• 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.

Overall maximum teraflops for next-gen launch consoles?

  • 8 teraflops

    Votes: 43 1.9%
  • 9 teraflops

    Votes: 56 2.4%
  • 12 teraflops

    Votes: 978 42.5%
  • 14 teraflops

    Votes: 525 22.8%
  • Team ALL THE WAY UP +14 teraflops

    Votes: 491 21.3%
  • 10 teraflops (because for some reason I put 9 instead of 10)

    Votes: 208 9.0%

  • Total voters
    2,301
Status
Not open for further replies.

Amir

Member
Jun 7, 2018
337
More Navi:
https://www.reddit.com/r/hardware/comments/br4wiv/amd_navi_rumor_interview_with_sapphire_technology/
  • Confirmed to release watercooled Poison Toxic for Navi. Toxic for Vega was canceled due to restricted margin.
  • Navi currently has two versions, one is priced at $499 and the other is $399.
  • (Assuming talking about the top version) it is stronger than 2070
  • Denied the existence of socalled "Big Navi" with 5120SP
  • In terms of hardware raytracing acceleration: maybe wait for AMD's next GPU architecture, although the arch is not finalized yet.
  • Navi will be presented at Computex on May 27th, and the launch date is July 7th.
 
Feb 10, 2018
17,534
More Navi:
https://www.reddit.com/r/hardware/comments/br4wiv/amd_navi_rumor_interview_with_sapphire_technology/
  • Confirmed to release watercooled Poison Toxic for Navi. Toxic for Vega was canceled due to restricted margin.
  • Navi currently has two versions, one is priced at $499 and the other is $399.
  • (Assuming talking about the top version) it is stronger than 2070
  • Denied the existence of socalled "Big Navi" with 5120SP
  • In terms of hardware raytracing acceleration: maybe wait for AMD's next GPU architecture, although the arch is not finalized yet.
  • Navi will be presented at Computex on May 27th, and the launch date is July 7th.

Hardware raytracing looking unlikely despite what some "leaks" are saying.
 

Azurik

Attempted to circumvent ban with alt account
Banned
Nov 5, 2017
2,441
Can console gamers tell the difference between 30 and 60fps?
What has that got to do with his question in regards to resolution? Bit pointless?

Yes, there is obviously a massive difference between 30 and 60 FPS which is clearly visible regardless of resolution.

However, seeing a difference in HD and 4K on a 50/ 55 or even 65 inch tv is not as easy. It depends on distance, quality of panel, eye sight etc.
 
Last edited:

Jeffram

Member
Oct 29, 2017
3,924
During the investor Q and A, Sony said the average lifetime spending of year 1 PS4 owners is sitting at $1600.

I've been saying Sony is going to be in a position to take larger losses on PS5 than they did on PS4, and this only solidified it in my mind as something that will happen.
 

tusharngf

Member
Oct 29, 2017
2,288
Lordran
https://www.reddit.com/r/hardware/comments/br1864/ps5_dev_kit_pcb_rumored_specs/

monolithic die ~22.4mm by ~14.1mm
16 Samsung K4ZAF325BM-HC18 in clamshell configuration memory vrm seems like overkill with multiple Fairchild/ON Semiconductor FDMF3170 power stages controlled by an MP2888 from MPS
3 Samsung K4AAG085WB-MCRC, 2 of those close to the NAND acting as DRAM cache (unusual 2GB DRAM per 1 TB NAND)
4 NAND packages soldered to the PCB which are TH58LJT2T24BAEG from Toshiba
PS5016-E16 from Phison

Not sure if this was posted here before.
 

modiz

Member
Oct 8, 2018
17,813
well then, i missed some news while i was sleeping haha. that spiderman demo is absolutely insane. i actually cant believe they shown that. i was sure that this will be a demo that will only be shown to journalists.
 

gofreak

Member
Oct 26, 2017
7,734
So from the IR day, it confirms the SSD solution is custom?

I think it was also interesting that they name-checked ray tracing more than once - although still not proof of how it's going about support for that.
 

Jeffram

Member
Oct 29, 2017
3,924
Wow that ps5 loading demo video is crazy. But I can't understand why they showing this tech so early ...
I imagine it's because they believe Microsoft won't match it. It's fairly well believed Microsoft will blow out their console first, so might as well have some of the conversation be around what it can't do.

