PS5 and Xbox Series speculation |OT11| Cry Havoc and Let Slip the Dogs of War [NEW NEWS, NEW THREAD - CHECK OUT THE STAFF POST]

Basarili · Feb 15, 2020

I got a question about these SSDs we're going to get. If both companies make a custom storage space. How will we be able to expand the storage?
I have a 1,5TB HDD in my PS4 since day one. So how will this go with the next-gen consoles or do we have no info about this yet?

chris 1515 · Feb 15, 2020

jelly said:
Why does nobody else make a nand factory, is it that specialist and expensive with little return, kinda weird though?

Aren't SSDs without DRAM worse?

This is for cost, The Xbox Series X SSD has no DRAM but everything for managing the SSD will be in main RAM.

Sony patent has SRAM but part of the things needed for the SSD will be in main RAM

anexanhume · Feb 15, 2020

gaffertape said:
I don't get your first part. We don't know about the Ps5 SSD solution yet.
SSD speeds are not possible with DRAM? SSD speeds are generally faster with DRAM.

I do not refute the fact that Microsoft will be affected, too if those types of memory solutions are scare.
I just don't understand what all the other parts of your post are supposed to say us.

The patent, explained well by gofreak details that the solution is based on SRAM and is without DRAM.

Anthony Hopkins said:
One thing that maybe what Sony are waiting for is confirmation on the existence of Lockhart, if lockhart is $299/$349, they will want to change lower if they only have one next gen machine.

I doubt Sony can go lower then $449 though, if BOM is $450, when u include shipping + packaging + R+D + labour, each ps5 will probably cost Sony about $500

Console makers have taken $100/200 losses as recently as two gens ago.

ovbm · Feb 15, 2020

modiz said:
those articles are old, i think NAND prices spiked very recently.

They are projected to rise by ~40% by Q4 2020, but as of now, they are still relatively low. I also wonder when companies usually secure their contracts. I would assume sony already secured their contracts for initial production and thus won't be affected by the price hike. Here's a more recent report showing DRAM prices still practically rock bottom as of end of Dec 2019: https://seekingalpha.com/article/4323846-tracking-dram-ready-for-contract-magnet

Edit: Now that I think of it, they probably have to do their cost predictions based on an average price, not only considering capacities they can get now, but future capacities. If they go with 1TB of nand now - because it's cheap - and set their price at 400$, it could seriously hurt their profitability if the next price of the batch of NAND they secure rises by 50%. Makes BOM much more interesting, as it's probably more variable than we think.

Axel Stone · Feb 15, 2020

III-V said:
In relation to the article:

Even Albert Penello said that some of the smaller details don't make sense, and I agree.

I don't think the author is way of base with final analysis but I do think this story is more of an extrapolated story than source-based fact-based reporting.

I think it's more likely to be source-based but including some errors myself, simply because I think that's more likely than this story being essentially a complete fabrication.

III-V · Feb 15, 2020

anexanhume said:
The patent, explained well by gofreak details that the solution is based on SRAM and is without DRAM.

Console makers have taken $100/200 losses as recently as two gens ago.

Any chance you have gofreak write-up handy?

Deleted member 10747 · Feb 15, 2020

one said:
It might not be controlled by the Sony PR, but might be controlled by some faction in Sony. Remember it's a huge corporation with so many stakeholders and it is much more complex than you imagine unlike those Era insiders/leakers.

Sure sure. I'm going to leave you and your tinfoil hat at the door. lol wth....

III-V · Feb 15, 2020

gaffertape said:
I do not refute the fact that Microsoft will be affected, too if those types of memory solutions are scare.
I just don't understand what all the other parts of your post are supposed to say us.

Axel Stone said:
I think it's more likely to be source-based but including some errors myself, simply because I think that's more likely than this story being essentially a complete fabrication.

