sounds interesting, can't wait for the first reveal

Sony has lied about performance every gen, so why not now? Remember how the PS2 could launch a missile?

There are even 20Gbs versions.... but yh thats not what we will see in these consoles. However there are some things to consider..

Right now 1GB of GDDR5 from 6Gbs -8Gbs all have a bulk pricing (2000 units) of around $6/chip. For 1GB GDDR6 that range from 12Gbs to 14Gbs is between $9 - $11/chip. 

Now even if soy and MS aredoing "super bulk" orders (literally in the millons) that will only further bring the prices down by say about 20 - 30%. But then there is also the fact that sony/ms won't be using 1GB modules but will be using 2GB modules, which cost more than a single 1GB module but less than two 1GB modules.

Basically, if we assume that doing million scale type ordes for 14Gbs 2GB GDDR6 ram means they spend around $12/chip, then 16GB/20GB/24GB will cost them $96/$120/$144 respectively........ just for RAM alone. To put this in context the PS4 RAM budget back in 2013 was around $88. At around $12/2GB chip, this is why it would make more sense adding 4GB of DDR4 ram to 16GB of GDDR6 to have a total of 20GB of RAM.

Technically even the PS2 could launch a missile....... 

I believe 16GB GDDR6 and 4GB DDR4 is feasible. Separate memory pools make sense from a performance point of view as well, since the latency is lower for the DDR4 memory and the bandwidth used by the CPU in a unified memory system nonlinearly reduces the bandwidth available to the GPU. As long as the performance characteristics of the GDDR6 bus is similar to that of the GDDR5 bus of the PS4, there would basically only be pros with seperate memory pools similar to this setup.

Edit: By similar performance characteristics, I ofcourse don't mean similar performance as for the PS4, but the same kind of characteristics i.e. about 10-15% of the GDDR6 bandwidth available to the CPU if the developers chose to access the GDDR6 memory from the CPU.

Also, I think Sony is prepared to take an initial loss on the PS5, something they not were on the PS4, thanks to their much improved financial situation and considering how important the PS business has become to them. I expect them to spend more money on memory for the PS5 than for the PS4.

https://www.theverge.com/2015/1/15/7551365/playstation-cpu-powers-new-horizons-pluto-probe

Even PS1 chips were used on this type of stuff.

And considering we had missile during the 50's and stuff, I have no issue believing a PS2 chip could launch a missile.

Also this is a leak, not Sony claiming anything. And would you disprove PS4 or PS4Pro information Sony gave?

@bolded: I said that before already. However, the way they are organised only allows for 64CU. And that's compute engines, not shader engines, which is something different entirely (basically a rebrand of the GCA, the Graphics and Core Array, introduced with GCN2). GCN5 only has 4 of them because they can only 4 of them reliably with instructions. Technically they could go past the 64 with more compute englines, but it wouldn't actually increase the performance as the CU would be idling half the time because they don't get any instructions. Hence why it's agreed that 64CU is the limit.

I know Vega is quite efficient around 1150Mhz (I love undervolting my hardware). But to reach those 12-14TFlops and with the practical limitation to 64CU like discussed above, the only way is to clock the chip way past it's sweet spot.

The proposed 80CU @1.2Ghz would probably not beat a Vega 56 because, like I said, the GCN architecture couldn't feed that many CU with instructions. The scheduler, from which the instructions are coming and  then dispense them to the SE, is the limitation. And that can only be removed by a complete redesign. Hence why AMD never even showed a successor to Navi in the roadmaps, they probably knew already back then that Vega was as far as they could go.

@italic: What I heard was that they could have done so, but decided against it because it wouldn't have removed the problem.

I agree that we need to see Navi to be sure about everything. Since it's so long since Vega, it's possible that they found a way around the problem, rendering our discussion here moot.

@underlined: English ain't my first language (or second or even third for that matter), so sorry if I was confusing you. But I calculated the size in 7nm in a previous answer; there I was just proving his assumption wrong that Vega is just so big due to his HBM memory controller and showed him that the One X, which he drew in as comparison, would be just as large with 64CU (if made in the same process of course) despite having a GDDR5 memory controller.

Can do! I've launched a multitude of various missile types with my ps2slim. Upgraded (two ps2 slim stacked) it will be capable of launching even hypersonic missiles. So far none of them were armed, test launches only so to say. But soon the time will come...

That is not how GCN works. Each CU has its own instruction scheduler. The command processor can process commands at a far higher rate than the front-end can process them and the same applies for the front-end to the back-end (SEs/CUs). It's very trivial to saturate the CUs with wavefronts and that happens all the time in the current design. That said, the processing power of some parts of the front-end probably should be increased if the CUs are increased to keep the architecture balanced.

I'm not aware of any technical limitations to go above 64 CUs but they probably exist. I suspect the biggest hurdle is to increase the bandwidth of the L2 cache, which should scale with the number of CUs to keep the system balanced.

Those are some really good specs. I'm pretty hyped.