#virtual-reality
1 messages ยท Page 75 of 1
well when people like McLarren and BMW want to have multiple people on their team collaborate together it can become important
?
It's a bug
They need to fix the bug
Not really hard logic there
The fact that people have to manually replicate or change source or use plugins, definitely means that the issue warrants the attention it deserves.
I mean you can just send a function call to the clients every tick and let them update the position
Yeah, there are plenty of workarounds
yeah seems like something thats an easy fix
But, what I am saying is, there shouldn't be.
I don't even think the engine should support it out of the box
Especially after trying to make VR super important...
thats the problem with big organizations. their "Triage" system is a little bit flawed
I mean, the bug has been around for over a year and plenty of workaround and PRs about and they can't accept a PR? Or fix it themselves?
simple stuff should be implemented in the game, not in the engine
no need to "fix" anything
yeah I agree about pr
yeah tough 1 to figure out which PR's to use and not use when you are aiming for specific feature releases. There is no proper automated testing in place when you do a PR though right?
Nope, they put it under review, test it out themselves
Usually people upvote PRs
make a big stink about it however they can
The current process is obviously showing its age
yeah thats the thing that slows them down a lot. with web stuff we always have to run it through unit tests and automated tests so you know the stuff compiles and works properly. makes it much easier to approve a PR
Right, and people are actually testing this though
The other community members
And will state usually if it's good or bad
yeah but there is no robot sitting there with a big red or green sign going yep its good to go.
Well, from the PRs that I've been paying attention to anyway
people only test that their specific feature is working. regression testing is something thats a lot more complex to do as a human
Right
I don't know, I think I am just annoyed that they make a big deal about stuff, but then abandon it
when i was doing enterprise grade stuff we would have to ensure 90% code coverage with tests. meaning 9/10 lines of code needed to be touched with a unit test. then you have integration tests and user acceptance tests. because you never know that maybe 1 line you change could break somehting completely unrelated a mile away
yeah thats why I joined up on the UDN. get much better answers and assistance much faster. plus helpful intro's. they just got me in touch with the HTC guys to try and get some of the vive trackers last night ๐
I just rather they fix stuff before moving onto other bigger features and leaving a trail of bugs and half ass implementations
Nice
@pearl tangle and how did you get access to udn?
yeah. pay them some money per seat essentially
is it possible for a single developer?
Yea
i went through the enterprise side of things since thats more the area im playing in with the events and other software projects but they will have different stuff for regular game development type stuff
you can buy as many seats as you want
is it fixed price?
doubt im allowed to talk about the pricing of it. you would have to get in touch with epic directly and they will let you know what suits your need
s
ok
I just wonder how much money you have to make with a game for it to be worth paying that money to epic for better support
I just think that if you have a game on steam and player ask about some bug that you can't fix because its some deep engine bug, then I think it would look reall unprofessional from the perspective of the players if they could follow the whole conversation you have with epic about that bug on the answerhub...
its kinda lame that epic wants 5% of your money and then not even gives you a private way of reporting bugs once you have a game released. doesn't really sound like a good system.
Considering you are getting acsess to a system that cost many many millions to develop.. I'd say it's pretty decent ๐
I don't think a player is going to care about how you fix a bug also
and anything that is that broken will normally get fixed pretty fast
yeah 5 odd years ago it was a 7 figure licensing fee
doesn't really matter how it was 5 years ago
for anyone having a game released there should be some private way to report bugs to epic
for "free"
I don't have a game released yet
its just that when I would have one released, I definitely wouldnt want players of the game to follow me on the answerhub and see all stuff I report to epic about what ue4 bugs I see in the game
well here's the thing.. there are tens of thousands of ue4 devs... all working with the system... if a massive bug comes up.. it'll most likely get reported... you should also spend most of your time testing as best you can. Also.. I really don't think players care that much.. they also DO not understand the game bugs are not the same as engine bugs...
massive bugs are fixed fast, yeah, but there are often bugs that no one else reported before
I see those all the time
especially when you do a game that isn't some sort of FPS
yeah man.. I get ya.. but I think your making a fuss of things that are not really an issue... you could also use a different name on the forums lol
well its not related to #vr anyways ๐
we have a bug right now that is minor and probably will never get fixed
it just happens
one persons bug is a feature to another (j/k sorta)
Some many extra optimizations for VR ๐ฆ
@full junco most bugs are better off being in public space for people to see and maybe somebody will give you an answer about it. when you are using the engine for free I don't think you should expect to have a customer support rep
@pearl tangle its not free, its 5% of revenue
you are developing with it for free, probably 90% of the stuff made with it doesn't pay epic anything actually
yeah, but I talk about when someone released a game somewhere and gives epic the 5%
yeah even when you release a game, majority of them released don't make over $50k so don't pay anything to Epic. if you are making enough money from it then you pay for dedicated support
why $50k? you pay epic 5% once you are over $3k
oh they changed it. used to be $50k. I guess they realised they weren't making anything on most stuff
it was never $50k for UE4
ah maybe i was thinking of udk
either way you aren't paying them enough for dedicated support like UDN if your game makes $50k a year thats going to be less than UDN access
if they need so much money just for giving you a way to privately report bugs then thats way too expensive, having something like the answerhub just in private shouldnt cause them any relevant amount of cost, so they should also offer it for free to anyone who gives them money through the 5%
thats my opinion, you can disagree
its also paying to have dedicated support which is monitoring your questions with a proper response time and assignment to specific staff responsible for those fields, assistance with publishing and pr stuff and other things which you get for paying for a service.
they dont need to offer that stuff for free, I only talk about a way to privately report bugs
i don't know that anybody has ever tried to bring that up to them before. since the answerhub is for the community to provide you support, not really epic staff since they don't monitor that or the forums so they would need some other kind of thing in place purely for bug reports i suppose
also.. free for VR if I remember right
the answerhub bug report sections is managed by epic
@restive blade whats free for VR?
