DF Direct Q+A: What Is Unreal Engine 6? If Steam Is A Monopoly, Why Do We Like it?

Added: 2026-06-04

[00:00:02]Hello there and a warm welcome to the latest edition of the DF Direct Q&A show. Lots to get through this week, particularly in terms of the surprise reveal of Unreal Engine 6. Joining me on the panel, first of all, Will Jut.

[00:00:15]Hello.

[00:00:16]>> Hello. Happy to be here in this lovely sweltering bedroom.

[00:00:21]>> Yes. Yeah, I was just saying, you know, basically locked myself in a in a room, incredible heat, turned on the powerful lights, closed the windows. It's going to be a hell of a ride.

[00:00:32]Hello there, John Linham.

[00:00:34]>> Hello, Rich. I know the the uh viewers andor listeners can't see it, but your profile pick in Discord, I think, says it all about the heat right now.

[00:00:43]>> Yeah, definitely.

[00:00:44]>> Pretty much.

[00:00:46]>> Yeah. And I'm very pleased to see that you're continuing the tradition of uh references at the very beginning of the show that will have no uh >> I must the idea >> video users have no understanding of whatsoever.

[00:00:58]>> It can drive uh video engagement. There we go.

[00:01:01]>> Oh, okay. That's what it's about. Okay, fair enough. Well, um before we move on, let's take a look at this. DF Direct Weekly is brought to you in partnership with Alienware. The latest state-of-the-art Alienware Area 51 delivers a flagship reborn with a focus on extreme highquality design, cuttingedge silicon and thermal precision, and it's available now. In addition to top-end GPUs, the new Area 51 also supports AMD's Ryzen 9000 X3D processors from the 9800 X3D up to and including the new 9950 X3D2. Ryzen 9000X3D processors deliver by far and away the best CPU gaming performance we've tested to date and are a key recommendation for a high-spec gaming PC. And to keep performance sustained at the highest level, Alienware has overhauled the internals with a new gasket architecture. The system uses a combination of 180 and 140 mm fans to create positive pressure, moving 25% more air while remaining 45% quieter than prior generations. Combined with a 1500W Platinum PSU and enough thermal headroom to handle the 600W powered draw of the RTX 1590, the new Area 51 is built for the absolute limit of modern PC gaming.

[00:02:27]Okay, well, let's just crack on with our first news topic or rather our first question, but it is a news topic. Um, basically, it looks like uh at a Rocket League event, it was revealed that Rocket League is going to be the first game, or rather the first game revealed to be running on Unreal Engine 6, and we saw some trailer footage there showing considerable graphical upgrades, but that's possibly not really what Unreal Engine 6 is about. Uh, obviously, tons of uh questions here. We're gonna kick off with this one from um Alexander Zionov. Uh very very big apologies if I've got the pronunciation wrong there, which inevitably I probably have. Um hello DF directors. What's your thoughts regarding the announcement of Unreal Engine 6? In particular, why do you think it was announced so far in advance of a new console release? Uh Marcus asks, "Hi, I found you exclamation point with Rocket League now showing on Unreal Engine 6. What should we take EU6 to mean for traditional games?" From Tim Sweeny's comments, it sounds less like a simple nanite/ lumen style rendering leap and more like Unreal becoming a broader platform. U5 UFN verse Fortnite style development, Fortnite style deployment rather, standalone games and deeper engine architecture converging.

[00:03:37]The key gaming hook seems to be fixing Unreal's longstanding singlethreaded simulation limits. So, is UE6 really a nextG engine or Unreal's quote unquote forever platform moment? Better multi-core scaling, easier gameplay programming, larger simulations, and cleaner deployment, or less or less elegantly. What the f is Unreal Engine 6 actually supposed to be? Sorry, Rich.

[00:04:00]Uh, Stew Monster, greetings, notable experts of Foundries Digital. With Unreal Engine 6 teased via the new Rocket League demo, what are you most excited about technically? Do you expect iterations such as major multi-threading improvements to fix UE5 stuttering and CPU bottlenecks or more innovations such as advances in large scale simulation networking, nanite/ lumen or verse integration? Perhaps something novel, new and exciting they haven't done before. Um John, a very much a weird way to announce Unreal Engine 6, would you say? Um what do you think's going on?

[00:04:32]>> Um I mean to me it seems like they're evolving the engine to suit their specific service game needs a little bit more. Um, and just I guess I I don't yet understand why they decided to throw the six on there yet. That may occur in time. But it does seem like >> what we know thus far and we don't know that much is that this seems more like a scalability leap. Something designed to sort of like solve some of the long-standing issues with the engine while also making integrations with certain like multi-user experiences perhaps easier.

[00:05:08]Right.

[00:05:09]>> Um I mean there is there's definitely no showcase for new visual features, right?

[00:05:15]Like aluminum or nanite or that kind of stuff.

[00:05:18]>> And so my guess is that they want to make something that uh simply performs better while perhaps living up to the promise of what Unreal Engine 5 was all about, right? Because >> um while they have made progress in certain areas, there's definitely been limitations. Specifically, one of the big things is the multi-threaded simulation, right? Uh which there's still a lot of singlethreaded issues with Unreal Engine 5 games and that has caused problems in terms of performance.

[00:05:50]So, by fully uh parallel parallelizing uh the engine, >> thank you. The gameplay simulation, they can make better use of modern style CPUs, which focus obviously on more cores as opposed to faster single cores.

