Star Citizen Live: Engine & Graphics Q&A

6 May 202201:03:111,758 linesWatch on YouTube ↗

Transcript — every row opens the video at that moment

00:00:02
hello everybody welcome to another episode of star citizen live engine and graphics q a i'm your host jared huckabee uh welcome once again back to my half ikea half still unpacked luggage uh a home office uh i'm jared huckabee uh creative content lead for cladded brain games robert space industries and if you've never seen star citizen live before it's where we take about an hour out of our uh a week and uh chat with some developers uh usually about their work usually about the stuff that they're they're working on or the stuff that they're about to work on or the stuff that they've worked on in the past that people still have questions and and stuff about uh joining us on the show this week is a uh uh is a menagerie of engine and
00:00:52
graphics uh programmers directors and and sorts so let's uh let's take a few moments as we usually do and introduce our our codery of developers this week starting uh to my immediate right with the lovely blue gradient background jonas who are you and what do you do for star citizen oh hi i'm jonas i'm starting i'm working for star citizen since one and a half years roughly and i was mainly doing co-engine work like doing profiling tools or internal stuff one of the fruits of that you could likely see in the last patch with the performance improvement we found that due to the new profiling tools or improved profiling tools and i'm also supporting on the john 12
00:01:41
work to get that done are you taking credit for all the performance improvements of the last match no of course not that's likely something for body to take credit all right uh sylvan who are you and what do you do for star citizen yeah hi everyone so i'm sullivan i joined cig three years ago um i am an engine program as well um mainly i do graphics work and helping the graphics team um specifically gen 12 and yeah in the future most likely vulcan as well and usually i do all sorts of things like wherever i'm needed i'll just jump in and you know fix box or whatever it's required well i'm there fair enough and a man who probably needs
00:02:31
no introduction but we're going to do whatever you want anyway because every show is somebody's first show mr ali brown who are you and what do you do for hey i'm ali brown i'm the director of graphics engineering uh so cover the graphics department and the vfx programming department and then work very closely with the guys in the engine department in frankfurt and i've been here since the beginning pretty much yeah you have you you're one of the you get you got to be close to your 10 year at this point right i think the uk studio has been going eight and a half last movie yeah all right so uh today's show uh is plainly labeled as engine and graphics a q a uh we we have we're basically going to answer questions related to
00:03:21
uh uh engineering graphics engineering gen 12 renderer all that you know ray tracing all that good stuff as usual you can submit your questions live in the chat during the show by prefacing your question with the word question in capital letters surrounded by brackets and we spent the week collecting questions as we usually do letting you guys vote up which ones you want to see uh answered most and then going through there stripping the ones that you guys voted up that were still off topic for this group and then finding the ones that were actually appropriate for our our esteemed colleagues today so we're going to start writing with that um i think the biggest topic for anything engine and graphics is probably gen 12 vulcan stuff so we'll just we'll jump into the uh
00:04:09
the more heavily traffic stuff um this question says uh what are the differences between gen 12 and vulcan and how are they related to get more optimization in the game uh yeah that's probably a question for me um probably the best question to start with since um yeah i'm working on it pretty much full time and for the people who still don't know what the difference is there i mean that's pretty much um like we've already published a lot of videos about it and other people talked about it but gen 12 is basically our internal engine structure rendering architecture
00:04:58
for how rendering works and vulkan is the underlying graphics api which one day hopefully gonna not hopefully but which which one day replaces um the directx um graphics api and yeah vulcan the benefit of vulkan is pretty much performance it scales much better on multiple cpus and threats so well yeah performance we all have performance and we all seen what it can do in 3.17 yeah if everything goes to plan we'll it performance will get even better the
00:05:47
coming patches so i'm very excited for that especially after we shipped one of our first milestones in 3.17 about gen12 and yeah i'm working hard on it to get it done what was that milestone that just shipped in 317 um so gen12 is so you can kind of split it up in two parts like all the scene rendering like all the objects you see on screen and then all the other rendering stuff like post-processing and compute passes and pretty much everything else and the scene rendering is especially difficult to what's especially especially
00:06:34
difficult to do because we have like hundreds of shaders and thousands of shader permutations and thousands of materials and we need to make sure that everything works um exactly as a legacy rendering as the legacy renderer and yeah we have a lot of issues and edge cases which you don't see during development so just once you submit your code and it crashes somewhere and then you just take a look and see oh okay that's what happens okay i forgot this too i forgot to add this so i'm gonna add this edge case yeah and so the first milestone was getting part of the scene rendering done and for 3.17 that was the brushes the brushes is what we call like all the
00:07:24
non-interactive static objects you see in game this is these are the brushes and yeah they're going through 10 12 already and and the next step would be um characters ships like like yeah dynamic objects and the planets didn't make it 3.17 as well though also the um objects on the planet they made it in 3.17 as well with gen12 so happy about that um yeah so once we have that um gen 12 is almost complete we still need to do all the transparent objects which is still a huge task to to tackle but yeah we started on that and i'm very
00:08:13
optimistic uh i i i know you we we we've hit the uh the performance uh thing you know kind of hard there are there benefits to uh the switch to a gen 12 renderer that aren't uh performance related i mean do do we get does does does the action does the quality of the render improve does i mean are there better visuals all around or is this strictly just a performance improvement kind of thing um gin 12 is strictly just performance but of course audio also mentioned that in previous talks pretty much it gives us more room for more fancy stuff like we can use all the new fancy features and vulcan like rage racing and what else there is so
00:09:04
