Well, now you've almost done a 180. Your initial statement which I took issue with was that local caching was not needed when shaders are distributed via network. That would indeed mean that every time you exit a game, all related shaders would be deleted instantly only to be redownloaded upon the next game boot, which is pretty irresponsible from an environmental point of view as I pointed out before, but also very annoying for most users in my opinion. So yes, according to your earlier statements, changing settings and rebooting the game would mean another multi-GB shader download, which I find horrible.
Now you're saying local caching is fine, but it shouldn't be that large of a local cache, or it should be configurable. And with that very last part (configurable shader cache size) I actually completely agree.
And regarding your 3 seconds remarks:
First of all, I had a math error as even your mighty 1Gbit/s connection would need at the very least 24s for 3GB of data (realistically, it would probably take around 30s or longer if integrity checks are done after downloading). But even if the 3s were accurate, game load times on PC are often way too long anyway for my taste, so I definitely wouldn't want to add another delay to that at all. And why would I? It would be completely unnecessary with a local cache except if there was a game or driver update in the meantime. Oh, and since you asked: in those saved 3s I would play the game instead of watching a loading screen. Obvious, isn't it?
Lastly, don't forget that not everyone has such a fast internet connection and for most people, those times would be a lot longer, which adds up quickly.
"Heck, I'd be fine with 30 seconds even if it was a game that had a 10+ minute shader burn that needed to be run."
But that isn't the question at all?! This whole article is about Advanced Shader Delivery, and our entire discussion revolves around whether or not a local cache is useful or needed for this approach or not. Locally burning shaders for 10mins is not part of the options here, rather it's:
1. download shaders and cache them locally or
2. download shaders without caching
I think that only number 1 is a viable option, while you proposed option 2.
And now, in your latest post, you finally agree on option 1, so I think we're in agreement, right?
I'm completely with you in terms of the need for a better solution than what we have right now. But that better solution in my view is Advanced Shader Delivery in combination with a local cache.
Let's pretend for a moment that downloading 3GB of shader data would indeed only take 3 seconds with your internet connection (it would most probably be longer). That's still 3s delay to EVERY SINGLE game launch. Why would you want that? What do you gain except to save a little local storage space?
What about the times where I'm having trouble with a game's settings and frequently change them and restart the game? There are days where I restart the same game ten times. That would be 30s and as many GB wasted.
If a new graphics driver comes out once a month (as an example, but that feels realistic) and you play your current favorite game every day, that's again a complete waste of at least as much time and data.
Also, as I mentioned earlier, it would be irresponsibly wasteful as downloads are far from free. Running entire server farms all over the world comes at a cost. Studies have shown that if you watch a movie more than once, buying the disk is better for the planet than streaming as the high energy consumption of streaming exceeds the combined costs for pressing, shipping and storing the physical version. For music albums, the turnover was seven times if I recall correctly. So if you listen to an album more than seven times, it's better to buy the CD.
Again, in all other respects I'm completely with you. But that single detail of either having or not having a local cache makes a world of difference in my opinion.
@Darren1967 We actually agree on most points, but the two where I strongly disagree are:
1. While it is true that you need to be online to purchase and install basically all PC games these days, it's simply wrong that you need to be online to play "most of them" as you put it. Games like Halo Campaign Evolved, which actually enforce an always-online system for a single-player game are the exception to the rule. Of course you need to be online for multiplayer games, but by far most singleplayer games are perfectly playable offline.
If that wasn't the case, then the entire PC handheld category would basically be gone by now. Also, I didn't mean that a PC gamer takes his gaming PC offline while he is at home, but rather something unexpected like an outage or some trouble when moving from one place to another. If you then cannot even play singleplayer games on your PC without a stutter fest, that would be pretty sad.
Oh, and PC handhelds are actually the perfect example why deleting all shaders upon exiting a game is the worst idea in that regard.