ue4
nah epic doesn't have anything different for VR. it's steam that you can bypass greenlight if you are doing vr
@restive blade maybe you mean that oculus stuff
it's not supported by them. they dont monitor the regular answerhub. as in nobody is specifically assigned to monitor it or keep track of it. its meant to be more community managed than epic staff managed
the bug report section is
oh they look at it but there is no accounts guy monitoring it or anything to assign to a particular staff member like they have with UDN
yes there is
if you report a bug on the AH then it gets assigned to someone at epic
then they trie to reproduce it etc until its entered as a bug in their JIRA
yeah but they dont have any guaranteed response time or anything to it or a specific engine dev etc assigned to give you a response on it
yep but usual response is "yep we can reproduce it, it's added to the backlog" with the UDN it will be like yep we reproduced it, here's a way to get around it
if theres a way around it, they also mention it on the AH
if it's a known thing they that can fix quickly sure. the benefit with the UDN is they will assign somebody to the task to give you a fix straight away. even if it takes them a few days to do it. definitely worthwhile if you are doing some weird complex stuff with tight timeframes but as an indie dev 90% of the stuff is probably things other people have already ran into i suppose. just bring it up with them if you feel it's a big problem to have things public and maybe they will introduce something. they only started the enterprise stuff less than a year ago off the back of user requests
what "enterprise stuff"?
so they have an entire team setup for enterprise use now. so non gaming stuff. like the project they did with McLaren.
ok
since I do events based stuff and whatnot thats more what I needed since it's very weird problems you might come up against that nobody else does so they started that entire department off the back of a few requests
Service level agreements are expensive. And that's what you're paying for with UDN.
Answerhub is great. But that's for epic's convenience, not yours ๐
Your convenience is access to answerhub and forums and a multi-million dollar engine to build stuff off.
forward renderer doesnt support ambient cubemap, skylight I find too expensive and its annoyingly ugly too, so I found a better solution for getting a minimal brightness in shadowed areas
just directly adding it there in the shader
any comparison images you got there @full junco
@pearl tangle it looks 100% same like an ambient cubemap
but its probably 100 times faster
ambient cubemap actually was something like 0.2 ms per eye I think, this should be almost not measurable
could be pretty interesting fix then
its not a "fix"
its more like a new feature
UE4s .usf shader have to be very general-purpose
this is a solution for a very specific purpose
so they would never add something like this in there, but adding it manually is great when it makes sense for the game
I've never really looked through the .usf files before, so it actually took me a few hours to figure out how to get the result I want
but now I think theres actually a lot of stuff that can be removed there in the shaders to make it fit a very specific purpose (my game)
something like this was sugested for photogrammetry captures.. to help intergrate shadows into them properly
good to know. going to be doing some full scene photogrammetry stuff soon for a project
Well if u have any questions @pearl tangle just ask.. I've done quite a lot of it ๐
full scene photogrammetry or you have done it for objects to place in a scene?
my min light thing was buggy, since it was used for point lights too
@uti@zenith charm @graceful junco If any of you guys downloaded my demo project when i posted it, pls redownload. I made an update because i found that sometimes you would spawn in the calibration room without a body. Looking forward to hear if you find additional bugs
sent you pm @pearl tangle
was it here or another section where someone was mentioning a source control system that allowed you to do blueprint comparison between versions?
it was here. plastic or something i believe it was
nah haven't tried it before
git with git lfs is what I use but its far from perfect. If you are working on a project alone then it's not so bad but once you have a large team it's tricky
svn somewhere is another alternative but neither are really made for large assets, just code really
tried git shortly, didn't like it
tried Perforce, liked it but was frigging buggy and kept timing out
tried svn, cant recall why i didn't continue with that
that Plastic stuff sure looks promising
especially with the blueprint compare plugin thingie
git works fine if you know how to use it. just use sourcetree or something and bitbucket or gitlab and you are fine. there is already a diff engine inside unreal directly for most stuff but yeah the plastic plugin looks like it does a better job. i have just never used plastic before so can't really comment on it as a vcs in general
i'll play around with them all again a bit i think
in the end, i suppose it's worth finding a good solution for
yeah unreal still struggles to have a highly efficient version control system integration. perforce is probably the best but licensing is expensive for indies
thought perforce was free for up to 5 members?
atleast i know i didn't pay a dime for using it
aha
but again, for me atleast it kept timing out during updates
could be because i was hosting it on my labtop next to me
quite possibly
i mostly do events based stuff. but have a few augmented reality games potentially for a client and some enterprise stuff with photogrammetry scenes coming up
working through showreel at the moment
https://www.youtube.com/watch?v=9HJfLT-41Vc 360 showreel
I do bits and pieces but I have a 3D guy in my team who does all the complex modelling and I have 1 other programming guy
Well yeah they come in handy if you are building games..Who would have thought
@eternal inlet got the link again?
thx, checking it out soon.
cool thx
btw is there any obvious reasons not to use ie. a plane with a translucent material on top with "disable depth test" instead of stereo layers? as far as i can tell, they pretty allow you to do the same
any news about Zeni vs Oculus ?
@eternal inlet. Just played your demo. Here are my thoughts. The full body tracking is interesting. First time I've been in a game which does that. The downwards view of the chest is fine. The legs obviously don't follow. Not sure, but not animating the legs might be better than animating them. I couldn't see my right leg move, just seeing the left leg a little bit. I kinda felt like it's ok to never see the legs at all when walking. Maybe try hiding the legs and just use them for shadow casting. That might not be an option though with a mirror or multiplayer though.
Not sure what the mirror was about. Didn't work properly. Scale and position of the mirrored mesh were completely off.
The hands of the mesh are following the controllers, but they need to be rotated and shifted a bit. I made a fist and it felt like the position of the hands in game is where my controllers are, not my real hands. The real hands are a bit further down on the controller, not on local 0,0,0 of the controller component.
@wintry escarp http://www.forbes.com/sites/gordonkelly/2017/01/23/samsung-galaxy-s8-release-date/#5297429620bc
And the hands need a little bit of rotation too. Maybe everyone grabs the controllers differently, don't know. Felt just a tiny bit off.
But in general I like the tracking of the hands/arms.
When walking into a mesh, the screen turn black, which is fine. But when moving the head out again it seems like the player is teleported a bit, moved outside the mesh. That feels a bit weird cause if you want to move your head into the mesh again you need to move further than before. I wouldn't move the player.
The climbing grips are hard to grab at first. Maybe need a better indicator. At first I thought I'd just have to overlap my hand with the yellow meshes, but moving the hand too far into the mesh doesn't work. It seems like the finger tips need to be close to the edge on top.
Climbing is fine.
I'm not sure if anything can be done to make it feel more like climbing. Sometimes I had the feeling I'm pulling down the tower/wall, instead of pulling me up. That might be a level design issue though. Maybe seeing things move in the background helps. I'm not sure.