not just performance has that battle work ally that you you get all these performance improvements and then you start adding more uh more and more features and more and more uh uh uh visual elements and that start taking up those performance gains that you just that you just made for yourself how do you how how do you balance that well ironically a lot of the new technology we have isn't actually saves you performance so it initially it might even go the other way like it enables a whole bunch of features with things like the mesh shading and the variable weight shading and and sometimes not even technology but more just like uh not hardware technology but internal like because we've restructured the renderer we can now do certain things much easier than we could before so there's lots of opportunities now for
00:09:53
optimizations that we may not have had before so it might actually be like a double win after this can implement uh some improvements and there's those features that i know that there's questions about later about the lss or fsr and these type of technologies which were were type of problematic to implement halfway through you know when we're doing the conversion to gen12 so then they become on the table again once we've settled on a single new renderer so i don't think it's the end of the performance improvements once we've uh got to gen 12 but yeah there is also a few that would go the other direction um and and losers performance um but future battle we'll see how fast we can make it first before we worry too much about how much we'll slow it down in the future um a big focus we want to do is to make the engine scalable so hopefully these any new shiny things we add can be scaled up and down
00:10:42
uh ideally so that we don't have to you know you can keep the game running on the on the lower spec but push it hard on the high spec and completing gen 12 is just the first step to a series of more optimizations we already planned out so expect much more to come after yeah after 10 12 yeah we've we've got it as a on the milestones as if like gen 12 and vulcan are like you know a concrete thing that is not done and then one day is done that's not quite true it's i mean it's it's useful for scheduling to type of make that type of appear true um but you know that it's not as precise as that um yeah this continuous set of features and even vulcan you know that that will introduce new performance um or it might you know detriment forms in some ways
00:11:30
might improve performances somewhere and as we learn more we'll be able to get more performance out of vulcan so yeah it won't just be uh we're done and we never look at this thing again but the milestones were mentioned in a type of are a point of confidence when we think yeah that's like 99 that's most of what we were gonna the biggest chunk of work has done i think that's probably a perennial uh concern with all aspects of star citizen because this is the this is not just a make it until one day and it's done kind of thing that the development of star citizen goes it just continues it's like any other mmo you're continuously iterating and developing stuff there is it's it's i think that i understand you know having come from the community so long ago this desire to
00:12:19
just put a bow on something and say it's done uh it's just not the way uh uh this project works it's this project is constantly about pushing those boundaries and seeing what's possible and stuff it's been in our dna since the beginning so it's uh it's always a like everything it's a delicate balance between between not wanting to just lead people on forever you know just you know but also you know just reporting things accurately and honestly and effectively one of the questions here uh is from somebody who's clearly looked at our uh uh i can't remember what it's called right now i'm blaming the name the thing that shows the the performance for all of our players and stuff that cool graph and i'm blanking on the name it says i see many
00:13:07
folks with older graphics cards like the 970 and 1080 cards those 1080 cards are really good though will those by chance see an fps boost after gen 12 and vulcan telemetry thank you people elementary good think of the word yeah we'll try to answer that answer that so the goal of gen 12 in the first place is to improve cpu performance so in case you run an older cpu with your old gpu you might see an improvement because we better you can better utilize all the cost of your cpu but the more likely case is that your bottleneck by the gpu and then um gen 12 won't really bring a benefit there welcome might bring a benefit there but you never know we have to have to
00:13:56
measure that it might bring benefits in one case but there's a drawback in another case that that's hard to tell but uh especially when you when you're rendering with uh in high resolution when you are trying to do 4k and then the older graphics cards graphics cards will most likely be a bottleneck anyway and and perform type there ali you and i were talking uh earlier in the week about common misconceptions that that gamers have just where they come from not just for our project for all gaming project like like everybody like everybody's got 12 different things that they all call lag you know and stuff like this uh there's there's this assumption that gen 12 and vulcan it's like this are video card related things and and as we
00:14:46
just discussed there they're actually more cpu related oh if you had to guess where that misconception comes from this isn't the show we're just going to clear up a bunch of misconceptions but i'm curious what you're thinking is on on on why those things all often get conflated into video card issues and not cpu issues yeah i mean it's stupidly confusing um we're working on a renderer and a graphics api but somehow they don't make the rendering or graphics any faster um it makes no sense um yeah but the reality is a lot of the work is feeding this these uber powerful type of processors we have these gpus with work and it's hard to keep up like to give them enough work um and that's why it's it's a lot of effort for the cpu to keep them fed and
00:15:35
these that's why like there's so much work to do on the on the gen 12 and the vulcan side these graphics apis and the renderer um but i mean like we said earlier that gen 12 will open doors to optimizations then optimizations will and can benefit the gpu but just the milestone we're mainly focusing on is improving the cpu performance because that has by far been the biggest bottleneck for most players um and that's why somebody with like a 3080 or 3090 might not be in the frames they want um so um yeah so for most people it's it is the cpu bottleneck i mean the the range between the cards mentioned there the 970 and 1080 is quite large uh the 1080 is going to be significantly faster then obviously resolution like like jonah said like is a big factor as well
00:16:23
um if you're running a 970 then hopefully you're targeting 1080p and not trying to push too much beyond that um otherwise you will struggle yeah but the um the i i can see there's many reasons why things would be confused in this area um yeah i don't know what we can do about that um it is just confusing i just embark on a crusade to re-educate the entire games industry why not we got time as we're talking about performance here this question says how come since the gen 12 renderer was implemented my performance actually dropped i get an average of 13 to 17 17 fps no matter the quality or resolution of the game well it's easy to answer and