2. The other point I strongly disagree with is the lack of a local cache in general. If we take the example from The Last of Us Part 1 with its 2.9GB of shaders, why would I delay each and every game boot and re-download all those shaders every single time? Would you like to add something like 2 minutes of time to each and every game boot? I know I wouldn't. And even if not all games have that much shader data, it's completely unnecessary to do this every time.
Also, while you would save on storage space, you would add a lot of wear to your SSDs in the form of write cycles. Probably not catastrophic, but definitely noteworthy.
As I said, I'm completely in support of advanced shader delivery, and you are absolutely right that it is a lot more efficient and environmentally friendly to have servers burn the shaders that tens of thousands of gamers can then use. But I also insist that a local cache is crucial for this concept for the same reasons.
I would even go so far as to propose to not automatically update the shaders for every installed game after a driver update, as many games are probably not being played until after the next driver update and those downloads would therefore have been another useless waste of resources. So I would love to either have some control over the process or to see at least some smart algorithms in place.
Anyway, a 2-minute download beats the hell out of a 20-minute shader burn, but only if it's not done on every game boot.
@Darren1967 Maybe I misunderstood you, but you wrote this: "but it also does away with the need to store them in cache since they can just be redownloaded each time you start the game"
I understand this in a way that the moment I start a game, shaders are downloaded. The moment I close the game, those shaders are deleted. This would be extremely wasteful and inconvenient, as I argued in my last post (and by the way, there is a difference between digital games and always-online games).
If you meant something different, then please explain. I'm totally on board with advanced shader cache and the likes, just to be clear. I think that's awesome, but the shaders should still be cached on the local machine in my opinion and only re-downloaded if they are invalid due to a game or driver update.
@Darren1967 I sure hope your "non-cached shader cache" dystopia never comes to fruition. If for any reason you have no internet access, each and every game on your machine becomes a stutter fest.
Not to mention the useless waste of resources for downloading the same data on a daily basis, which would cost a surprisingly high amount of energy as there are whole server farms involved on a global scale.
Personally, if a game has a lengthy shader compilation step, I simply don't update graphics drivers until I'm done with it.
I do understand the problem, but your proposed solution would actually be worse.
@JetjoO Thanks for the tip with Uncharted 4, a game I know pretty well and absolutely adore. I've given DLSS 5 another shot, this time on my 77-inch OLED (though admittedly only via YouTube). And I have to say, I seem to get where you're coming from. I still think your statement is exaggerated (as of now), but the potential is definitely huge.
I don't like that every character seems to age at least a decade in U4, it seems DLSS 5 exaggerates all wrinkles and impurities in the skin. Also, the more games I see, the more they look the same, like the same filter being applied to them. A bit like Unreal engine games. But still, overall very impressive at times I have to admit.
I "only" have a 4090 and value high-fps experiences on a 4K screen, so I don't have the GPU horsepower to spare for DLSS 5 right now. But I hope you're right in that those technical restraints will be solved or alleviated sooner rather than later.
As for VR losing it's charm: I'm only about two years into it, so not nearly where you are. However, I'm also long past the initial honeymoon phase. For example, yesterday I played GT7 for the first time since February. The difference (so far) between me and you is, however, that I never even considered playing the game on my TV, it never crossed my mind. When I don't want the hassle of setting everything up (in the case of GT7 that includes a wheel/pedal setup on a wheel stand), I simply play something else.
Again, I can't vouch for the next ten years, but so far this is how I feel and it doesn't seem to change over time.
Anyway, thanks for the discussion, let's have fun gaming either way
@JetjoO My enjoyment of the medium doesn't depend on sales numbers, so I don't care that VR hasn't taken the world by storm.
I also never expected it to replace flat gaming, at the very least due to comfort and the impossibility of social interaction during a VR session. It is definitely something very specific and was always destined for a niche rather than mainstream. The issues around motion sickness in a lot of people don't help either.