I couldn't climb the overhang unfortunately. I just couldn't find a spot for my hands to grab. I managed to grab one overhanging yellow mesh with one had, just couldn't grab a second one.
Falling feels good.
Not sure about flashing the screen black on impact. Might need to experiment with the impact effect to see what feels best.
Cool observations ๐๐ผ
I will add a sound or something to convey the fact that you actually get teleported out of a wall if you stay more than 1 sec in it
One more thing about the climbing that was bugging me. When you grab with one hand, then grab with the other, the first hand lets go although you're still holding the trigger button. It would be nice, if you let go of the second hand, the first hand attaches again, so you don't fall down, while physically still holding your hand and trigger button where you grabbed with that hand.
Yah, another issue with two hand climbing.. i actually had an implementation at first where both hands kept holding. And if you moved hands apart in real life, one of the hands got pulled off (would let go)
Didnt re implement that after i refactored it all the last weekend
Your suggestion is however definetely worth a try tho.. i like the idea
Have you played The Climb?
About grab locations. Good points on the difficuly to find the correct spot to hold. Will revisit it
Nope i should buy it hehe
But i saw videos of it
I played it at my borther's. I don't remember exactly anymore how they did the climbing but both hands were definitely holding. You didn't fall, when letting go with one hand and still holding with the other.
So does the overhang climbing actually work? Was I just too stupid to figure it where to grab? Or is that indeed impossible atm?
As far as i could tell atleast
It is possible, but can just hold the ledge
The holds are leftover holds from when i had collision volume implementation
I changed it last week to look at angles of what your hand hovers over
And only on meshes tagged with "climbable" to administer what i want to allow the player to actually climb
Will however try.. with my feeble 3d modelling skills to mold a few actual climbing holds ๐
๐
Wish me luck! Im gonna need it
You could use the handles of some marketplace furniture ๐
Thats too easy ๐
I'm sure there are some meshes you could use from the free epic example projects too.
But i'll keep it in mind when i tear my hair out once i get to the uv unwrapping lol
How did the flying feel btw?
flying?
Oh, didn't know. I'll test that right away
Also remember u have a nice little menu behind ur left shoulder
Oh, also didn't notice that ๐
I am looking for either broken beyond repair and RMA Touch controller. Alternatively, can someone with Touch controller do photogrammetry ? I need to 3D print Touch so I could use it to test my accessory design
there are 3d models of a touch, real life sized, in ue4
with buttons and all ? I need to make exact replica to be able to fit accessory.
@eternal inlet K, more feedback. Didn't know there was walking and running in there too. The walking/running animation is nice. Seeing your feet move in that case is cool. I guess it looks a bit odd when physically moving through the room because the speed is so low?
A few issues though with the hand movement. When you move your hands back and forth to run, you speed up and slow down all the time, which makes the movement not very smooth.
Excellent point. Noted! ๐๐ผ
Also, moving your hands sideways changes speed, which is weird
I only just got the S7, I wouldn't be buying an S8 right away anyway, if at all.
i like to get a few years out of a phone
That also means, when you move your hand in a circle, like from your side to front, then the other side, you move forward and at a strange speed.
@wintry escarp well, if you want to develop, you probably should get it - whole new level of power in S8
I think I also wouldn't rumble the controllers when running. Keep that for the jetpack, makes it feel like you're actually using an engine.
yes
Thats a bug
Hang on, how are you supposed to run?
Did u have jet sound playing too?
I held down grip and moved my arms
The rumbling is a bug
The swinging is supposed tocallow u to run/walk yes
Zaptruder had that jet bug too when running/walking when he tested it for me
I cant reproduce it myself for some reason
Ah, didn't notice the sound the first two times. Indeed, the sound is nice.
Ah, I see.
At first, just hold grips and moving arms = running. No sound, no rumbling. If you use the jepack once though, then just holding grip triggers the sound and rumbling
Also, when using the jetpack and touching the ground, the walking animation is played. It shouldn't in that case imo.
I'll check in later. Otw to some social rl stuff. But thx for taking the time to test it out @graceful junco
Np, You're welcome.
Also, specifically in this situation your logic fails to correctly check if the character is falling or not.
Your menu is pretty btw. Really cool that the hand blends to pointing a finger when getting closer. I like that a lot. The pressing of buttons is really nice the way you did the menu.
@mighty carbon no matter what you do you won't be able to ensure exact 1:1 with a touch controller w/o having one.
even the 3d models could be exact only to the mm
and you may need more
@normal thorn a model made using photogrammetry and dimensions taken from actual Touch controller will be precise enough to make an accessory that attached well to the actual Touch controller (I'd 3D print controller shell first, making sure it's the right size, then deal with accessory)
I can't justify spending $200 on Touch without having Rift and VR-ready GPU
(and Oculus isn't willing to provide broken and gutted Touch; or at least my contact at Oculus was not willing to ask on my behalf or get me in touch with someone who could answer my question)
quality content makes money
and it got frontpaged on both oculus and steam vr
plus zombies still sell
and its quite cool
has anyone gotten the pc mirror to be non stereo?
@south dove Non-stereo? Do you mean widescreen?
just not the double eye mirror
You're talking about the vive, right?
https://www.youtube.com/watch?v=YrQOi6r3cbs << what kind of PC does one need to run this kind of visuals o.O
and that gives me single cam, but its still tunnel vision
Oh, no idea about rift, sorry.
well i mean its the same stuff
theres no difference in what im doing between vive and rift
I know steamvr has a setting for one eye, two eye or flat crop mirror. That's independent of UE4.
@wicked oak ^^
do you think it's a teaser to get people's attention and it won't look like that in VR ?
mode 3 is not warpsed
mode 2 is for a monoscopic eye , but it is a bit warped
mode 3 has NO distortion, but its quite low fov, be VERY careful
this is mode 2
this is mode 1
this is all on mode 3
mode 3 didnt work for me
@mighty carbon I'm working on a cave scene with a bunch of foliage in it. I can tell you, that it looks better on the PC screen than in VR. Foliage, done with meshes, looks really aliased unless you turn screen percentage and AA up a lot (which is hard). And foliage done with textures looks like textured planes, even if quite far away. But that video looks really nice, if performance works for them.
this are all on oculus
hmm i guess if the fov is small on the 3 Ill just stick with the distorted 2
thaks
thanks @wicked oak
@graceful junco I see.. I don't expect for foliage to look realistic in VR, but as long as it's "believable", I am cool with that ๐ I am just pointing out that video has sooo many actors, geometry and materials in it that I can't see it performing well in VR
Nah, seems possible. With FXAA.