00:17:12
in fact so if um whoever's watching so i posted like a gen 12 update and progress like two weeks ago on the spectrum and initially like i didn't even plan to do that because i didn't expect any performance benefits of gen 12 in 3.17 due to the uprising amounts of threats and people talking about these magical improvements in 3.17 and people were wondering where that actually came from i felt like i had to i had to give an update so uh yeah that's what i did um the thing is gen 12 right now only the brushes um [Music] they're rendering in general the big caveat is that
00:18:03
our rendering right now we're mostly limited due to the main thread and not the render thread so they're like two two threads and it goes very deep into engineering i would already say so for everyone who can't follow me right now i'm sorry but basically that's like a render threat which um as ali said previously it's like um this kind of um thing feeding the work the rendering work into the gpu and that with the old legacy renderer is really slow because constantly every frame um check every state and see what objects we can render and trench 1012 uses a different approach by just pre-baking the state
00:18:52
in like in a small object and then when we render we just have the entire state ready and can just say gpu here another subject so the render thread with i tested it myself on the performance was like with gen12 it's more than twice as fast but if you're not limited by the render threat which was at least on my machine was mostly the case um like it doesn't give you anything that's why i didn't expect any benefits and didn't feel the need to give an update but yeah but in the future if with full gen 12 that's that's another thing right now because we have this mixed mode we have
00:19:40
objects rendering with gen12 and other objects going through the legacy path and there is actually a cost to that on the main thread which is the bottleneck which means or what i would have expected gen 12 would have made it actually a little bit slower that didn't even happen so i'm happy about that but once we can get rid of all the legacy stuff and have full chain 12 then we'll see the full potential and i'm very excited about your story is a very common one throughout all of game development it's one of the reasons quite honestly most games don't do the level of content and interaction that we do anything because if there's a reason it's called game
00:20:27
development not game construction we don't always know how the thing is going what the puzzle is going to look like once we finish putting it together uh and it's it's why when when you know when folks ask questions about how much gen 12 is going to improve performance versus how much server meshing is going to improve versus how much you know object you know container streaming is going to stuff like this it's it can often sound like our developers are being cagey and it's and and it's less about being caging just want not wanting to over promise uh on on on something it's it's because there's always these pleasant surprises there's always the we're doing this hoping that this will do the thing we want it to do and if it
00:21:15
does then awesome and if it doesn't then we're going to do something else trying to hope it does the thing we do and if that that does great and if it doesn't we're going to do something else to try it's it never you don't ever you don't ever give up you know not on a project like this anyway maybe on some other ones uh i've worked for but it's it's it's it's it's hard to make these it's hard to always speak in these definitive things ahead of time because a lot of this is being discovered along the way when uh when you're building something that's never been built before how do you say how fast it's going to be or or how precise it's going to be and so it's a it's one of the combination best things and sometimes worse things about
00:22:02
working on a project of this size scale scope openness let's see if we've got a couple more gen 12 vulcan stuff here we'll just keep them all grouped together here's the forever question uh linux support linux support after vulcan integration is this still planned well i i can give a short update on that as well um so for the people who don't know um only works on windows and vulkan is um created by the chronos group which also created opengl so this is like a different opengl is like more an old graphics api and vulkan is like the new modern graphics api
00:22:51
these kind of graphics apis they also run on various other operating systems and yeah vulcan is like known to run on windows and on linux and even on apple with an additional wrapper it's called molten os i believe molten vk that's the name so and christopher balti like he's very keen on on the linux port and so it's up top on his priority list so all i believe one day we'll see linux port uh the other thing i'd add is um this isn't firsthand knowledge but second or third has knowledge but the um people can already run it using the uh the various layers you got called proton and wine on linux they can run the game now i think there's a there
00:23:40
was an issue with the easy enter cheat type of uh blocking that i don't know if that's been resolved yet but the um that was on uh epic i believe to resolve that but um that does work but going through these layers that emulate windows has a major performance cost um apparently it's playable some people say but um the use of vulkan i've been told allows it to skip that layer that it has to apply to translate direct text into something usable so even if we just had vulcan and did implement native support for linux we would still get much better performance on the rendering by using these layers because it wouldn't have to emulate directx anymore um it would just have to emulate what the other there's a you know any other windows calls or with the input and the sound and things like this but obviously the graphics tends to be a huge processor hog
00:24:29
uh that not have to do a translation uh between windows code and limited code uh we'll save quite a lot there we're hoping so it makes that boot more viable until we get a an actual official you know pure linux native linux client i know i think the the folks who had it running on mac i think that ended with when they switched to the m1 like you still do it on the old intel machines i think the m1s in one switch they did um i'm still waiting for the voodoo 3dfx port personally uh thanks jackaroshi for the joke yeah i'm sorry guys my my not my mic is quiet i'm quiet i have a sore throat i'm under the weather right now it's why i'm sweating like a hug and everything my throat's all scratchy so i'm i'm just
00:25:16