[00:06:07]uh which could help with some of the performance issues. Although looking at it, I mean when I look at what Unreal Engine 5 has achieved and I was discussing this with somebody re recently, Unreal Engine traditionally has been something that focuses on a future that doesn't quite arrive and developers essentially have to uh find ways to make up for it as Epic scrambles to get it more performant on the current consoles, right? Like Unreal Engine's origins back in the 90s was very PC centric, but then once we got to the second generation engine, that started to appear more in consoles and there was definitely a lot of performance limitations at the time. And but Unreal Engine 3 is where they really tried to push things out and that's when they sort of took over the industry, but on consoles especially, the performance was always absurdly poor. I would say 30 FPS was generally like the target, but even that was not that common. Um, and when you look at first generation software, especially like Gears of War, uh, Mass Effect 1, stuff like that, they run terribly on Xbox 360, right? Like really, really bad. Like that's that level of performance would never have been acceptable today. And image quality is not great either. Like they just don't hold up that well. And things did get better. But then again, it happened with Unreal Engine 4 where they were sort of like promising these like big changes to um you know, the lighting pipeline and all kinds of other things and most of it didn't really bear out and performance targets were still somewhat limited. Uh all of these engines, by the way, had major issues with stuttering, specifically on PC. Uh that was a that was a problem. And on the console side with Unreal Engine 4, most most games, not all, but most were either 30 FPS target where they successfully hit that much more uh reasonably or sort of like an unlocked experience with just a few upper end games or simple games targeting 60.

[00:08:05]Unreal Engine 5 actually mostly solve this on the console side. Performance is not perfect, but it's way better than any other Unreal Engine in history, right? But the introduction of the technologies such as you know lumen nanite VSSM obviously went too far for what the hardware was capable of. And even though developers were able to mostly reach the 60 fps promise they gave up so much in terms of image quality with low internal resolutions and just huge issues with temporal like effects, right? Like the ray gathering and all the other problems that occur like it just creates this noisy smeary look. So even though we got good performance, we gave up image quality.

[00:08:47]So reading this then if if we take what we know is like some of the big focus here, we could just be looking at a future where they're trying to make Unreal Engine 6 uh bring all the features that currently exist together, optimize things, and make something that's perhaps hopefully more performant on game consoles of the era while also delivering good image quality this time.

[00:09:11]Maybe we'll finally get there. I don't know. But that part is interesting. I'll let you guys talk more about the what is it like meta metaverse light connected ecosystem stuff. I have no interest in that stuff. Uh I'm not sure what what this is going to mean for the development side, but on the game side, I mean, I get it. That's what EP that's the types of stuff that Epic makes. And I think that's why something like Rocket League makes sense as an announcement similar to like a Fortnite, right? They own it. So it's like a playground for new features, >> right? Yeah. Mhm. Yeah, that kind of makes sense. Uh Will, what's your take?

[00:09:46]>> I mean, yeah, it feels like one of those uh operating system releases where they say like, "Yeah, we're just really working on polish. We really want to have a good experience." You know, there's no big standout feature that is going to, you know, move rendering forward in a novel way. But instead, it's more about just trying to get everything that you've already got, you know, running in a way that means the people can actually access it. Because I think the biggest issue so far this generation has been, you know, we've got all these amazing features in lumen and nanide and yet the only games that really kind of perform well and we're really happy with the image quality and the performance are the ones that kind of don't use them or use them in a very limited way. So, if there was just a new version of Unreal Engine 5, like 5.9 or whatever, that had all of these features and you could just enable them at 60 fps and it was just fine and chill because a lot of stuff was done in the background to make, you know, more modern CPUs uh be used optimally and stuff like that.

[00:10:47]And that would still be like a huge step forward, right? It would mean a lot of people would actually get to see the UE5 features used to their full effect. So, it seems to be more about that. Um, in terms of like metaverse stuff, I also don't really know much about the metaverse. I'm not a Fortnite stand that logs on every week to see what else is possible, but it seems like you'll just be able to say, you know, okay, instead of exporting for Windows or Mac or whatever, you can just export to Fortnite. And that is quite an interesting development option for people especially who are kind of getting into the industry and you know presumably Unreal uh sorry Epic really want to be able to capture those sort of first-time developers right and if you have a button that just says export to Fortnite and it just does that in a really simple and easy way then that means that you're getting people who are just have one really good game idea you're getting them into the Unreal ecosystem right away. So, I think, >> you know, it's going to be more about kind of shoring up kind of the meta stuff rather than, you know, kind of pushing the boat out. Or at least that's what we've seen so far, right? Like I was kind of expecting this announcement to have a ton of detail and be very advanced, but it is mostly, you know, like about a minute of people cheering, like logos appearing on a screen and then a few bits and pieces of a car. So, you know, maybe there is more coming.

[00:12:10]Maybe there is a huge um novel feature that we just haven't seen yet, but it does kind of feel like it's more about, you know, kind of leveraging this and making sure that it's like actually accessible in a way that UE5 never really panned out to be.