Maybe with a 1080 or Titan X even with high screen resolution.
I'll try to imlement those bloom changes now to reduce it to 1 low res blur
This runs at barely 90 fps on my GTX 970. At 100 % screen resolution and FXAA. I'm not finished optimizing yet. A big part of the rendering cost is due to the water. They don't seem to have water in that video.
nice, but imagine you add AI and effects and 3D HRTF sound... Probably all that will eat at performance too
Yeah, but AI and sound doesn't run on GPU ๐
@raven halo did you already implement those bloom changes?
I don't think I will get to test it
why?
I have turned off mobileHDR
ah
I did the stuff now and it appears to work in 2D at least: http://puu.sh/tywjp/f4534aa51b.png
that looks very good!
and wow, my game just looks so much better with bloom enabled again!
already forgot how awesome bloom is
had it disabled for so many months now
bloom is very awesome indeed
bloom is always a post process
with the change bloom now takes 0.24 ms (both eyes together)
protip for VR
disaable all postprocess you can
also it hates translucency
i had a lot of issues with the particle fx in my game
due to overdraw
and i mean stuff lik 5ms to render a fireball
this kind of bloom doesn't require HDR
I know this for a fact, because that's what we implemented in Storm Engine 2 (Doom 3 BFG) a way before we added HDR
also, this kind of flares would perform really well in VR https://simonschreibt.de/gat/doom-3-volumetric-glow/
have you played Doom 3 BFG in VR ?
btw, is there a way to make actor render only in one eye ?
Newb question, 4.14.3, max fps seems to be 50 FPS, on a 1070, that's pretty much a blank map, I used the optimizations from the BP Template Project, are there any other recommendations?
something is wrong there
reinstall engine, create new project
check yoyu arent on a laptop with power saving on
Reboot first, before starting to reinstall software :). Several times I've had performance issues go away by rebooting.
doesn't it switch to a mobile optimised bloom for mobile platforms?
Nah
No idea what it was tbh
I was getting sub 1FPS readings soemtimes lol
Like .75 FPS
POINT 75
I had enabled Forward Rendering(shading), so not sure if there was something in memory
causing conflicts
it's at 90FPS now, so whatever ยฏ_(ใ)_/ยฏ
If it happens again, I'll try to properly replicate and submit bug
@full junco bloom is awesome in 4.15?
@granite jacinth nothing changed to 4.14 regarding bloom
You were saying bloom was more efficient
Was wondering what version you were talking about
@full junco
I talked about a source modification
The doc about bloom says:
https://docs.unrealengine.com/latest/INT/Engine/Rendering/PostProcessEffects/Bloom/
I would like to reduce the amount of Gaussian blurs to 1 to save performance. How to do that? I did not find any settings that affect the performance of the bloom. All settings in the PP volume are purely cosmetically it seems with no impact on performance (apart from setting Intensity to 0). Then there is r.bloomquality, 0 disables bloom and 1 is default. And r.bloom.cross,
@granite jacinth
Deus Ex VR and Arizona Sunshine with slidey locomotion. Today is a good day in VR land.
DeuxEx VR?
Yeah. It's a 'VR experience' i.e. a look around environments affair.
do the characters even animate or is it just a static render?
Got no idea.
But I love Deux Ex environments.
So happy to have a perv on deus ex environments simulator.
sounds like its just static scenes and you walk around with the joystick, no teleport or other movement. all the negative reviews are people being sick
lolol
Tell you what - the Golf VR Simulator is pretty darn good.
Still needs work obviously, but it's actually pretty solid.
which 1 is that?
Dayum.
Deus Ex VR super rough around the edges.
Scenes were pretty. But well, not exactly optimized for VR.
Real problem is that they implemented a 2014 gamepad style control scheme for VR.
No capsule movement while moving around in roomscale.
Motion relative to head not hands.
No motion controllers visually available (although you can use them).
Both simultaneously immersive and unimmersive.
Pretty environments, but ultimately, couldn't get over the super-meh interactions.
i.e. movement, collision, no hands
Also shiny shaders need a lot more detail for VR. That shit just looks like cling wrap around things.
More detail or more polygons.
yep sounds like they just threw in the most basic possible implementation of VR without even spending a day to add controller schemes
Yeah. Pretty disappointing. Like thanks, it's fun to be able to see awesome environments in 3D. But maybe think about spending a bit of extra time to make it a worthwhile experience, instead of something that'll detract from your own promotion?
'Thanks for viewing! NOW BUY OUR GAME! WE PROMISE IT WON'T BE AS JANKY AS THIS VR EXPERIENCE. Maybe.'
seems like a lot of guys will jump on that for VR just to show stuff off. Would not take them too much time to implement a few extra pieces to it that make it a really interesting thing for VR.
Yeah really. I mean, there's already some work put in to get it to that point.
It doesn't take a shit load more to remove unnecessary collision (if they don't want to move the capsule around)
Or make things more friendly for the intended control scheme (i.e. motion controller + roomscale)
Or take out effects like sun shafts that don't play well in VR.
It's like it was made by a dude sitting at his desk, and his testing amounted to - ok, it runs in display mirror. Excellent.
Part of the problem with developing for VR currently is the fact that all the heavy development tools are still bassed in a 2D realm.... and graphics look very different untill you get them into VR systems. The itteration time is far longer
Kinda true. Once you get a sense of what works and what doesn't work in VR, you're quicker to pick up on them while in 2D creation too.
i.e. you know that players are more willing and able to stick their heads up close to things
and can scrutinize from more angles.
but even anybody that has spent any amount of time with a vive or a rift knows that you should have some different control schemes to that
But yeah, that.
Oh Arizona Sunshine.
Setting aside the nice graphics and that it's kinda the first big campaign shooter of decent quality.
There are a lot of things wanting. Like the menu. It's abominable.
Guess how you open the menu in that game. And what happens when you do.
Well, if you haven't played it, you couldn't possibly guess at the extent of how wretched it is.
When you're in game... you get a watch on your wrist.
That lets you know your health and stuff. Cool. Put your gun away and press the menu with a short range widget interaction coming from your finger. Also not bad.
Click on the menu - game asks if you're sure you want to go to the menu. Uh, sure.
