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

Forums - Sony Discussion - PS5: Leaked Design And Technology [RUMOUR]

Trumpstyle said:

Hmm my math gives 13.3 Teraflops for 80CU clocked at 1300mhz, you need more Mhz.

About the CPU I believe an 8 core zen2 makes most sense with no hyper-threading. It's easier for backwards compatibility and with threading disable it will allow for higher clocks and better yields which will save a lot of money for Sony and microsoft.

My first choice for storage is a 2TB mechanical drive with 128gb super fast soldered flash storage. This gives both the big storage they will need for next-gen and with super fast flash storage it will make up for the lack of Ram increase. My second choice is 1TB ssd drive, the cheapest 1TB ssd drive on newegg.com is 130$ and flash storage prices is expected to drop 50% next year making 1TB ssd easily affordable.

You are right. Need a little more Mhz. But we are in the ball park and I am already grossly understating what can be possible. 

As for CPU, I don't think BC would be a design consideration to that extent. Reason being that even if they went with 6 cores a 12 threads clocked at 2.5 - 3Ghz, Ryzen 2 architecture is still so much more powerful than the jaguar architecture in the PS4 that they would run circles around the jaguar CPU. And the next CPU being x64 is more important than any number of cores i has as far as BC is concerned.

With regards to storage thats my thinking too. I don't think there will be any kinda hybrid mix and match splution though as that just complicates things. I think they would either solder the storage onto the board or use an M.2 drive. Especially when you consider that as you pointed out nnd flash prices are dropping. By around july 2020 (which is when i expect the PS5 to go into mass production) i wouldnt be surprised if sony can throw in a 1TB sata based M.2 drive for under $30. The real question is if they want to build in future proofing and give their users the option to upgrade their ssd and even put in a nvme drive if they want or if they opt to just save as much money as possible and solder the storage directly onto the PCB. 



Did early (design/tech) leaks of console/controller ever turn out to be true?



Hunting Season is done...

atoMsons said:

This doesn't make sense for this argument. Strange blanket statement.

It makes perfect sense.

atoMsons said:

A CPU only provides a bottleneck in severe cases and there isn't one on the PS4, or the XBO.

Depends, I can point to a ton of games where the CPU is a bottleneck on the Xbox One and Playstation 4.
The CPU bottleneck will shift depending on the game itself and sometimes even the scene that is being displayed on the screen.

atoMsons said:

It's majority of the GPU to produce frames for a video game, 3D pipeline rendering.

The CPU assists in preparing those frames you know.

atoMsons said:

A CPU never provides 60 frames. A CPU is terrible at rendering 3D pipelines.

The CPU assists at rendering in many game engines... It was common especially in the 7th gen.
Shall I point out the rendering techniques the CPU was doing?

atoMsons said:

You clearly haven't any idea why a GPU bottleneck happens.

That is a bold assertion.
I was obviously "dumbing down" my rhetoric to make it more palatable for less technical persons that frequent this forum, if you would like me to stop, I would be more than okay to oblige and start being more technically on point?

atoMsons said:

The CPU is responsible for real-time actions, physics, audio, and a few other processes. If the bandwidth can't match that of the GPU, a bottleneck happens and you lose frames that you can actually use. Think of a partially closed dam. All of the sudden the data can't flow fast enough through the dam(CPU) because of a narrow channel. 

Yawn.
The CPU is responsible for more than that... And you should probably list them, otherwise it is a little hypocritical if you are going to complain about my statement not being fully fleshed out and you go and do the same.
 

atoMsons said:

Now, 60 FPS is a GPU issue. That simple. This isn't a E8500 running a 1080 Ti. 

It is a GPU and a CPU issue. - Sometimes even a RAM issue.

atoMsons said:

PS: Flops ARE everything. It gives a good baseline for performance, even outside of similar architecture in comparison. Just not on a 1:1 ratio in that case (per say NVIDIA/RADEON).

Bullshit it's not everything.
FLOPS or Single Precision Floating Point Operations... Is a Theoretical number.

By that admission alone, Flops is irrelevant... Not only are they irrelevant.. But Flops tells us absolutely nothing about the hardwares actual capability, it doesn't tell us the amount of bandwidth a chip has, it's geometry capabilities, it's texturing capabilities, whether it employs any culling to reduce processing load, whether it has various compression schemes like S3TC or Delta Colour Compression, it tells us nothing of it's quarter floating point/double floating point/integer capabilities... It tells us absolutely nothing.
It's just a theoretical number that is calculated by taking the number of pipelines * instructions per clock * clock.

