Quad core CPU's have taken the role of the old dual-cores in terms of performance.
I.E. AMD's Ryzen's low-end are Quad-Cores.
Intels i3 is now Quad-Cores.

So it goes without saying... Regardless of the amount of cores that is chosen (8-Jaguar cores were roughly comparable to an Intel i3 Dual Core), having CPU performance around that level is probably something we might expect.

The Xbox 360 had SMT, but it did bring with it some caveats that I won't get into here.

In short though, SMT would not prevent backwards compatibility, but when leveraged right can allow you to extract more performance out of each core.

Not even a factor when we start talking about monolithic semi-custom SoC's.

They will go with whatever is cheapest. - Even the Xbox One X isn't using the biggest, fastest mechanical disks and that was a "premium" console.

SSD's/m.2 drives still command a premium and will do so even in a few years from now.
Plus the demand for NAND is increasing, so costs could actually increase, especially if NAND fabs retool to start taking advantage of higher DRAM prices.

The Xbox One X outselling the Pro is an inconsequential issue as the Pro isn't the bulk of Playstation sales anyway.

DDR5 isn't going to happen. JEDEC has only just finished demonstrating the DRAM and hasn't even finished ratifying the standard... And is thus probably still a few years away from a market release.
It does promise to double the bandwidth and density of DDR4, which means it's not going to be enough for next gen anyway unless you implement a stupidly wide bus.

It will be GDDR6, it will be what will offer the greatest benefits, higher capacities and good enough bandwidth/latency.
GDDR6 is currently ramping up production right now, so costs will be leveled out by the time 2020 hits.
 

But it proves that most PS4 owners don't deem it worth it to fork over 400 bucks to upgrade. New owners are still going with the base model. The Pro, unless it sees a price drop to $349 or even $299, sales will stay stagnant compared to the more powerful One X.

While the DDR5 standard is supposed to be published this year, don't count on it for the next console. DDR5 is late, it was supposed to get finalised in 2016 already, with market release in 2020. In other words, they are 2 years late on that plan already, and from final specs to the released product there are always a couple of years ( due to how lithography works, that can't be shortened by more than a few months if they're lucky). Plus, any new standard is much more expensive than the established one, making them more unsuitable for consoles.

Finally, GDDR gets developed from DDR, and a GDDR standard deriving from DRR5 will not come before DDR5 is released because GDDR is basically DDR modified for high bandwith. Just have a look how long it took for GDDR6 to be developed from DDR4. While there can be an improved GDDR7 in the next years, that one would still be derived from DDR4, not 5 (just as GDDR2 and 3 are derived from DDR2 and GDDR4 and 5 are derived from DDR3)

In short, don't count on DDR5 before 2021 earliest and any GDDR derived from it for about 2023, and pricely viable for a console not before 2025.

Actually GDDR6 is going to release with hardware before DDR5 does ... (JEDEC also never intended for the final spec to release in 2016 either so there was no delay and we don't know if future GDDRX technologies will be derived from DDR again as JEDEC never mentioned similarities between GDDR6/DDR4) 

If a memory standard becomes available then you can bet you'll see it in products next year after that ... 

I don't know if you understand german, but hey: https://www.golem.de/news/arbeitsspeicher-ddr5-naehert-sich-langsam-der-marktreife-1608-122737.html In the last paragraph they are talking about that JEDEC supposed to finalize DDR5 before the end of that year, so yes there is a 2 year delay. Availability was only expected for 2020, so if the same delay applies then commercial availability would not be before 2022.

Of course GDDR6 releases before DDR5. After all, GDDR6 is based upon DDR4, not DDR5. Any GDDR version that would be derived from DDR5 will not come out before 2023.

Maybe late 2012 because of the Wii U, but 2011? I honestly don't remember it.

I don't get why people are throwing around HBM like its the holy grail of memory. And in the process underselling GDDR6. The fastest GDDR6 ram (aka as highest capacity) will have 24GB and a bandwidth of 864GB/s. In comparison the next best thing on the HBM front will have a bandwidth of around 901GB/s. We are talking about a bandwidth difference of less than 50GB/s. And a price difference of more than 50% and thats not factoring in the significantly lower yeilds. 

I think we need to stop looking at the tech and more at the performance when it comes to this.... anything higher than 500GB/s bandwidth will be more than sufficient for 4k gaming.  There is also that possibility that console makers can even go with a custom "extra wide bus" which would result in even more bandwidth from their chosen ram.

