• Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
  • We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Status
Not open for further replies.

dgrdsv

Member
Oct 25, 2017
11,879
RT generally scales linearly with resolution, so a Pro version might simply spearhead RT at higher resolutions, eg full RT guaranteed at a native 4K or higher.
Scaling RT this way is the land of totally diminishing returns though. Even going to more than a couple of light bounces tend to produce nearly invisible results on practice - unless we're doing full path tracing and the light is actually rendering the whole scene for us but this isn't likely to be the norm on next gen.
 

褲蓋Calo

Alt-Account
Banned
Jan 1, 2020
781
Shenzhen, China
I'm no expert but voltage/heat reduces life. If you can achieve higher clocks on stock volts, life span would not be reduced. If Sony's cooling is good and the apu works as navi/zen2 do, drops in temps will allow higher clocks on stock/lower voltage.
Higher voltage is required to achieve higher clock, this is rather independent of heat dissipation efficiency. Basically high voltage makes electrons move more freely inside transistors, so that the transistors can switch faster. Otherwise due to propagation delay the circuit will run into instability issues. So there is no higher clock on lower voltage, unless it's a change of processing node.
 
Oct 26, 2017
6,151
United Kingdom
From the reddit comments:

Hmm... intriguing.

I suppose the big pros are that it's a fairly simple representation - no connectivity information to worry about (between points, as in triangles), no uv mapping to worry about. They can be a good representation for certain effects - particle systems are basically point clouds I guess, so for things like smoke, clouds etc. Things like CSG are very easy with point clouds compared to polygons. Dreams uses a point cloud representation for example (alongside an implicit representation). Certain things can be done pretty easily with point clouds compared to polygons - dynamic LoD is a matter of just dropping points. Certain effects might come relatively 'free' like DoF, depending on how you render your point cloud. I guess one of the biggest benefits is that getting a point cloud representation of real objects, from either LIDAR or photogrammetry or the like, is 'easy'. You can readily scan objects into point clouds and get super highly detailed, 'photo-real', models - if you can render them. And that scanning is cheap compared to manual production.

Which leads to the downside - without an implicit companion representation (as in Dreams), they're huge data hogs. Rendering them has challenges - you need to do hole filling. GPUs are obviously not targeted at point rendering specifically. So much middleware is built around polygons - like physics for example. So even if you're using point clouds, you'll probably want another representation on hand for certain other tasks (like Dreams does). Getting decent polygonal models out of point cloud data for that task could be tricky in some cases.

I can't see games giving up polygons any time soon, in general. Though we may see more getting experimental. It doesn't have to be all or nothing, you can mix representations in a single scene or game. I think that capture/render tech could be perfect for something like a racing game or a football game or the like, where you want to reproduce real world locations, or other games that want to bring in 'real world' assets. Data management is the trick though.

Thanks. This is very helpful. Very interested to see how Sony devs approach these challenges with this new tech available to them.

I'm no expert but voltage/heat reduces life. If you can achieve higher clocks on stock volts, life span would not be reduced. If Sony's cooling is good and the apu works as navi/zen2 do, drops in temps will allow higher clocks on stock/lower voltage.

Is it the voltage that reduces the processor lifespan or the heat?

Important question, as the former would mean having a better cooler doesn't make a difference, whereas the latter would allow you to maintain longevity with a higher clocked, higher voltage processor so long as you can control the thermal output.
 

Ploid 6.0

Member
Oct 25, 2017
12,440
Higher voltage is required to achieve higher clock, regardless of heat dissipation efficiency. Basically high voltage makes electrons move more freely inside transistors, so that the transistors can switch faster. Otherwise due to propagation delay the circuit will run into instability issues. So there is no higher clock on lower voltage, unless it's a change of processing node.
Not always, undervolting + overclocking is a thing. You can push it further with more voltage but you can overclock with a undervolt. I do this on GPUs, it's probably not a thing on CPUs since I don't undervolt that, but you can overclock some on stock voltages, same for memory, just not as high when trying to push OCs to the limit.

www.youtube.com

9900K Overclocking & Undervolting - Fighting The Molten i9

Check prices below i9 9900K: http://geni.us/czbBx i7 9700K: http://geni.us/Guo838 i7 8700K: http://geni.us/T3RzA5V NCASE M1 Mods: https://optimumtechmedia.wi...
 