********************

I will try and keep this as simple as possible... But lets take the Geforce 1030.

DDR4: 884.7Gflops.
GDDR5: 942.3Gflops.

That is a 6.5% difference in Gflops... And you said flops is everything.
And yet we get to the crux of the issue. Gflops doesn't tell us everything else about a GPU, only a theoretical component.
In short... The DDR4 version is often less than half the speed of the GDDR5 version.

But don't take my word for it: https://www.techspot.com/review/1658-geforce-gt-1030-abomination/

****************

Or hows about a different scenario? (There are so many examples I can do this all day.)

Hows about we grab the Terascale based Radeon 5870 that operates at 2.72 Teraflops? It should absolutely obliterate the Radeon 7850 that operates at 1.76 Teraflops, right? That's almost a Teraflops difference huh? Both AMD based.
And yet... Again... Flops is irrelevant as the Radeon 7850 often has a slight edge.
But don't take my word for it: https://www.anandtech.com/bench/product/511?vs=549

Do you want some more examples of how unimportant flops are? I mean, I haven't even started to compare nVidia against AMD yet. Flops is everything right?



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

fatslob-:O said:
EricHiggin said:

Ryzen 2 will be the 3000 series on 7nm, based on the Zen 2 arch. We've had Ryzen 1000 and Ryzen+ 2000 so far. Zen 3 is supposed to be on 7nm+. 

If I remember correctly, Ryzen APU's came out considerably later than Ryzen CPU's did at launch, but while they were based on the Ryzen+ 2000 series, they were still on the 14nm process, not the 12nm process Ryzen+ CPU's were shortly transitioning to. Based on this, PS5 could be Zen 2, but not necessarily on 7nm.

I just meant Samsung isn't the same company as TSMC. I am also not totally convinced of a 2019 launch, that's just what the rumor was based on. Could be 2019 or 2020.

Oh, you'd best believe it's real since Samsung is going to live launch a product based on their logic node utilizing EUV like their next iterations of Galaxy phones ... 

Samsung may not be a pure-play semiconductor foundry like TSMC is but their definitely competing with each other since both of them participate in logic node fabrication ... 

I was referring to your PS5 2019 remark.



Intrinsic said: 
EricHiggin said: 

What about before the PS3 to PS4 transition? What about PS2 to PS3 or PS1 to PS2? Could PS4 have been more powerful? Could they have launched at a $499 price point? Could PS4 have been given a larger case and better cooling? Were PS and MS begging AMD to find time to work with them and not price gouge them? It's not as simple as what can be done in terms of just hardware.

Yes, its not as simple as what can be done with just hardware. But since we are all making educated guesses here, thats the best way to go about it. The begging thing and price gouging thing aren't quantifiable. 

As for launch prices, we ca also make very informed guesses at what to expect with a $400/$500 box.

No they are guesses as well, and any company who's doing better now than they were before, especially because they have the best overall product available in their market, isn't likely to make such a good deal this time around. It still may be a good deal based on the options available, but to assume PS5 parts are sold to PS at the same rates as PS4, is being pretty darn optimistic.

Informed based on what we can typically assume, but do we go by humbled 2013 PS or the world is our's 2006 PS? Do we assume selling at cost remains the name of the game, with MS looking stronger and stronger behind the scenes, and PS burying money since their vaults are full, or do we assume another PS3, or maybe in between? What will the new PS management decide? PS4 is still $299 and Pro is still $399. How would a streaming console from MS or PS effect the dedicated hardware in terms of specs and price? I see way more questions that need to be answered to get a strongly informed idea of what might be coming next gen, and when exactly?

Intrinsic said: 
EricHiggin said: 

With Nvidia and their RTX line, what are the odds AMD has no idea that was in the works, and isn't aiming for something similar? Didn't Cerny mention ray tracing being the holy grail or something like that? What if they try to partially implement that and use a fair amount of resources to push that, instead of everything else they could? Would that also fit under next gen? We all know they like their buzz words. 4k, HDR, so why not ray tracing?

Another question would be how much more powerful than 4.2TF or 6TF is really needed to make a worthwhile jump, in comparison to previous gens? If 4.2TF was worthwhile after just 3 years, why assume they would jump to 14TF after another 3? 10TF would be another 2.3X jump.

