• 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.

When do you think the PS5 reveal will take place?

  • January

    Votes: 6 0.3%
  • February

    Votes: 1,172 65.7%
  • March

    Votes: 273 15.3%
  • April

    Votes: 81 4.5%
  • May

    Votes: 116 6.5%
  • June

    Votes: 48 2.7%
  • Later

    Votes: 89 5.0%

  • Total voters
    1,785
  • Poll closed .
Status
Not open for further replies.

vivftp

Member
Oct 29, 2017
19,754

source-2.gif
 

vivftp

Member
Oct 29, 2017
19,754
Do we think that next month we're getting a PS5 focused State of Play?

One possibility is a SoP where the press are simultaneously at a location playing around with the hardware for their articles. In that scenario Sony could have already sent out invites but requested the press remain silent until a certain date.

The only thigh that concerns me is that SOPs usually get announced close to the actual date. I need 3 weeks notice to book days off of work!!! 😩
 

modiz

Member
Oct 8, 2018
17,831
no disadvantage other than "harder to make", but it's a complete win for customers
None. The IO die just makes their chips easier to make as their overall yields ar better.

shit..beaten lol.
so a CPU with less cache but no separate IO die could perform on par with the desktop parts technically? i remember reading that latency is actually one of the bigger problems with zen 2 desktop CPUs.
 

c0c0suma

Banned
Jan 20, 2018
79
so a CPU with less cache but no separate IO die could perform on par with the desktop parts technically? i remember reading that latency is actually one of the bigger problems with zen 2 desktop CPUs.
yes, should be performing better and reducing the space where the original large IO die would take
 

anexanhume

Member
Oct 25, 2017
12,913
Maryland
I've completed remeasuring the different functional blocks in Navi, based on the new Fritzchens Fritz die shot. Here's the new figures. I've also updated my predictions based on this, and to account for the likely need for additional ROPs and cache in the higher-WGP PS5. If any questions arise, let me know.

Mecha Meister , I saw that someone had asked my prior post be threadmarked. If you decide to do this, please mark this post instead. Of course, you may also determine that neither is of sufficient significance.

naviblocksizes2y9j75.png


nextgenpredict27vjge.png
Nice work. What I/O and periphery circuitry are you accounting for?
If it's not Phison, then the reddit leak which had the die size 315mm2 and Phison ssd controller is wrong. We can easily discard that reddit rumor now
It makes it less likely, but it's not unfeasible Sony would use a COTS solution if their device wasn't ready.
Maybe this is where the DS lightbar has gone (if, indeed, it is gone)? A Sony patent application for an inside-out tracking system. Looks a lot like a variant of Steam Lighthouse tech, projector(s) flood the environment with beams of invisible light, photosensors on the controller or HMD detect those beams and work out position/orientation.

Systems and methods for using multiple MEMS projectors to determine a position of a photosensor of an HMD or another controller

Good catch. Appears to be exactly what it is.

__________________
credit to AbsoluteBeginner at B3D. Looks like 5600 XT consumes 24W less than 5700 in similar conditions. Even accounting for memory consumption, it sounds like N7P gains could be in play.


Sony also looking into NFC tags. Kratos Amibo here we come!

 

Hey Please

Avenger
Oct 31, 2017
22,824
Not America
Liabe Brave Thank you kindly for explanation as well as the new diagrams. Certain aspects became crystal clear hence.

I do wonder, however, what does having a physically larger front end area mean in terms of what constitutes this extra size (in 5500) over its more potent sibling, 5700/XT.
 

Thera

Banned
Feb 28, 2019
12,876
France
One possibility is a SoP where the press are simultaneously at a location playing around with the hardware for their articles. In that scenario Sony could have already sent out invites but requested the press remain silent until a certain date.
A thing that big you have leaked throught smaller journalist / influencers.
I think we will have a main event, I don't know in which form, and I can see what you are describing one or two months after that.

I can see everyone just waiting the date while, as a french, I am waiting the event to NOT be at 3 AM here. I trust you Jim !
 