However, when you can stomach it, it's like nothing else. I could never go back to playing e.g. realistic racing games without VR. I've been an avid Gran Turismo fan all my adult life, but if GT8 doesn't support VR, I'm out. The experience is on a whole other level, playing it flat isn't fun anymore, it feels like child's play.
The same happened to my beloved Ace Combat 7 once I tried the VR mods, I just cannot go back to the flat version anymore and will only buy AC8 once a VR mod becomes available. The intensity, immersion and emotions have to be experienced to be believed.
I will never forget how I frantically laughed during my first VR flight in AC7, the joy burst out of me and I kept grinning from ear to ear for an hour straight.
DLSS5 is a nice layer of polish, but it's not even in the same league as a good VR experience. And nothing I've seen so far justifies your "the greatest jump since 3D" phrasing. The most impressive shot of DLSS5 for me was the controversial Grace scene from RE9, but everything that's available now looks a lot more dialed down and muted, very minor changes comparatively. What impresses you the most currently? Which game is the perfect DLSS5 demo in your opinion?
@JetjoO "This is a genuine revolution for gaming, on a scale we haven’t seen since the arrival of 3D."
Have you never experienced VR? Honestly, making graphics look a bit more photorealistic absolutely pales in comparison to actually entering your favorite game worlds instead of watching from the outside.
DLSS5 is evolution, VR is the true revolution since 3D in my opinion, and it's not even close.
@Godzookie I've read about a study that came to the conclusion that streaming a movie twice is worse for our planet than producing, shipping and using a Disk to watch it twice. For music albums, the turnover point was around seven times listening if I recall correctly. So if you stream the same album more than seven times, a CD would be better for our planet.
They haven't looked at games specifically in that study, but due to the similar data sizes they should be comparable to movies. We tend to underestimate the enormous amounts of energy that is needed for running servers all over the world to stream Gigabytes of data across the planet.
So even from this perspective, physical Disks are the better solution and their doom actually a loss to the planet.
@MemuAccount I‘m not sure I understand your first sentence, maybe because English is a foreign language for me. Could you please rephrase or explain what you're implying there? Is it that DF is disrespectful towards their audience and therefore you mirror that sentiment?
As for your main point of why discussions about certain technical features only gain traction once they get implemented into consoles: I think that's due to the history of DF, because when I first came into contact with them during the PS360 era, there was no PC coverage at all if memory serves. They were rooted in consoles from the get go, and if I recall correctly they hired Alex specifically to be able to dive more into PC topics as interest rose over the years.
Also, I‘m pretty sure console gamers are still the vast majority of DF's audience, but I can't be sure of that. But assuming that's true, it makes absolute sense to tailor their content for their main audience and e.g. SSD load times analyses would have been pretty pointless back in the day when they only covered consoles and those consoles didn't support SSDs.
As for the 720p conflict, I assume that's a combination of differences in image quality and somewhat of a downregulation of standards, but I'm just guessing here. For the first, I guess that early Series S games at 720p didn't have sophisticated upscaling solutions, so 720p back then maybe looked significantly worse than games on the main consoles in more recent years. And for the second, I guess at some point we all realized that the consoles achieve native 1440p or higher increasingly rarely and that the reality will be lower resolutions going forward, so it necessarily became more accepted.
Put differently, we stopped making fun of 720p native resolutions because they became more common and the ridicule would get dull at some point. And also because actual results were not as dreadful as they had been even a few years prior due to better upscaling.
Don't get me wrong, I agree with you on some points, but overall I think you overshoot the target with your criticism and especially tone.
My biggest personal gripe with DF lately is that they are not critical enough in my opinion. Everything is great according to them, as if they are afraid to step on the toes of some publishers.
For example, Forza Horizon 6 without RT doesn't look significantly better than its predecessor apart from a few sharper textures in my opinion, but it is significantly heavier on resources. And yet, DF praised the great optimization.
An even worse offender is the latest Bond game. DF praised its scalability and called it well optimized if I recall correctly, but performance on even a 5090 is so poor that I don't see the headroom for the planned path tracing option. All the while it doesn't look particularly spectacular or does anything extraordinary technically.