Either that, or they felt on the back foot of the power conversation and wanted to change opinions.
 

gofreak

Member
Oct 26, 2017
7,734
The rumor is dead no custom SSD. This is official the SSD solution inside PS5 is custom...

Well, that post doesn't make it official - that's a rumour/leak. But the rather more official presentation material from the IR day does specify a 'customized broadband ssd'.

So unless the reference to customization is purely in the realm of the software stack, it does indeed seem like a custom solution.
 

'V'

Banned
May 19, 2018
772
To be honest is that demo impressive or what we should expect as a standard? We're comparing what, a 5400RPM hard drive in the PRO to a NVMe (possible even custom) SSD in the PS5? Wouldn't that be the performance you'd expect anyway from that upgrade or am I missing something completely here.

We're talking a ~7 second reduction in loading times which is what I would have expected even using a regular SATA SSD. Am I wrong in thinking that?

Either way I'll reaffirm that its nice to see the baseline adopting this tech. I'm intrigued to see how it compares in games that take really long to load on the PS4 like battlefield 4 multiplayer for example (can take like 30 seconds from what I remember).
 

modiz

Member
Oct 8, 2018
17,813
To be honest is that demo impressive or what we should expect as a standard? We're comparing what, a 5400RPM hard drive in the PRO to a NVMe (possible even custom) SSD in the PS5? Wouldn't that be the performance you'd expect anyway from that upgrade or am I missing something completely here.

We're talking a ~7 second reduction in loading times which is what I would have expected even using a regular SATA SSD. Am I wrong in thinking that?

Either way I'll reaffirm that its nice to see the baseline adopting this tech. I'm intrigued to see how it compares in games that take really long to load on the PS4 like battlefield 4 multiplayer for example (can take like 30 seconds from what I remember).
remember that for the fast travel they did in the wired article it took 15 seconds on the pro and still 0.8 seconds on the next gen playstation. which is like 19x the jump in speed. and this is on low speed dev kit.
 

'V'

Banned
May 19, 2018
772
remember that for the fast travel they did in the wired article it took 15 seconds on the pro and still 0.8 seconds on the next gen playstation. which is like 19x the jump in speed. and this is on low speed dev kit.
Do we have any numbers we can compare it to? For example a comparison between a HDD similar to the one in the PRO, a slow SATA SSD, a fast SATA SSD and an NVMe SSD. If such a comparison existed across a range of games it would be possible to make a ballpark prediction of what kind of SSD the PS5 is sporting (and possibly make more accurate predictions regarding the custom modifications, if any).
 

gofreak

Member
Oct 26, 2017
7,734
Do we have any numbers we can compare it to? For example a comparison between a HDD similar to the one in the PRO, a slow SATA SSD, a fast SATA SSD and an NVMe SSD. If such a comparison existed across a range of games it would be possible to make a ballpark prediction of what kind of SSD the PS5 is sporting (and possibly make more accurate predictions regarding the custom modifications, if any).

Googling around, I'm not sure about comparisons with a 5400rpm hard drive, but ssd/hdd benchmarks seem to indicate improvements of up to 2-3x.

So, no, at face value an improvement of 10x and sub-second loads does not seem to be typical.

But of course it's possible this example was cherry picked for dramatic impact. And of course this isn't an A:B test, the CPU and memory speed differs too. But taking it as a raw bench of load time speed up, it is possibly better than might be expected I guess.
 

Sekiro

Member
Jan 25, 2019
2,938
United Kingdom
Could it be possible that Sony's hired another company to make their hardware ray-tracing component, rather than AMD, maybe Sony made one themselves in their internal R&D department?
 

Heretic

The Fallen
Oct 25, 2017
1,925
I'd like to see the comparison of a next gen game running off an HDD vs SSD. I'd rather pay less for a console even if it has slower load times 🤷‍♂️
 

modiz

Member
Oct 8, 2018
17,813
I'd like to see the comparison of a next gen game running off an HDD vs SSD. I'd rather pay less for a console even if it has slower load times 🤷‍♂️
the difference is not only in load time though, as you can see in the second demonstration in the spiderman video, once you break the speed limit on the ps4 pro the console starts to lag, get stuck, have a lot of pop in etc, but on the PS5 all of the assets load in time for the super speedy camera to render them and not be stuck at all.
that means that developers can do a lot more with their game streaming with the usage of SSD and lead to new design choices that were not possible with an HDD.
 