Deleted member 17402

User requested account closure
Banned
Oct 27, 2017
7,125
How much smaller can they get beyond 5nm die? Where does technology go? I am not well versed so I apologize if the way I word my question is odd. Do things get stacked? What are they going to do to continue progression? You can't get smaller forever. Where will it stop and what after will they have to do?
 
Oct 27, 2017
4,642
man what the hell is going on at intel?

AMD just need to add TB3 / USB4 and go in for the kill now...
 
Last edited:

gremlinz1982

Member
Aug 11, 2018
5,331
Sorry man but i vehemently disagree..if Sony buys Zenimax and keep Elder Scrolls from being 3rd party that's not how the industry works, if that happens then there's no stopping games like Destiny, Assassins Creed etc. going exclusive to a platform holder. Lets just hope it doesn't happen...Sorry for being off topic
Let whoever has money buy whoever wants to be bought out.


Click play and educate yourself a little on the benefits of reconstruction techniques.



So you're trying to tell me that scaling is better than reconstruction now?



Interesting that actual developers don't share your opinion.
You seem to be on a mission trying to convert me to the message that reconstruction is best. I have not only told you, but I believe someone else that I really do not care whether games are native 4K or reconstructed. I know what both are.

What I am trying to tell you and you fail to understand is this; most developers do not change unless they are forced to. Most will simply stick to whatever solutions they have been going through simply because it saves time, it saves money, and being a business, this is what the bottom line is all about.

Been in this thread and there have been people crying themselves hoarse about how developers ought not waste resources chasing 4K, at least not native. They will chase that because that is what they have always done, and when you are putting games on what is not a closed box environment like PC, chances are that there will be excess power on console. You think they are simply going to take their time reconstructing stuff when they can ship it as is?

Lastly, I look at evidence. You will take a few games where there has been reconstruction. Those are the outliers. Most developers have simply not bothered outside the PS4 Pro, and they only bothered with checkerboard because there was a console with excess power that they could essentially play around with. Otherwise?
 

vivftp

Member
Oct 29, 2017
19,754
A thing that big you have leaked throught smaller journalist / influencers.
I think we will have a main event, I don't know in which form, and I can see what you are describing one or two months after that.

I can see everyone just waiting the date while, as a french, I am waiting the event to NOT be at 3 AM here. I trust you Jim !

Yeah I'd prefer a traditional stage show as well. Hell, I'd like them to rent out a sports arena and stuff it with tens of thousands of screaming fans, but something like PSM 2013 would be fine. 😃
 
Dec 10, 2019
298
Last edited:

CatAssTrophy

Member
Dec 4, 2017
7,611
Texas
Hmm, that Godfall talk is a bit odd. I don't think that game looks *considerably* better than something like Warframe, so being told to keep our expectations in check concerns me.
 
Oct 27, 2017
7,136
Somewhere South
I don't think the Eurogamer dude is saying anything out of knowledge, but out of prudence. "It might end up looking better, or worse. At this point we can't say either with any certainty."
 

Papacheeks

Banned
Oct 27, 2017
5,620
Watertown, NY
What does this even mean? That other PS5 potentially could look worse than Godfall? What an odd comment in general

Sounds like Tom Phillip has a PS5 in front of him. /S

But in all honesty unless he knows the specs people shouldn't be making comments like that.

Only way I would make that comment is if I actually saw it running on PS5 hardware and PC.
 

CrispyGamer

Banned
Jan 4, 2020
2,774
You seem to be on a mission trying to convert me to the message that reconstruction is best. I have not only told you, but I believe someone else that I really do not care whether games are native 4K or reconstructed. I know what both are.

What I am trying to tell you and you fail to understand is this; most developers do not change unless they are forced to. Most will simply stick to whatever solutions they have been going through simply because it saves time, it saves money, and being a business, this is what the bottom line is all about.

Been in this thread and there have been people crying themselves hoarse about how developers ought not waste resources chasing 4K, at least not native. They will chase that because that is what they have always done, and when you are putting games on what is not a closed box environment like PC, chances are that there will be excess power on console. You think they are simply going to take their time reconstructing stuff when they can ship it as is?

