Next Generation Hardware Speculation with a Technical Spin [2018]

Status
Not open for further replies.
And it wasn't until months after release that Nintendo faithful accepted the Switch was what it was and didn't have any magic sauce left untapped.
 
To me, many people are taking the semiaccurate article too seriously, even supposedly affecting their PS4 Pro purchase decisions. A lot of the "insider" information is stuff someone paying attention could guess at even if it ends up right. Previous real leaks had codenames, which would never have been guessed.

Also, for PS4, XBox One, PS4 Pro, and XBox One X, the rumors with project codenames started floating around about GDC or soon after of their previous launch years.
 
Realistically, should be expect DF's first remarks to be about a year ahead of launch?
DF post articles when they are speculatively addressing popular rumours, so probably not. :nope: DF are obviously ultimately funded by ad-clicks so they'll produce articles that will prove popular. Here's their July 2011 article discussing a possible PS4 launch in 2012.

Their first nailed on accurate PS4 article (18 January 2013)was just one month before Sony's official unveiling (22 February 2013) with the console launching that nine months later. For PS4 Pro it was three months (June 2016) before Sony announced the machine (September 2016) when the developer guidelines leaked.
 
Before Digital Foundry, VGleaks was first to accurately report on PS4 architecture and codename.

"leaks" from that time were all over the place, many listing 2-4GB of RAM [nobody talked about 8GB, except Crytek] and for CPU both Jaguar and Steamroller were in contention.

What is interesting, the majority of leaks about Orbis and Durango happened not after GDC, but week after CES 2013.
 
Last edited:
Before Digital Foundry, VGleaks was first to accurately report on PS4 architecture and codename.

VGleaks were posting so much that, as BRIT said, "You throw enough darts and sometimes you hit the board."

"leaks" from that time were all over the place, many listing 2-4GB of RAM [nobody talked about 8GB, except Crytek] and for CPU both Jaguar and Steamroller were in contention.

We know PS4 was targeting 4GB. Some devs spoke up about 8Gb being a surprise.
 
VGLeaks got the docs early, giving exact details on Onion bus and Durango's custom hardware tweaks. It was an unprecedented leak for Durango and PS4 and can't be relied upon to happen again.
 
VGLeaks got the docs early, giving exact details on Onion bus and Durango's custom hardware tweaks. It was an unprecedented leak for Durango and PS4 and can't be relied upon to happen again.

I'm glad you posted this, Shifty. The above posts surprised me. They most likely got one time lucky (we'll see I guess) but it was always obvious their info was actual documents to me.
 
Stumbled on this article that was from a few months back:

http://www.patentlyapple.com/patent...e-year-with-production-kicking-off-in-q2.html

Another powerful growth driver for TSMC rests with its 7nm process. The company has won orders from more than 50 customers for fabricating chips on 7nm process to support smartphones, gaming consoles, processors, AI applications, bitcoin mining, and more, leaving Samsung Electronics lagging far behind in the race of 7nm volume production.
 
That is probably PS4 pro slim on 7nm and an Xbox one X slim also, die shrinks are usual, and mid gens are currently on large expensive dies.

Wouldn't 7nm design cost estimates rule that out? I can't see a enough ROI for an ~20% of base console sales to warrant it.
 
If the PS5 were to be fully backwards compatible with the PS4Pro, I could see Sony phasing out the Pro. I think it would be a console without a market then.

Just release a $199 PS4 super slim and a $399-499 PS5 next year.
 
That is probably PS4 pro slim on 7nm and an Xbox one X slim also, die shrinks are usual, and mid gens are currently on large expensive dies.
The wafers required for 7nm are more expensive than for 14nm and combined with the questionable (and certainly much lower) yields, I'd be surprised if it was cheaper to switch to 7nm yet.
 
If the PS5 were to be fully backwards compatible with the PS4Pro, I could see Sony phasing out the Pro. I think it would be a console without a market then.

It's an interesting dilemma. I also think Sony will offer two consoles and kill one and conventional wisdom it would make sense to keep the cheapest offering along with the highest performing, but it depends on Sony's future strategy.

Despite what they'e said publicly (old management have moved on), if Sony decide to go for something more 'generationless' like Microsoft, it would make more sense to kill the oldest, lowest-performing device (PS4) so developers can make Pro the new low bar, but but I can only see them doing that if they can sell PS4 Pro at the same price as PS4. Which could result in them keeping PS4 artificially high to make that possible.
 
Even though 7nm may be ready in 2019 I'm not sure Sony would launch in 2019. There is usually a ~6 month gap between announcement(or leaks) and release date. In the case of the PS4 it was announced in Feb '13 and launched in Nov '13. So that's 9 months. So you are actually "killing off" (seriously hampering sales) of the existing gen about a half a year before the launch date.

So a fall 2019 launch of a new gen would mean only a year (or less) from today of PS4 being the current main platform. A fall 2020 launch would mean about 2 more years from today with PS4 being the prominent platform.

To me a fall 2020 launch seems most realistic. Perhaps spring 2020 launch.
 
The readyness to go on 7nm surely depends on what deal Sony or MS have struck with TSMC.

There's been a few comments in the thread that mobile sucks up capacity as they have high volume. Don't the larger console chips mean that in waffer terms, Sony and MS end up as a significant book of business?
 
And there's no chance of the '7nm game consoles' referring to Nintendo Switch lineup?
 
Status
Not open for further replies.
Back
Top