I am not saying it is a complete fabrication, just that it reads more like an opinion or analysis piece than an actual report. The entire article is a jumble of "sources" and I think this is intentionally done to make up for lack of actual hard sources. Thats all I am saying. I think it is poor reporting with a lot of hand-waving to intentionally blur the lack of hard sources.

ovbm said:
They are projected to rise by ~40% by Q4 2020, but as of now, they are still relatively low. I also wonder when companies usually secure their contracts. I would assume sony already secured their contracts for initial production and thus won't be affected by the price hike. Here's a more recent report showing DRAM prices still practically rock bottom as of end of Dec 2019: https://seekingalpha.com/article/4323846-tracking-dram-ready-for-contract-magnet

I agree the pricing would have been agreed upon a while ago. Honestly seems like a great Albert Penello question - when do these prices for bulk purchase get hammered out? 1 yr prior to launch?

chris 1515 · Feb 15, 2020

Phison Previews Next-Gen PCIe 4.0 SSD Controllers: Up to 7 GB/s, NVMe 1.4

www.anandtech.com

There is a great table inside the Anandtech article with comparison between all different Phison controller

The E19T is here, it is the fastest Phison controller mainstream SSD controller. It means the SSD ARM CPU is less powerful than typical high-end SSD it means more pressure on the Xbox X86 Zen 2 CPU and no DRAM, all things form the SSD will be inside RAM. For example, the translation address table is typically 10% of the SSD size here it is 1GB.

For example this SSD at 114 dollars at launch now more expensive use the Phison E12 controller:

Addlink S70 1TB NVMe M.2 SSD review

How good (or bad) is a cheapo M2 SSD bought from Amazon? We received an email from Addlink claiming they offer an SSD (Addlink S70 1TB NVMe PCIe 3x4 SSD) that often gets compared to fast as Samsung... Introduction

www.guru3d.com

III-V · Feb 15, 2020

chris 1515 said:
Phison Previews Next-Gen PCIe 4.0 SSD Controllers: Up to 7 GB/s, NVMe 1.4

www.anandtech.com

The E19T is here, it is the fastest Phison controller mainstream SSD controller. It means the SSD ARM CPU is less powerful than typical high-end SSD it means more pressure on the Xbox X86 Zen 2 CPU and no DRAM, all things form the SSD will be inside RAM. For example, the translation address table is typically 10% of the SSD size here it is 1GB.

For example this SSD at 114 dollars at launch now more expensive use the Phison E12 controller:

Addlink S70 1TB NVMe M.2 SSD review

How good (or bad) is a cheapo M2 SSD bought from Amazon? We received an email from Addlink claiming they offer an SSD (Addlink S70 1TB NVMe PCIe 3x4 SSD) that often gets compared to fast as Samsung... Introduction

www.guru3d.com

Why do you think more CPU pressure? And how much of a drain?

Anthony Hopkins · Feb 15, 2020

anexanhume said:
The patent, explained well by gofreak details that the solution is based on SRAM and is without DRAM.

Console makers have taken $100/200 losses as recently as two gens ago.

I would be surprised if they did that again.
I don't think sony would want to lower the price of the ps5 to much, because it will still be in the same power bracket as Xbox's premium console.
And if Sony go crazy and do a $399 12+tf ps5 ms would probably counter with a $249 lockhart and $399 XsX.

modiz · Feb 15, 2020

dgrdsv said:
GDC 2020 sessions are online for those interested:

Game Developers Conference (GDC)

GDC is the game industry's premier professional event, championing game developers and the advancement of their craft. March 21-25, 2022.

schedule.gdconf.com

So basically only Epic will have a talk about next gen.
i bet devs are dying to talk about it more openly.

anexanhume · Feb 15, 2020

Anthony Hopkins said:
I would be surprised if they did that again.
I don't think sony would want to lower the price of the ps5 to much, because it will still be in the same power bracket as Xbox's premium console.
And if Sony go crazy and do a $399 12+tf ps5 ms would probably counter with a $249 lockhart and $399 XsX.

You would be surprised if they did that again, but then assume Microsoft would take an even bigger loss on XSX?

III-V said:
Any chance you have gofreak write-up handy?

PS5 - a patent dive into what might be the tech behind Sony's SSD customisations (technical!)

This will be one for people interested in some potentially more technical speculation. I posted in the next-gen speculation thread, but was encouraged to spin it off into its own thread. I did some patent diving to see if I could dig up any likely candidates for what Sony's SSD solution might...

www.resetera.com

Deleted member 12635 · Feb 15, 2020

chris 1515 said:
This is for cost, The Xbox Series X SSD has no DRAM but everything for managing the SSD will be in main RAM.