@MemuAccount As much as I dislike your disrespectful tone, you got a point here. While I loved the way John represented the game and its history in the video, there was shockingly little actual technical analysis in it. A rough resolution estimation and that was it in terms of image quality analysis, very poor by usual DF standards.
There was one level where shimmering bothered me (I think sector X), but the rest of the game looks too good to have no AA. It‘s a bit soft, yes, but I‘m pretty sure they have some form of image treatment in there (that maybe even contributes to the softness). I‘d love to know what they do exactly, and I usually get this kind of info from a DF analysis, but not this time sadly.
The only way for a 60fps mode would be frame generation in my opinion. No way they get to 60 native, but FG from a 30fps base could be feasible at least on the Pro. If input lag doesn't get too bad (and GTA is not a twitchingly-fast-paced game), I would consider playing it this way.
If it turns out to be 30fps only, I'll wait for the PC version, but if it‘s 40fps or 60fps I might play it close to launch on Pro.
My gut tells me that going with an LCD panel when targeting frame times of 1ms is stupid, this would only ever make sense with an OLED panel. Is LCD capable of reaction times below 1ms? Never heard of it.
Comments 13
Re: Advanced Shader Delivery Adds Nvidia GPU Support Later This Month To Banish Shader Compilation Stutter
Well, now you've almost done a 180. Your initial statement which I took issue with was that local caching was not needed when shaders are distributed via network. That would indeed mean that every time you exit a game, all related shaders would be deleted instantly only to be redownloaded upon the next game boot, which is pretty irresponsible from an environmental point of view as I pointed out before, but also very annoying for most users in my opinion. So yes, according to your earlier statements, changing settings and rebooting the game would mean another multi-GB shader download, which I find horrible.
Now you're saying local caching is fine, but it shouldn't be that large of a local cache, or it should be configurable. And with that very last part (configurable shader cache size) I actually completely agree.
And regarding your 3 seconds remarks:
First of all, I had a math error as even your mighty 1Gbit/s connection would need at the very least 24s for 3GB of data (realistically, it would probably take around 30s or longer if integrity checks are done after downloading). But even if the 3s were accurate, game load times on PC are often way too long anyway for my taste, so I definitely wouldn't want to add another delay to that at all. And why would I? It would be completely unnecessary with a local cache except if there was a game or driver update in the meantime. Oh, and since you asked: in those saved 3s I would play the game instead of watching a loading screen. Obvious, isn't it?
Lastly, don't forget that not everyone has such a fast internet connection and for most people, those times would be a lot longer, which adds up quickly.
"Heck, I'd be fine with 30 seconds even if it was a game that had a 10+ minute shader burn that needed to be run."
But that isn't the question at all?! This whole article is about Advanced Shader Delivery, and our entire discussion revolves around whether or not a local cache is useful or needed for this approach or not. Locally burning shaders for 10mins is not part of the options here, rather it's:
1. download shaders and cache them locally or
2. download shaders without caching
I think that only number 1 is a viable option, while you proposed option 2.
And now, in your latest post, you finally agree on option 1, so I think we're in agreement, right?
Re: Advanced Shader Delivery Adds Nvidia GPU Support Later This Month To Banish Shader Compilation Stutter
I'm completely with you in terms of the need for a better solution than what we have right now. But that better solution in my view is Advanced Shader Delivery in combination with a local cache.
Let's pretend for a moment that downloading 3GB of shader data would indeed only take 3 seconds with your internet connection (it would most probably be longer). That's still 3s delay to EVERY SINGLE game launch. Why would you want that? What do you gain except to save a little local storage space?
What about the times where I'm having trouble with a game's settings and frequently change them and restart the game? There are days where I restart the same game ten times. That would be 30s and as many GB wasted.
If a new graphics driver comes out once a month (as an example, but that feels realistic) and you play your current favorite game every day, that's again a complete waste of at least as much time and data.