[00:12:25]>> Yeah, I think there's a lot to this and um I think how can you say it? The the whole concept of metaverses hasn't done particularly well. Obviously, Meta's own metaverse isn't on particularly uh firm footing at the moment, but then you look at something like Roblox, which is doing incredibly well. And then obviously, I think, you know, um Epic are eyeing something similar with Unreal. And um I think that's part of what this is all about. Uh I was just reading up on this um they've got some sort of social engine framework that they're they're looking to integrate. um uh a kind of how I'm just looking at this metaverse and interoperable creator economy. I mean, you know where this is all going, right? And I think this is probably a good direction for Epic itself to be uh moving in as long as it doesn't lose uh focus on the stuff that actually matters for gaming. And I don't think they will, you know, I think um moving on to the next generation of consoles. Um, I mean, you can kind of see what's happening with Unreal Engine 5 on PC now, where generally Unreal Engine 5, as long as you're not running on epic settings, >> as long as you're using optimized settings, the UEI5 experience on PC is actually a lot lot better in terms of stuff like image quality and performance than it is on the consoles at the moment. So, you know, in a s in a sense, the um uh the move to next generation should solve those problems and I dare say there will be graphical upgrades, right? Um I'm I'm dare say that at at some point Unreal Engine 6 will reveal some sort of um uh path tracing push for example which is kind of makes sense right >> um but at the same time I think you know looking at Rocket League um it's quite interesting right because the existing game still runs on Unreal Engine 3 so >> so moving to Unreal Engine 6 I think is quite interesting because it's got to be performant across a massive range of hardware, right? Um, so the optimizations that they're looking to push out to um make uh Unreal Engine 5 more performant, just kind of like naturally move into Unreal Engine 6. And they've decided, right, well, we can do this now with Rocket League hopefully. I haven't got too much more to say about it really. Um, I do think there probably will be some kind of um uh uh re-reveal or new feature. I mean this is just a tease once there is an actual next generation console reveal. Uh John >> I I want to ask you Rish if you think I I thought about a comparison right against all generations of Unreal.

[00:14:55]>> Uh Unreal Engine the original is like Windows 95 right it's like an attempt to sort of create this standardized technology. Unreal Engine 2 is like Windows 98 where it takes what the original had and builds on it successfully and it's pretty good and wellreceived. Unreal Engine 3 is like Windows 2000 where they made a big shift to a like new kind of architecture and like change the way they do lighting, but it still wasn't always that well suited or performant for a lot of tasks.

[00:15:26]And this is where I think Unreal Engine 4 could be considered Windows XP, which even though it has issues, it was pretty well-liked and did a lot. And um you know it was a very capable moment for the engine where it kind of became like the engine to use. But then we come to Windows Vista also known as Unreal Engine 5 where it tried to rewrite everything again and do things in a different way and it was slow and hated even though you know at a lot of things and now that gives us hope that Unreal Engine 6 could become Windows 7.

[00:16:01]>> Wow that's beautiful. That's yeah that is beautiful man.

[00:16:07]>> No I I have not uh smoked anything before this uh question and answer session.

[00:16:12]>> I just think you know ultimately if we go back you know the the whole thing about Unreal Engine 5 image quality and performance. Um I think um Epic were kind of expecting that they could probably roll out a 30fps engine for the most part and it would look great and it was kind of borne out by the demos and the various quality modes that we've seen. But you know basically consumers demanded 60 frames pers and um I still think Epic has been sort of um scrambling to deliver that on the engine side and where they haven't developers have had to make do and some have been better equipped to do it than others.

[00:16:46]But you know, >> you know, I think Rich, one of the things that uh if I recall, Unreal Engine 5's like lumen system was designed under the idea that consoles wouldn't have a sort of hardwarebased ray tracing features or wouldn't be well suited to it, right? And they kind of built like an alternative kind of side concept with their softwarebased ray tracing solution. So, it feels like they kind of were focusing in the wrong direction early on and had to kind of course correct a bit. And I do wonder if that had some impact on the way things turned out.

[00:17:19]>> Interesting. I do think we should take on board that this is just a tease though. It's just like Unreal Engine 6 is coming when they're actually going to make a further announcement. I mean, we got Unreal Unreal Fest coming soon. But from what I understand, that's just going to be concentrating more on the optimization side uh on in some respects that's happening with Unreal Engine right now. Uh which we'll talk about shortly. Um, Unreal Engine 6. I'm curious when Rocket League 6 is going to be coming out and the extent to which it can offer, you know, something new and exciting, I guess, over the Unreal Engine 3 original. Probably not difficult as to whether it's going to sort of be something that couldn't have been done on Unreal Engine 5, just cuz we'll have to wait and see.

[00:17:59]>> You're going to have to stop playing Biomutin on the Bio Richie channel and switch over to Rocket League Rich, >> I guess. So, >> uh, let's move on. Okay, sticking with Unreal Engine, perhaps inevitably, Darj asks, "Lads, uh, triple, no, double exclamation point. What do you think to the new medium lumen setting that has been added to Unreal Engine 5, which will quote unquote support handhelds that run at 60 on consoles? Is UE5 finally going to be delivered fully featured games on Switch 2 that aren't gutted beyond recognition? Is this the true beginning of the next forever cross period?"

[00:18:35]Okay. Um, yeah. I mean basically um when we're talking about lumen here um medium lumen as things stand you engage that on PC and it basically turns off uh lumen as as far as I understand um the features that make lumen lumen uh disappear and you go back to a drop back sort of situation. If you look at um what uh Blackmmith Wukong did in its um performance mode on Xbox, which was later brought to PlayStation, it kind of just sort of ruins the look of the game because it was authored with lumen in mind and then they offer up a performance mode that turns it off. Um so the concept of a medium lumen that is actually lumen while offering um performance increases and uh increasing the viability of lumen on stuff like Switch 2 and other handhelds. Obviously, this is a good thing, right? Um potent uh John I mean potentially they probably wouldn't need to put too much effort into the fallback solution at that point if it if it runs well on switch 2, right?

[00:19:34]>> Yeah, this definitely just seems like a way to make the cheaper version of the implementation actually look okay cuz previously going for that sort of lower setting as you noted essentially disabled all the benefits of lumen, right?

[00:19:48]>> And it really ruins the visuals. It also would have made it difficult for the art team to have to target these different like visual stylings, right? Because it does drastically alter the look of the game because it basically is missing most of the actual GI bounce. Uh so I guess the question is how how is this going to look in motion? Like what are what are the uh actual like downsides to it, >> right? And that's kind of what we need to see, I guess. But it makes sense to >> focus in on on supporting platforms like the Switch 2 because thus far has there any has there been much I mean there was the um >> I think Layers of Fear >> too that used it it was a it's a very slow you know kind of slow paced quiet game of course.