Sony patent has SRAM but part of the things needed for the SSD will be in main RAM

How do you know the first part for certain?

chris 1515 · Feb 15, 2020

III-V said:
Why do you think more CPU pressure? And how much of a drain?

Because here it is only one core A5 inside the SSD see the table inside the Anandtech article. On Sony patent, they want to have a beefy secondary CPU probably a multicore ARM one to manage fully the SSD and it is corresponding with what Remedy told.

For the BOM of Sony SSD I think the Sony controller will be fully custom. For calculating the BOM it will be useful to add the price of all component ARM CPU, hardware decompressor and other components NAND Flash itself and so on.

PlayStation 5 Control Devs Talk About the Console's "Stand Out" Feature

Here's everything we know about PS5 thus far.

www.techtimes.com

"It's the new SSD that really stands out; essentially streaming will become something that we don't really have to worry so much about, and it will free up some extra CPU bandwidth in the process," Donnelly added.

Anthony Hopkins · Feb 15, 2020

anexanhume said:
You would be surprised if they did that again, but then assume Microsoft would take an even bigger loss on XSX?

No I expect ps5 and XSX to have a very similar cost.

But if Sony did something crazy like $100-200 loss, ms would respond, it would be delusional to think otherwise.
Which is why I don't think either company will take big losses because they know there competition will respond and it will mean less profits which are unnecessary.

BitsandBytes · Feb 15, 2020

chris 1515 said:
For the BOM of Sony SSD I think the Sony controller will be fully custom

Interesting as I was wondering about how custom Sony will go with the SSD. Do you think it will be fully custom but built for Sony by Phison/Samsung/other or custom made by themselves?

Axel Stone · Feb 15, 2020

III-V said:
I am not saying it is a complete fabrication, just that it reads more like an opinion or analysis piece than an actual report. The entire article is a jumble of "sources" and I think this is intentionally done to make up for lack of actual hard sources. Thats all I am saying. I think it is poor reporting with a lot of hand-waving to intentionally blur the lack of hard sources.

The article consists of 15 paragraphs (and a sentence) 9 of which refer directly back to these unnamed sources within Sony. Seems pretty clearly sourced to me.

disco_potato · Feb 15, 2020

GUNDAM STARDUST said:
Yeah, that's a downer if true. We may have to wait until mid gen refreshes for real RayTracing.

I think even if AMD were all in for RT doing as best as they could, whatever we got in consoles would still be a subpar implementation. 95% of AMDs's eggs are in the ryzen basket. What they did with the ryzen line is probably not something they can do with radeon/navi at this point so to expect their RT implementation to rival nvidia doesn't sound likely.

Axel Stone said:
Agree with this. Why would you deliberately put the idea of boardroom disagreements into the public domain?

It falls inline with the narrative of internal turmoil that has been around for a few months.

III-V said:
The windowscentral article implies that XSX will be using a flash based memory:

We have to keep in mind the windowscentral article is entirely based on the rumor that the phison e19 controller is indeed what's in xsx. Since that is a DRAMless controller MS wouldn't be affected by the dram price hike, for the ssd. They still need ram chips and nand, and both are going through price hikes which are likely NOT affecting either console.

This is my issue with the idea that dram prices/shortages happening now, affect consoles. Remember years and years of confirmation that sony/ms get their contracts and prices set long in advance? That means whatever market fluctuations happen, they would be unaffected as their price/quantity is already locked in. We know TSMC and likely other foundries/manufacturers set allotment of chips long ahead, right? So whatever sony/ms need, that quantity is guaranteed for them unless they underestimated their needs, at which point yes, they're affected.

Axel Stone said:
The article consists of 15 paragraphs (and a sentence) 9 of which refer directly back to these unnamed sources within Sony. Seems pretty clearly sourced to me.

/s?

There isn't a single original thought or idea in the article and the few times a source is mentioned, it goes back to something written elsewhere.

chris 1515 · Feb 15, 2020

Colbert said:
How do you know the first part for certain?

Because of the Linkedin Phison employee and the Phison controller. It is the same for Sony the only difference there is SRAM they can use as a cache of for having part of the translation table inside the SRAM. But it is normal Sony or MS will not break the bank on the SSD like I told before the two SSDswill probably have BOM between 40 to 60 dollars and have a high BOM/performance ratio.