Click on the menu... and you're booted out of the game and loaded back into the lobby trailer.
Where you have to go and put a virtual game cartridge named 'settings' into their little loader.
How do you exit the menu? You eject the settings cartridge and put the campaign cartridge back in. And then it loads the game from the last checkpoint.
Like... I get that you're proud of your cool gimmicky menu. But maybe that's not the best way to change basic game settings.
Locomotion mode that they implemented today. Copies Onward locomotion. Sensible right? Well, I guess so. But despite how frequently people talk about it, it's kinda poop. Touch to walk, press to run. Easy, what could go wrong?
Well for one thing your forward movement vector is all over the place.
Ideally, I should be able to hold my hand in a relaxed position and locomote easily like that. Nope. If you hold your hand down by your side, the forward vector is now pointing backwards.
Also, the touchpad is a big surface to rest your thumb on.
i haven't tried it on my vive yet, was just playing with touch and felt too restricted with the forward facing and teleport needing to focus on direction constantly.
keen to play on my vive, was enjoying the game itself in single player
actually one of the biggest problems with VR is teliporting... its far to reliant on it
yeah not a great solution thats for sure, but at least with vive you don't care about the direction change with teleporting so much. like il teleport and then manually turn around. whereas with the touch I can't do that, you have to rotate the teleporter very precisely each time and make sure you are facing the right way
@restive blade Definetly trying to change that
Just tried to put my mother through some VR climbing. Didn't go well ๐
But at least she didn't get motion sick despite constant falling.
She didn't quite understand the analogy of using hands to pull yourself around in VR. "mum, grab that corner" "But it's moving towards me now!" "Because you're pulling it towards you!"
"Why am I dropping when I let go?"
To be fair, a good system would train users up to do what I was trying to get her to do.
I kinda just threw her at the climbing wall and said: Goooo!
@clever sky p00p or not, Arizona Sunshine made $1.4M
@mighty carbon It was a great game. It just has a lot of sub-optimal quirks relating to its design choices.
I see
A lot of things that could've been done better. Still good fun.
can interfaces not have more than one function attached to them?
For some reason only one of my functions is showing up when I'm calling the event
yeah seems like those gimmicky things are good fun to develop, but maybe not so practical
in the new shitty yahoo episodes of community the dean gets a VR system and has to like run around and climb a filing cabinet to get the serial number. then he wants to delete it so he grabs it and goes down to a water fountain and drowns the file to delete it
@pearl tangle Climbing is super useful, ironically not for climbing, but for navigating past waist and chest high obstacles.
hah its not useful if you needed to climb a wall to choose a level then run across the level to press the start button
Are you talking about Run VR? ๐
Just probably had 40+ people try our game
And like hundred saw it
If you are a dev and have a local@meet up group.
GO
It was awesome
Also they will find bugs.
whats your game @odd garnet
@clever sky i was ttalking about gimmicky interactions for the sake of the cool gimmick but it not actually being efficient
there is a reason that we don't put all our stuff on floppy disks anymore, it's easier to just click a folder hah
@pearl tangle But I like my open and enter doors to enter a new level metaphor ๐
Seriously though, there's room for well designed physical interactions to perform game functions. But I think they should be used sparsely with good reason.
Like... you don't want to reduce everything to efficiency for efficiency's sake, because the great thing about VR is interaction immersiveness.
So something like Solus project which reduces interactions to hold in hand and point and click to interact with world kinda falls short of that duality (although to be fair that was less of a design choice and more of a practical system limtation).
e.g. in Solus, you get a torch in your hand and you can light things on fire with it... but you don't do so by putting the flame close to the object to light on fire, but by pointing the widget interaction pointer at it and clicking with the trigger.
@odd garnet Did you have to bring all your gear? Sounds great though... wish I was in a bigger city for this kinda thing. @pearl tangle Does Singapore have this sort of VR meetup thing going on?
Yeah Zap... this other guy came in late and fucking took the main stage and we had to change our lighthouse setup for him too
He was a decent guy tho. Is making a game called Molecular where you put elements together to for molecules
Education deal
Ah sounds cool.
How many hours of playtesting did you guys get?
Difficult to be in my position. My stuff needs a lot of testing, and basically I've only been able to infer from text based feedback and a couple of uploaded videos.
As well as testing from 10 or so in person.
Still... worked pretty well for a good portion of people!
Like 4?
Pretty solid!
We gave out 3D printed gems if people got to the end
Overall great!
I had one guy thank me for having only trackpad movement, one complain about it.
Rest were smiling faces when they took of the headset.
It was cool cause this one girl straight up understood how to play the game and wrecked.
First person to beat our boss
Nice.
Yeah, VR is a funny thing. Some people get it without having to learn.
And others are like... this is strange and unusual. But it's like a computer right, so lets apply that computer learning to this ๐
"It's all ones and zeros maaaannnnn"
It was at a bar.
If anyone wants to make some easy money, offer to bring your computer to bars for a cut.
@clever sky taking a break but when I get back; onto climbing mechanics
@clever sky yeah there is a VR meetup im going to be demoing some stuff at next week actually, seems pretty popular but I have never had time to go before
@odd garnet do you know if that molecular guy has a demo up? I have built something relatively similar to that actually
Nice. Let me know how many turn up? ๐
there are 1000 odd people in the meetup group, i think about 70 have registered for this event
im just going to be showing our showcase stuff in gear vr's and maybe give away a couple dozen cardboard headsets or something. don't really feel like manning a proper Vive setup in a small crowded space
with the skiing game i built that was at a 2 events we had around 100 people a day for 6 days in it. I have go pro footage of nearly all of it, some funny stuff in there hah
Nice ๐ Must be fun. Also exhausting ๐
ah i don't stay there and do it. Just hand over to the clients and sit back in the hotel room watching the analytics to make sure its running ok ๐
@pearl tangle Fair call. Sounds better ๐
yeah you don't want ot be stuck maning a booth for a week straight thats for sure. watch the videos in timelapse and use the captured audio from the staff for blackmail later ๐
so in VR it's a good practice to use stationary lights over dynamic
I mean static
I wonder if stationary are actually better than dynamic performance-wise
stationary lights give better performance than static in some instances.
