die is built affects the behavior of the components on that die: one process makes better capacitors, another makes faster transistors, etc. Even more interesting, the layers may be built at different process nodes – that is, the electronic components may differ in size. This affects the cost, complexity, and performance of each layer. It is even possible to stack layers that are built of different materials. All of these possibilities mean that a 3D IC can combine the best of each process, node, and substrate without compromising some components to accommodate others. In fact, a multi-die stack can contain combinations that are flatly impossible to achieve on a 2D chip."

"How is it more power efficient? And how would it effect latency?"

Does it make sense now?
its like even if you have a gas-effecient car, the longer the road you travel, the more gas you'll use.
3D stacking = shorter roads. 
Also long road vs short road = less time travelled.


"how does any of that even matter?"

Because consoles are limited in scope by power consumption?
The more power something uses the bigger the cooling unit, the more bigger the fans or the more noisy (usually).

This allows for more performance in a smaller size, less cooling issues/noise.

----------------------------------------------------------->

Now onto the Variable clockrates thingy you asked about.

What is overhead?
We'll if you "lock" a clock rate, you actually dont lock the power consumption of a thing.
Because depending on workloads, even af same clock speeds, differnt tasks will require differnt amounts of power consumption.

So... how do you fix that? If you want to lock in a clock speed, and you know your power consumption is going to jump up and down, even at same clocks?

Well the "fix" (so its not unstable or crashing) is to design it around always haveing access to more power than it normally needs.

So its always just "wasteing" power, to make sure, it can run stable, at a locked in clock rate.


If you design around a locked power consumption, this issue is avoided.
Instead you will have your clock speeds go up and down, based on workloads, to stay within a given power consumpion.

This means you dont have to "waste" a certain amount of power consumption, when not needed.
This means you can design a chip thats more power effecient, and thus also easier to cool.
Also because its load is always locked, you can design the cooler around it, so it wont ramp up and down and be crazy noisy on some games.

What are the downsides to this? Like cerny said, sometimes it ll power down like 2% or so in cpu/gpu performance, but saveing upto 10% powerconsumption.
Developers are going to have to accept that they cant always use 100% of both cpu+gpu at the same time, they might have to take 2% from one or the other, if it reaches the power consumption cap.

What are the upsides to this?
Well you can maximise performance to a higher level, because you get around the "bouncy" powerconsumption levels like this.
Another could be, you could use it to have a more effecient chip. (higher performance/watt, because no wasted overhead).

---------------------

About the SSD and bottlenecks.

Do you know for a fact that the Xbox series X has dedicated hardware units, to offload the cpu/gpu(cashe), and do everything Sony mentioned, when it talked about SSD and bottlenecks?

Suppose Xbox needs to do some of these funktions, but instead of offloading it another unit, runs it on the CPU?
That could mean the extra mhz the Xbox series X has, is used on that, and the playing field, in CPU terms are closer than speeds might suggest?

or some of these tricks used by sony, even if not taxing for the cpu, might just be done at slower speeds, compaired to sony's solution.
(because they might lack such units)

Why did sony get rid of the "overhead" (waste of power) & use 3D stacked chips (energy effecient) ?

I think its to drastically reduce the size of the chip.
The difference was already quite large (36 cu vs 52 cu), the power effeciency was used to run the chips at higher clock rates.
This is why the PS5 chip runs at such high speeds for its GPU.  Its the gains of smart techniques (turned into savings on the chip costs).

This allows for a smaller and probably much cheaper chip.

Last edited by JRPGfan - on 24 April 2020

the only thing this implies is that the ps5 actually might be flat



 "I think people should define the word crap" - Kirby007

Join the Prediction League http://www.vgchartz.com/predictions

Instead of seeking to convince others, we can be open to changing our own minds, and seek out information that contradicts our own steadfast point of view. Maybe it’ll turn out that those who disagree with you actually have a solid grasp of the facts. There’s a slight possibility that, after all, you’re the one who’s wrong.

