5 ms·

I guess, I'm a bit divided on this. While software is abstract (and deterministic), the experience of running the software is not. Meaning, as you become part o

I guess, I'm a bit divided on this. While software is abstract (and deterministic), the experience of running the software is not. Meaning, as you become part of a feedback loop, the sensorical side of things matters. And this is really hard to recreate. E.g., not only do emulators introduce lag, also modern systems tend to be laggier than vintage ones. Graphics taylored to the specific attributes of a display technology won't convey the original feeling on modern hardware. Or, interacting with a game using modern controllers, by D-pad or thumb stick, which are operated by fine motor control, feels much different from using joysticks that involve the entire body, by this establishing a rhythm. (This is also true the other way round, e.g., Pong, meant to be controlled by fine motor controls using a paddle, is half the fun using the brute means of a digital joystick, either jammed up or down.) This may sum up to the effect of the user entirely missing the point of a game. (There are some VCS games, I always deemed to be noteworthy, which are generally rated low in modern reviews. I had a hard time understanding why reviewers would miss the rather obvious qualities of those games – until I tried them using a D-pad controller.) P.S.: This doesn't mean that emulators aren't worth it (to the contrary, it's great to have them), but it may be worth keeping in mind that you are not necessarily having the original experience.

byuu 7y ago
> E.g., not only do emulators introduce lag, also modern systems tend to be laggier than vintage ones. Some emulators have run-ahead now, which removes internal processing frames from games. Combined with an adaptive sync monitor, 1000hz USB polling, and other tricks, emulators can actually respond to inputs faster than the original gaming consoles did on CRTs. It's pretty much the closest thing we have to magic. Here's a video demonstration I made: https://www.youtube.com/watch?v=1AvOa8yt6Vc https://www.youtube.com/watch?v=1AvOa8yt6Vc An an article of mine explaining the concept: https://byuu.net/input/run-ahead https://byuu.net/input/run-ahead
- tombert 7y ago
  While I agree with you on this for the most part, aren't we going to start hitting a ceiling with what's possible to do run-ahead with? For example, I know in that article you mention bsnes, but wouldn't it be exponentially more difficult to have that feature in something like a GameCube or N64 emulator (let alone something like a PS3 with its funky Cell architecture)?
  - DCKing 7y ago
    The system setup of old devices was quite different from what we're used to now. CRT TV's, no OS in the way. Just inputs and near instant display of what was going on. The more modern your game system gets, the harder it is to perform runahead. But it's also required less and less. Newer consoles got more 'laggy' by default, with OSes, USB/Bluetooth input processing, triple buffering and so on. So runahead requirements have become less and less to recreate experiences. As to you examples, the PS3 was meant to be played on LCD TVs much like our current setups. I don't know how RPCS3 compares to a real console, but I expect the difference isn't huge. Dolphin (GameCube) and Mupen (N64) can emulate their consoles so efficiently that you should be able to achieve 1 frame of runahead on somewhat powerful systems for many games.
  - byuu 7y ago
    Yes it definitely would be. Every frame of run-ahead requires emulating another instance of the system. So a run-ahead of four frames means needing 5x the processing power. Generally speaking you'd only want 1-2 frames of run-ahead, however. And in practice you can usually skip video rendering during the run-ahead frames to reduce overhead a bit.
- whateveracct 7y ago
  > Some emulators have run-ahead now, which removes internal processing frames from games. Combined with an adaptive sync monitor, 1000hz USB polling, and other tricks, emulators can actually respond to inputs faster than the original gaming consoles did on CRTs. It's pretty much the closest thing we have to magic. Super Smash Bros Melee has a little bit of polling lag on console (the tournament standard) due to poor coding. By removing this lag (or rather, making it configurable), you can "hide" network latency behind this lag people are used to, and online play can feel like local play. Like you said, magic!
deltaryz 7y ago
I address this near the beginning - and you are correct that there are some examples that can have a significant impact on the resulting experience of the software. Emulation is not a 1:1 recreation of the original experience including all of its intricacies. But I'm thinking extremely long-term here, far beyond "what our kids will think about the games we played". I'm thinking about what historians of the distant future could learn from our creations, due to how well they can be preserved compared to other forms of media. Ideally people would have an intuitive understanding that input methods and displays make a tangible difference in the experience, and that an emulator is never going to be a full replication of that - so that hopefully they can study and judge a piece of software with those differences in mind. Technically this is true of anything that can be represented digitally, but software's interactive nature gives it a much more profound ability to represent our culture and elicit a strong and personal reaction in those who experience it.
- masswerk 7y ago
  I totally consent with this. However, doing as much as we can to reproduce the experience is of some importance. E.g., there has been a basic Spacewar! emulator in Java since 1997, but it only recreated the game on an abstract level, missing the intricacies of the display. Many, including myself, somewhat shrugged their shoulders on this. It has only been, since there have been efforts to recreate the display experience on a more or less analog level that the game is appreciated for what it is, a quite perfect beginning, not to the least for its esthetics. (Disclosure, the web emulation of the PDP-1 running Spacewar! is by me.) Based on this experience, I'd say, there's need to constantly update emulators to the current state of technical media representation. We are never on the safe side… This will only become trickier as any original hardware and/or media recordings of the software running are lost. Much will be lost in translation, since software has context.
  - deltaryz 7y ago
    > there has been a basic Spacewar! emulator in Java since 1997, but it only recreated the game on an abstract level, missing the intricacies of the display Did this version actually emulate PDP-1 machine code, or was it a purely mechanical clone? Because if it's the latter, then that isn't an emulator. Either way though, Spacewar is definitely one of those examples where the hardware directly impacts the experience. This is also why I am saddened by the fact that the only truly good CRT emulation I've personally witnessed is from Super Win The Game!, and since I believe that game is closed source, it unfortunately cannot be repurposed for other games or emulators. Unless you happen to know of a particularly good reshade implementation or something. For a game like Super Metroid, the "CRT experience" essentially boils down to "a greater sense of atmosphere and an enhanced sense of depth in some of the sprites". Does this arguably improve the experience? Yes. But the core experience of Super Metroid is not lost if you can't experience it this way - I would argue Spacewar does lose a core part of its experience without that iconic display, and I absolutely believe we should make an effort to preserve that experience.
mntmoss 7y ago
It's the difference between events and artifacts, and there absolutely are differences between the two as time goes on that make even the very best preserved materials different in their essence. An event is experienced once and remembered as its artifacts; an artifact retains some substance allowing similar events to reoccur. But on some level all artifacts are just echoes of the event and therefore are prone to decay and lose their meaning. Making it a digital artifact just allows some of those axes of decay to slow down, like an anti-aging therapy that can't treat dementia. And so it is also the case that physically realized games are more likely to play similarly in the past and present - sports have changed, Chess has changed, and both have seen reshaping via tech, but neither one unrecognizably so in the way that video games communicate phenomena with orders of magnitude more detail today versus 30 years ago. The question, "which is better to favor, events or artifacts?" is a worthy one to ponder. Is it better to have fresh coffee available at all times, or a perfect coffee date once a month? Is it better to be able to listen to a recording 1000 times, or to attend a single live practice session? What technology has done, beyond the basic needs aspects of Maslow's hierarchy, is let us pick and choose those options more often with fewer compromises. Its biggest failing tends to be in only solving a surface part of the issue: For example, having Facebook lets you message your friends and view their postings in some ambient fashion, without direct engagement. But it doesn't organize the party or send the invitations or prepare for guests or otherwise do the real work of a friend relationship. And so in that sense Facebook is a friend artifact that produces few useful friend events, and converts all those events that occur within its domain into proprietary artifacts of the company. But these artifacts aren't exactly worthy of emulating either. Anyone who controls them is probably sitting on blackmail.