what WHAT
stationary in my imagination is like static+movable performance-wise
now I hear they might be EVEN better than static given the right situations
my mind is officially screwed
yeah i cant remember the situations off the top of my head but i recall thinking that was weird too
and stationary you can't move but you can change their colour
and stationary can have dynamic shadows which static cant. so you should be using a mixture of those pretty much and never have overlapping movable lights or it kills performance
static all the times you can
VR hates stationary lights with a passion
becouse they are "dynamic"
they calculate all the specular and stuff at runtime, they only have their shadows baked
static lights have zero runtime cost
one thng you can do is use static lights liberally
for fill lighting and that kind of things
and use stationary lights when you need some light to be dynamic and lit dynamic objects well
given the massive resolutions in vr, dynamic lights absolutely murder it
DWVR is 100% static lights
dynamic lights should be fine with forward renderer
problem is dynamic lights with shadows
or shadows in general
Epic said that stationary lights aren't that taxing
I used it in Gear VR
(1 light)
hi people! Do you have any idea how to define collision for VR character? I see that character collision area is roomscale's center
And VR pawn doesnt allow child objects collision
btw, I don't know if reddit is full of trolls, but there are a lot of people claiming conventional FPS locomotion doesn't make them sick.. Vocal minority ?
Ali, I'd personally say a sphere collider on the camera for head collision, The way I have it setup in Unity is basically a sphere collider as a trigger, when it overlaps with anything that has the tag "Environment" it fades your view to black to stop you looking through walls. For body collision, not sure yet. Maybe the average location of the character? So it doesnt freak out and lock to players heads forcing them to have to learn gymnastics to play your game
vocal minority
fps locomition makes 95% of the community barf or something like that
the 5% gets super annoyed becouse they prefer FPS movement and games dont implement it
so yup, vocal minority and a massive annoyance
Your talking Onward styled thumbpad motion?
the different kinds
it fits in some games, and doesnt fit in others. The goddamn fps movement squad wants fps movement in EVERYTHING
and complain if its not there
yeah i spend hours a day in VR and back in DK1 days had no issues with that kind of locomotion. but now with proper positional tracking can't really do it
im not into teleporting around everywhere so much but its a better solution than using a joystick for most people
I dont mind it as long as I can move with it
like in H3VR, Ill walk with it and I wont projectile vomit
https://forums.unrealengine.com/showthread.php?133621-Unreal-Engine-4-15-Preview&p=656186#post656186
PREVIEW!
A Preview of the upcoming 4.15 release is available now on the Launcher and Github. We have made this Preview available so that our developer-community can help us catch issues before the final release. As fixes are implemented, we will release updated previews throughout the development cycle. Please be aware that the preview releases are not fully quality tested, that they are still under heavy active development, and that they should be considered as unstable until the final
Preview #2
Btw, just read in a new leak that new Gear VR is coming with S8
I wonder what new features it will bring
I think I fixed the biggest issue a lot of people had with my locomotion system ๐
Walking sensitivity is high!
Doesn't take a lot of movement to 'walk' around now.
๐ fair enough!
@pearl tangle surprisingly enough Galaxy S8 related leaks are abundant, but there is literally no leaks about new Gear VR ๐ฆ
Not sure if good thing
oh yeah but figured everybody knew it was coming
there are some interesting new pieces coming along for gear vr
hah could be a few bits and pieces
positional tracking too?! ๐
(outside-in I assume)
if that's indeed a thing, I sure hope we don't have to wait for UE 4.99 to get support for all of that
Probably best to dial back your expectations lest they be dashed ๐
Surely positional controllers and positional tracking is the goal of mobile VR... but it doesn't seem like an early mobile VR cycle thing.
I just would rather not spend $100 on Nolo VR and put it toward S8 and new Gear VR
As if any VR nowadays is convenient
none is, so why mobile VR should be different :/
Yeah. For some reason I just never thought to use it on the comment bubbles themselves.
I've been using it on the smaller comment nodes, which has been useful.
so how long does it usually take for GearVR to display the overheat warning?
is this to be expected after 10 minutes?
or should it not happen at all?
depends
S6 overheats quickly on complex scenes
S7 - not so much (and sometimes doesn't overheat)
waiting for funny videos of people crashing stuff around when using these trackers ๐
motorstep, is there anything that can be done to avoid it?
yeah - use minecraft style art
lol
seriously..
there is only so much you can do for S6
supposedly you have to be able to run for 45 min before overheat warning
that's Oculus standard
I am sure that if you don't meed that you still can release in Gallery
holy crap
my game is overheating after 10 minutes
and there is nothing...
it's an S6
well, it's UE4 :/
I haven't tested my game for over 5 min yet, so I can't say much
Does anyone know payout schedule for Oculus? Is it monthly as on Steam or something else ?
@pearl tangle not that I know of. Sorry
running VR on a passively cooled device is, well, just stupid design ๐
@full junco get a cooler. They sell them for Gear VR
CUVR Cooling Cover Fits Samsung Gear VR Fan Cooling Mod
I don't need one, I don't have a gear vr and my vive is not overheating
your Vive has no CPU/GPU/RAM/batter inside
yeah
Make your own cooler using a arduino of raspberry pi zero
Doesn't the vive have an additional USB port for that kinda stuff?
this thing is still not in 4.15: https://github.com/EpicGames/UnrealEngine/commit/af9d6262473492e4e9850133a128c95663ebb5be
poop
but theres a new interesting cvar in 4.15 for vive: https://github.com/EpicGames/UnrealEngine/commit/508b231845c94268e71a2506d709f6c2c2103e23
I wonder why all these performance optimizations always are for vive and not for rift
has anyone been in touch with microsoft about VR?
they sent an email around a few days ago for those in xbox ID
asking to fill out a form
vr.SteamVR.UsePostPresentHandoff seems to be very interesting, it reduces render thread time by more than 1 ms I think! http://puu.sh/tzVs8/6780dc9142.png
there I switched from 0 to 1
and you see render thread time got significantly less
on the microsoft thing
it has LOADS of GPU power
but just its shitty CPU but overclocked
good luck to do 90 fps VR (oculus/vive) on a 2 ghz Jaguar cpu
I don't see how CPU can be a bottleneck, I mean, it can, but most games will be surely GPU bound.
it's the main difference between a non-VR game and a VR game
it's render thread (due to the 2x draw calls) and gpu
render thread is CPU, and only runs on 1 core
doesn't instance rendering help in this case?