Pemalite said:
JRPGfan said:

1) "They are patents. They aren't actual products. " - Permalite

I know that, but this appears to for the PS5, and will likely be for a real product.

Sony has a patent for cartridges. Didn't mean the Playstation 5 will be using carts.

And how do you know it will be for the Playstation 5? Gut feeling? Need more than that to assert that something is true I am afraid.

JRPGfan said:

2)  "Variable clockrates that aren't developer controlled/influenced isn't a good thing..."

Theres nothing that isnt good for something.  This is fantastic for reduceing overhead, and increaseing power effeciency.
Thats why you see it on the PC, mobil, laptop, smartphone side.  Which in turn could mean you get a small chip to run faster, and waste less power.

Is it fantastic for the developer? maybe not, but overall I'd say its a good thing (for the hardware).

It doesn't reduce overhead. (Whatever that is supposed to mean?)

PC and mobile uses it because not all components are taxed at 100%, thus there is headroom to drive up clockrates or reduce power consumption and thus heat.

Consoles are innately different in that regard where we can assume that the developer will be using every component to it's maximum potential in order to drive overall simulation quality.

JRPGfan said:



3) "We have been stacking NAND chips for years."

That form of stacking is differnt than this one right? Those are layers, to reduce the size of it
(if your layout was side by side, the SSD would be thin as a sheet of paper but gigantic):
They dont really take any advantage of the fact that they are ontop of one another (afaik).

It's still stacking, it's just a variant of it.
I also mentioned HBM which stacks memory on top of a Silicon Interposer and logic.

https://www.anandtech.com/show/15469/jedec-updates-hbm2-memory-standard-to-32-gbps-samsungs-flashbolt-memory-nears-production



It's nothing new, Sony didn't go out and invent the wheel again.

And you are correct, that if an SSD had all it's NAND laid out in a planar form factor the size would be massive, hence the need for stacking, there isn't the same need for stacked processors, otherwise AMD would have done the same with it's chiplets.

JRPGfan said:


4)  "Controller is just an evolution of past designs, it's different, but it's not going to be playing games any differently,..... ."

"Wired were also given the opportunity to play a version of Gran Turismo Sport on the PS5 dev-kit, which allowed them to experience the improved haptic feedback, which they described as "Driving on the border between the track and the dirt, I could feel both surfaces."

This sensation, Wired explains, "disappeared entirely" when they replayed the same track using the DualShock 4, which emphasises exactly how much the haptic feedback has been improved in the DualSense." - Wired

Theres been other journalist and tech guys, that have tried it, and says it plays very differntly than the past controllers.

It's still a controller that is based on past controller designs, it's nothing new, novel or different.

It still uses the same button layout, it still uses two analogue sticks, still uses a D-Pad, it even copied the ergonomics from Microsoft. - It has a few extra features bolted-on in a new shell, it's NOT a new or revolutionary change in how you play games with a controller, it's an evolution of past designs... Let's not try and frame it as anything else other than that... Otherwise that is just being disingenuous.

JRPGfan said:

5)  "Please elaborate on those bottlenecks."

They have a custom memory controller, that allows for upto six levels of priority when reading from the SSD (when normally its ~2).

example:  "While textures are being loaded, an enemy might be shot and have to saying some dying words. That interrupts the texture loading."

On sony's SSD that npc that dies, and says some words, wont effect loading speeds of other things needed to be loaded.
While on Xbox it probably does.  Will this actually be something noticeable ingame? Who knows.

https://www.youtube.com/watch?v=ph8LyNIT9sg

Watch from 12m onwards.

Check-in (copying data from one place to another) takes roughly 1 next gen CPU core. And thats just the tip of the iceberg, if all the overheads get 100 times larger, that will cripple the player as soon as that massive stream of data comes off the ssd.

It is all well and good to parrot information you heard elsewhere... But you didn't really elaborate on what those actual bottlenecks are.

Because the Xbox Series X and PC also removes those "bottlenecks" by moving to an SSD from a mechanical disk and it doesn't use that priority level system.