Addlink S70 1TB NVMe M.2 SSD review

How good (or bad) is a cheapo M2 SSD bought from Amazon? We received an email from Addlink claiming they offer an SSD (Addlink S70 1TB NVMe PCIe 3x4 SSD) that often gets compared to fast as Samsung... Introduction

www.guru3d.com

The BOM will not be higher than the addlink SSD (114 dollars at retails probably 50 to 60 dollars BOM).

The Xbox Series X budget PCIe 4.0 SSD could be half the speed of the PS5

The Phison contoller can only deliver last-gen speeds, but still far in excess of today's consoles... though maybe not the PS5

www.pcgamesn.com

Same for Sony, SRAM SSD controller are not new and are cheaper than DRAM SSD:

SSD Controllers | Unleashing High-Performance Edge to Cloud Data Storage - Marvell

An industry leader in the disk drive controller market, Marvell is uniquely positioned to provide performance solutions to the fast-growing solid state disk (SSD) market.

www.marvell.com

DRAM-less SSD Controllers

For having performant SSD other part is software or having some hardware customization built for purpose SSD are like this:

SSD and All flash storage

A growing number of enterprise datacenters requiring high data throughput and low transaction latency previously reliant on Hard Disk Drives (HDD) in their...

www.purveytech.com

A growing number of enterprise datacenters requiring high data throughput and low transaction latency previously reliant on Hard Disk Drives (HDD) in their servers are now migrating to Solid-State Drives (SSDs) as a viable storage solution to increase their datacenter performance, efficiency, reliability and lowering overall operating expenses (OpEx).In today's market, SSD and NAND Flash memory consumption are split into two main groups:e" />

SSDs for Enterprise Data Storage.SSDs encapsulate flash storage within a platform that can take advantage of existing disk architectures. This design offers benefits for the manufacturer, such as compatibility with existing platforms and a faster route to market.

Purpose-Built All-Flash Architecture.In contrast to SSDs, a purpose-built architecture can be designed to deliver the maximum performance that flash can provide. The flash can be designed to connect directly with the PCIe bus, removing the additional latency of first going through a SAS interface as an SSD must. Also, since the array vendor is able to design the array to connect directly to the discrete flash modules, management functions can be controlled to minimize the impact to performance. The manufacturer can also apply advanced data protection algorithms designed for their architecture, which in a flash environment, can deliver higher overall availability and efficiency than the traditional RAID levels used in SSD arrays.

anexanhume · Feb 15, 2020

Anthony Hopkins said:
No I expect ps5 and XSX to have a very similar cost.

But if Sony did something crazy like $100-200 loss, ms would respond, it would be delusional to think otherwise.
Which is why I don't think either company will take big losses because they know there competition will respond and it will mean less profits which are unnecessary.

I think you have the logic backwards. The existence of Lockhart would remove the necessity for price-parity of the XSX and PS5.

Deleted member 12635 · Feb 15, 2020

chris 1515 said:
Because of the Linkedin Phison employee and the Phison controller. It is the same for Sony the only difference there is SRAM they can use as a cache of for having part of the translation table inside the SRAM. But it is normal Sony or MS will not break the bank on the SSD likt I told before the two SSD will probably have BOM between 40 to 60 dollars and have a high BOM/performance ratio.

Addlink S70 1TB NVMe M.2 SSD review

How good (or bad) is a cheapo M2 SSD bought from Amazon? We received an email from Addlink claiming they offer an SSD (Addlink S70 1TB NVMe PCIe 3x4 SSD) that often gets compared to fast as Samsung... Introduction

www.guru3d.com

The BOM will not be higher than the addlink SSD (114 dollars at retails probably 50 to 60 dollars BOM).

The Xbox Series X budget PCIe 4.0 SSD could be half the speed of the PS5

The Phison contoller can only deliver last-gen speeds, but still far in excess of today's consoles... though maybe not the PS5

www.pcgamesn.com

Same for Sony, SRAM SSD controller are not new and are cheaper than DRAM SSD

SSD Controllers | Unleashing High-Performance Edge to Cloud Data Storage - Marvell

An industry leader in the disk drive controller market, Marvell is uniquely positioned to provide performance solutions to the fast-growing solid state disk (SSD) market.

www.marvell.com

Thanks.