Also, as I mentioned earlier, it would be irresponsibly wasteful as downloads are far from free. Running entire server farms all over the world comes at a cost. Studies have shown that if you watch a movie more than once, buying the disk is better for the planet than streaming as the high energy consumption of streaming exceeds the combined costs for pressing, shipping and storing the physical version. For music albums, the turnover was seven times if I recall correctly. So if you listen to an album more than seven times, it's better to buy the CD.
Again, in all other respects I'm completely with you. But that single detail of either having or not having a local cache makes a world of difference in my opinion.
Re: Advanced Shader Delivery Adds Nvidia GPU Support Later This Month To Banish Shader Compilation Stutter
@Darren1967
We actually agree on most points, but the two where I strongly disagree are:
1. While it is true that you need to be online to purchase and install basically all PC games these days, it's simply wrong that you need to be online to play "most of them" as you put it. Games like Halo Campaign Evolved, which actually enforce an always-online system for a single-player game are the exception to the rule. Of course you need to be online for multiplayer games, but by far most singleplayer games are perfectly playable offline.
If that wasn't the case, then the entire PC handheld category would basically be gone by now.
Also, I didn't mean that a PC gamer takes his gaming PC offline while he is at home, but rather something unexpected like an outage or some trouble when moving from one place to another. If you then cannot even play singleplayer games on your PC without a stutter fest, that would be pretty sad.
Oh, and PC handhelds are actually the perfect example why deleting all shaders upon exiting a game is the worst idea in that regard.
2. The other point I strongly disagree with is the lack of a local cache in general. If we take the example from The Last of Us Part 1 with its 2.9GB of shaders, why would I delay each and every game boot and re-download all those shaders every single time? Would you like to add something like 2 minutes of time to each and every game boot? I know I wouldn't. And even if not all games have that much shader data, it's completely unnecessary to do this every time.
Also, while you would save on storage space, you would add a lot of wear to your SSDs in the form of write cycles. Probably not catastrophic, but definitely noteworthy.
As I said, I'm completely in support of advanced shader delivery, and you are absolutely right that it is a lot more efficient and environmentally friendly to have servers burn the shaders that tens of thousands of gamers can then use. But I also insist that a local cache is crucial for this concept for the same reasons.
I would even go so far as to propose to not automatically update the shaders for every installed game after a driver update, as many games are probably not being played until after the next driver update and those downloads would therefore have been another useless waste of resources. So I would love to either have some control over the process or to see at least some smart algorithms in place.
Anyway, a 2-minute download beats the hell out of a 20-minute shader burn, but only if it's not done on every game boot.
Re: Advanced Shader Delivery Adds Nvidia GPU Support Later This Month To Banish Shader Compilation Stutter
@Darren1967
Maybe I misunderstood you, but you wrote this:
"but it also does away with the need to store them in cache since they can just be redownloaded each time you start the game"
I understand this in a way that the moment I start a game, shaders are downloaded. The moment I close the game, those shaders are deleted. This would be extremely wasteful and inconvenient, as I argued in my last post (and by the way, there is a difference between digital games and always-online games).
If you meant something different, then please explain.
I'm totally on board with advanced shader cache and the likes, just to be clear. I think that's awesome, but the shaders should still be cached on the local machine in my opinion and only re-downloaded if they are invalid due to a game or driver update.
Re: Advanced Shader Delivery Adds Nvidia GPU Support Later This Month To Banish Shader Compilation Stutter
@Darren1967
I sure hope your "non-cached shader cache" dystopia never comes to fruition. If for any reason you have no internet access, each and every game on your machine becomes a stutter fest.
Not to mention the useless waste of resources for downloading the same data on a daily basis, which would cost a surprisingly high amount of energy as there are whole server farms involved on a global scale.
Personally, if a game has a lengthy shader compilation step, I simply don't update graphics drivers until I'm done with it.
I do understand the problem, but your proposed solution would actually be worse.