Is the only bottleneck the storage? What about the rest of the system?
I am not asking these questions because I don't know, I am asking these questions to see if you know and actually understand them.

JRPGfan said:


1) Custom flash controller (to handle that) and custum units inside the APU.

2) Kracken (compression)  (turns those 5,5GBS (raw) -> 8/9 GBS (typical) and can go upto 22GB/s (max theoretical))
(in terms of cpu performance (for compression), this kracken chip is like 9 cpu cores, in the PS5)

3) Dedicated DMA controller (which again is like 1 cpu core worth saved)

4) I/O Co-proccessors that direct the custom hardware parts (again offloading the cpu)
(1 of them, is for SSD IO, lets sony by-pass traditional file IO, as it bottlenecks when reading ssd)
(1 of them, is for memory mapping, mapping and remapping as part of file IO, and this too as a bottleneck)

5) Coherancy engines (to assist the co-proccessors, mainly data in the gpu caches, flushing gpu cashe when ssd is read, is a "unattractive option" and can really hurt the GPUs performance, so sony implimented a "gentle" way of doing things, were coherancy engines talk to the gpu and inform it of overhead ranges and scrubbers... bla bla bla)

honestly just go watch Mark Cerny again, he talks about these bottlenecks and how they delt with them in the video.

Points, 1, 2 and 3 all relate to the same thing.

What are those custom units you speak of?

What Kraken chip? You aren't confusing things are you? Because the Kraken decompressor isn't actually a chip, it's a fixed function unit on the APU.

DMA controller isn't anything new.

I/O Co-Processors isn't anything new.

Coherancy engines? Again, nothing new. Playstation 4 gamers talked up coherency last generation... Still waiting to see the results.

And I don't want to re-watch Cerny's video, I know exactly what he is talking about, I am trying to see what your perception of his technical explanations are.

JRPGfan said:

6) "Seems? Elegant solutions? We haven't even seen the console yet!
It's all well and good to use buzzwords to hype something up, but don't use them before we have even seen the hardware."

Fair enough, so far its just a patent that points out how the PS5 APU might be cooled, and that its useing 3D stacking.
Its not a actual product yet.

"might" being the key operative word.

JRPGfan said:

7) "Chip stacking has a ton of caveats... And like you alluded to, cooling is a big one... Implementing an additional cooling layer between stacked chips increases complexity and reduces yields and is entirely unnecessary in scenarios where you have enough space (I.E. PC and Console) when you can simply go the chiplet route and interface those chips with a larger and more efficient surface area to draw heat from. - Basic thermodynamics comes into play."

Yes you have more heat, in a smaller area, makeing cooling abit harder.
The upside is ; its more power-effecient (which means less heat overall) and has lower latency when moveing data between parts.

The idea that they just cool the 3D chip from both sides, kinda off-sets that draw back though right?
You just doubled the area were heat can disspate from.

Also this isnt a "additional cooling layer between stacked chips"
The ilusstration shows that its cooled from the front of the chip, and the back side of the chip.
Theres no cooling "inside" the layers of the chip or such as you worded it.

How is it more power efficient? And how would it effect latency? And how does any of that even matter?

I will dispute the DualSense copied the ergo from Xbox controller. Someone have posted a gif comparing the visuals from DS4 to DS and Xbox, and it is much much much closer to DS4.



duduspace11 "Well, since we are estimating costs, Pokemon Red/Blue did cost Nintendo about $50m to make back in 1996"

http://gamrconnect.vgchartz.com/post.php?id=8808363

Mr Puggsly: "Hehe, I said good profit. You said big profit. Frankly, not losing money is what I meant by good. Don't get hung up on semantics"

http://gamrconnect.vgchartz.com/post.php?id=9008994

Azzanation: "PS5 wouldn't sold out at launch without scalpers."

DonFerrari said:

I will dispute the DualSense copied the ergo from Xbox controller. Someone have posted a gif comparing the visuals from DS4 to DS and Xbox, and it is much much much closer to DS4.

Huh?

I would like to see this Gif.

Last edited by Azzanation - on 25 April 2020