[00:20:36]>> Yeah.

[00:20:36]>> But most Unreal games have had to cut that.

[00:20:39]>> So I mean this seems like this will be a good thing for Switch too.

[00:20:45]>> Yep. Mhm. Um yeah, basically as I understand it, just uh looking up here, um medium lumen is basically a lighter weight version of lumen using irradiance fields and probe occlusion, >> reduced trace resolution, uh capped light bounces, proximity prioritization.

[00:21:01]The I'm just reading here, the engine dynamically prioritizes calculating lighting for objects closer to the player, aggressively culling or simplifying calculations for distant uh geometry. Uh the engine heavily downscales the resolution at which lighting traces are calculated strict and uh yeah capped light bounces strictly limits the number of indirect light bounces allowed in a scene. Um yeah just kind of makes sense, right? I guess as John says though um will proof of the pudding is basically in in the tasting or rather whether it's going to look decent or not. You'd kind of hope so.

[00:21:34]>> Yeah, absolutely. I mean I think you know the Switch 2 is fundamentally a much more powerful device than the Switch one. So, you know, we kind of need to reccalibrate our expectations for the platform. And, you know, the fact it has a modern CPU, it has DLSS, it means that it's already in a decent um place in if developers want to go after these more advanced effects. So, this seems like it'll be another useful toolkit uh part of the toolkit, but you know, we kind of need to see it in person to to see if the trade-offs that it's making are actually reasonable and worthwhile or whether they kind of sacrifice too much. Like it sounds very promising, but we'll have to see it in a shipping game and to to really judge it.

[00:22:13]>> Yeah. Cuz fundamentally, Will, like the the issue I think most people have with Unreal Engine graphics these days is just the fizzling and noise that you get when using Lumen at low resolutions that we often see on consoles, right? They they really need to ensure that that is solved >> or improved. Yeah.

[00:22:30]>> Going forward.

[00:22:32]>> Yeah. I think it's basically I mean I really want to take a look at to to look at this new medium lumen to see what the trades are because there's no such thing as a free lunch in rendering but at the same time we kind of need that scalability with Lumen u because the future is all about running as many games as possible on as many different systems as possible. And the the gulf now in hardware performance between something like a Steam Deck and a 5090 is like like astonishingly high. But we still need to be able to run the same games on this level of hardware, uh, on this range of hardware. Um, so yeah, looking forward to seeing how that all turns out in actual shipping games. Um, next question. This one comes from Jose and uh, he says, "Hello, Jets."

[00:23:18]Exclamation point. What Unreal Engine 5 game got closest to the Matrix demo?

[00:23:24]Um, he's he hasn't actually uh, posed this as a question. There's no question mark there as an exclamation point which pretty much sums it up. Um will I'm tempted to say none.

[00:23:37]>> Yeah, I think that's kind of the default answer, isn't it? Like nothing really has come close to that demo in terms of just sheer wow factor of like oh wow this is actually all running in real time and you know look at all these systems working in harmony and and so on. I would put forward um Stalker 2 as a game which is openw world, nanite, lumen. It's using, you know, most of the core text. It's reasonably performant.

[00:24:02]It works on consoles. On PC, it's incredible. I really enjoyed playing through that game and seeing how everything worked. That felt like a real kind of nextgen moment for me. But yeah, that demo is just was so good that it's kind of hard for any shipping game to be like, "Oh, yeah, we we can just equal that and layer on an actual working game on it as well."

[00:24:22]>> Mhm. Yeah, John. Um it's it's a bit unfair to say >> uh that no game has got close to it because I don't think any game has tried to. They've all basically got their own targets, right?

[00:24:34]>> Yeah. They kind of have their own visual target so it's not that comparable.

[00:24:37]Actually, the closest thing I would say in terms of like quality, even if the character models aren't quite there, is Cyberpunk 2077 with path tracing, but that's not Unreal. So, just uh but in in all seriousness, the one that I would put out there, and again, artistically, it's completely different.

[00:24:55]It's Hellblade 2. I actually think Hellblade 2 is probably the best like visual example we've seen just in terms of like utilizing the engine to create like near photorealistic visuals at times like just what they did with the characters, the environments, lighting.

[00:25:11]It's a great use of the engine.

[00:25:14]Absolutely stunning game. I might also put um >> when it's running on a powerful PC, I do think Blacksmith Wukong is quite impressive to behold, right? Uh, but that's one where I think none of the console versions really hold a candle to what you know that Matrix demo and certainly the PC version. They made too many weird sacrifices that don't really hold up. But yeah, Hellblade is probably the one I got to go with here. Even though I mean I guess part of that comes down to the fact that it is a very very very limited game in terms of scope. It is like an evolution of this kind of like >> uh almost it's like a modern FMV game.

[00:25:55]>> Oh, and one one other I would say that gets there on in one area specifically is Robocop Rogue City. Like character modelwise, it's not even close. It's it's awful in that regard. And I think that's just a matter of budget. But the actual scenes, the sets that they built, the lighting in those sets, I think is absolutely like amazing and is actually not far off from what the Matrix demo does. So, if you just look at the environments, I think Robocop actually does a pretty darn good job there. I think one thing we need to bear in mind also is that the Matrix Awakens demo was actually quite a poor performer. Uh, frame rates weren't particularly great.

[00:26:38]cut scenes were capped to 24 frames pers with letter boxing. It was all >> intended to produce this cinematic look, but as a sort of example of a viable game. I don't think it really is one. Um and uh yeah, you know, something worth bearing in mind there. It was kind of like just a sort of vision piece really as opposed to a game. And yeah, it's interesting that you point out stuff like um um Hellblade 2, John, which now runs at 60 frames per second on the base consoles and is looking pretty awesome.