褲蓋Calo

Alt-Account
Banned
Jan 1, 2020
781
Shenzhen, China
Not always, undervolting + overclocking is a thing. You can push it further with more voltage but you can overclock with a undervolt. I do this on GPUs, it's probably not a thing on CPUs since I don't undervolt that, but you can overclock some on stock voltages, same for memory, just not as high when trying to push OCs to the limit.

www.youtube.com

9900K Overclocking & Undervolting - Fighting The Molten i9

Check prices below i9 9900K: http://geni.us/czbBx i7 9700K: http://geni.us/Guo838 i7 8700K: http://geni.us/T3RzA5V NCASE M1 Mods: https://optimumtechmedia.wi...
That's when the stock voltage is already more than enough.
 

Ploid 6.0

Member
Oct 25, 2017
12,440
Is it the voltage that reduces the processor lifespan or the heat?

Important question, as the former would mean having a better cooler doesn't make a difference, whereas the latter would allow you to maintain longevity with a higher clocked, higher voltage processor so long as you can control the thermal output.

Voltage is more likely to kill/degrade components than a temperature that's allowed to skyrocket with no fail safe in place to throttle.

More on that, timestamped @ 3:50 if link doesn't put it there.
youtu.be

Ask GN 92: What Kills CPUs - Heat or Voltage? Used Mining GPUs Safe?

This Ask GN talks about heat degradation of CPUs vs. over-volting to death, used mining GPU safety, AIDA vs. Prime for stress testing, and more. Ad: EVGA's 1...

That's when the stock voltage is already more than enough.
So when are they not enough on a product that should last a long time outside of extreme overclocked conditions (manual, maybe something by the manufacturer to show off something, a clock binned and overclocked for one event not mass produced). Even overclocked cards are shipped with a lot of room for increasing frequency out of the box, they have to be. A console can't ship with bleeding edge voltage + clock, that thing will degrade and need more voltage to hit it's original shipped numbers.
 

褲蓋Calo

Alt-Account
Banned
Jan 1, 2020
781
Shenzhen, China
So when are they not enough on a product that should last a long time outside of extreme overclocked conditions (manual, maybe something by the manufacturer to show off something, a clock binned and overclocked for one event not mass produced). Even overclocked cards are shipped with a lot of room for increasing frequency out of the box, they have to be. A console can't ship with bleeding edge voltage + clock, that thing will degrade and need more voltage to hit it's original shipped numbers.
Welp the stock voltage on PS5 is just gonna be higher than normal cards then.
 

Ploid 6.0

Member
Oct 25, 2017
12,440
Welp the stock voltage on PS5 is just gonna be higher than normal cards then.
It's a consumer product, the voltage on it's components are going to be safe and tested to last. The main issue for consoles is to prevent RROD situations.

The thing that caused RROD is likely the lead-free solder balls used which "when exposed to elevated temperatures for extended periods of time becomes brittle and can develop hair-line cracks that are almost irreparable". There's no way one of the console makers would not be sure to avoid a situation like this, it was devastating for Xbox 360. That's why R&D is important, the PS4 pro is loud, but it's a blower type cooler which are known to be loud, and it's apparently doing the job of not having a RROD situation, not even close to it as that would make big news.

That said, this is why overclocking, or overclocking+undervolting is so nice on PC. All the components that can be overclocked has to ship with some room, even after AMD launched RX 5600 XT with last minute bios updates to overclock them in response to a Nvidia card there was still room to overclock it. There has to be or a lot of people would have problems with a product they bought that wasn't working at stock settings.

www.techradar.com

AMD’s RX 5600 XT gets a speed boost to answer Nvidia’s big RTX 2060 price cut

Revised firmware lets GPU manufacturers ramp up clock speeds, reports claim
 
Last edited:

褲蓋Calo

Alt-Account
Banned
Jan 1, 2020
781
Shenzhen, China
It's a consumer product, the voltage on it's components are going to be safe and tested to last. The main issue for consoles is to prevent RROD situations.

The thing that caused RROD is likely the lead-free solder balls used which "when exposed to elevated temperatures for extended periods of time becomes brittle and can develop hair-line cracks that are almost irreparable". There's no way one of the console makers would not be sure to avoid a situation like this, it was devastating for Xbox 360. That's why R&D is important, the PS4 pro is loud, but it's a blower type cooler which are known to be loud, and it's apparently doing the job of not having a RROD situation, not even close to it as that would make big news.
Yeah? I didn't say the voltage was gonna be so high that it reduces the lifespan, obviously they know the characteristics of the silicon and they will do their best balancing it. I was merely pointing out that you cannot get high clock for free simply with better cooling. And even if the "normal" voltage range is enough (extremely unlikely) they will leave some headroom for stability.
 