byuu 7y ago

> E.g., not only do emulators introduce lag, also modern systems tend to be laggier than vintage ones. Some emulators have run-ahead now, which removes internal processing frames from games. Combined with an adaptive sync monitor, 1000hz USB polling, and other tricks, emulators can actually respond to inputs faster than the original gaming consoles did on CRTs. It's pretty much the closest thing we have to magic. Here's a video demonstration I made: https://www.youtube.com/watch?v=1AvOa8yt6Vc https://www.youtube.com/watch?v=1AvOa8yt6Vc An an article of mine explaining the concept: https://byuu.net/input/run-ahead https://byuu.net/input/run-ahead

tombert 7y ago
While I agree with you on this for the most part, aren't we going to start hitting a ceiling with what's possible to do run-ahead with? For example, I know in that article you mention bsnes, but wouldn't it be exponentially more difficult to have that feature in something like a GameCube or N64 emulator (let alone something like a PS3 with its funky Cell architecture)?
- DCKing 7y ago
  The system setup of old devices was quite different from what we're used to now. CRT TV's, no OS in the way. Just inputs and near instant display of what was going on. The more modern your game system gets, the harder it is to perform runahead. But it's also required less and less. Newer consoles got more 'laggy' by default, with OSes, USB/Bluetooth input processing, triple buffering and so on. So runahead requirements have become less and less to recreate experiences. As to you examples, the PS3 was meant to be played on LCD TVs much like our current setups. I don't know how RPCS3 compares to a real console, but I expect the difference isn't huge. Dolphin (GameCube) and Mupen (N64) can emulate their consoles so efficiently that you should be able to achieve 1 frame of runahead on somewhat powerful systems for many games.
- byuu 7y ago
  Yes it definitely would be. Every frame of run-ahead requires emulating another instance of the system. So a run-ahead of four frames means needing 5x the processing power. Generally speaking you'd only want 1-2 frames of run-ahead, however. And in practice you can usually skip video rendering during the run-ahead frames to reduce overhead a bit.
whateveracct 7y ago
> Some emulators have run-ahead now, which removes internal processing frames from games. Combined with an adaptive sync monitor, 1000hz USB polling, and other tricks, emulators can actually respond to inputs faster than the original gaming consoles did on CRTs. It's pretty much the closest thing we have to magic. Super Smash Bros Melee has a little bit of polling lag on console (the tournament standard) due to poor coding. By removing this lag (or rather, making it configurable), you can "hide" network latency behind this lag people are used to, and online play can feel like local play. Like you said, magic!