quieter than normal um let's switch to some graphics stuff so some stuff uh you knew there was going to be a dllss question an fsr question alex let's just let's just jump those ones right away uh what can you tell us if anything about dlss or fsr at some point any point in development uh must be someone that didn't watch the last uh star citizen live every show somebody's first um yeah um i mean upscaling is a very common feature now you know every single console game pretty much uses it a huge percentage of the aaa pc games use it um it's it's the i don't know it's the future but it's it's the current like this is the expected way if you want extreme
00:26:03
resolutions unless you have like the the fastest graphics card um and even when you do there's actually reasons that the lss and fsr can exceed the quality of native resolution in many scenarios they the compromises sometimes there's artifacts in some situations but on a stable frame you can actually get higher quality because it's actually super samples the image beyond the number of pixels you actually have so it's like a perfect anti-airlift image or it can theoretically can achieve a perfect anti-list image which a native render cannot so um yeah they're definitely the future um or in the next few years the way to get to 4k and beyond um in terms of the implementation we need to just finish gen12 first and there's no point this is like by far the biggest performance improvement is the cpu gains we need to make from gen 12 um but the
00:26:50
ls as an fsr or an equivalent are type of top of the list uh in terms of the general rendering improvements we have lots of other workflows you know the pu and squadron that we need to get on with in terms of general render improvements these are the top of the list um so yeah definitely um um i like i said last time though i'm not gonna promise one over the other or timelines for them or whether we even come up here and one you know epic came up their own one that um it's more similar to a hybrid of the two who knows um but you know the existing apis for dlss and fsr are theoretically going to be easier to implement than rolling your own one so that's the most likely scenario and most likely would be to support both but we have to see when we get to it and one thing that makes them a little bit awkward to implement our engine is
00:27:37
because traditionally you render your scene you upscale it to 4k or whatever and then you end your ui on top of that however our ui is very much in the world a lot of cases we have holograms in the world we have mfds on ships we have floating holograms uh holographic displays in the ships as well uh mobile glass all these things so they're not a flat 2d image projected on top of the render they're actually part of the world which makes it problematic because if you try and upscale very intricate text and ui these algorithms tend to destroy it and do some evil things for for what you really want to be crystal clear so we already have a solution which type of splits are rendering so that certain parts of the ui type of they have when they get to a certain size on screen or distance they type of swap to a different mode that renders them and make sure that the post effects
00:28:25
don't interfere with them so we're hoping we can capitalize on that and use the same technique so we can do fsr or dlss to the main portion of c then some objects get removed from that and composite it on top but that does make some things complicated for us so that's that's a little like oh we have to see how it's going to work um it'll probably won't be bug-free in that sense so we'll have to tweak that till we get it right but yeah it's definitely something we need upscaling is important and for the vast majority of our users i see as the common group you know some people want to a native but i think once people have it and they see yeah i can well especially once they have the cpu gains which you need um but once you've got cpu gains and you can get 60 and you're like ah i can have native 30 fps so i can have you know upscaled 60fps you know maybe i'll you
00:29:15
might be an easier choice once they see it i'm imagining the kotaku article now star citizen developer says feature probably not bug free let's see while we're on graphic let's let's just do our greatest hits here uh how is hd hdr support coming along i want to sell them uh yeah um i i talked to our developer who works on that and right now he other priorities took over um that task so it's not hard right now yeah he once he he's doing home office right now a lot of the time but and and he has like a super fancy hdr monitor in the office so once he comes back which
00:30:05
hopefully soon and we even gonna move into the new office soon which i'm super excited about um yeah hopefully then he can um take it on and finish it because i'm i'm very much interested in that as well for everyone who doesn't know what hdr is it it means a high dynamic high dynamic range basically it means um colors can or colors from a monitor can be much more brighter and vivid than a standard monitor like everyone's used to so and rendering always or rendering in physically based rendering which most engines these days do and they already render an hdr and then as a post-process effect all those colors get like compressed into what we call low
00:30:54
dynamic range ldr and that's what your monitor after some other fancy post-processing effects like um gamma needs to be applied but yeah that's the story for something uh for something else but yeah so i think hopefully so a bunch of sport i think has been done i think uh the last i looked the the tasks that were meaning to make that fully finished is three of them i think so we have um color processing is something they are to supply where like you know they they want to set the mood of a scene and they are on a blue tinted sky and they don't make the you know they won't tent the actual sky when they want to achieve a mood that you know they they went to the sky physically accurate and then they just postponed you know if you imagine uh bad example that every night time scene in a film is filmed in
00:31:41
daylight then they just make it blue you know it's a exaggeration but that's kind of the approach um so sometimes the artists use these things usually quite subtly but that process is currently mapping one low dynamic range color into another low dynamic range color the process becomes a bit more complicated when you're dealing with high dynamic range colors you know when they say they want to turn a white sky sky to blue or you know blue sky to white it gets more complicated there's now millions of shades of white millions of shades of blue like how you know did you just mean the very bright blues or the ultra bright blues or the extremely bright blues like so that process needs to be changed and it involves changing the art pipeline a little bit and how they offer the color correction and the two other issues are the we have what we call the game post effects which are like not the critical rendering ones like uh color correction
00:32:29