Ploid 6.0

Member
Oct 25, 2017
12,440
Yeah? I didn't say the voltage was gonna be so high that it reduces the lifespan, obviously they know the characteristics of the silicon and they will do their best balancing it. I was merely pointing out that you cannot get high clock for free simply with better cooling.
It depends on how it's made to work, how the throttling works. We will have to see how they made their device. For example, Nvidia make sure you can't kill your card by having it throttle based on a safe temperature. If you liquid cool your Nvidia card (Pascal or younger) and keep it in the 40 degrees you can get some awesome overclocks that don't throttle down, the cooler the better. On AMD Polaris cards the throttle wasn't so aggressive, what you got was what it was until temperatures got dangerous and it pull things down.

And even if the "normal" voltage range is enough (extremely unlikely) they will leave some headroom for stability.
I was with your clarification, but this confused me. Normal? Maybe you're saying it's enough for some overclock numbers they posted? We will have to wait and see I think, unless everything to know about their device is known, the TDP, the amount of voltage they will use, the controller quality and more (safe voltages for it, the controller will likely die before the chip).

The only reason I first quoted you is because you mentioned it's not possible to overclock on a undervolt. It's possible, and if this is something Sony is claiming to do I'm sure they have tested it enough to claim it. Maybe they picked or designed their chip based on what they wanted to do.
 

ShapeGSX

Member
Nov 13, 2017
5,224
Is it the voltage that reduces the processor lifespan or the heat?

Both voltage and temperature affect reliability, but voltage affects it more. Electromigration is the main (eventual) killer of chips. A higher voltage means that the electrons move more freely and can literally slam copper atoms out of the way. And heat accelerates this. But this is all simulated. Literally every single wire on the chip is analyzed for reliability. If a wire or a via is too thin for the current and the self heating and overall temperature would bringthe lifetime of that particular wire below a certain time threshold, you just make the wire wider or use a bigger via or two vias. This takes extra design time to fix, which can affect time to market. The resistance goes down and the self heating of the wire decreases and the electrons can travel more slowly and won't knock as many copper atoms out of the way over time.

Electromigration shouldn't be an issue for the lifetime of a product. It is simulated and even tested under accelerated conditions after the chip is made.

Btw, electromigration HAS actually been an issue on products in the past, before they could simulate it. The Commodore 64 had a huge problem with electromigration.

Now, if you operate the chip outside the normal parameters, all bets are off. :)
 

Mathieran

Member
Oct 25, 2017
12,861
It's gonna be really nice having vr with almost no load times. Load times with vr is the worst cause you can't even look at your phone or do anything really
 

褲蓋Calo

Alt-Account
Banned
Jan 1, 2020
781
Shenzhen, China
It depends on how it's made to work, how the throttling works. We will have to see how they made their device. For example, Nvidia make sure you can't kill your card by having it throttle based on a safe temperature. If you liquid cool your Nvidia card (Pascal or younger) and keep it in the 40 degrees you can get some awesome overclocks that don't throttle down, the cooler the better. On AMD Polaris cards the throttle wasn't so aggressive, what you got was what it was until temperatures got dangerous and it pull things down.


I was with your clarification, but this confused me. Normal? Maybe you're saying it's enough for some overclock numbers they posted? We will have to wait and see I think, unless everything to know about their device is known, the TDP, the amount of voltage they will use, the controller quality and more (safe voltages for it, the controller will likely die before the chip).

The only reason I first quoted you is because you mentioned it's not possible to overclock on a undervolt. It's possible, and if this is something Sony is claiming to do I'm sure they have tested it enough to claim it. Maybe they picked or designed their chip based on what they wanted to do.
Assume we have a perfect cooling solution that can always keep the circuit at room temperature, you'll still need to increase voltage to get higher frequency. Since the processors run at such high frequency, the time it takes a transistor to switch state becomes nontrivial, if the switching time is too long, at the end of the clock cycle, the sequential circuits that make up our processor will have wrong result read from their combinational logic paths (propagation delay), which will result in instability like glitches or shutting down. Unless you can cool it down to some super low, sub-zero temperature (where the conductivity becomes much higher and even superconductive), this remains true.