Anthony Hopkins · Feb 15, 2020

anexanhume said:
I think you have the logic backwards. The existence of Lockhart would remove the necessity for price-parity of the XSX and PS5.

No it won't, ms will still want to compete on the high end level, ms won't price XsX more then ps5 if ps5 is more powerful.

anexanhume · Feb 15, 2020

Anthony Hopkins said:
No it won't, ms will still want to compete on the high end level, ms won't price XsX more then ps5 if ps5 is more powerful.

Who says PS5 will be more powerful?

Anthony Hopkins · Feb 15, 2020

anexanhume said:
Who says PS5 will be more powerful?

Klee, Reiner

chris 1515 · Feb 15, 2020

After it does not means the Xbox SSD will be slow it means the SSD will partially use software-defined flash like SSD in Open channel initiative in the datacenter and they are faster than the same SSD in a PC.

And the PS5 SSD will too have the translation table for everything linked to the OS in main RAM.

Axel Stone · Feb 15, 2020

disco_potato said:
/s?

There isn't a single original thought or idea in the article and almost every time a source is mentioned, it goes back to something written elsewhere.

No /s. It seems to me that the article is clearly presented as being based on information given by sources within Sony. If it isn't based on these sources, then the article is wilfully misleading readers. I think you need a slightly higher standard of proof to say that an article from a publication such as Bloomberg is intentionally misleading its readers than has been provided so far.

Which isn't to say that the article is 100% accurate necessarily, journalists certainly do make mistakes, but in most reputable publications, protecting that reputation is more important than clicks, so the idea that this is just made up or extrapolated from elsewhere while lying about sources seems to me to be unlikely. Not impossible, but unlikely.

III-V · Feb 15, 2020

cc: Colbert ???

anexanhume said:
You would be surprised if they did that again, but then assume Microsoft would take an even bigger loss on XSX?

PS5 - a patent dive into what might be the tech behind Sony's SSD customisations (technical!)

This will be one for people interested in some potentially more technical speculation. I posted in the next-gen speculation thread, but was encouraged to spin it off into its own thread. I did some patent diving to see if I could dig up any likely candidates for what Sony's SSD solution might...

www.resetera.com

Thank you! brb need to go read it

chris 1515 said:
Because here it is only one core A5 inside the SSD see the table inside the Anandtech article. On Sony patent, they want to have a beefy secondary CPU probably a multicore ARM one to manage fully the SSD and it is corresponding with what Remedy told.

For the BOM of Sony SSD I think the Sony controller will be fully custom. For calculating the BOM it will be useful to add the price of all component ARM CPU, hardware decompressor and other components NAND Flash itself and so on.

PlayStation 5 Control Devs Talk About the Console's "Stand Out" Feature

Here's everything we know about PS5 thus far.

www.techtimes.com

That would be insane if true!

disco_potato said:
I think even if AMD were all in for RT doing as best as they could, whatever we got in consoles would still be a subpar implementation. 95% of AMDs's eggs are in the ryzen basket. What they did with the ryzen line is probably not something they can do with radeon/navi at this point so to expect their RT implementation to rival nvidia doesn't sound likely.

It falls inline with the narrative of internal turmoil that has been around for a few months.

We have to keep in mind the windowscentral article is entirely based on the rumor that the phison e19 controller is indeed what's in xsx. Since that is a DRAMless controller MS wouldn't be affected by the dram price hike, for the ssd. They still need ram chips and nand, and both are going through price hikes which are likely NOT affecting either console.

This is my issue with the idea that dram prices/shortages happening now, affect consoles. Remember years and years of confirmation that sony/ms get their contracts and prices set long in advance? That means whatever market fluctuations happen, they would be unaffected as their price/quantity is already locked in. We know TSMC and likely other foundries/manufacturers set allotment of chips long ahead, right? So whatever sony/ms need, that quantity is guaranteed for them unless they underestimated their needs, at which point yes, they're affected.

/s?

There isn't a single original thought or idea in the article and the few times a source is mentioned, it goes back to something written elsewhere.