adds 1ms to the GPU
seems doable if you know the limitation from the beginning
in my game stuff like occlusion culling takes way longer than draw calls I think
and instanced stereo is absolutely not worth the cost here
but some cvar like vr.SteamVR.UsePostPresentHandoff is great of course, 1 ms less RT time for no additional cost
in the commit they write "this call is NOT safe for scenes that have frame-behind GPU work, like SceneCapture components and Widget Components"
but I tested it in a scene with a widget component and it all works fine
ue4 game code is all on 1 thread
so good luck if you have some AI or some complex-y gameplay code with stuff happenning
my 10 npcs with character movement are bottlenecking the game in my ps4 devkit
that's why Gear VR rules - larger market
yup, the largest
still a boring market
how so ?
you are too limited by hardware and you dont really get much feedback from players
limited hardware means easier to develop for (unless you are trying to make it look like DOOM or Dishonored 2)
no, it means harder to develop for
no
gameplay ideas on look + 1 button are hard
optimization is what takes most time
well, if you worked with Quake, then developing for Gear VR shouldn't be an issue
never played quake or doom
I'm not playing many games
@wicked oak not really. You can play FPS with 1 button. Just slow down pace.
any non-roomscale vr is kinda boring
(well, mechanics are going to have to change slightly too)
I think if a game worth it, people will purchase gamepads
I wish there were stats for MineCraft on Gear VR. It requires gamepad.
anyhow, I don't have good gamepad for my phone simply because there are no games i consider worth investing my time.
but if there was a game, I'd buy gamepad in a heartbeat
Dreadhalls made decent amount of money on Gear VR and it requires gamepad (I talked to the dev, that's how I know)
lol, look at the GPU requirements for that VR game
GF 600 series
how is that possible ?!
(aside from devs potentially just putting shit into specs section)
Hey! Anyone know where the documetation is for the Rift Audio Plugin? My google-fu seems to be failing me.
go to oculus.com, Developers, Learn, Documentation
thanks
np
anyone tried gearvr minecraft?
The "5 millions headsets shipped" number is not really very helpful.
Daily active users is the more useful number. Or total daily hours of use
The most profilfic VR tech is still cardboard. There are at least 100million headsets out there.
@full junco I know you've been obsessed with CPU/GPU profiling, have you thought about doing a wiki/tutorial on your findings, especially in regards to VR?
I've been obsessed with CPU/GPU profiling?
can't remember that, @wicked oak talked a lot more about it recently I think ๐
Here or forums
Yeah, probably so but, the problem is, there's crap for docs on it
A few videos, but they weren't VR related
what docs do you need about it? you only need to know how to open the profiler
Right, but what everything actually means
What those crazy sounding, sometimes abbreviated, names mean
etc etc
How to fix. etc etc
its usually clear what something means
tell me an example where it isnt clear?
and where google doesnt tell you what it means
I only thought about the GPU profiler, the CPU one isn't as clear, yeah
but you can't write docs or tutorials about that because everyone uses different stuff of the engine, so everyone sees different stuff there thats expensive
and if you dont know what something means, just ask in a fitting channel here and someone surely knows it
I guess more figuring out what exact things are causing the problems is the thing that the profiler doesn't really explain so much
It's not like it yells out that your mesh is too dense or you have overlapping lights etc
It'd be nice if it gave an itemized breakdown of cost based on object in the world.
This BP costs this much. The landscape costs this much. Your skysphere costs that much.
Ahh newbie question
Triggering an event to update a variable doesn't seem to impact the event tick checking for said variable.
So something is "untrue" then I trigger my controller to set it to true (via an interface call, if that matters) it triggers the bool and sets it to true. However that doesn't seem to matter unless I "release" my trigger.
This is odd.
put some breakpoints on your blueprint nodes and see where its falling apart
@clever sky @pearl tangle actually...
There's the Blueprint Profiler for BPs ๐
I've only used it a few times, but it tells you what functions are costing you
@granite jacinth Really? Cool stuff...
Is there documentation on its use?
@granite jacinth I have never had any problems with blueprints slowing things down at all I don't think, but handy to know for the future
the only blueprints I do that are crazily slow to do stuff are blutility stuff that I run in the editor
Yeah, it's very niche for usecase
But it works fine if you're trying to track down something that's slowing your game down
There's a lot of stuff that's barely ever mentioned
Like Network Profiler
that's another good one
Other than those four, I don't know of any other "Profilers"
But I'm sure there's tons, other than the Optimization Viewports
the gpu profiler was just recently expanded in 4.14 right?
no...
It's been around since forever
Oh expanded, my bad
Yeah, the live GPU profiler
I actually was trying that out yesterday when I was getting SUB 1FPS
I was like, wtf is causing me to get .75 FPS
hah. wtf was causing you to get that
NO IDEA
I just took someone's advice and restarted my PC
I had turned on Forward Shading (rendering)
Not sure if there was something bad in a cache somewhere
But yeah, it was bad
ah yeah the forward renderer can do some screwy stuff
Gnomon : Epic session on UE4 VR experiences , December 14th ( 214 mins ) โ
[ https://www.youtube.com/watch?v=s5qBd2ulBXw ]
--
GCAP '16 : Epic UE4 VR quick start session , December 20th ( 60 mins ) โ
[ https://www.youtube.com/watch?v=RWIQVTB6km0 ]
Nice stuff @noble cradle
Hey no worries nice to be here ..!
Haven't watched that last video but saw Chris Murphy here last week ..
They did a breakdown of "Showdown" and VR optimisation that was pretty good !
Ah nice
Also got into parallax occlusion mapping , and tessellation โ
Which was nice given it was meant to be a introduction to VR type talk ..
They came to SCAD this week
Need to read back on previous posts here โ
Though regards blueprint performance , "cooking" ( compiling ) them down to C++ is now out of experimental phase ..
Meaning Epic considers that feature shippable and complete. Chris mentioned that it's working well and you should get pretty close to native code performance from cooked blueprints ..
Yup
I have pretty much everything turned on now
Everything builds just fine, which is always good
Until I forget to check out my binaries
Ha -ha had a nightmare project recently โ
Where I built the project off an old UE4.13 prototype with some custom widget stuff ;
That wouldn't compile properly and basically remade it all in 4.14 to get it all packaged ok ..
there are still bugs with BP nativization, I see fixes all the time on github
Hmm ok
and c++ is cooler anyways ๐
Yeah , got to find some time to learn that !
What are you guys working on re: UE and VR ?
