This week’s Digital Foundry Direct Weekly delivers another ‘war story’ content coverage as we try to do our best to cover Star Wars Jedi: Survivor, with no PC code arriving until the day before release and Despite there being no console review code until launch day. Attempts to get it through retail outlets earlier helped, but from a manufacturer’s perspective, I can’t help but feel a little disappointed that we didn’t have more time with the console versions of the game. However, it’s the other topic on the docket that I’m going to focus on in this blog – the CMA’s blocking of the Microsoft/Activision deal because of the alleged unfair advantage it would give Microsoft in the cloud space.
There has been much analysis on whether CMA’s reservations are valid, but there was one piece of information contained in the documentation that bothered me – that ‘rival’ – Sambhavna Soni – Contacted CMA to inform them that latency challenges facing cloud gaming have been overcome. The CMA was lacking in terms of detail here, saying only that ‘more powerful graphical processing units (GPUs)’ could solve the problem. The immediate reaction online was to dismiss the idea because you can’t overcome the laws of physics. Latency will always be a problem that cannot be conquered.
I’m speaking as a journalist who has covered ‘the power of the cloud’ since its inception, when OnLive claimed to deliver an experience ‘as good as local gaming’ back in 2011. It couldn’t possibly work, I said. It certainly worked, with a functional gaming experience possible, but lag was terrible and image quality was even worse. By the standards set by OnLive, it will never replace local gaming – and that perception remains today.
The thing is, assuming that the CMA correctly defines Microsoft’s ‘competitor’, I think this opinion has some weight.
- 00:00:00 Introduction
- 00:00:45 News 01: Star Wars Jedi: Survivor Impressions
- 00:23:57 News 02: Microsoft Activision buyout hits roadblock
- 00:34:30 News 03: Microsoft pushes for Xbox energy sustainability
- 00:48:21 News 04: ASUS ROG Companion Chips Benchmark
- 00:55:50 Pro Q1: If RT becomes a bigger rendering focus, can future GPUs return to rasterization performance?
- 01:00:36 Pro Q2: Are 8GB GPUs Doomed to Sloppy Textures on PC?
- 01:07:33 Supporter Q3: How did Don like his first video?
- 01:11:15 Pro Q4: Will you be making a GTA5 retro time capsule video when the title hits its 10th anniversary later this year?
A lot has changed since 2011 and of all the cloud gaming start-ups, it’s really only Nvidia that has provided a system that can be said to address the issues. And yes, a lot of this is down to the ‘more powerful GPU’, along with an incredibly robust and convoluted approach to transfer and reducing latency with an end-to-end pipeline. Check out Cyberpunk 2077 RT Overdrive on the GeForce Now RTX 4080 tier and not only are you looking at an experience that may well represent the next generation of consoles, you’re also getting perfectly reasonable latency. With the mouse and keyboard you can feel the difference in response – but it’s subtle. Go to a controller—an inherently backward form of input—and you’d be hard-pressed to tell.
So how was this achieved? There are advances galore. Systems from OnLive to first generation xCloud to Stadia all have external encoders installed, which means the GPU will finish its work before sending it to the proxy ‘display’ (i.e. encoder). GeForce Now and xCloud integrate the encoder into the main processor, reducing lag. Secondly, where possible, Nvidia aims to reduce latency in games either through its own driver-level functions or through its bespoke latency-saving technology, Reflex. Latency can be further reduced by running the game with the frame-rate unlocked. There is no V-sync off ‘tearing’ as full frames are pulled from the framebuffer for video encoding.
Lastly, GeForce Now also supports 120Hz and even 240Hz streaming technology. The faster the stream updates, the lower the latency. I’m underestimating the many challenges that have been overcome here, but the point is that GeForce Now is delivering 4K HDR 120Hz streaming, and it’s a class apart. I’m still trying to find time to complete my GeForce Now RTX 4080 review, but here are the latency numbers put together by my colleague, Tom Morgan, using Nvidia’s LDAT system, to measure the end is a fanciful but near-idiosyncratic method of—the lag from button press to on-screen response, milliseconds to the end.