I agree with your analysis but I do think the XSX Physon controller is accurate: see chris 1515 posts.

TheThreadsThatBindUs · Feb 15, 2020

Axel Stone said:
No /s. It seems to me that the article is clearly presented as being based on information given by sources within Sony. If it isn't based on these sources, then the article is wilfully misleading readers. I think you need a slightly higher standard of proof to say that an article from a publication such as Bloomberg is intentionally misleading its readers than has been provided so far.

Which isn't to say that the article is 100% accurate necessarily, journalists certainly do make mistakes, but in most reputable publications, protecting that reputation is more important than clicks, so the idea that this is just made up or extrapolated from elsewhere while lying about sources seems to me to be unlikely. Not impossible, but unlikely.

But it becomes increasingly likely when the same author has been caught with his pants down before.

See: previous posts from Dave.

Axel Stone · Feb 15, 2020

Was the author caught inventing sources before? Because that's a massive journalistic no no. Writing articles which turned out to be incorrect are one thing, inventing sources is an entirely different kettle of fish.

Anthony Hopkins · Feb 15, 2020

I think ssd performance on both platforms will be practically identical for real world use, because even with a current gen nvme (3gb/s read + write) it's such a huge leap over current gen anything better its returns will be diminished. 3gb/s can fill 16gb ram in 5 seconds, so it won't matter if one console has 5gb/s and can fill the ram in 3 seconds it's not going to make that much difference when current gen takes about 50 seconds to fill the ram from the hdd.

SublimeAnarky · Feb 15, 2020

ovbm said:
They are projected to rise by ~40% by Q4 2020, but as of now, they are still relatively low. I also wonder when companies usually secure their contracts. I would assume sony already secured their contracts for initial production and thus won't be affected by the price hike. Here's a more recent report showing DRAM prices still practically rock bottom as of end of Dec 2019: https://seekingalpha.com/article/4323846-tracking-dram-ready-for-contract-magnet

Edit: Now that I think of it, they probably have to do their cost predictions based on an average price, not only considering capacities they can get now, but future capacities. If they go with 1TB of nand now - because it's cheap - and set their price at 400$, it could seriously hurt their profitability if the next price of the batch of NAND they secure rises by 50%. Makes BOM much more interesting, as it's probably more variable than we think.

Albert Penello - could you clarify something for me from the text I've quoted please?

When planning a console launch/subsequent cycle - what is the norm when it comes to contracting components? Do platform holders ask vendors for an average price with component price range triggers over a multi-year period? Or would they engage in smaller validites at fixed costs?

How does that math/projection normally work?

Otakukidd · Feb 15, 2020

So thinking about expanding storage. What would you guys think of allowing external USB storage but instead of using it more like a hard drive We play games from like now. What if it's like a local backup. It would allow us to install games to it and when you want to play them, it would ask what games on the ssd to delete or backup to the USB drive and then move the game you want to play to the ssd. It would at the minimum be faster than redownloading it.

TheThreadsThatBindUs · Feb 15, 2020

Axel Stone said:
Was the author caught inventing sources before? Because that's a massive journalistic no no. Writing articles which turned out to be incorrect are one thing, inventing sources is an entirely different kettle of fish.

There isn't even anything meaningful that the author attributes to sources in this recent articles. The rest of the stuff that make sure up the prevailing bulk of the article is stuff that's been reported on elsewhere and simply cobbled together here to form a narrative.

It's no different to what 70% of the folks here do.

disco_potato · Feb 15, 2020

III-V said:
I agree with your analysis but I do think the XSX Physon controller is accurate: see chris 1515 posts.

Ok. I wasn't sure if the e19 was confirmed.

TheThreadsThatBindUs said:
But it becomes increasingly likely when the same author has been caught with his pants down before.

See: previous posts from Dave.

And when bloomberg has been sued and exposed for fabricating stories as recently as the past year.