deltaryz 7y ago

I address this near the beginning - and you are correct that there are some examples that can have a significant impact on the resulting experience of the software. Emulation is not a 1:1 recreation of the original experience including all of its intricacies. But I'm thinking extremely long-term here, far beyond "what our kids will think about the games we played". I'm thinking about what historians of the distant future could learn from our creations, due to how well they can be preserved compared to other forms of media. Ideally people would have an intuitive understanding that input methods and displays make a tangible difference in the experience, and that an emulator is never going to be a full replication of that - so that hopefully they can study and judge a piece of software with those differences in mind. Technically this is true of anything that can be represented digitally, but software's interactive nature gives it a much more profound ability to represent our culture and elicit a strong and personal reaction in those who experience it.

masswerk 7y ago
I totally consent with this. However, doing as much as we can to reproduce the experience is of some importance. E.g., there has been a basic Spacewar! emulator in Java since 1997, but it only recreated the game on an abstract level, missing the intricacies of the display. Many, including myself, somewhat shrugged their shoulders on this. It has only been, since there have been efforts to recreate the display experience on a more or less analog level that the game is appreciated for what it is, a quite perfect beginning, not to the least for its esthetics. (Disclosure, the web emulation of the PDP-1 running Spacewar! is by me.) Based on this experience, I'd say, there's need to constantly update emulators to the current state of technical media representation. We are never on the safe side… This will only become trickier as any original hardware and/or media recordings of the software running are lost. Much will be lost in translation, since software has context.
- deltaryz 7y ago
  > there has been a basic Spacewar! emulator in Java since 1997, but it only recreated the game on an abstract level, missing the intricacies of the display Did this version actually emulate PDP-1 machine code, or was it a purely mechanical clone? Because if it's the latter, then that isn't an emulator. Either way though, Spacewar is definitely one of those examples where the hardware directly impacts the experience. This is also why I am saddened by the fact that the only truly good CRT emulation I've personally witnessed is from Super Win The Game!, and since I believe that game is closed source, it unfortunately cannot be repurposed for other games or emulators. Unless you happen to know of a particularly good reshade implementation or something. For a game like Super Metroid, the "CRT experience" essentially boils down to "a greater sense of atmosphere and an enhanced sense of depth in some of the sprites". Does this arguably improve the experience? Yes. But the core experience of Super Metroid is not lost if you can't experience it this way - I would argue Spacewar does lose a core part of its experience without that iconic display, and I absolutely believe we should make an effort to preserve that experience.

mntmoss 7y ago

It's the difference between events and artifacts, and there absolutely are differences between the two as time goes on that make even the very best preserved materials different in their essence. An event is experienced once and remembered as its artifacts; an artifact retains some substance allowing similar events to reoccur. But on some level all artifacts are just echoes of the event and therefore are prone to decay and lose their meaning. Making it a digital artifact just allows some of those axes of decay to slow down, like an anti-aging therapy that can't treat dementia. And so it is also the case that physically realized games are more likely to play similarly in the past and present - sports have changed, Chess has changed, and both have seen reshaping via tech, but neither one unrecognizably so in the way that video games communicate phenomena with orders of magnitude more detail today versus 30 years ago. The question, "which is better to favor, events or artifacts?" is a worthy one to ponder. Is it better to have fresh coffee available at all times, or a perfect coffee date once a month? Is it better to be able to listen to a recording 1000 times, or to attend a single live practice session? What technology has done, beyond the basic needs aspects of Maslow's hierarchy, is let us pick and choose those options more often with fewer compromises. Its biggest failing tends to be in only solving a surface part of the issue: For example, having Facebook lets you message your friends and view their postings in some ambient fashion, without direct engagement. But it doesn't organize the party or send the invitations or prepare for guests or otherwise do the real work of a friend relationship. And so in that sense Facebook is a friend artifact that produces few useful friend events, and converts all those events that occur within its domain into proprietary artifacts of the company. But these artifacts aren't exactly worthy of emulating either. Anyone who controls them is probably sitting on blackmail.