"Of those, two people said they were directly familiar with plans for Sony’s new console. Those two people both told me that the next PlayStation is unlikely to release in 2019"

This is from the article.

https://kotaku.com/sources-the-playstation-5-is-still-a-ways-off-1825152206

Also from reading the article there is no information that devs have devkits either, which Jason should know as he is suppose to have the best inside information of all gaming reporters.

Five years is "so soon" to you? If you were in Jr. high school when PS4 launched, you'd be getting ready for college when PS5 released... IF it released this year. Five years is fucking forever in tech.

exactly, only a handful of studios (imo mostly 1st parties) will even be able to make use of the available power (with added detail/ complex physics; complex AI for much more life-like worlds / highly detailed, fluid animations etc) of next gen consoles within a sane dev cycle of 4-5years

that's why most games will mainly offer better image quality and higher framerates, which means there won't be much of a "generational leap" in side-by-side videos

VR could profit a lot though, more high fps games could mean more games get a VR mode and higher res is desperately needed aswell

No need to worry, we can wait for built-in hardware acceleration for it like the PowerVR Wizard architecture ...  

It's all in the good days work for Microsoft (well I guess IHVs also deserve credit too) standardizing DXR ... 

This, there could be some hardware related bugs that some software takes advantage of or an undefined behaviour related to hardware implementations ... 

Five years used to be a long time in tech, it is going slower, and slower now. If PS4 does launch with the specs shown, in 2020, PS4 to PS5 will be by far the smallest gap between PlayStations preformancewise in the history of the brand. Then say you want a PS6 in 2025, the will be an even smaller gap between platforms.

Unless some new form of fabrication becomes viable for mass market, new generations should be taking longer and long. There is no reason to split up userbases, and force consumers to go to new platforms if their is no justifiable performance improvement. Your average Joe consumer, already questioned why they had to buy a PS4/XBO, when they can't hardly tell the difference between PS4 and PS5, what makes you think they are going to upgrade in droves.

Kind of. At best PS5 will be as powerful as today's high-end PCs. In two years time, it will be considered mid-range. Apart from 4k60fps capable machine, I don't think it will look like generational leap in terms of pure visuals. I'm more interested in what devs (especially 1st party) can do with utilizing that Ryzen CPU to its full potential. We could see some very impressive (or even never-seen-before) physics and environments. So it might not look like a leap, but it will most definitely feel like it.

I have been on the 2020-2021 window for a long time =p

I think graphics have gotten to a point where it is pretty much impossible to be impressed by any kind of leap in terms of visuals. I mean, look at gow or a fully modded gta5 on pc. We're getting extremely close to photorealism (or actual animation, depending on the visual style).

Next will be physics, AI and world density, but I doubt those things will be able to provide the same level of amazement when simply looking at a game. What I mean to say is, we don't need to go much further in terms of visuals. We're "there" already. What we need, now, is consoles able to provide 4k/60fps gameplay (+HDR). Also, many devs are not able or willing to keep up with the current state of the art visuals. Developing a game that can take full advantage of today's hardware simply cost too much, we don't need to put devs under more pressure simply because hw could theoretically allow them to develop photorealistic games.

GCN can only accomodate 64CU's, custom would be too expensive. Loving the TF numbers people are throwing about. PS5 will not be a massive leap, the pro proved that.

2020 could be the first year when a huge performance leap will be viable and cheap: Zen2 or even Zen3 for CPU cores, mature, powerful and cheap Navi tech or even Next Gen for the GPU (if not delayed Next Gen will already be there, but it's not sure if models with the right performance, consumption and price for consoles will already be available), and last, but not least, maybe by then RAM manufacturers will have caught up with additional demand due to cryptomining, or maybe the whole cryptocurrency bubble will have burst, making RAM dirt cheap thanks to sudden excess supply.
By then the necessary power to do VR at at least 1080p@60fps per eye will be commodity, and most probably even 1440p will be cheap enough, so for VR to finally become mainstream it will only be matter of good headsets and well devised input devices to exist and be cheap enough, and above all of SW making good use of the available HW resources.