France's AMF watchdog fines Bloomberg €5 million over Vinci hoax

France's markets watchdog AMF said on Monday it had fined U.S. news agency Bloomberg News 5 million euros ($5.5 million) for publishing a hoax press release relating to construction group Vinci <SGEF.PA> in November 2016.

www.reuters.com

iTWire - Bloomberg reporters stand to earn bonuses if chip story holds up

The Bloomberg story, claiming chips are being implanted by a Chinese contractor on server motherboards sold by US firm Supermicro Computer and being used to spy on some companies, will benefit reporters Jordan Robertson and Michael Riley — who wrote it — if it holds up under scrutiny. The...

www.itwire.com

anexanhume · Feb 15, 2020

Anthony Hopkins said:
Klee, Reiner

PS5 being more powerful is inconsistent with XSX being substantially more expensive as postulated by Zhuge. If XSX is indeed 50+ more than PS5, the likelihood it's less powerful is very low.

Anthony Hopkins · Feb 15, 2020

anexanhume said:
PS5 being more powerful is inconsistent with XSX being substantially more expensive as postulated by Zhuge. If XSX is indeed 50+ more than PS5, the likelihood it's less powerful is very low.

Well it depends on who you believe, klee + reiner or Zhuge.

TheThreadsThatBindUs · Feb 15, 2020

anexanhume said:
PS5 being more powerful is inconsistent with XSX being substantially more expensive as postulated by Zhuge. If XSX is indeed 50+ more than PS5, the likelihood it's less powerful is very low.

Zhuge's latest tweet specifically says that XSX is more costly but "not significantly so".

Why has that now become, "substantially more expensive"?

2Blackcats · Feb 15, 2020

Otakukidd said:
So thinking about expanding storage. What would you guys think of allowing external USB storage but instead of using it more like a hard drive We play games from like now. What if it's like a local backup. It would allow us to install games to it and when you want to play them, it would ask what games on the ssd to delete or backup to the USB drive and then move the game you want to play to the ssd. It would at the minimum be faster than redownloading it.

This has been mentioned 100s of times at this point. I'd be happy with that as a solution. Especially if games are more modular and we can start when the first required section is copied.

I'll go one further though and would like to have support for a HDD on the local network. Stick a HDD in your router rather than having it hanging permanently off your console.

Carmelozi · Feb 15, 2020

More powerful does not mean it performs better than PS5. We know both consoles are close so whatever we'll get, it'll be good for gamers.

chris 1515 · Feb 15, 2020

Other interesting things the Xbox Phison controller can work with SSD at M2 format and CFX.

I think Microsoft and So y will propose two solutions one using HDD or SSD via USB as external storage with copying the data to the internal SSD and expansion using a port. Like the rumored CFX Express Port on the Xbox Series X.

CosmicBolt · Feb 15, 2020

anexanhume said:
PS5 being more powerful is inconsistent with XSX being substantially more expensive as postulated by Zhuge. If XSX is indeed 50+ more than PS5, the likelihood it's less powerful is very low.

ZhugeEX clarified XSX BOM is not significantly more than PS5.

I'm not saying PS5 is more powerful but not significantly behind also.

TheThreadsThatBindUs · Feb 15, 2020

CosmicBolt said:
ZhugeEX clarified XSX BOM is not significantly more than PS5.

I'm not saying PS5 is more

Cheers.

I wanted to post this but didn't know how to embed a tweet.

Anthony Hopkins · Feb 15, 2020

CosmicBolt said:
ZhugeEX clarified XSX BOM is not significantly more than PS5.

I'm not saying PS5 is more powerful but not significantly behind also.

PS5 could still be more powerful with a slightly lower BOM, the price differences could be things like power supply, bd drive, case etc. MS like to make very modular consoles with each part labelled, that could drive price up a few bucks.

TheThreadsThatBindUs · Feb 15, 2020

Anthony Hopkins said:
PS5 could still be more powerful with a slightly lower BOM, the price differences could be things like power supply, bd drive, case etc. MS like to make very modular consoles with each part labelled, that could drive price up a few bucks.

You're right, and the evidence points to a likelihood that either way they'll be close in performance.

So everybody wins.

(Reason to optimistic I think, Cyborg )

III-V · Feb 15, 2020

TheThreadsThatBindUs said:
Zhuge's latest tweet specifically says that XSX is more costly but "not significantly so".

Why has that now become, "substantially more expensive"?