[00:27:10]So yeah, I think if you're talking about sort of technological achievement with Unreal Engine 5, um it's different ways of looking at it, right?

[00:27:18]>> Mhm.

[00:27:18]>> Okay. Um let's move on. Uh got a question here from Molson. Hello DF exclamation point. recent retrospectives on the 1080 Ti, I couldn't help but think of another card that might join the quote unquote fine line fine wine collection in due time. Uh, the RTX 4090. Hear me out. Uh, if I told you that it's eye watering $1,600 price tag at launch, assum assuming you could get one at MSRP was good value, you would have called me crazy. But what if I told you that going on four years later, it would still be the second most powerful gaming GPU in existence and likely to remain that way for up to five total years as new release as new GPU releases see further delays. I kind I kind of see where he's coming from on this one, Will. Um because the 4090 is still a beast. I guess it is still the second most powerful GPU money could buy because, you know, it kind of sits in that uh position between the 5080 and the 5090.

[00:28:15]Um, fine wine though. Is that pushing it?

[00:28:20]>> Yeah, I guess maybe we need to wait for uh, you know, a few more years after after the next card, next generation of cards come out to to really see if it deserves the the name. But, I mean, yeah, it's a kind of unprecedented for the 4090 to still be pulling ahead of the 5080 in so many games. Like, I'm still using that as my primary GPU. I had to, you know, send it off to be repaired by a specialist service in Germany after it had the 12vt high power issue. And that wasn't cheap, but it's like still it's still really really holding up. I think you said in the initial review that it kind of solves rasterized rendering. Like you just any game you just boot it up and you know it's going to be fine, which is just such a cool experience. And yeah, it was ridiculously expensive. And if you got like one of the third party cards where you know it was maybe £500 or even more above the RRP, then you know that's not great. But if you got one of those Founders Editions ones that was actually available at RRP for a while and kept coming into stock, then I think you're doing pretty well. Like it's it's really held up. And you know, I I think it's going to be interesting to see if we get like a 60 series, like where does the 4090 fall into that? like is it still gonna be like the third best card because that wouldn't be too bad actually.

[00:29:48]>> Yeah, there's this there's this weird thing where basically I think Nvidia pushed hard on making massive GPU dies, right? So they did it for the 4090, they did it for the 5090 remarkably on the same process node. Um but AMD didn't follow suit. They didn't produce their own massive GPU, which basically means that with only one player in town, I guess Nvidia could feel comfortable in producing a 5080 that wasn't as fast as a 4090 and was basically an iterative step over the over the 4080 with um dynamic frame gen added essentially. Um yeah, it's kind of weird and fine wine, you kind of need to wait a few more years, I think, before we actually make that determination. But I think, you know, when you're talking about I don't know about that because we were talking about the 2080 Ti as fine wine, but it was again that was massively expensive at launch as well judged by the um the price points of the time.

[00:30:44]>> Um yeah, it's going to be interesting. I think going into the 60 series though, I wouldn't be surprised if um Nvidia is leaning more into um the machine learning side of it. Um, I think they kind of have to be in a situation where the more mainstream cards are able to do the higherend ML stuff they want to do.

[00:31:05]So, I'm wondering how that works out on an economics perspective, whether they actually cut back on rasterization.

[00:31:12]Um, interesting. I'm not quite sure what's going to happen there. One thing I did note though, Will, because there is this sort of lack of um a high-end die from um uh AMD, when I was doing all of the reviews for um RDNA4 and for for Blackwell, just seemed remarkable how many GPUs seem to be delivering um performance that was sort of like an RTX 5090, give or take. that seemed to be where we actually sort of topped out on those on those uh kind of mid uh mid to higherend cards and uh which in turn is kind of like not that much faster than an RTX 50 uh 3080 even. So there has been that kind of like um you know sort of like a um an invisible barrier almost um unless you're going to go big. Uh John at 4090 fine wine.

[00:32:04]>> I think >> I think in another five years we may be looking back at this and saying remember when they sold us GPUs and we could buy them for money rather than renting space on a data server that's been uh placed in our front yards that we don't actually own. And also the future where instead of playing video games ourselves, we just have an AI agent do it for us.

[00:32:26]Yes, those those are the days, you know, man. The thing is, there are crazy people out there that actually want that for some reason. It's like, you know, those those stories of Elon Musk hiring somebody to play Path of Exile for it.

[00:32:39]It's like, what's the point?

[00:32:40]>> I know. I I'm really worried about uh the future of this stuff because it really feels I'm there is this feeling that the PC market is in big trouble and companies are realizing that wait like this is peanuts compared to we could be investing in more AI and data centers >> and giving everybody what they clearly want and uh gaming is being deprior deprioritized.

[00:33:06]I hope that's not the case, but like Nvidia is at a weird point, right, where like they're the they're the only one doing high-end gaming GPUs right now.

[00:33:14]They're also they own most of the market share, but they've also shifted almost everything over to this like AI and data center business, right? like like I I just I I don't I don't know where things will go and I I just worry about it this like the future of of everything around this hobby just because of these mad men at the top that are just obsessed with more AI stuff like and I I don't know I don't know like what do you think do you do you think something's gonna break here and it's going to just like they'll realize that they were wrong?

[00:33:50]No, I I think that as long as there is a market for gaming, uh there will be um uh people making devices for gaming. I think that's the sort of bottom line.