or type of lighting and blurs and things like this but just like when you've been shot blood appears on the screen or you're blacking out and these types of effects they'll run in low dynamic range so they'll need uh to convert into html and higher note range so they don't lose the quality otherwise they'll appear very dull if we don't do that and then the final one is there's a the more open-ended one uh when people say this game has a good hdr implementation and this game has a bad one that's type of we like someone said we already rendered the game in hdr the problem is even the html monitor isn't actually high but it's not infinite dynamic range um we would stop there if we had infinite bank range we'll just give you the image we rendered but that's not the case so what you have to do is you have to query the monitor find out what range it's got and type of do a best guess of how bright you think you should render
00:33:18
the image and how you should kind of correct the image and then probably also give some sliders for the user so they can kind of tweak it themselves but that can be a hugely subjective point of view and like say different games get either praised or or beaten for how amazing or terrible their hdr implementation is but it's really just that last part uh in the pipeline that people uh get upset over so yeah with that we need to just balance that and make sure we expose the right knobs and tweaks to get everyone to get the image they want on their monitor it doesn't help that there's probably a dozen hdr standards or a half dozen hdr yeah lots of color spaces lots of different ranges different ways of interpreting things it's a bit of a minefield yeah uh you mentioned well you sort of mentioned night vision but you got close enough that we can't avoid it now the question that comes up
00:34:06
for every other team and they keep putting it i always put it off to the graphics team what can you tell us if anything about night vision tools uh uh any option for players for to to improve their their experience flying or fighting at night last time this came up you promised you were going to lynch todd pappy um so i i think you should follow through on that promise and get in and uh because we need uh we need i have technical ideas and thoughts on that but we need via design like is is night vision in the future the same as now is it infrared is it was it based on you know some other forward technology but um yeah what i've mentioned in the past is that if you want to do it really well then you use you know you emulate what we call spectral rendering which is an
00:34:53
incredibly complicated field but there's elements of it you can you can simulate um where you can type of identify the materials that have a non-standard way of reflecting infrared light uh which would be like you know anything well anything that glows in the dark or spheres to glow in the dark on an infrared camera um which is type of you know like reflective gear like you know your i don't know i can't think of any good examples right now but um uh there's certain materials that type of don't reflect as you'd expect um so they need picking out especially otherwise they don't doesn't look right in night vision um otherwise it's type of a filter you need to type of take your um your the colors and kind of uh it's a computer topic it's hard to ever break it down but what we see is red green and blue is obviously not the
00:35:40
true picture of what happens in reality in reality there's there's millions of wavelengths of light infinite wavelengths in a longer spectrum and then infrared is just looking at one portion of that spectrum but we don't the artist doesn't model what happens in that different spectrum they just model what happens in these very three specific spectrums of the human eye can see so we have the type of map or guess how do we get from the artist or what colors to the these type of the infrared spectrum or whatever spectrum our imaginary night vision might be um and there's ways of approximating that or you can offer custom textures if you want to like say if the artist can tell this is exactly what it looks like in infrared but we clearly can't go and we make every texture in the entire game to show how the infrared would work we double the textures in the game so it's you normally do like some type of automated mapping that guesses a best guess approach and then you have to override the particularly special
00:36:29
case ones but yeah it's all going to start with todd in the request so you did good you did good the next time next time i ask the question will be to todd uh he's actually i don't think he's actually been on a show since the last time we had you on he's managed to avoid it thus far how is the graphic system going to accommodate large fleet battles uh greater than 50 fighters plus multiple capital ships the lighting budget alone would seem too crazy for that scenario as things are right now what you got for me jonas yeah yeah right now you will very likely be cpu bound in this situation because so many ships they all need to go through the render
00:37:15
threat alone but with upcoming gen12 changes it's hopefully the changes that we make will very likely improve that at least from the rendering side because we can distribute your distribute the work more evenly across your cpu course and that should help there but of course there are other systems than just rendering that you need to keep in mind for example ai or game code or physics or whatever that also takes apart on the cpu so um j12 will of course improve it but uh yeah it's hard to say how much sorry all right i'm gonna say on the lighting side because i was mentioning the question as well um the lighting tends not to be a big issue in space obviously space is mostly pretty dark
00:38:03
not too many not too much light going around there we obviously can potentially have lights on the lasers but they're quite easy also to type of lod out if we were particularly struggling for performance ironically we're in a big ship battle one of the things that hurts us quite hard is the interior of the spaceships um we don't have a great way of not rendering the interiors of spaceships so obviously if you can't see a window then we call them out or if they're very far away we call them out but if you can see in the spaceship or you have like the 600i with all its glass you can't just not render it you need an approximation of what's inside that group and that's quite difficult to make you know you can't just fade it to black because like this say 600 is a very bright interior it just fades to black looks really bizarre so um one thing we were planning is to use type of we have these cube maps we call them
00:38:50
environment probes which type of they used in the lighting system to type of capture the environment and then reflect it and as a substitute type of more advanced fade racing reflections so we were planning on using them to kind of render a low quality version of the room the type of would be indistinguishable at the distances we're talking about but would be the correct colors and have the correct type of geometry in there and so that will give us a way of basically culling all the interiors of the ships when we've got a massive fleet battle going on and we don't really care exactly how the interior looks just as long as the right color and you know rough shape of the geometry inside so that would help and that's that would save us the lighting cost of all this interiors um which was yeah but i think largely it's a cpu issue like you and i said um gen 12 it'd be that definitely are part of the problem we'd have to see profile what the likes
00:39:38