Re: Will AMD Deliver Its Own Take On DLSS 5?
@JetjoO
Thanks for the tip with Uncharted 4, a game I know pretty well and absolutely adore. I've given DLSS 5 another shot, this time on my 77-inch OLED (though admittedly only via YouTube). And I have to say, I seem to get where you're coming from. I still think your statement is exaggerated (as of now), but the potential is definitely huge.
I don't like that every character seems to age at least a decade in U4, it seems DLSS 5 exaggerates all wrinkles and impurities in the skin. Also, the more games I see, the more they look the same, like the same filter being applied to them. A bit like Unreal engine games. But still, overall very impressive at times I have to admit.
I "only" have a 4090 and value high-fps experiences on a 4K screen, so I don't have the GPU horsepower to spare for DLSS 5 right now. But I hope you're right in that those technical restraints will be solved or alleviated sooner rather than later.
As for VR losing it's charm: I'm only about two years into it, so not nearly where you are. However, I'm also long past the initial honeymoon phase. For example, yesterday I played GT7 for the first time since February.
The difference (so far) between me and you is, however, that I never even considered playing the game on my TV, it never crossed my mind. When I don't want the hassle of setting everything up (in the case of GT7 that includes a wheel/pedal setup on a wheel stand), I simply play something else.
Again, I can't vouch for the next ten years, but so far this is how I feel and it doesn't seem to change over time.
Anyway, thanks for the discussion, let's have fun gaming either way
Re: Will AMD Deliver Its Own Take On DLSS 5?
@JetjoO
My enjoyment of the medium doesn't depend on sales numbers, so I don't care that VR hasn't taken the world by storm.
I also never expected it to replace flat gaming, at the very least due to comfort and the impossibility of social interaction during a VR session. It is definitely something very specific and was always destined for a niche rather than mainstream. The issues around motion sickness in a lot of people don't help either.
However, when you can stomach it, it's like nothing else. I could never go back to playing e.g. realistic racing games without VR. I've been an avid Gran Turismo fan all my adult life, but if GT8 doesn't support VR, I'm out. The experience is on a whole other level, playing it flat isn't fun anymore, it feels like child's play.
The same happened to my beloved Ace Combat 7 once I tried the VR mods, I just cannot go back to the flat version anymore and will only buy AC8 once a VR mod becomes available. The intensity, immersion and emotions have to be experienced to be believed.
I will never forget how I frantically laughed during my first VR flight in AC7, the joy burst out of me and I kept grinning from ear to ear for an hour straight.
DLSS5 is a nice layer of polish, but it's not even in the same league as a good VR experience. And nothing I've seen so far justifies your "the greatest jump since 3D" phrasing. The most impressive shot of DLSS5 for me was the controversial Grace scene from RE9, but everything that's available now looks a lot more dialed down and muted, very minor changes comparatively. What impresses you the most currently? Which game is the perfect DLSS5 demo in your opinion?
Re: Will AMD Deliver Its Own Take On DLSS 5?
@JetjoO
"This is a genuine revolution for gaming, on a scale we haven’t seen since the arrival of 3D."
Have you never experienced VR? Honestly, making graphics look a bit more photorealistic absolutely pales in comparison to actually entering your favorite game worlds instead of watching from the outside.
DLSS5 is evolution, VR is the true revolution since 3D in my opinion, and it's not even close.
Re: "Don't Kill the Disc": Petition Against Sony PS5 and PS6 Digital Media Plans Reaches 150,000 Signatures
@Godzookie
I've read about a study that came to the conclusion that streaming a movie twice is worse for our planet than producing, shipping and using a Disk to watch it twice. For music albums, the turnover point was around seven times listening if I recall correctly. So if you stream the same album more than seven times, a CD would be better for our planet.
They haven't looked at games specifically in that study, but due to the similar data sizes they should be comparable to movies. We tend to underestimate the enormous amounts of energy that is needed for running servers all over the world to stream Gigabytes of data across the planet.