Thats actually what I meant by "next gen architecture". There would be no doubt modifications or additions to what makes up each compute unit in the GPU. But those modifications wouldn'gt need to come at the cost of raw old school GPU power. For example, even though the CUs in the PS4 to PS4pro  GPUs doubled from 20 to 40, half of the CUs in the PS4pro are noticeably bigger than other half.   

Lol.... this thing again. Every gen we look at graphics and people say we dont need much for the next gen or that this is as good as its gonna get then boom.... we see horizon or GOW and minds are blown. Anyways, all going from 1.8TF to 4.2TF did just after  3 years was just bump up the resolution. You really don,t think the only dfference between PS4 and PS5 games will be higher a rez do you?

RTX has separate hardware features for ray tracing, and those take up space that could be filled with the typical old skool GPU power. AMD may have their own approach, but it will likely take up some die space regardless.

People were also saying the PS4 APU was weak overall in terms of it's specs (not for an APU specifically though) and many of them have been fairly surprised and impressed by what the devs have been able to do with that 'weak' APU. Now if PS wants to really make some more leaps instead of upgrades, they will likely need more than just 10TF next gen, but do they want to make that large of a leap? I myself don't believe they called the Pro a "mid gen upgrade/refresh" by accident.

Intrinsic said: 
EricHiggin said: 

High yields is quite important for a cheaper high volume product like a console. The larger and more complex, the worse the yields. Making sure the fab can fill the demand that product will have is also as important, whether it be yields or capacity. That last thing PS wants is a PS5 flying off shelves, with people constantly complaining they can't get one. If you forecast 10 million sales, but will only be able to produce 5 million due to the fab, that's a pretty big problem. There are other ways, but the CPU/GPU/APU are the prime factor. It's no coincidence that Pro and slim came out when they shrunk from 28nm to 16nm. Will PS celebrate their 25th anniversary?

This is not how yeilds work. As funny as it may sound, it can only impact price not volume.

If your yields are better, you will end up with more usable dies, which lowers the price per die, per wafer. If your yields are poor, you end up with less usable dies, which raises the cost per die, per wafer. That would mean if they didn't have enough volume due to poor yields, they would either need to install more production lines and eat the cost, or tell PS they couldn't make enough to fill the forecasted orders, so use another higher yield process instead. The companies don't swap all fabs at once to the new node. As the new node yields and orders increase, they install more and increase production to offer a higher volume of cheaper product.



There was recently an interview with a AMD rep and two question were interesting for us regarding next-gen console. It looks like ps5 and Next Xbox will have a chiplet design instead of a SOC/APU.

 

IC: With chiplets connected via IF on Rome, if a customer wanted a semi-custom design with different IP, such as a GPU or an AI block or an FPGA, would that be possible? (Say for example, a console?)

MP: Our semi-custom group is wide open to talk to customers to brainstorm! What excites me about the chiplet approach is that I think it’s going to disrupt the industry. It’s going to change the way the industry dreams of different configurations. Some might be right, and I can guarantee that someone will conjure up ten other ones that we didn’t think of! Honestly I think it is a disruptive force that is just nascent, just starting right now.

IC: With IF on 7nm, it offers 100 GB/s GPU to GPU connectivity. One of your competitors has something similar which allows both GPU-GPU and CPU-GPU connectivity. Currently with Rome, PCIe 4.0 has been announced from CPU to GPU but not IF. What has AMD’s analysis been on that CPU to GPU link?

MP: We haven’t announced applying the IF between the CPU and GPU and while it is certainly feasible, it is likely just dependent when workloads could truly leverage that protocol being applied, when the full coherency is required across both CPU and GPU. It is certainly feasible, but we haven’t announced it at this time.

 

Basically what this means is that Sony and Microsoft can pump in more Gpu cores into their next console. Something like a 88CU clocked at 1400mhz giving 15,7 Teraflops seems very feasible. Probably Vivster or Pemalite can give a better explanation what chiplet design is.

 

The interview:

https://www.anandtech.com/show/13578/naples-rome-milan-zen-4-an-interview-with-amd-cto-mark-papermaster



6x master league achiever in starcraft2

Beaten Sigrun on God of war mode

Beaten DOOM ultra-nightmare with NO endless ammo-rune, 2x super shotgun and no decoys on ps4 pro.

1-0 against Grubby in Wc3 frozen throne ladder!!