Lastly, I look at evidence. You will take a few games where there has been reconstruction. Those are the outliers. Most developers have simply not bothered outside the PS4 Pro, and they only bothered with checkerboard because there was a console with excess power that they could essentially play around with. Otherwise?
You seem to be on a mission trying to convert me to the message that reconstruction is best. I have not only told you, but I believe someone else that I really do not care whether games are native 4K or reconstructed. I know what both are.

What I am trying to tell you and you fail to understand is this; most developers do not change unless they are forced to. Most will simply stick to whatever solutions they have been going through simply because it saves time, it saves money, and being a business, this is what the bottom line is all about.

Been in this thread and there have been people crying themselves hoarse about how developers ought not waste resources chasing 4K, at least not native. They will chase that because that is what they have always done, and when you are putting games on what is not a closed box environment like PC, chances are that there will be excess power on console. You think they are simply going to take their time reconstructing stuff when they can ship it as is?

Lastly, I look at evidence. You will take a few games where there has been reconstruction. Those are the outliers. Most developers have simply not bothered outside the PS4 Pro, and they only bothered with checkerboard because there was a console with excess power that they could essentially play around with. Otherwise?
 

Nachtmaer

Member
Oct 27, 2017
347
so a CPU with less cache but no separate IO die could perform on par with the desktop parts technically? i remember reading that latency is actually one of the bigger problems with zen 2 desktop CPUs.
This has been brought up before when Raven Ridge launched, but the extra cache the discrete CPUs have over AMD's APUs are mostly there for their server platform. This way you can keep more data locally and you don't have to cross over to other CCXs/chiplets/CPUs over the IF and you save on power and latency. Gaming results between a 1500X (which has only four cores but has the 16MB L3 enabled) and AMD's first Zen APUs weren't that different, everything else being equal. I bet when Renoir's desktop SKUs launch and people compare em to a 3700X or whatever with normalised clocks, the difference won't be huge either.

And it's not so much the I/O die causing the latency as it is the way AMD built their memory access in general. Hell, overall latency went down compared to Summit Ridge. But yes, technically a monolithic SoC should have lower memory access than doing a chiplet design because you lose the extra hop that has to go off die.
 

modiz

Member
Oct 8, 2018
17,831
This has been brought up before when Raven Ridge launched, but the extra cache the discrete CPUs have over AMD's APUs are mostly there for their server platform. This way you can keep more data locally and you don't have to cross over to other CCXs/chiplets/CPUs over the IF and you save on power and latency. Gaming results between a 1500X (which has only four cores but has the 16MB L3 enabled) and AMD's first Zen APUs weren't that different, everything else being equal. I bet when Renoir's desktop SKUs launch and people compare em to a 3700X or whatever with normalised clocks, the difference won't be huge either.

And it's not so much the I/O die causing the latency as it is the way AMD built their memory access in general. Hell, overall latency went down compared to Summit Ridge. But yes, technically a monolithic SoC should have lower memory access than doing a chiplet design because you lose the extra hop that has to go off die.
thanks for the correction and clarifications, very interesting.
 

gremlinz1982

Member
Aug 11, 2018
5,331
Having problems with the quote but i think gremlinz1982 you quoted me by accident.
My bad, I added a quote but forgot to reply. What I wanted to say was that in the grand scheme of things, developers get bought out and others step up. You may not see the big teams, but there is always someone to step up and fill that void eventually.

I think that people worry too much about something being taken away from them when this is how gaming ecosystems, especially consoles are built. There will be a lot of great games going forward regardless of who is bought up by any existing player. And for me if there is a game that I really feel I must play, then I get the console where it is at.............you are not being denied anything, it is just that choice is shrinking when it comes to a single developer/publisher.

But these are the things that drive competition.
 
Oct 27, 2017
7,136
Somewhere South
You're right. Can you elaborate a bit more on the color to process connection?

It's less that there's a connection between color and process and more that different colors likely mean different things. Processors are usually capped with some sort of ceramic for increased mechanic resistance and heat dissipation properties - different ceramics have different formulations and, thus, very different properties. Use some sort of metallic dopant and you might have increased heat dissipation at a slightly higher cost to manufacture, or less mechanics resistance, or increased electromagnetic noise, etc.

If AMD saw fit to change the ceramic formula used to cap their stuff, there's some reason to it.
 
Status
Not open for further replies.