Osaragi

Member
Oct 27, 2017
173
Germany
I'd like to see the comparison of a next gen game running off an HDD vs SSD. I'd rather pay less for a console even if it has slower load times 🤷‍♂️

Sure the loading speed was impressive. But that's not what this is about. Just the ridiculous difference in traversal speed should show what a game changer this is. We can't even imagine what devs will come up with.
 

foamdino

Banned
Oct 28, 2017
491
I think that it's likely that the custom SSD solution (be it entirely hardware customisations or mostly very clever software + caches) that Sony has is something that will not be matched in the new XBox.

I believe this to be the case as Sony have now promoted this as the centre of their next-gen plans twice: the wired article was mostly about the SSD solution and now this investor relations presentation.

I think this will lead to different experiences (primarily created by Sony first-party).

I'm also still not sure that MS won't be using a Vega-derived GPU (my original speculation/guess way back).

Looking forward to next-gen now after seeing this demo.

I think what will happen is:
XBox -> more TFs or more compute - leading to better visuals or better framerates on 3rd-party / cross-platform games
PS -> fewer TFs/compute, stupidly fast ram (I'm guessing now 32GB) - leading to different games entirely enabled by this custom solution
 

GhostTrick

Member
Oct 25, 2017
11,297
I think that it's likely that the custom SSD solution (be it entirely hardware customisations or mostly very clever software + caches) that Sony has is something that will not be matched in the new XBox.

I believe this to be the case as Sony have now promoted this as the centre of their next-gen plans twice: the wired article was mostly about the SSD solution and now this investor relations presentation.

I think this will lead to different experiences (primarily created by Sony first-party).

I'm also still not sure that MS won't be using a Vega-derived GPU (my original speculation/guess way back).

Looking forward to next-gen now after seeing this demo.

I think what will happen is:
XBox -> more TFs or more compute - leading to better visuals or better framerates on 3rd-party / cross-platform games
PS -> fewer TFs/compute, stupidly fast ram (I'm guessing now 32GB) - leading to different games entirely enabled by this custom solution



SSDs are not some dark magic.
If the Next Xbox has a fast drive (even 1gb/s), dont expect "different games only existing thanks to the secret sauce of the SSD from Sony".
 

Doctor Avatar

Member
Jan 10, 2019
2,579
Honestly, I don't buy that AMD MS is using components from the GPU after Navi that AMD has not even named it.

But didn't you know that MS is using a post-Navi GPU architecture and Zen 3? They invented a time machine, that is their secret sauce!

It is funny that when people suggest the PS5 is going to be competitive with Anaconda they get laughed out of the thread (and told they are having fanboy dreams) but when people say MS will use future tech that doesn't exist in the Xbox Two there are people that take it completely seriously...

Unless MS has invented time travel the XB2 does not have Zen 3 and some post Navi GPU arch. Especially since MS are announcing and likely launching before PS5.
 

'V'

Banned
May 19, 2018
772
But of course it's possible this example was cherry picked for dramatic impact. And of course this isn't an A:B test, the CPU and memory speed differs too. But taking it as a raw bench of load time speed up, it is possibly better than might be expected I guess.
Yeah that's why I'm more interested in seeing other games which really took a long time to load.

Sure the loading speed was impressive. But that's not what this is about. Just the ridiculous difference in traversal speed should show what a game changer this is. We can't even imagine what devs will come up with.
This goes back to a discussion that was had before. Will 3rd party developers really design their games to fully take advantage of this? I can't see them alienating the millions of PC gamers who will still be using HDDs. I expect faster loading times across the board and less pop-in, which is what you'd expect anyway from an upgrade to an SSD.

1st party devs will very likely take full advantage of it though.

I don't want to start that whole discussion again but just as something to think about (if you missed the previous discussion), assuming it is using a custom SSD solution, is the added cost of that hardware worth the sacrifice in raw hardware specs (assuming this custom solution has a significant cost to it) e.g. a cheaper GPU and a reduction in CU count?
For me I'd say no and would prefer a standard NVMe SSD + faster GPU. It's something to think about anyway.
 

foamdino

Banned
Oct 28, 2017
491
SSDs are not some dark magic.
If the Next Xbox has a fast drive (even 1gb/s), dont expect "different games only existing thanks to the secret sauce of the SSD from Sony".