| Destiny 2 Average Latency (ms) | native pc | Native Xbox Series X | GeForce Now 3080 (PC App) | GeForce Now 4080 (PC App) |
|---|---|---|---|---|
| 60 Hz | 49.0 | 85.0 | 81.7 | 82.2 |
| 120 Hz | 31.8 | 40.6 | 59.5 | 60.6 |
| fortnite average latency (ms) | playstation 5 | Native PC (V-sync off) | GeForce Now 4080 (V-Sync Off) |
|---|---|---|---|
| 60 Hz | 96.8 | 35.4 | 78.3 |
| 120 Hz | 48.0 | 27.4 | 53.3 |
| 240 Hz | , | , | 33.1 |
There’s also another factor at play: infrastructure upgrades. As you can see above, the truth is that Nvidia solved the latency challenge when it launched the GeForce Now RTX 3080 tier. High frequency streaming has improved – 4K120 and 1080p240 are new additions. However, the usefulness of cloud gaming – especially in the ADSL era – was limited due to the fact that contention over the same home connection would mar the experience. If I was playing xCloud on the ADSL line, while my wife loaded up Netflix and watched some videos, my gaming would be seriously compromised. It would be even worse if something like a bandwidth-sapping Steam download kicked in.
The advent of FTTP (Fiber to the Premises) has brought the bandwidth for cloud gaming to run concurrently with other standard Internet activity – although Steam downloads with unrestrained download caps can still prove problematic. The point is that, for the most part, modern infrastructure allows GeForce Now to coexist with standard Internet access within the home. client to server latency? For me, it’s usually 5-7ms. Not everyone has fttp, of course, but within ten years? Absolutely.
GeForce Now isn’t a total replacement for local consoles, not yet. One of the ways to address latency is to pipe frames to the user as quickly as possible, meaning ‘frame pacing’ could look like this. However, it stands as a practical example of how good cloud gaming can be and is perhaps the only cloud service that has aggressively made good on the other great cloud promise: that servers be regularly upgraded with the latest hardware. Will go That’s how you can play Cyberpunk 2077 RT Overdrive right now at 4K max at around 50-75fps – the servers were upgraded with the latest Ada Lovelace GPUs.
I’m not going to say that CMA blocking the Microsoft deal has anything to do with the viability of cloud gaming or that the ‘power of the cloud’ gives Xbox an unassailable edge because xCloud itself is fine, but clearly something Generation behind Nvidia’s technology. Where Microsoft is well positioned is that it can provide the level of infrastructure needed to make cloud gaming work – it owns Azure. On top of that, the path to profitability with cloud gaming is easier for Microsoft than for anyone else now that Google is out of the market. There’s no reason why xCloud Server blades can’t hold their own as standard Windows servers when they’re not calling for gaming.
Although right now, xCloud is far from the killer app it needs to be. Image quality is still problematic, there isn’t enough resolution or bandwidth to replace the local experience. Latency is far more noticeable, although 60fps games respond well enough to be reasonably priced. Upgrading to this, especially in terms of lag reduction, is only possible if Microsoft decides to pursue Nvidia-like strategies – and it’s not as easy as you’d naturally call the PC approach to gaming. Can
However, Microsoft’s ‘competitor’ mentioned in the CMA document suggests that the challenges of cloud gaming are being actively investigated and may be progressing. And if you really want to have your mind blown, dig into id Software’s patent for Bethesda’s in-Limbo Orion streaming system, a latency-masking system using user input to affect video stream motion vectors. Thoughts like It seems a bit like Stadia’s claims of negative latency were made a lot of fun of. I think the point is that engineers in the gaming space love impossible challenges — making cloud gaming work among them.
This expands a lot from my thoughts in this week’s DF Direct Weekly, but there’s a lot of other good stuff too – discussion on the Xbox’s remarkable stability drive (which I hope to write in more depth soon) In this article, we discuss ROG Reveal’s collaborative benchmarks, as well as our thoughts on the struggles facing owners of 8GB graphics cards. As always, consider joining the DF Supporter Program to tune in to the show, talk tech with our team, and get early access to Direct and a bunch of other cool stuff. In the meantime though, if you can sample GeForce Now’s RTX 4080 tier, I wonder what you make of it – and whether cloud gaming has really been solved in the major ways I think it is. Has been done