By "normal" I was referring to most RDNA GPU we have seen so far, most of them operate at sub 2.0 GHz clock frequency. If you look at this Reddit user's analysis, you can see to increase the sustained clock from 1.75 GHz to 2.07 GHz, the voltage increases by about 30%. I guess "normal" voltage thereby refers to V <= 1000 mV.

Welp I guess I should've phrased it better. I think they will test each chip individually to assign it a best base voltage.
 
Last edited:

Pheonix

Banned
Dec 14, 2018
5,990
St Kitts
Assume we have a perfect cooling solution that can always keep the circuit at room temperature, you'll still need to increase voltage to get higher frequency. Since the processors run at such high frequency, the time it takes a transistor to switch state becomes nontrivial, if the switching time is too long, at the end of the clock cycle, the sequential circuits that make up our processor will have wrong result read from their combinational logic paths (propagation delay), which will result in instability like glitches or shutting down. Unless you can cool it down to some super low, sub-zero temperature (where the conductivity becomes much higher and even superconductive), this remains true.

By "normal" I was referring to most RDNA GPU we have seen so far, most of them operate at sub 2.0 GHz clock frequency. If you look at this Reddit user's analysis, you can see to increase the sustained clock from 1.75 GHz to 2.07 GHz, the voltage increases by about 30%. I guess "normal" voltage thereby refers to V <= 1000 mV.

Welp I guess I should've phrased it better. I think they will test each chip individually to assign it a best base voltage.
That defeats the purpose of having a fixed peak power solution.

As sony has said, their GPU is "capped" at 2.2Ghz. They built a cooling solution around the peak power that they KNOW they can pass to their APU. They have designed a chip (hence the numerous amount of steppings) for the specific purpose. They have built a cooling solution that they have indicated is admirable to accommodate said chip. They have spent years and millions of dollars doing all this.

Maybe we should just err on the side of caution and assume sony actually know what they are ding?

Has this ben posted?
 
Last edited:

褲蓋Calo

Alt-Account
Banned
Jan 1, 2020
781
Shenzhen, China
That defeats the purpose of having a fixed peak power solution.

As sony has said, their GPU is "capped" at 2.2Ghz. They built a cooling solution around the peak power that they KNOW they can pass to their APU. They have designed a chip (hence the numerous amount of steppings) for the specific purpose. They have built a cooling solution that they have indicated is admirable to accommodate said chip. They have spent years and millions of dollars doing all this.

Maybe we should just err on the side of caution and assume sony actually know what they are ding?
Did I say they didn't know what they are doing? I thought I said the exact opposite.

Each silicon chip that rolls off the production line is tested, to see whether it's fully working, an each chip will have some inevitable variations in its electrical properties, but it will stay in some range so that on the macro level all of them behave the same. This testing process can be taken advantage of to customize the P-state profile to get the best stability and power efficiency. You might want to revisit The "Hovis Method" that was employed in current-gen consoles.
 

foamdino

Banned
Oct 28, 2017
491
Both voltage and temperature affect reliability, but voltage affects it more. Electromigration is the main (eventual) killer of chips. A higher voltage means that the electrons move more freely and can literally slam copper atoms out of the way. And heat accelerates this. But this is all simulated. Literally every single wire on the chip is analyzed for reliability. If a wire or a via is too thin for the current and the self heating and overall temperature would bringthe lifetime of that particular wire below a certain time threshold, you just make the wire wider or use a bigger via or two vias. This takes extra design time to fix, which can affect time to market. The resistance goes down and the self heating of the wire decreases and the electrons can travel more slowly and won't knock as many copper atoms out of the way over time.

Electromigration shouldn't be an issue for the lifetime of a product. It is simulated and even tested under accelerated conditions after the chip is made.

Btw, electromigration HAS actually been an issue on products in the past, before they could simulate it. The Commodore 64 had a huge problem with electromigration.

Now, if you operate the chip outside the normal parameters, all bets are off. :)
Thanks for this - I learn something new every day!
 