The numbers ZhugeEx has thrown out ($10-70) correspond to a 2-14% difference. Now we get into the semantics of the phrase "not significantly so" and I don't want to go there ;)

Anthony Hopkins said:
I think ssd performance on both platforms will be practically identical for real world use, because even with a current gen nvme (3gb/s read + write) it's such a huge leap over current gen anything better its returns will be diminished. 3gb/s can fill 16gb ram in 5 seconds, so it won't matter if one console has 5gb/s and can fill the ram in 3 seconds it's not going to make that much difference when current gen takes about 50 seconds to fill the ram from the hdd.

Chris 1515 has pointed out that the Sony patent solution actually saves CPU cycles as well due to a difference in accessing lookup tables and having a dedicated small cpu for the SSD. Also, the comments from Remedy concerning "additional destruction" also seem to bolster this claim. It is all described very well in go freak's write up.

TheThreadsThatBindUs said:
You're right, and the evidence points to a likelihood that either way they'll be close in performance.

So everybody wins.

(Reason to optimistic I think, Cyborg )

Everyone should be thrilled

Kapten · Feb 15, 2020

anexanhume said:
PS5 being more powerful is inconsistent with XSX being substantially more expensive as postulated by Zhuge. If XSX is indeed 50+ more than PS5, the likelihood it's less powerful is very low.

Well, Xbox One BOM was $457 in comparison to PS4's $372. According to some graph. And we saw how that turned out.

Though. Kinect was incorporated into that price at $75.

So BOM costs do not show if a console is more powerful or not.

chris 1515 · Feb 15, 2020

III-V said:
The numbers ZhugeEx has thrown out ($10-70) correspond to a 2-14% difference. Now we get into the semantics of the phrase "not significantly so" and I don't want to go there ;)

Chris 1515 has pointed out that the Sony patent solution actually saves CPU cycles as well due to a difference in accessing lookup tables and having a dedicated small cpu for the SSD. Also, the comments from Remedy concerning "additional destruction" also seem to bolster this claim. It is all described very well in go freak's write up.

Everyone should be thrilled

There is a dedicated CPU in every SSD it would simply means the PS5 will be powerful enough to do full SSD management. The interesting part is MS seems to be the customer for the Phison controller and was created for the next generation Xbox family. And there is an ARM CPU inside the Xbox SSD just less powerful than in high end PC SSD.

If you read the table many Phison SSD controller have a dual core or a triple core A5, the Xbox SSD controller is a single core A5.

Axel Stone · Feb 15, 2020

TheThreadsThatBindUs said:
There isn't even anything meaningful that the author attributes to sources in this recent articles. The rest of the stuff that make sure up the prevailing bulk of the article is stuff that's been reported on elsewhere and simply cobbled together here to form a narrative.

It's no different to what 70% of the folks here do.

Are we reading the same article? The majority of claims in that piece are attributed to these sources. A lot do line up with information elsewhere, but that doesn't mean that the information didn't come from the sources claimed.

disco_potato said:
.

I am aware of that Bloomberg article and it's definitely not a good look, but it is one bad article out of god knows how many.

anexanhume · Feb 15, 2020

TheThreadsThatBindUs said:
Zhuge's latest tweet specifically says that XSX is more costly but "not significantly so".

Why has that now become, "substantially more expensive"?

CosmicBolt said:
ZhugeEX clarified XSX BOM is not significantly more than PS5.

I'm not saying PS5 is more powerful but not significantly behind also.

"Significant" is a matter of interpretation given he gave an exact number range. The top-end $70 is 15% more. That's significant to me.

Kapten said:
Well, Xbox One BOM was $457 in comparison to PS4's $372. According to some graph. And we saw how that turned out.

Though. Kinect was incorporated into that price at $75.

So BOM costs do not show if a console is more powerful or not.

We know this doesn't have Kinect, so how is this relevant? We also know how they prioritized die size, which we have good reason to believe is not how they did it this time around.

PS5 and Xbox Series speculation |OT11| Cry Havoc and Let Slip the Dogs of War [NEW NEWS, NEW THREAD - CHECK OUT THE STAFF POST]

Which team are you on?

Double Team (1997)

Team Walnut

The A-Team

Team "No One Can Stop Mr. Domino"

Sports Team

"I'm a loner, Dottie. A rebel."

Team Margarita

User requested account closure

User requested account closure

User requested account closure

The cutest v-tuber

Self-Requested Ban