The point isn't that SSDs in general are special - it's the fact that Sony have gone out of their way to design their entire system around this, where people are just convinced it will lead to 'quicker loading', I think it will completely open up new game design options. I also suspect that they will sacrifice peak GPU performance for silly amounts of fast ram - I guess 32GB - again because to me it makes sense if design your entire system around fast storage.

Sony have prior experience of stupid bandwidth in the PS2 days with rambus - it wouldn't be the first time they've trodden this path.

But again just my opinion/speculation
 

DrKeo

Banned
Mar 3, 2019
2,600
Israel
This been posted here yet? That loading time comparison video of the PS5 vs the Pro got released online


Did anyone notice the difference from the Wired article? In Wired they compared the PS5 with the PS4, now it's compared to the Pro.

Wired:
PS4 - 15 seconds
PS5 - less than 0.8 seconds

New demonstration:
PS4 Pro - 8.1 seconds
PS5 - 0.84 seconds

The Pro has faster memory and CPU so loading times had improved but I wonder why did the PS5 slow down from less than 0.8 seconds to 0.84 seconds. In addition, seeing the reduction in loading time from the PS4 to the Pro, from 15 seconds to 8.1 seconds, makes me wonder how fast could the PS5 do this fast travel without an SSD considering it has at least x4 more powerful CPU and much faster memory and IO. Obviously a lot slower than with an SSD but still it seems like the SSD is giving them ~4-5 times faster loading times, not 19 times like the original article claimed.

Does anyone here have a PS4 or PS4 Pro with a copy of Spider-man and is willing to do a fast travel test? Does the game even have fast travel?
 
Last edited:

blackthone

Member
Oct 27, 2017
129
Seoul, South Korea
Did anyone notice the difference from the Wired article? In Wired they compared the PS5 with the PS4, now it's compared to the Pro.

Wired:
PS4 - 15 seconds
PS5 - less than 0.8 seconds

New demonstration:
PS4 Pro - 8.1 seconds
PS5 - 0.84 seconds

The Pro has faster memory and CPU so loading times had improved but I wonder why did the PS5 slow down from less than 0.8 seconds to 0.84 seconds. In addition, seeing the reduction in loading time from the PS4 to the Pro, from 15 seconds to 8.1 seconds, makes me wonder how fast could the PS5 do this fast travel without an SSD considering it has at least x4 more powerful CPU and much faster memory and IO. Obviously a lot slower than with an SSD but still it seems like the SSD is giving them ~4-5 times faster loading times, not 19 times like the original article claimed.

Does anyone here have a PS4 or PS4 Pro with a copy of Spider-man and is willing to do a fast travel test? Does the game even have fast travel?
I looked back at the Wired article and they actually compared PS5 with PS4 Pro.
 

modiz

Member
Oct 8, 2018
17,813
So, it seems like the hbm rumor finally got a contradiction.
In the ir presentation you can see the ps4 wont lower its MSRP this fy, which means no super slim. In the hbm leak it was said that sony will release later this year a 199$ ps4 super slim on 7nm.
 

Gamer17

Banned
Oct 30, 2017
9,399
So, it seems like the hbm rumor finally got a contradiction.
In the ir presentation you can see the ps4 wont lower its MSRP this fy, which means no super slim. In the hbm leak it was said that sony will release later this year a 199$ ps4 super slim on 7nm.
It was dash line,no ? meaning they will not talk about price now
 

gofreak

Member
Oct 26, 2017
7,734
Did anyone notice the difference from the Wired article? In Wired they compared the PS5 with the PS4, now it's compared to the Pro.

Wired:
PS4 - 15 seconds
PS5 - less than 0.8 seconds

New demonstration:
PS4 Pro - 8.1 seconds
PS5 - 0.84 seconds

The Pro has faster memory and CPU so loading times had improved but I wonder why did the PS5 slow down from less than 0.8 seconds to 0.84 seconds. In addition, seeing the reduction in loading time from the PS4 to the Pro, from 15 seconds to 8.1 seconds, makes me wonder how fast could the PS5 do this fast travel without an SSD considering it has at least x4 more powerful CPU and much faster memory and IO. Obviously a lot slower than with an SSD but still it seems like the SSD is giving them ~4-5 times faster loading times, not 19 times like the original article claimed.