[00:34:00]>> There's still a huge amount of money in it. The question is the makeup of those devices and whether it's kind of like de prior uh prioritized where I'd say there is, you know, that's there's genuine concern there. I still think, you know, the PC market is huge. Um Nvidia sold millions upon millions of Blackwell cards. I was looking at Steam hardware survey um a while back, you know, I was looking at that and and um I was surprised to see that the RTX 5070 was the number one Blackwell card um on the Steam hardware survey and it was something like 4% um of of the um of the of the marketplace. 4% of the Steam user base is a gigantic amount of GPUs that have been sold and that's just one card.

[00:34:46]So, I think, you know, it's not really a question of, you know, are we going to see um graphics cards deprioritized? I think that's probably likely to happen.

[00:34:54]We aren't going to be getting new GPUs this year. Um new architectures for sure. Maybe next year. Um but I still think the market is big enough to to to actually make it happen. I think that's that's the bottom line. And if they don't, well, I guess it's going to be happy days for AMD, right? Um or somebody else that will step in to do it. That's just the nature of the market.

[00:35:17]>> Chinese graphics cards, baby.

[00:35:20]>> The gold rush has got to come to an end at some point though, I'm pretty sure.

[00:35:24]>> I hope so.

[00:35:24]>> Um, yeah man, it's a crazy stuff though.

[00:35:28]Um, let's move on.

[00:35:32]Question here from uh Dots Finest 82.

[00:35:36]Hello Jens. Exclamation point. Something that keeps getting neglected when discussing Sony pulling their PC sports is the ports rather is the waning quality of each subsequent release after a rocky launch of Horizon Zero Dawn in 2020. It seems like Sony got things together with great ports of Spider-Man and Miles Morales as well as as well as God of War and Days Gone. But each port thereafter, Last of Us Part One and two Returnal and The Biggest Sinner, Spider-Man 2, have diminished greatly in quality. If part of Sony's reasoning for pulling back these ports is slowing sales, they're throwing stones in a glass house. Exploration point. To this day, Spider-Man 2 stutters like crazy on all but the beefiest setups with no updates since May 2025.

[00:36:19]Um, well, it's been quite interesting from my perspective that the reaction from PC gamers to to to Sony basically putting out has kind of like been almost like a a shrugging of the shoulders.

[00:36:31]Maybe this is one of the reasons. And maybe another reason is that the amount of uh first part uh party um single player narrative driven games has kind of like diminished quite drastically to the point where it's like well if I'm only missing a couple of games a year what's the big deal?

[00:36:47]>> Yeah. Yeah, for sure. I mean I think yeah like most of the games that are on that list like The Last of Us Part One and two and Returnal they're not terrible ports. Like we've seen much worse ports than these. Well, Spider-Man 2.

[00:37:03]>> Yeah.

[00:37:03]>> Yeah. Last part one wasn't great, but yeah, they put in the time. Um, part two was okay, but you know, Alex got quite a lot of flack for pointing out, hey, this was running on a base Play PlayStation 4 and uh it's running on this PC and it kind of should be running better.

[00:37:19]>> Yeah, absolutely. I mean, we recently um covered this on the website and as part of that, we had a poll where we kind of looked at what people's sentiment was about, you know, the news. is like, "Do you agree with Sony pulling back from from PC in this way?" And I think we had 29% were outraged of some description.

[00:37:38]Uh 32% didn't like it, but they kind of understood the reasons for it. 19% were in favor of um Sony withdrawing, and 21% didn't care. So, it's clear it's quite a split, you know, reaction, I guess, based on whether or not you were one of those people that was kind of lining up to buy these day one on PC. And yeah, I mean obviously narrative single player stuff isn't in the renaissance that it has been, you know, a few years ago perhaps, but it's still like a massive publisher which was doing relatively good work and especially all the Nexus ports were really, you know, technologically really stellar. So, it's a huge loss for for PC in that way. And, you know, it's, you know, strategically, I totally get it. you know, Project Helix is coming out and you don't want to be making sure that your uh PlayStation games run really well on that. And there are lots of other reasons as well, but um yeah, it's uh it's obviously uh yeah, a mixed a mixed bag.

[00:38:39]>> What do you make of this, John?

[00:38:41]>> Uh I don't think that the slowing sales determined why they stepped away from PC or want to step away from it. I really think it's more to do with what we discussed last time regarding the market changing and shifting more towards a model where the latest Xbox hardware could conceivably just play their games, right? I don't think I don't think they want that. But it it is it is odd though that things kind of ended the way they did. And yeah, Spider-Man 2 situation, that's uh that's pretty disappointing from from my perspective given that the console version was so optimized. It ran really well on PlayStation 5. And so you would hope that the PC version would receive the kind of care it deserves, but alas, I actually haven't revisited it lately, but hearing that it hasn't been updated since May of last year, yikes. I mean, it kind of sounds like they probably made that decision internally probably around that time and it's only now that we're kind of hearing about it and it's being kind of publicly or at least, you know, in Sonywide meetings >> being if you're selling a product, you should at least support the product, right? And it just feels like it's been abandoned, >> which kind of sucks. And it's especially weird because it's it's Marvel Spider-Man, right? It's not just a Sony property here. So, you don't really want I mean, I can't imagine Marvel would love having a nonoptimal version of something out there, but then again, you look at some of the older stuff that had the Marvel name, and maybe it doesn't matter, but uh >> yeah, fair.

[00:40:15]>> Yeah, I don't know. It's it's it's a shame that the the PC legacy of Sony will be an interesting one to look back on and think about.

