DF Direct, our weekly news and Q&A show, recently celebrated its bicentennial—though not because it’s 200 years old. Instead, it hit an impressive milestone: 200 episodes. All four of Digital Foundry’s non-holiday hosts came together to mark the moment, trading stories about the earliest installments and other highlights from the program’s almost four-year run so far.
Because I was one of the members on holiday, it was especially enjoyable to return to the show. If you’re a fan of the channel, I’d strongly suggest giving at least the opening segment a go—it’s packed with thoughtful reflections from Rich, John, Alex, and Oliver.
It also made me look at the series through my own lens, since it may have shaped the DF team’s workflow more than many people might assume. For example, the weekly broadcast serves as the cornerstone of the DF Clips YouTube channel, which launched in 2023 and is now closing in on 100,000 subscribers. As DF Direct has stretched out over time, the chance to recycle portions into more easily shared clips has been incredibly useful—helping us spread the most important information without having to produce a fully scripted standalone video.
With five full-time Digital Foundry contributors handling testing, scripting, recording, and editing their own segments—Rich, Tom, John, Alex, and Oliver—we usually manage only around three or four topics each week. That simply doesn’t cover the volume of game releases and updates we want to address. Having Direct means that one or two people can quickly review those smaller items, letting us tackle a much broader range than we could otherwise.
The Direct is also one of our go-to ways to stay connected with the community by using the questions submitted for each episode. Like the show’s news section, the Q&A has kept expanding over time, with viewers consistently raising the bar—both with smart, thought-out questions and with plenty of funny memes. Everyone on the team looks forward to choosing the week’s questions, often discovering there are more questions available than there are slots for news.
The questions featured in this week’s Direct are, as always, top-notch. I was particularly taken with Michael Giles’ idea for building a DF Latency Rating—one that, in his view, could be based on millisecond cutoffs. That approach would give viewers a straightforward way to judge whether the experiences offered by different frame generation levels come with input latency that’s good, acceptable, or intolerable.
- 0:00:00 Introduction
- 0:02:07 News 1: DF Direct celebrates episode 200!
- 0:32:01 News 2: Nintendo sets Switch 2 price expectations
- 0:46:17 News 3: Next Battlefield shown in brief teaser
- 0:55:45 News 4: Monster Hunter Wilds beta, benchmark tested
- 1:09:05 News 5: RTX 5080 overclocking tested
- 1:26:52 News 6: Epic acknowledges Unreal Engine has a #StutterStruggle issue
- 1:34:17 News 7: Star Wars Outlaws updated with new PSSR version
- 1:41:07 Supporter Q1: Could you test the 5090 without AI features vs. the 5080 with all AI features?
- 1:47:23 Supporter Q2: Could DF create a latency rating system?
- 1:54:41 Supporter Q3: Would enhanced resolution and AA quality suffice for Nintendo Switch 2 games?
- 1:57:53 Supporter Q4: Is a Bloodborne remaster on the horizon?
- 2:03:32 Supporter Q5: Could 1080p actually be preferable to 4K for Switch 2 games?
It’s an engaging concept, and it may have an answer that’s just as interesting. Ideas like this—and others along the same lines—have come up more than once as possible ways to sort different input latency levels in an intuitive way at steady frame-rates, and there’s definitely something to be said for that. Still, the hurdle for any such system is that what counts as “acceptable” input latency can shift a lot depending on the control method, the game itself, and even personal preference. For instance, an action-adventure title played with a controller might feel perfectly fine with 4x frame generation, while a mouse-and-keyboard shooter could feel completely intolerable even when running at standard 2x frame generation.
Looking back over the years, we’ve repeatedly seen fairly large swings in input latency from one console generation to the next. Rich, for example, called out the PS3 as a notable step down from the PS2. That outcome traces back to a mix of influences: a broad industry move toward 30fps games instead of 60fps, game engines becoming more elaborate to support increasingly advanced visuals, and early LCD TVs gradually replacing CRT displays. Taken together, these changes produced a major rise in input latency—we recorded more than 200ms in Killzone 2, even after factoring out display latency—but because we didn’t yet have the right tools to measure it properly, this shift didn’t receive much attention.
Today, we’d likely benefit from more specific tools that help pinpoint where the line is for excessive input latency. I like Rich’s thought about a company such as the makers of 3DMark building an input latency test that covers a range of game genres—so users can randomly vary latency and flag the point where the experience becomes unresponsive.
With those two points wrapped up, I’d like to end on one final thought: Digital Foundry wouldn’t be where it is today without the support of its audience. We truly value everyone who takes the time to share feedback, questions, and good humor each week—or who simply reads and watches what we publish. Thanks to you all.
If you’d like to become part of our incredible community, take a look at the DF Patreon. You’ll get some excellent benefits, including access to our private Discord server with wonderful people, weekly updates on our projects, and high-quality video downloads for every video we’ve released over nearly a decade. Higher membership levels include everything mentioned above, plus extra perks like early access to non-embargoed videos, behind-the-scenes material, and—at the top tier—early access to DF Retro episodes along with exclusive retro bonus content.
As always, thank you again for watching and supporting Digital Foundry.