Does anyone here have a PS4 or PS4 Pro with a copy of Spider-man and is willing to do a fast travel test? Does the game even have fast travel?

The wired article said '0.8 seconds' (less than one second, not less than 0.8), which sounds like a rounding of the same figure here.

The comparison with the base PS4 in the Wired article was fair - the PS4 represents the base bandwidth devs have to work with. Insomniac discussed at GDC how the base PS4, plus the variety of HDD and user upgraded HDDs limited their IO bandwidth budget and the fundamental knock ons that had for game design. So comparing to the base PS4 seems OK to me given that it is the fundamental limiter for PS4 games, and this will be the new baseline for PS5.
 
Last edited:

'V'

Banned
May 19, 2018
772
So, it seems like the hbm rumor finally got a contradiction.
In the ir presentation you can see the ps4 wont lower its MSRP this fy, which means no super slim. In the hbm leak it was said that sony will release later this year a 199$ ps4 super slim on 7nm.
What about that slide saying "$199 promotional pricing"?

Edit: just realized it says 2018 not 2019
 

modiz

Member
Oct 8, 2018
17,813

lynux3

Banned
Oct 28, 2017
569
I looked back at the Wired article and they actually compared PS5 with PS4 Pro.
It's weird because the writer mentions PS4 Pro, but Cerny appears to talk about the original PS4, as interpreted by the writer.

On the original PS4, the camera moves at about the speed Spidey hits while web-slinging. "No matter how powered up you get as Spider-Man, you can never go any faster than this," Cerny says, "because that's simply how fast we can get the data off the hard drive."
 

modiz

Member
Oct 8, 2018
17,813
It was dash line,no ? meaning they will not talk about price now
On the dot of 2019 it is still on 299$ the question is if the dash line after 2019 dot means the duration of this fiscal year or the next i guess.
Its 2018 sale, not an official msrp price drop in 2019
Look at how the price goes down on the 2016 dot to 299, yea i am pretty sure it confirms no msrp drop in 2019
 

DrKeo

Banned
Mar 3, 2019
2,600
Israel
A link to the presentation:
https://www.sony.net/SonyInfo/IR/library/presen/irday/pdf/2019/GNS_E.pdf

Some highlights with interesting stats regarding some discussions we've had here:
- over 96.8m PS4 worldwide (end of March 2019).
- over 36m PS+ subscribers (end of March 2019).
- over 90% of PS4 machines had connected to the internet at some point.
- Average lifetime device spend exceeds $700
- Average launch year lifetime device spend exceeds $1,600 (very important to the "lost leader" argument).
------------------------------------------------------------------------------------------------
Next-gen:
- Spider-man fast travel drops from 8.1 seconds on the Pro to 0.83 seconds on the PS5.
- We will leverage Backwards Compatibility to transition our community to Next-Gen faster and more seamlessly than ever before.
- What we have said so far:
  • All new CPU GPU
  • SSD
  • BC
  • RT
  • 8K
  • Disc support
  • 3D Audio
- What we have not spoken about yet:
  • Date
  • Price
  • Games
  • User experience
  • Country rollout
- PS4 Will remain the engine of engagement and profitability for the next three years.
- 5.6m users use remote play (FY18).

- PSNow:
  • 15 Data centers
  • 19 countries
  • 780 games
  • 700K PSNow subscribers (March 2019)
  • The yearly growth rate is 40%
PSNow future plans:
  • support 1080p and beyond
  • Improve the quality of content
  • Strengthen marketing support
  • Increase yearly growth rate to 50%
 
Last edited:

DrKeo

Banned
Mar 3, 2019
2,600
Israel
The wired article said '0.8 seconds' (less than one second, not less than 0.8), which sounds like a rounding of the same figure here.

The comparison with the base PS4 in the Wired article was fair - the PS4 represents the base bandwidth devs have to work with. Insomniac discussed at GDC how the base PS4, plus the variety of HDD and user upgraded HDDs limited their IO bandwidth budget and the fundamental knock ons that had for game design. So comparing to the base PS4 seems OK to me given that it is the fundamental limiter for PS4 games, and this will be the new baseline for PS5.
Yeah, you are right, it is 0.8 seconds. My bad.

But regarding the comparison, what I wonder is how much of that improvement is because of the SSD and how much is because of the faster CPU and memory.
 
Status
Not open for further replies.