Trumpstyle said:

There was recently an interview with a AMD rep and two question were interesting for us regarding next-gen console. It looks like ps5 and Next Xbox will have a chiplet design instead of a SOC/APU.

Basically what this means is that Sony and Microsoft can pump in more Gpu cores into their next console. Something like a 88CU clocked at 1400mhz giving 15,7 Teraflops seems very feasible. Probably Vivster or Pemalite can give a better explanation what chiplet design is.

While that sounds like a good idea at first, it really isn't practical for something like a console which has to be cheap to manufacture. Chiplet designs require much more time to design, it requires interfaces between the chiplets (and this at very high speeds) more testing individual components, and some other, more obscure things. In the end, the price will be too high so I expect a monolithic chip, again. One chip still is the most cost effective solution for products that have to be cheap to manufacture.



drkohler said:

While that sounds like a good idea at first, it really isn't practical for something like a console which has to be cheap to manufacture.

The whole idea of the chiplet design is to reduce manufacturing costs.
Things like I/O, memory controllers and so on... Don't scale down in manufacturing nodes very well... So it makes sense to keep them on an older, more mature process that is cheaper.

drkohler said:

Chiplet designs require much more time to design, it requires interfaces between the chiplets (and this at very high speeds) more testing individual components, and some other, more obscure things.

AMD has a team dedicated to that very task.
But now that Infinity Fabric is a known quantity... It's very easy for AMD to build that out and take full advantage of it very quickly.

drkohler said:

 In the end, the price will be too high so I expect a monolithic chip, again. One chip still is the most cost effective solution for products that have to be cheap to manufacture.

I am not willing to make an assumption either way until I have more information.

Trumpstyle said:

There was recently an interview with a AMD rep and two question were interesting for us regarding next-gen console. It looks like ps5 and Next Xbox will have a chiplet design instead of a SOC/APU.

It's a "possibility". - Awhile ago there was a thread where people were making guesses at what the next gen consoles would have... And Ryzen+Infinity Fabric was mentioned as one approach that could be taken.

The big advantage is scalability... Microsoft and Sony could with relative ease build an entire lineup of consoles with differing GPU capabilities whilst keeping the CPU+Chipset+I/O+Ram identical.


Trumpstyle said:

IC: With chiplets connected via IF on Rome, if a customer wanted a semi-custom design with different IP, such as a GPU or an AI block or an FPGA, would that be possible? (Say for example, a console?)

MP: Our semi-custom group is wide open to talk to customers to brainstorm! What excites me about the chiplet approach is that I think it’s going to disrupt the industry. It’s going to change the way the industry dreams of different configurations. Some might be right, and I can guarantee that someone will conjure up ten other ones that we didn’t think of! Honestly I think it is a disruptive force that is just nascent, just starting right now.

This was always the case, Anandtech just got clarification.

Trumpstyle said:

IC: With IF on 7nm, it offers 100 GB/s GPU to GPU connectivity. One of your competitors has something similar which allows both GPU-GPU and CPU-GPU connectivity. Currently with Rome, PCIe 4.0 has been announced from CPU to GPU but not IF. What has AMD’s analysis been on that CPU to GPU link?

MP: We haven’t announced applying the IF between the CPU and GPU and while it is certainly feasible, it is likely just dependent when workloads could truly leverage that protocol being applied, when the full coherency is required across both CPU and GPU. It is certainly feasible, but we haven’t announced it at this time.

Obviously playing coy. I don't see no reason why that approach couldn't be taken.


Trumpstyle said:
Basically what this means is that Sony and Microsoft can pump in more Gpu cores into their next console. Something like a 88CU clocked at 1400mhz giving 15,7 Teraflops seems very feasible. Probably Vivster or Pemalite can give a better explanation what chiplet design is.

Because the individual chips are smaller, they can get better yields, which means lower costs on the latest manufacturing process.
So in theory it should mean more CU's.



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

wonder when sony and microsoft will reveal the next generation...one thing is sure, we will hear many rumors about in the next months



shikamaru317 said:
ZODIARKrebirth said:
wonder when sony and microsoft will reveal the next generation...one thing is sure, we will hear many rumors about in the next months

I'll be surprised if at least one of the 2 of them doesn't announce their next console by E3 at the latest personally. 

very likely, because the big guns will always be shown at e3 and i don't believe they want wait for 2020 for reveal