of the physics and other threads are doing i'm sure they'd have optimizations to do as well for everyone who's interested how these interiors could be rendered just check out the some footages from from the latest spider-man games on playstation how they do it that's an example on how you can do it how's the tech coming which replaces the windows of buildings with the 3d looking interior the new window shader how's that coming i think that's exactly what jonas mentioned i believe is that do they do that in the spider-man jonas staring at the question when you when you hit it so i was like i'll just ask it now yeah so
00:40:24
we already have um this kind of we call it interior shader in in our engine just as a standalone shader right now and i actually didn't know bef actually didn't know about it like a month ago before i've seen it in game because i had to fix a buck for 1012 on it so um i've seen it live in the engine and i was i was already impressed it looks pretty damn cool and we're already experimenting right now in engine um with it and so i'm optimistic that we gonna see it very soon and but yeah most it's mostly art art site now who has to implement it and yeah test and iterate on it and yeah make sure that it looks good and but yeah
00:41:12
from the technology side everything's working looking good so it's at the point where it's ready for artists to start playing with it and then they're going to discover their bugs and all the problems with it and then they'll come back to you to fix it exactly yeah yeah i mean they're already playing with it i'm excited to see it in engine it's definitely gonna give some more life to those buildings when you can actually see inside with all the great optimizations seen in do you have plans to have ship lights cast shadows in the near future also helmet lights i liked how they buttered you up with some compliments at the beginning and then asked something not related to it whatsoever but ship lights casting shadows that's a
00:42:00
big one um we've come back and we've gone back and forth a few times internally um i don't uh it's controlled by the art setup so i never know for sure what it has and hasn't been set up on each spaceship um but the problem with with the feature is the cost doesn't uh it's quite difficult to control the cost of this so one spaceship floating around in space if you put shadows in them lights no problem at all you park 50 of them all together pointing at each other in a you know a really busy location and you all get out your spaceships and jump around in front of them stack up boxes and all the crazy things people do in the pu um it's gonna start costing a serious amount of uh gpu to do that um so we have to figure out how we can add it yeah keep it under control in the more difficult scenarios um
00:42:47
and it's been suggested as we can like you know just fade these shadows out when they're too costly but that actually becomes much more complicated than you think because what you don't want to happen is have you moved your camera and look around the scene things are popping on and off all the time you want a stable consistent visual but we've got some ideas of how we might handle that and to try and uh avoid the issue um and we might apply the same idea for all type of dynamic lights from like ships your helmet light your flashlights um and muzzle flashes on guns there's another request to have shadows um all these things the muzzle flashes especially is something that it'd be great to have shadows from the muzzle flashes but you know if 50 people are firing in a room then maybe you can say oh we don't care so much um but it's things we could also just you know uh mostly we could provide to the user to say you know do you want these how many dynamic lights do you want to support
00:43:35
um let them choose if they want to pay that cost but there is also some internal budgets we couldn't um like we have basically like a texture pool where all the shadows go into once we've exceeded that there's just no more space um without sacrificing the quality in the rest of the level so yeah i'm hoping we can come up some type of system know how many dynamic lights do you want maybe like you know eight might be a a default or four or something like that and then you can scale up whatever you want if you want to take that hit when you have a really busy scenario um yeah so hopefully um but yeah the shift ones are particularly problematic because they can cast thousands of meters and cover huge areas um which means you know if you cover a whole space station there might be a few thousand draw calls in that space station which is like a instruction for the gpu you just double that by putting your lighter and you know 50 people putting 50 lights
00:44:24
at it you know 50 times the cost for that space station that's a big deal yeah uh besides the the the ship lights and the shot casting another bit another graphical tech thing that comes up quite often i've seen it a couple times in the chat already today and it was in the thread as well uh rtt's for to use as exterior cameras uh for ships um ideally one at the bottom to make landing a whole lot easier uh for folks uh is there anything you could you can you can tell us uh i'm also gonna point you to uh a todd pappy or someone again um or another design director maybe uh jungkook um i i've actually been pushing for this i want this um but with one giant gotcha i don't want a photo realistic camera
00:45:14
like you know like a i'd prefer um mainly because of cost i think you'll just you'll pay a frame rate cost when that happens um if you play a racing game like uh you know any fours or something like this they'll have like their wing mirrors and their review camera but they have heavily optimized their game to make that work and they've heavily optimized what happens in that review camera you'll notice they never look as good we have a hugely complicated game there's much more than just a roadway so we know we have orbit with gas clouds we have all sorts going on so to render all of that again periodically in a external camera it you know a worst case scenario could half your frame rate um which we don't want more sensible for me is that we do some kind of like lidar depth camera i think you'd actually be getting cars in real life and i think that'd be far more
00:46:02
useful to be honest because um imagine if you're landing a giant spaceship on like a space station you don't really have any sense of scale like it's an enormous beast and you're landing on what might be something that's like concrete or just tiled metal panels you don't know how close you are so i think something that really kind of gives you a great sense of depth even if there's not a lot of interest there or let's say their example might be landing on a snowy mountain you you won't be able to tell anything if we just show you a white picture underneath the spaceship so yeah i think a depth camera or lighthouse or some type of holographic you know some type of nice imaginative non-photoreal uh render would be very cheap and be cool so yeah i'd love that as long as design don't ask the photo bill then i'll hide lots of folks in chat
00:46:51