Oct 26, 2017
6,151
United Kingdom
Both voltage and temperature affect reliability, but voltage affects it more. Electromigration is the main (eventual) killer of chips. A higher voltage means that the electrons move more freely and can literally slam copper atoms out of the way. And heat accelerates this. But this is all simulated. Literally every single wire on the chip is analyzed for reliability. If a wire or a via is too thin for the current and the self heating and overall temperature would bringthe lifetime of that particular wire below a certain time threshold, you just make the wire wider or use a bigger via or two vias. This takes extra design time to fix, which can affect time to market. The resistance goes down and the self heating of the wire decreases and the electrons can travel more slowly and won't knock as many copper atoms out of the way over time.

Electromigration shouldn't be an issue for the lifetime of a product. It is simulated and even tested under accelerated conditions after the chip is made.

Btw, electromigration HAS actually been an issue on products in the past, before they could simulate it. The Commodore 64 had a huge problem with electromigration.

Now, if you operate the chip outside the normal parameters, all bets are off. :)

Yep, ShapeGSX , this is pretty much exactly what I was looking for. Thanks for the explanation. It's posts like these that make tech threads on Era worth it. 😁
 

Piggus

Member
Oct 27, 2017
4,698
Oregon
It's a consumer product, the voltage on it's components are going to be safe and tested to last. The main issue for consoles is to prevent RROD situations.

The thing that caused RROD is likely the lead-free solder balls used which "when exposed to elevated temperatures for extended periods of time becomes brittle and can develop hair-line cracks that are almost irreparable". There's no way one of the console makers would not be sure to avoid a situation like this, it was devastating for Xbox 360. That's why R&D is important, the PS4 pro is loud, but it's a blower type cooler which are known to be loud, and it's apparently doing the job of not having a RROD situation, not even close to it as that would make big news.

That said, this is why overclocking, or overclocking+undervolting is so nice on PC. All the components that can be overclocked has to ship with some room, even after AMD launched RX 5600 XT with last minute bios updates to overclock them in response to a Nvidia card there was still room to overclock it. There has to be or a lot of people would have problems with a product they bought that wasn't working at stock settings.

www.techradar.com

AMD’s RX 5600 XT gets a speed boost to answer Nvidia’s big RTX 2060 price cut

Revised firmware lets GPU manufacturers ramp up clock speeds, reports claim

I think both companies have learned that a strong bracket is also key to helping prevent that type of failure. The early PS3 and 360 boards would flex a lot under load, which the cheap weakened solder eventually couldn't take and the chip would partially separate from the board. Newer systems like the PS4 Pro have a pretty beefy bracket to help mitigate that flex.
 

Vimto

Member
Oct 29, 2017
3,714
That defeats the purpose of having a fixed peak power solution.

As sony has said, their GPU is "capped" at 2.2Ghz. They built a cooling solution around the peak power that they KNOW they can pass to their APU. They have designed a chip (hence the numerous amount of steppings) for the specific purpose. They have built a cooling solution that they have indicated is admirable to accommodate said chip. They have spent years and millions of dollars doing all this.

Maybe we should just err on the side of caution and assume sony actually know what they are ding?

Has this ben posted?

Is that topic worthy?

Even if the cover is legit, it doesn't mean anything. The magazine straight up spits misinformation and repackages old interviews as new information. I don't know how they got the official title.
 

Cthulhu_Steev

Member
Oct 27, 2017
2,387
That defeats the purpose of having a fixed peak power solution.

As sony has said, their GPU is "capped" at 2.2Ghz. They built a cooling solution around the peak power that they KNOW they can pass to their APU. They have designed a chip (hence the numerous amount of steppings) for the specific purpose. They have built a cooling solution that they have indicated is admirable to accommodate said chip. They have spent years and millions of dollars doing all this.

Maybe we should just err on the side of caution and assume sony actually know what they are ding?

Has this ben posted?


With magazine lead times being what they are, this probably means info at the end of this week, or sometime next week.

Even if the cover is legit, it doesn't mean anything. The magazine straight up spits misinformation and repackages old interviews as new information. I don't know how they got the official title.

HYPE THRUSTERS DOWN
 

Binabik15

Member
Oct 28, 2017
4,615
Random Tuesday!

6 weeks since road to PS5 presentation
3 weeks since the Dualsense reveal

Hoping we get something today, anything! I'm Thirsty...

True, true, who isn't thirsty at this point if you like games in general.