[00:40:24]>> Mhm. Yeah. Interesting, I think, is the right word. I'm not really sure what I can really say about this further except that Spider-Man 2, I remember when it first came out, you know, we did a segment on it in DFX um to provide a kind of snapshot of how it was running and how it was running was not particularly good. The aim was that, you know, a couple of weeks later we were expecting the key issues to be resolved and then we could probably do um a full review, you know, say, okay, you know, we've gone past the issues of the launch. here's Spider-Man 2 and you know optimistically we would have said hey this is a really good port but you know it never felt as though those optimizations came those improvements came it always felt sort of a bit strange and weird um not great not great at all and I think Alex sort of um put together like a list of um Nixis issues that he had particularly in terms of like VRAM management that he was hoping to see improved over time and I don't think those improvements ever came um we did get a pretty port of Death Stranding 2, but again there are issues with that one at launch which did um ultimately see improvements. Uh the extent of that though I'm wondering whether you know there is much more oversight if you like from Kima Productions do care very very much about the quality of the games >> and um even after we put our review out which was quite late where we thought a lot of the issues were resolved updates still kept coming >> um to address further consumer feedback.

[00:41:52]Um, yeah, I think but I do think that is probably it at this point. It's probably the last we're going to see of Sod stuff and I think that's a bit of a shame, but I kind of see where uh the posters the the question asker is coming from this because yeah, initially, you know, things there was there were issues with Spider-Man at launch, but they did tend to be mitigated, improved, and it turned into a pretty decent port. I had a great time playing it on Steam Deck, for example. You know, that's that's pretty awesome.

[00:42:17]>> And there were games that out the gate actually seemed to be pretty good. you know, Ghost of Yote, no, um, Ghost of Tsushima, rather. Ghost of Yote is never coming to PC at this point.

[00:42:26]>> Yeah.

[00:42:27]>> Um, but, um, yeah, the quality did did reduce and, uh, yeah, to the point where Returnal actually looking at that, it's, you know, I've never man actually managed to get a decent experience on a handheld because the stuttering was so bad.

[00:42:40]>> Less of an issue on desktop, but handheld, not not >> not good. I'm just disappointed we never get Gran Turismo 7 cuz I love GT7.

[00:42:47]Gorgeous on console already, but The PC could have supported ultra wide resolutions, even further visual refinements, better support for like a wider range of steering wheels and like other custom configurations. Like it really could have been something special.

[00:43:04]>> But alas, alas, >> and it GT7 was actually the uh game that disproved that finally brought down Jensen's Prophecy.

[00:43:13]>> That's right.

[00:43:14]>> The great the great leak that was saying all of these Sony games were coming to PC. I bet it was planned at some point though. That's my guess is that they were working on GT7 and then somebody made the call to not bring it over.

[00:43:26]>> Mhm. Yeah. Okay. Well, let's move on to our final question. This one comes from Lexington 87. Hi there, Crimson Gods exclamation point. I don't understand that reference.

[00:43:36]>> Crimson Desert. Anyway, >> something something maybe, >> possibly. I don't know. Um, as a society, we're generally against monopolies, right? Competition is key for a well functioning consumer market.

[00:43:48]I'm pretty sure we all agreed on that.

[00:43:51]So why do why do we all, me included, love Steam so much? Why do we want to marry Gabe? I'm not sure I go that far.

[00:43:59]Why do we want to marry Gabe and have feelings for Valve the corporation?

[00:44:05]Valve is a PC gaming monopoly. If you're on Steam, you're in trouble. Not only that, we all want them to be more of a of a monopoly. No other publisher apps on my PC, please.

[00:44:17]Um, well, he's got a point, right? I think it's basically the the the situation is that Valve have actually done a great job of delivering a great service.

[00:44:30]>> Yeah, >> I think that's the bottom line.

[00:44:31]>> Yeah.

[00:44:32]>> They're kind of like the closest we're going to get to a benevolent dictatorship.

[00:44:37]>> Yeah. We're all living in fear for the for the day when Gabe decides to step down. But hopefully he's done enough to ensure continuity. That's the main thing. But yeah, I mean, you know, ultimately we put up with it because it's just so much better than any alternative. It was kind of the first, you know, to really tackle digital releases on PC and it was rough going early on, but they turned the ship around really fast and, you know, they've just made a product that people actually like. Like if you >> go on, you know, the Epic Games Store, you get so many free games. It's incredible. But also, people don't hang out there. Like there's no community aspect to it. Like nobody cares if you have achievements there. like Steam have just, you know, Valve have just made Steam like an actual place, a community, has all the features you want and it also just doesn't break in weird and interesting ways, which has been my experience with pretty much every other launcher or digital storefront. Like, you know, I keep banging on about this, but I'm hoping eventually one day it'll change. Like, the Xbox app is made by the people who make Windows. Like, why can't those two things like work properly together? You would think that that would be the easiest thing in the world is that you have like vertical integration and everything is amazing because you know you can line up what different teams are doing to work together better. But yeah, at the end of the day, Steam is just amazing and therefore people are willing to put up with, you know, this monopoly that we probably wouldn't do otherwise. Another small thing is that, you know, Valve as a company seems approachable. Like you can email Gabe and get a response. you know, if you reach out to their PR people, like they have real answers to your questions. Like they have a lot of I don't know, a lot of humanity about them. They're not faceless and corporate. So, I think that makes a huge difference.

[00:46:33]>> Mhm. John.

[00:46:34]>> Yeah. So, I think I agree with what you've been saying there, Will. And even as somebody I was initially against Steam, I didn't like the concept and I didn't want things to go down this digital rabbit hole. But in the end, they've proven themselves as the most capable service. Uh, but I think a lot of this has to do, if if I may, I feel like it may be connected to the fact that Valve remains a private company as opposed to a company traded on the stock market.

[00:47:03]>> Because as we know, you know, these publicly traded companies that are driven by investors to continuously increase the line. Things must always line must go up. You must always be exceeding what you've done in the past.