X is not selling because of exclusives. Switch is a great example, though it is a hybrid and that could be the reason for high sales I think exclusive games is an important factor. I guess people will buy 500$ console if it has the games and obviously BC for PS4.

If native 4K is too demanding then IMO it will not be bad idea to keep using checkerboard 4K but we get huge upgrade in shadow, lighting, special affects, stuff on screen at 60fps. I am a 1080P user but honestly I think gfx is more important than native 4K. Having said that I think native 4K will be given the priority because of marketing purpose.

an 11 tflop 1080ti costs over $1000 right now and the prices are going up.

In terms of memory bandwidth, it's absolutely top notch and I don't wish for 4K, I wish for physically based dynamic global illumination ... (BW might become a severe bottleneck when doing ray traversal in ray tracing so I want this mitigated as soon as possible for next generation) 

Also the difference is much larger than 50 GB/s. The fastest HBM 2 memory module can let us achieve rates as high as 1.25 TB/s on a 4096-bit bus width while the maximum a GDDR6 standard memory module can achieve on a 512-bit bus width will net 1 TB/s. GDDR6 has a massive 20% BW deficit compared to the fastest HBM 2 memory module ... 

That may have used to be true in the past but now I'm not so sure anymore since there's nothing stopping their recent momentum. (industry on the verge of transitioning to EUV, Rambus reuniting with JEDEC(!) and higher demand than ever for higher performance DRAM) Before memory standards used to have a much lower turnover rate but currently their bringing out new memory standards faster than they did in the past. GDDR3 standard wasn't developed by JEDEC since they adopted it from ATi Technologies at the time, the specifications for the GDDR4 standard was released in 2006 by JEDEC, GDDR5 standard was released in 2007, GDDR5X standard was released in 2016 and the GDDR6 standard released a year after that. Consequently the original HBM standard was adopted in 2013, HBM 2 standard was finalized in 2016 but HBM 3 standard could get finalized by as early as the end of this year ... 

The final DDR5 standard is about to be published this year too ... 

That’s highly unrealistic to assume, just for the fact the new CPU will be much faster for sure. GPU will also be at least Vega based. The X1X doesn’t even use Vega features - the Pro does though.

But it proves that most PS4 owners don't deem it worth it to fork over 400 bucks to upgrade. New owners are still going with the base model. The Pro, unless it sees a price drop to $349 or even $299, sales will stay stagnant compared to the more powerful One X.

And didn't even know about GDDR6. In that case, I can't wait. As I stated though, don't expect to even hear about a PS5 until the PS4 hits 100 million which is easily doable at the end of this holiday season.

Remember when people said the no console would ever sell 100m again because of stupid mobile phones and tablets (which turned out to just be just a huge fad btw)? Aah, those were the days...

In terms of memory bandwidth, it's absolutely top notch and I don't wish for 4K, I wish for physically based dynamic global illumination ... (BW might become a severe bottleneck when doing ray traversal in ray tracing so I want this mitigated as soon as possible for next generation) 

Also the difference is much larger than 50 GB/s. The fastest HBM 2 memory module can let us achieve rates as high as 1.25 TB/s on a 4096-bit bus width while the maximum a GDDR6 standard memory module can achieve on a 512-bit bus width will net 1 TB/s. GDDR6 has a massive 20% BW deficit compared to the fastest HBM 2 memory module ... 

Eventually yes, but I don't think anything higher than 4K headsets will be viable for ps5, and spread over 110 degrees that compares to 720p in the center at the very most. Probably less as that likely won't be 4K per eye. Besides expecting next gen to render native 4K x 2 at minimum 60fps without any hitches is still a tall order. Memory speed will need to go up a lot as well to use 24GB effectively or those high res textures will just be a bottleneck.

Yet to future proof for a ps5 pro, 32GB would be better. Or perhaps 16GB for games, 8GB for the OS. That's still over 3 times more than games get now.

One thing one has to consider though...Even in 2-3 years not even 50% of people will own a 4K TV or will have the possibility to get one, additionally to their shiny new 4K console.

Are 12-15 Tflops really needed for a highly detailed 4K/30 performance?! 

I still think 10-12 seems possible for PS5, but higher Tflops would be more needed for VR than for 4K imho.