Spencer tweeting about the "feel" of games makes me want to know more about DS5, that could be a real game changer in how we experience games...especially if devs give us more gyro assisted aiming. Both could transform our interaction a great deal compared to twin sticks+rumble and I have the slight feeling that this comes as a preparation for PSVR2 and new controllers for that. Games would already have some sort of feedback built in - and hopefully gyro and/or pointer controls - for a glove controller or at least upgrade Move controller to make use of once released. Win-win, makes the games better with DS5 and future proofs them for VR a bit. Same with an emphasis on 3D audio.

Or am I completely crazy thinking this?
 

Cyborg

Banned
Oct 30, 2017
1,955
I would like to see a PS2 kind design for the PS5. In my opinion PS2 is the best looking PlayStation design in history.
 

Webbo

Member
Nov 27, 2017
1,755
United Kingdom
True, true, who isn't thirsty at this point if you like games in general.

Spencer tweeting about the "feel" of games makes me want to know more about DS5, that could be a real game changer in how we experience games...especially if devs give us more gyro assisted aiming. Both could transform our interaction a great deal compared to twin sticks+rumble and I have the slight feeling that this comes as a preparation for PSVR2 and new controllers for that. Games would already have some sort of feedback built in - and hopefully gyro and/or pointer controls - for a glove controller or at least upgrade Move controller to make use of once released. Win-win, makes the games better with DS5 and future proofs them for VR a bit. Same with an emphasis on 3D audio.

Or am I completely crazy thinking this?
Not al all!

I'm also super excited by all the improvements and innovations Sony are doing to better the whole experience, DS features, 3D audio, no loading times etc... these things together should elevate the whole experience over just improved graphics (which will be awesome too).

I just can't wait to see some demos for all this stuff now.
 

MrKlaw

Member
Oct 25, 2017
33,056
I don't think a Pro model is a given this time around. Next-gen consoles feel like they're more aggressively spec'ed this time around so it might be back to the grand tradition of trying to get the costs down over the next 6+ years rather than introducing a mid-generation variant. I don't doubt they're thinking about it but whether they'll follow through may be down to what the competitive landscape looks like down the road.

Agree. Don't see 8k being a driver like 4k was, and the SSD will take a while before 'normal' PC specs significantly outstrip consoles and create a potential demand (a factor Sony mentioned for the pro was keeping people From migrating to PC).

add to that expensive node shrinks and there is a strong case for trying to hold off if possible. Ps6/x2 on the best node (2-3nm)? for decent performance jump. Mid gen refresh on 5nm will limit options
 

Thera

Banned
Feb 28, 2019
12,876
France
Random Tuesday!

6 weeks since road to PS5 presentation
3 weeks since the Dualsense reveal

Hoping we get something today, anything! I'm Thirsty...
We have the "final" date of the last PS4 big games on a random Monday. That means the road is finally free of big PS4 communication.
For a random tuesday? I don't know. For a near future reveal? Probably.
especially if devs give us more gyro assisted aiming
I stopped to hope.
 

Binabik15

Member
Oct 28, 2017
4,615
We have the "final" date of the last PS4 big games on a random Monday. That means the road is finally free of big PS4 communication.
For a random tuesday? I don't know. For a near future reveal? Probably.

I stopped to hope.


They're tracking near everything from Days Gone if you look at their statements about Hordes killed and camps cleared. Surely they know how much better people play with the gyro option and tell their buddies at ND and GG, right 😭
 

SharpX68K

Member
Nov 10, 2017
10,518
Chicagoland
A pair of 36CU chiplets (and maybe a pair of 8c/16t chiplets for the CPU?) plus HBM3 or 4 and a bigger SSD sounds glorious for a PS5 Pro.

A pair of 36CU chiplets, yes, but I highly doubt they would double the CPU cores/threads. Doesn't make sense. Games would have to be built on 8c/16t.
What they would do is clock the CPU past 4 GHz. What I don't know though, is if it would make sense to still use Zen 2 architecture, or use Zen 4 architecture for the IPC boost. Either way, I think consoles will stick with 8c/16t until PS6.

HBM3 or HBM3+ would make sense, but HBM4 (it's due in 2022-2024) would be too new for a console, or not even available.

A bigger SSD, for certain.
 

Kunjiru

Member
Apr 14, 2020
1,483
Damn, I was expecting today's random Tuesday to be the one with the reveal event info. I guess we keep on waiting then
 

--R

Being sued right now, please help me find a lawyer
Member
Oct 25, 2017
12,777
Holy shit the Grubbinator is back at it.
 
Status
Not open for further replies.