have been asking about ultra wide support scaling for things like chat hud personal inventory etc uh some some of these monitors are getting super ridiculously wide now 32 9 and stuff like that um what are the challenges involved what what can you guys tell us about uh better ultra wide support so i myself using a ultrawide support monitor so i know about the the issues um with star citizen um unfortunately we as graphics and engine we don't have too much to do with that it's more like a question for ui and how they implement it especially in the
00:47:39
building block system and yeah just how the ui scales with all those different monitor resolutions i'm not sure maybe ali knows a little bit more about that if there are priorities on that or no no it's not easy we've actually got something with the ui team scheduled here in the next i think the next two months i think so um okay prepare lots of questions for them then um there's a few graphical things like of ultrawide that we sometimes need to fix or improve the performance of maybe like um or sometimes the quality of maybe our post effects aren't always amazing and super wide and also you know if you super wide
00:48:28
depending on how we deal with the aspect ratio kind of more objects to render and obviously you have more pixels to render so you know gen 12 and other optimizations like dlss and things we've talked about would all help there but yeah in terms of just making it you know a fun experience with you know where you positioned the hud and the chat windows and stuff but yeah it's all on the ui team the end and we don't really have any control of that um let's see we've only got 10 minutes to go let's see what more we can squeeze in here uh are there any plans in the future to to dynamically use more vram for cards that have it available for example the game seems to use six gigs on my 390 but i have 24 gigs available would you also would more vram usage allow for better frames or higher quality textures yes
00:49:18
okay so first thing i'd say is um you don't want to use 24 gigs just because you can like you want to use what you need so the question is is the quality improvements we could make which way we could use them 24 gigs at the moment the system is such that basically it assumes it wants to achieve the resolution you ask for so if you ask for 4k it's going to spend a lot of money achieving 4k it looks what memory it's got left on your graphics card and it says okay i'll use all that's rest left on the meshes and textures and it will scale up the texture budget up until about three or four gigs i think the internal limit is of how much you lose on textures um we could uncap that that three gig limit and put it to 100 gigs or whatever and it'll just try
00:50:06
and load all the textures in the game um for you um the only reason there is we haven't really uncapped that is it doesn't usually lead to any visual improvement like three gigabytes tends to get you every single texture we have on screen loaded at the highest quality we have available so um raising up is just risks a little bit we might get some streaming issues where your hard drive just starts churning away and maybe you've got a fast hard drive it's okay but maybe you don't have such a fast hard drive and you're wondering why the game runs slow and because it's trying to load the 24 gigs of textures you've asked for which you don't really need so um that's where the situation today where we want to be though with uh once we've finished 1012 is to expose some of these numbers to the player instead so you pick your resolution and it shows you how much money is left and you pick how much text resolution you want to use um and then you can download upwards and
00:50:55
then if you want to hang yourself you know put 24 gigs of textures and maybe it does look better maybe it doesn't but we'll give that to user to pick how they uh distribute the memory um one thing i would say is one thing it would be nice which we can't promise booze something would definitely is a future option is to have a high-res texture pack um we offer all of our textures at double the resolution we actually ship them at um but there's some several issues in trying to achieve at the moment uh mainly just the size of the build would would obviously go enormously high um and that has costs associated with as well on the back end side there was there's some money uh a cost for bandwidth um so we also have some some tasks in the pipeline to reduce the size of the build um so you know you know we're this is stuff that's a higher level decision than we can make in our team but
00:51:43
maybe if we have detectors available and we have technology to use them we also save a bunch of memory on the build maybe we get to a point where it's more viable uh to have a high-risk texture pack and then the people that do have uh 20 gigs or whatever memory will have a very real reason to use that memory and you know if you say double the resolution of every texture that would be four times the texture memory so our three gigatexture limit might go to 12 gigs instead um at least so and 12 gigs once you add on 4k and a few other things you each get to 16 to 20 gigs of memory so so yeah this is a um something in the future definitely um there is some technologies as well in the future that we might use that could eat up some of that memory things like weight tracing and sdfs uh distance fields that are used for various pieces of technology
00:52:30
these all consume extra memory we just have to make sure the game can still scale so that the people that have six gigs or whatever can still run the game you know sensibly you understand i have to put everything ultramax ultra ultra right just all the sliders got to go up alley yeah not worth playing if i can't max everything out well interesting a lot of old games used to let you max out beyond what your gpu could do um i think some games now have taken the approach that we're talking about here where if you maximize it it will not let you like it but if you want to run it your game at 8k and to have all the textures loaded in the world it's not it's not going to happen there's no space physically there yeah it will kind of force you to live within the constraints of your gpu it should be a whole new like difficulty level settings in graphic settings that
00:53:17
you like like the the demon souls version of graphics settings you know that just that you know you run the risk of murdering your computer we definitely hope to have that yeah that's a definite goal we want all right let's see if we can drop some uh more engine related questions in before we get going uh let's see is it planned to support direct storage on windows 11 and do you expect a performance boost from it we have at least taken a look at that but we already have a custom technology in place to improve the memory transfer to the gpu so it's unclear if we can gain a lot of performance from that
00:54:06
that's something we will have to take that people looking to measure and check where it makes sense if it makes sense and then it will probably be implemented the diplomatic answer here all right this next question's about lots and and distance uh what's being done with this uh the pop in is honestly dreadful and seems the render distance for ground foliage trees rocks and such is so low that it is really jarring visually compared to anything else will this change or increase due to gen 12 or vocal implementation or what can you tell us overall lot popping um yeah i i'm seeing that too the thing is that's mostly um decided by art how they want to set it