[00:47:18]And this this constant growth, this striving for infinite growth is a cancer on so many companies. And these investors don't actually care about the future of the company because they'll just trade it away when it's trash, right? They will leave wanting destruction in their wake on the way towards a profit. Uh and that has not been good for many companies within the industry. And by Valve remaining private, it has allowed them to likely make decisions that kind of benefit both the company and the user base, right?

[00:47:50]Like they have canceled projects that were deep in development before. You know, they thrown away good work when it's not working for them and basically picking and choosing what they do in a very different way than a lot of publicly traded companies might, which I think is kind of interesting. And it actually does kind of play out in terms of what they do release, right? Uh it also gives them like a ton of flexibility on like timings and such.

[00:48:17]Like I think if they had these plans for the Steam machine being blown up by RAM increases, it may have gone differently versus, you know, if it was a public company versus a private company in terms of how they choose to re-reveal it, what the price is going to be, you know, how they're going to handle all this logistics. they can just be very agile with it. So, I think that has paid off. And of course, like you said, Will, they do feel slightly more uh accessible. That's not to say there aren't issues with certain aspects of Steam and all that, but the overall experience, I think, is largely pretty good. Like, if you're going to do a digital storefront, that's the way to do it. And I my my my library on Steam goes back to 2003 when they first launched it, you know, so it's still there. It's easy to to trust that they will keep things running. We've already lost many other digital storefronts over the years that have just closed up shop and everything's gone, right? If not for piracy, we would have been in trouble.

[00:49:19]Uh and like you say, you know, even platform holders haven't been able to get this right on the PC. Microsoft has has they've done so many things since Steam opened up starting with like games for Windows Live stuff and then moving into the you know the modern era now and they they just can't seem to get it right. there's always issues with it that are deeply frustrating for the user, which just shouldn't be the case.

[00:49:43]And right, >> all those other launchers, I think the reason it annoys people is one, the launchers usually aren't as good as Steam, but two, uh, they're just for that publisher specific games, which means your library is usually inherently limited on that platform. Or if there are other companies on there, it's also a much more limited selection. The only one I think that maybe has gotten some traction as like a a seller of digital games is Good Old Games because they do specifically specialize uh in a a subset of game releases that nobody else is really doing and their whole like uh nonDRM focus I think has been positive as well.

[00:50:23]So, I mean, their launcher situation's not not on Steam's level, but still, I think they're genuinely wellliked and they're a good reseller of digital products, but the rest, man, I mean, Epic Games Store, like they just it it doesn't feel like it's really evolved at all over the last few years. You know what I mean?

[00:50:44]Like, you still get in there, it's still missing a lot of base features that you would want. It's just not that great to use. and everything else. I mean, a lot of companies have kind of come crawling back to Steam, haven't they?

[00:50:56]>> Right. Yeah.

[00:50:58]>> So, >> yeah, pretty much. Including Microsoft, of course. EA, >> but it is kind of just like, I guess, you know, Windows is the most widely used operating system despite its faults. It's not actually a monopoly, but it it could it feels that way to a degree. You know, Mac OS is out there, but the install base is limited by Apple hardware specifically. And then Linux is gaining traction, but it's still doesn't have the wide mass market appeal yet, right? So Valve is basically just like the Windows of the gaming UI space in a sense, except for they're a lot better than that, but you know, >> I think another thing to bear in mind is that um with Microsoft that, you know, this is like a gigantic company with with Strata, right? you know, there's like um you know, the bigger the company, the more difficult it is to manage, and I think the more difficult it is to get things done. Whereas Valve has managed to retain the fact that it's a small company with a limited amount of um of resources, but still gets stuff done. Do you know what I mean? I think you're right. There's also um that sort of mentality that it is a private company, that it is somehow, you know, on the side of the gamers or whatever >> um delivering. I think that's the key point they're delivering. And you're right, the the launcher itself is just kind of like it's kind of like become the standard almost. Nobody has actually managed to come up with something that's quantifiably better. Certainly not Epic.

[00:52:25]Um where you know I launch a game and I I never seem to get any sort of immediate feedback that the game's launching or not, which is right.

[00:52:34]Exactly.

[00:52:34]>> Crazy thing. And then you look at the Xbox game store, which, you know, there have been sort of improvements there.

[00:52:39]But, you know, the the how can a launcher be slow? I don't get it, you know, on a on a 14900 K or a 9800 X3D.

[00:52:49]Why why are we waiting? I think is is the bottom line. It's a lot of crazy stuff that's happening there. How how difficult is it to run a game? And, you know, I guess there's the other thing, of course, where um just um the legacy, you know, Steam were there first practically, right? So, um, when you start to build up a library that can run into the hundreds or even thousands of games, why would you want to have another launcher to have another bunch of games? You kind of that, you know, that the collecting mentality is that everything is sort of easily accessible with the same um with the with the same launcher, right? That's that's kind of >> makes sense, doesn't it?

[00:53:29]>> Yeah. I'm not sure I want to marry Gabe, though. Uh, I don't particularly have feelings for Valve, the corporation.

[00:53:38]>> No, >> they're a corporation ultimately, but they're doing a lot of things right.

[00:53:43]And, um, you know, the fact that they're actually launching their own hardware.

[00:53:46]The fact that they went there with Steam Deck, I think is is quite remarkable.

[00:53:50]Um, they're doing good stuff.

[00:53:53]Okay, that was the final question therefore the end of the show. Please do like, subscribe, share if you enjoyed it, ring bells for whatever. And uh DF supporter program um patreon.com/digitalfoundry.

[00:54:04]Um check us out there. Early access to uh DF Direct Weekly, early access to uh videos, highquality video downloads of everything we do. Adree video downloads, I should add, ad free website. Lots going on there. Please support the team.

[00:54:18]Um but that's all from us on this one.

[00:54:20]And I guess we'll see you in the next TF Direct.