hughj

The contributing factors to what caused this past generation to feel underwhelming on launch (the plateauing of Dennard Scaling and Moore's Law) are going to be felt even more so this coming generation, albeit in different areas. The tightening of the belts this past generation resulted in a low wattage CPU to focus more on GPU, while this generation looks to flip that. In either case it's still not going to feel like a generational leap as we were accustomed to during the 80s, 90s and early 00s where clock speeds, transistor budget, memory capacity, bus width, etc were all seeing an order of magnitude leap every cycle.

Going from the PS3 -> PS4 -> PS5 did not and will not revolutionize existing genres or make previously infeasible ideas suddenly feasible. The performance jumps on the order of 1.5x, 2x, or 3x that we're seeing now are great for refining the quality of experience of existing forms of content (it perhaps allows you to execute on a handful of things that weren't quite performant enough last go-round) but it's not a revolutionary jump that's effectively a blank cheque for a designer to do things that have never been done before. Those days of seeing across-the-board 10x to 100x increases in performance every 5 years are over, and with that the expectations people have about how the industry works, releases products, markets them, and ultimately generates revenue are going to have to adjust. Same thing goes for smartphones, PCs, and any other industry where revenue has been directly tied to what the semiconductor fabs were delivering. These circumstances are reflected by the waning enthusiasm for electronics trade shows like CES and E3 as the products showcased there no longer feel like a glimpse at the future. There is no equivalent of rubber ducks in a bath tub to get your imagination churning if the new hardware can only provide 25% more ducks.

In a backwards sort of way, the fact that the next batch of hardware is going to be another incremental upgrade is precisely why the coming generation will continue to break from tradition -- the industry has to adapt to this new norm by finding new ways to operate and generate revenue. This past generation we saw mid-cycle hardware upgrades, price tiered content SKUs, loot boxes, and games-as-services as new routine revenue streams, and there's probably going to be even more examples of that in the coming cycle.

Regarding data caps: I'd imagine for a lot casual gamers it might actually utilize less data to stream a few games for a few hours a week than to have to download multiple 100-200GB installations (especially factoring in patches, updates, and expansions), and it also removes the need for having very large HDDs.

Another thing to consider: There's nothing necessarily forcing the concept of streaming to be an all-or-nothing affair. Depending on the gameplay and environment, I can imagine cases where you could render and stream layers of skyboxes (or even light-fields) and then be transformed and composited with client-side foreground elements. Doing it that way would avoid having to render unique POVs for every connected client, as the most distant layers could be shared across multiple clients in a given area, and those distant layers wouldn't need to be updated at the full refresh rate either.

NES Double Dragon 2 would be perfect. It takes about the right amount of time to play through, the mechanics are simple but take practice to master, the environments are varied, it has some of the best music on the NES, and the platforming sections will add a lot of nice drama.

The industry will move to whatever medium can reach the largest market, and game genres that don't play as well under that larger umbrella will either adapt or gradually be sidelined with less publisher investment. We saw this with many of the PC franchises of the 90s and early 2000s as they pivoted to become multi-platform console games resulting in gameplay mechanics being streamlined, slowed down and loosened up to accommodate gamepads and living room TVs. If mobile and cheap streaming boxes represent the chance to grow the potential market by a factor of 10x, (which seems reasonable), then I'd imagine any franchises and genres that can be engineered to accommodate slower and sloppier input will see that happen.

The other factor here that I've not heard addressed by the GB staff are the industry ramifications from the gradual death of Moore's Law. The days of getting new iterations of console (or PC) hardware that can dramatically outstrip the previous iteration while maintaining the same price point are effectively over. Even reasonably modest jumps in performance are going to come at a growing price, so something has to give there -- the baseline price of a new console can't be too high or else your market shrinks, but the performance upgrade over existing hardware has to still be enough that people think it's worth buying. Streaming services coupled with a high-volume dirt cheap SKU seems more reasonable than expecting the entire market to be willing to buy a new $500 box just to get slightly more stable framerates in the same types of games.

Window's Mixed Reality might help people for getting over the MSRP hurdle for hardware adoption (although the Rift permanently being at $399 now seems to make that less of an issue), but it's still going to have a problem with the trade-off between having content that meets or exceeds user expectations and having that content actually perform well enough on the average desktop computer or laptop. Low latency requires low frame times, and you only get that with very stripped down content, or a moderately powered gaming computer. This initiative by Microsoft was done because the reception to Hololens was pretty lukewarm and needs to stay in the oven a while longer, and because MS wanted a stopgap platform that could also appease their numerous hardware partners that were unable to sell their own HMD hardware until now.

"In my opinion, VR is something that should've waited until the time and tech was right. I felt its release was too early, as for the hardware to run the device had to be top-notch, and the amount of content available was not enough."

Nothing has really changed in that regard, imo. The biggest hurdles for VR adoption were (and still are) the low visual fidelity, stiff hardware requirements, and the lack of killer app use cases on the same level that Lotus123, desktop publishing, or the internet were for PC. Bespoke made-for-VR games are cool, but the cost of game development is high, the value for dollar expectations from consumers are high, and the size of the market is small. VR games made by teams larger than 10 people have little hope of earning their money back, and games made by fewer people have a hard time justifying their price -- the 'chicken or the egg' problem of content on a nascent platform.

I think VR will probably continue to struggle with this problem until the HMDs have sufficient fidelity (pixels per degree) to be a viable display replacement for any/all existing tasks in the same way that the Iphone was able to leverage the common usage of cellphones, ipods, and the web. The fact that the Iphone had an app store was pretty incidental to how viable it was early on, where as the value VR brings right now is purely contained within their app stores. Having a $200-400 HMD that can give your PC or laptop as many virtual monitors as you like, or make the Netflix video you're watching on your phone look like an Imax is something that wouldn't need tons of GPU horsepower to run, and would be an easy sell when TVs and monitors cost several times as much. All of the available or announced VR HMDs though are still on the order of 1/10th the perceived resolution of a current TV/monitor though, so we're probably not going to see it anytime soon.