00:54:56
up of course it's measured based on performance so but 1212 and vulcan it'll improve another performance and i i think we we will scale it up at one point but yeah i'm not sure what what exactly is planned there if there's anything official yet so yeah this is the cpu um plans or not not concrete but definitely um discussions of the planet tech team how we can take things one level higher um at the moment basically each object is considered an incidence an individual object in the scene um this type of spawned and managed the position of it um we do have some small scales grouping
00:55:46
where like we might group five or ten objects together but what you need if you want to say render an entire rainforest is the ability to have millions of trees and just to have the concept exist of an individual tree means that that becomes beyond the reach of a sensible speaking cure you need to have the concept of it this is a whole you know this is a patch of forest that might be a kilometer wide or 10 kilometers wide and the type of the trees either never existed individually and that type of a group or they are being spawned extremely quickly and then merged into one type of uber club that we just type of can submit to the gpu or even nightmares with gpu but like type of just render on the gpu without any cpu involvement but yeah but the current approach of type of having one object you know type of constructed and sent to the gp reach frame it just doesn't scale quite to the to the level
00:56:33
we need but yeah gen 12 will give us like a big boost on what would be possible with the current system but yeah those there's plans for the extreme lods we'd call them like you know the rainforest where you could just sleep for miles and miles of trees that need something slightly different for that to work and and all these objects on the planets they're already using a more specialized path to be rendered much quicker than our usual objects like all those brushes uh but yes i pointed out um if we want to have like really dense light forests like you've seen a microsoft flight simulator um there's definitely a lot of room to improve and definitely something we're looking at aware of and yeah hopefully we'll implement something in the future about
00:57:24
oh we're just about out of time so let's let's end with something a little more general for graphics engine performance gen 12 what's the what's the what's the immediate future look like like like like what what do you got on your what you got on your agenda for the next three six months here no dates no it just just what are you working towards right now well as i already pointed out the beginning we want to finish 1012 so there are few more milestones we have to finish the transparent rendering that's the biggest one and yeah once that's done i expect all the other things which are missing which more like a few compute passes uh like planet atmosphere stuff and like
00:58:14
compute skinning and a few other things we're probably uh half of a look at um are finished um we have then gentle offs done and i really excited to see how um how much performance we gained from that yeah um after that we can walk in which is not um lacking too far far behind because like in general we're um developing 1012 and vulcan parallel and vulcan is just lacking a little bit behind so once gen 12 is done um it shouldn't take us too long to even get a first version of vulcan working okay anything you want to add to that ally jonas
00:59:01
uh now we've got like uh we've got these guys have 10 12 on the entry team we've got um about half the graphics team on gen 12 as well um the ones that aren't on that um are working on a number of features mainly there's a bit of a backlog for s42 type of the critical features we need for that they'll obviously feed into the pu as well there's very little that would be specific to squadron it's more just you know who who is the prime you know need for for each feature but yeah there's a few features in the works for them some research for longer term uh features like uh jump points um to the graphical implementation of that things like uh improving the water effects is something that's on the kind of near-term goals as well and unifying then we have like a planet uh ocean
00:59:50
shader we have a water volume shader we have a river shader um these things all need work and love and integrate them with the vfx as well we don't we've lost in the in the in the transition of our engine over to you know all the improvements with major planets and stuff we've lost like splashes and dynamic waves and stuff like this so we need to reintroduce these types of technologies and um that's ibm mainly of a half team on gen 12 and when that's done they'll be on vulcan until that is running butter buttery smooth um and even then that team might in then look at the dls uh fsr type approaches and only then will the type of join the rest of the team on the uh the general graphics features cool um yeah two or two more things i would like to say really quickly um like one
01:00:39
of the questions was could we pay a beer to christopher balti so i want to have a small shout out to balti for all the nice improvements he did for 3.17 he he claimed all all those improvements for himself internally so give them a big shout out and one last thing i've promised one of my friends okay there's your dab yeah that's it all right i think i think they might have finally found out the way to make sure i never invite them back on the show ever again dab we're going to see ian leon dab on the next show just to get out of it
01:01:27
everybody uh thank you so much jonas sylvan and ally for taking time out of the end of your week to be here on the show folks if you haven't check out yesterday's inside star citizen uh it was a special tools time episode uh detailing uh the work of two of our tools teams one in montreal and one in uk on some recent things that they've done um it's it's it's not the flashiest show but it's the kind of thing that only star citizen whatever touched explore so those tend to be my favorite and so thank you so much for indulging me uh in in those really really esoteric behind the scenes stuff i will be back next week with a all sprint report ish episode of inside star citizen so that you have all that so you can get all your your cool visuals and stuff back uh and then we're back here uh
01:02:16
next week on scl i think it's the modular content team i'm not sure it's my i don't my schedule's all upside down right now uh but take care i remember that invictus launch week was just uh announced yesterday the day before uh check that out for the full schedule on what what ship manufacturers are going on on what day and uh yeah we'll see you next week everybody take care bye bye don't you dab on my show you

Transcript from YouTube’s automatic captions — punctuation and Star Citizen jargon are approximate. The video itself is hosted by CIG on the official channel.