So even from this perspective, physical Disks are the better solution and their doom actually a loss to the planet.
Re: Review: Star Fox On Switch 2 Is a Refined Return to an Iconic Game - and the Rail Shooter Genre
@MemuAccount
I‘m not sure I understand your first sentence, maybe because English is a foreign language for me. Could you please rephrase or explain what you're implying there? Is it that DF is disrespectful towards their audience and therefore you mirror that sentiment?
As for your main point of why discussions about certain technical features only gain traction once they get implemented into consoles:
I think that's due to the history of DF, because when I first came into contact with them during the PS360 era, there was no PC coverage at all if memory serves. They were rooted in consoles from the get go, and if I recall correctly they hired Alex specifically to be able to dive more into PC topics as interest rose over the years.
Also, I‘m pretty sure console gamers are still the vast majority of DF's audience, but I can't be sure of that. But assuming that's true, it makes absolute sense to tailor their content for their main audience and e.g. SSD load times analyses would have been pretty pointless back in the day when they only covered consoles and those consoles didn't support SSDs.
As for the 720p conflict, I assume that's a combination of differences in image quality and somewhat of a downregulation of standards, but I'm just guessing here. For the first, I guess that early Series S games at 720p didn't have sophisticated upscaling solutions, so 720p back then maybe looked significantly worse than games on the main consoles in more recent years. And for the second, I guess at some point we all realized that the consoles achieve native 1440p or higher increasingly rarely and that the reality will be lower resolutions going forward, so it necessarily became more accepted.
Put differently, we stopped making fun of 720p native resolutions because they became more common and the ridicule would get dull at some point. And also because actual results were not as dreadful as they had been even a few years prior due to better upscaling.
Don't get me wrong, I agree with you on some points, but overall I think you overshoot the target with your criticism and especially tone.
My biggest personal gripe with DF lately is that they are not critical enough in my opinion. Everything is great according to them, as if they are afraid to step on the toes of some publishers.
For example, Forza Horizon 6 without RT doesn't look significantly better than its predecessor apart from a few sharper textures in my opinion, but it is significantly heavier on resources. And yet, DF praised the great optimization.
An even worse offender is the latest Bond game. DF praised its scalability and called it well optimized if I recall correctly, but performance on even a 5090 is so poor that I don't see the headroom for the planned path tracing option. All the while it doesn't look particularly spectacular or does anything extraordinary technically.
Re: Review: Star Fox On Switch 2 Is a Refined Return to an Iconic Game - and the Rail Shooter Genre
@MemuAccount
As much as I dislike your disrespectful tone, you got a point here. While I loved the way John represented the game and its history in the video, there was shockingly little actual technical analysis in it. A rough resolution estimation and that was it in terms of image quality analysis, very poor by usual DF standards.
There was one level where shimmering bothered me (I think sector X), but the rest of the game looks too good to have no AA. It‘s a bit soft, yes, but I‘m pretty sure they have some form of image treatment in there (that maybe even contributes to the softness). I‘d love to know what they do exactly, and I usually get this kind of info from a DF analysis, but not this time sadly.
Re: DF Weekly: The GTA 6 60fps Debate Returns: Is There a Route Forward on Current-Gen Consoles Like Series X, PS5 or PS5 Pro?
The only way for a 60fps mode would be frame generation in my opinion. No way they get to 60 native, but FG from a 30fps base could be feasible at least on the Pro. If input lag doesn't get too bad (and GTA is not a twitchingly-fast-paced game), I would consider playing it this way.
If it turns out to be 30fps only, I'll wait for the PC version, but if it‘s 40fps or 60fps I might play it close to launch on Pro.
Re: LG Unveils First Native 1080p 1000Hz Monitor, Equalling CRT Performance
My gut tells me that going with an LCD panel when targeting frame times of 1ms is stupid, this would only ever make sense with an OLED panel. Is LCD capable of reaction times below 1ms? Never heard of it.