hughj

Looks like Forza Horizon is pivoting more and more to arcade off-road. I didn't see a single stretch of road that I would actually want to drive a sports/super car on. I really don't get the point of having licensed cars that look and sound like the real thing, but require you to fantasy mod them to AWD and make them drive unlike actual cars.

The dirt road, off-road, and AWD+supercar modifcation aspect of FH4 is way too prominent as it is (to the point of the game being broken, imo) so I hope where ever they put it has a lot of twisty mountain and canyon paved roads to drive. Mexico isn't the first place I'd think of as a driving road destination for exotic cars, but obviously they have creative freedom to make it work.

I think the examples of people that stay in broadcasting well into their senior years are probably the exceptions rather than the norm. They also happen to be working in an industry that is highly competitive, so there's constantly pressure to be 'on' because there's always new people ready to take their job, or competing on other stations/networks. In the case of GB they're not really in direct competition with anyone, because what they're selling today isn't simply content, but a continued parasocial relationship they've built with their audience over many years.

Dan was and still is hungry. Whatever the contributing factors may be that inhibit GB's content quality and quantity (corporate/management overhead, technical workflow/pipeline, leadership, etc), they've clearly established a groove in terms of what they accept from themselves and I'd expect that to continue for as long as they're earning enough to pay salaries.

Yep. The building system/mechanics is probably the high point for me. The lack of having Minecraft-degree of control over the terrain encourages you to build structures which accommodate the terrain topology, rather than just carving a single level foundation into every hillside.

The other thing I discovered today is that wood panels actually degrade over time from being exposed to the elements, so using thatch roofs is more than just an artificial requirement for workbenches and passive buffs. It encourages you to have roofs overhang exterior decks/flooring, and after you've added the structural supports for that overhang.. voila! you've made a veranda. For an early access game to have this amount of variability with relatively few building components is really promising for the future.

@lionsy: I think it was 2000 that most of those big changes came into TFC. If I remember correctly, the 1.1.0.0 "TF1.5" patch was the UI, netcode, and conc jump change, which was in June 2000. The big changes in 2001 were the 1.1.0.7 and 1.1.0.8 patches which added the bunnyhopping cap and HLTV's release. I was always kinda bummed that the fastest era of TFC was never able to have HLTV demos recorded.

@whitegreyblack: I think what a lot of people had in mind with leveraging the "cloud" would be whatever benefits could be extracted from having the game clients in close proximity with a fast network/fabric interconnecting them. The scale/scope of multiplayer games really hasn't changed a whole lot over the last two decades, whether it be concurrent users in a given zone in an MMO, or maximum player numbers in FPS/action games.

I'd chalk that up to being stuck with the client-server network architecture where the clients can never be trusted and the latency between them varies from single to triple digits. Stuff like client-side hit detection and backward reconciliation would be way more streamlined (perhaps not necessary at all) if all the clients are trusted and have low and/or very predictable latency.

I'd like a Planetside-style game that can actually scale up to hundreds of players in a skirmish without exhibiting any of the telltale network jank (desync, stutter, lag, rubberbanding, etc). I'd like an MMO city/hub that's as overcrowded as a launch day, but actually have the infrastructure be able to handle it.

@noobsauce: I dunno, I feel like instances such as this carries an increasingly heavy cost with Google going forward. A lot of people, including businesses, likely didn't get involved with Stadia in the first place for the precise reason of Google's track record, and that's probably going to be felt doubly so for however they try to pivot and rebrand their Stadia tech. The next product/service/initiative that Google attempts to rollout is going to be received with even more skepticism than Stadia was, especially if there's competing offerings from the likes of Amazon, MS, Facebook, etc (not that these companies are perfect, they're just not quite as notorious as Google have made themselves.)

Internally I'd have to imagine that these situations don't do wonders for morale either, as people tend to care about seeing their years of work pay off into shipping products that actually have impact in the world. I recall one of Valve's past VR/AR engineers (Tom Forsyth) specifically citing this as a reason for motivating the mass exodus from Valve to Oculus/Facebook when the opportunity arose, as they were growing tired of an environment where Steam paid the bills and all other initiatives never having to amount to anything. In the case of Google, they probably have a bunch of people with CVs that look like the killedbygoogle.com meme, and sooner or later their best people will leave to work elsewhere.

I think GFN, Luna, Xcloud, and PSNow are in a more comfortable position as those services get to piggyback off existing content ecosystems that would exist with or without game streaming. Stadia is a software+hardware stack and platform unto itself and their position doesn't seem to allow much room to pivot.

If we were to look at streaming services like Youtube and Netflix, I would argue that the primary reason for success was the content being free (user made), or ready-made (back catalog of a half-century of television and film). Stadia doesn't share either of these features, and now that they're backing away from developing their own content, they've put their fate in the hands of third-parties that have no vested interest in Stadia's long term success.

@gundato:

"But it also doesn't really matter at the level of a datacenter/"The Cloud". They go through a LOT of hardware and are buying in bulk often times (at the google or amazon scale) either from the manufacturers themselves or from the toppest tier distributors."

If we were just talking about generic blades of CPUs and DRAM sitting in rack, then sure, but Stadia's hardware is relatively unique (Vega/RadeonPro with HBM memory.) If they had newer Nvidia GPUs then they would at least have something they could lease time for Tensorflow, but that's not the case. I don't know if there's really a business case for having millions of GCN/CDNA GPUs other than to run Stadia gaming sessions.