Me personally, I am working on two separate UE4 projects for school, One TechArt with Material Transitions and Particles
The other one is a 1v1 Vive Multiplayer Action-Arcade Platformer
Then I have my personal projects and one work project
One day, probably when I am dead, I will get a good night's sleep.
plus yyou have to actually play games and whatnot inbetween too right @granite jacinth thats where I tend to lose a lot of time hah
@pearl tangle sigh..
I wish, I try to, especially for research. The most games I play are during my game jam streams.
Oh, I forgot gamejams participation for other stuff I do in UE4
i haven't had time to participate in 1 for like 18 months, and i never actually had time to finish 1
There's just not enough time in the day
You should try to get back at it
I personally feel that each jam I grow more as a dev
Probably has to do with my crap long term memory ๐
My short term somehow packs it all in there, condenses it, and retains for future use in long term slots
ยฏ_(ใ)_/ยฏ
oh yeah i learnt heaps back when i did try them. I also have another developer and 3D modellers that can do stuff now in my team too. just whenever they come up seems to be right at the wrong time
It would be nice to see more VR games @pearl tangle
Sadly, I am usually doing jams with remote devs and they don't have VR setups
Which is what's going on with my 1v1 multiplayer thing at school anyway
the other 3 don't have a vive, one has a rift though
The school has a vive we can use to test MP parts so that's always nice
hah we have 5 vives here and 2 rifts so no worries there
Wheres the VR gloves already
there are a few different 1s around. Too annoying to use compared to a controller for a lot of things
anybody else do their vive tracker applications yet?
exept all gloves are going to fail
at least short-term
becouse they arent official, and due to that, no dev is going to bother implementing quality glove controls
something like tracked guns or a bat is much simpler to make in a game engine, but the implementation of gloves needs a few things
like gesture recognition and physics and all that fluff
and i can't use them at events and stuff because of hygeine and they take too long to take on and off
yup
they are cool, but until they are "official" maybe in gen 2-3, nobody is going to grab them
only arcades or things like investigation
where they make tailored aplications for a specific thing
think an app against parkinson, or an app to lean sign lenguage
but not general gaming
even the simple tracked guns or similar are going to have a hard time, and they are super cheap and super simple
if i made a vr gun tracker, i would pay the Onward guy to implement it
yeah for enterprise stuff I am looking into them, but they are also going to be used in combination with the vive trackers and custom arduino controllers and whatnot
a game like Onward would be incredible with a omni treadmill + gloves + tracked gun
pretty sure the US military would love that shit
VR gloves would be awesome for playing something like elite dangerous with a full custom cockpit that matches the vr
not quite
they have been doing that for 10 years or so now
well, if you combine them with a hotas, maybe
costs like $100k per kit though
using a air joystick feels SUUUUUPER bad
yeah im saying you make a full 100% replica of the VR cockpit
then its awesome
physical replica of it like a flight sim
but fairly infeasible
so you have to use the proper buttons and levers and whatnot
a possibility would be that the hotas manufacturers add a tracking puck
so you slap one of those and it appears in the exact same place while on the cockpit
it's doable. $100k and I could build it. maybe for an arcade over here i might convince somebody
but its still too much hardware
it would be pretty much perfect inmersion
for a thing like Elite Dangerous
there is no gravity in space anyway
well you should still have g forces
but its pretty much the dream setup for a space game
a moving cockpit, + tracked gloves + hotas on the same place as the ship controls + high quality VR
yeah thats where I am lucky in that I can build fully custom stuff for my things. working on a launch 1 for the S8 for huge event ๐
I just wanted gloves for a snowball fight game ๐
Its abit dangerous to get people to throw things when they are holding something they cant see, thats why those safety straps for motion controls exist
It doesnt really need to be a full articulated glove, a motion tracker in the palm could be enough for the basics
The world seems hung up on holding things with buttons, always with the buttons ๐
๐๐ป this is the international sign of the enter key ๐
I'll be truly happy when I (can spell) get a Blueprint wand so I can be a real wizard
Buttons allow for tactile feedback and motor control mastery.
They're very familiar to us. Easy to use.
And provide useful functionality.
You'd only get better with gloves once gloves provide robust haptics.
and even then, it's not entirely clear that people would prefer to slip on gloves than grab controllers.
I think we have a place where robust haptics is at its best, its called reality ๐
Yes, that doesn't really help in this situation though.
well if I had to put motor control mastery of button presses against say shooting a basket ball I know which one I would consider in higher regard ๐
Ok.
It still amazes me how Pool games are popular, its a reasonably simple game to copy but the haptics have never been anywhere near close, there are some games which do well but its far easier in the game to have a steady hand that in actually playing pool.
Why should it amaze? Pool is fun, computer game pool is fun. VR pool is as close a facsimile to the real thing as we can get currently. And of course, it's still fun.
Comparing VR to reality will leave you wanting for a long long time. But accepting that VR is immersive and compelling in its own right allows you to engage in the fun now.
I dunno, simulations of something I can do in reality dont really interest me
Well, that's fair enough. But the flipside is, even if you can play pool in real life, it's more convenient in many circumstances to play VR pool. A more obvious example would be racing cars.
I like car racing games because I can drink beer driving down the wrong side of the road singing my favourite tune at the top of my lungs while being chased by cops
I mean, I suppose I could do that in reality ๐
You can go race a car on a track in the real world, but it's expensive and time prohibitive.
And like you point out, you can do things in VR that don't have the same consequences in reality. For something as mundane as pool - don't need to have access to a pool table, or have it take up space in your home. Don't have to set up, or chase balls. Have access to AI opponents.
Yeah I understand it for that, or like say simulation of a space shuttle launch, that comes with a set of risks someone even with sufficient cash reserves may not be willing to take
Anyway, this conversation was tangential to the controller vs hand tracking discussion that spurred it.
Its amazing how complex the simplest games can be, like pong, in reality there are so many little things than can affect that ball, input is only the tip of the ice berg
Hand tracking is great, but at this point can only serve as supplementary, due to the lack of haptics
Due to the lack of muscle memory activated functionality. The lack of directional input for movement, etc.
Yeah its unfortunate that we will still end up with peripheral madness with VR
Maybe. We'll see how it goes.
Regarding pong/table tennis - it's pretty compelling in VR!
Well I dont think the touch controllers will be suited to everything, a racing game as an example is probably still best played with a wheel, stick and pedals even in VR
Of course.
But if it's a comparison between hand tracking and controllers, controllers win in that situation too.
those are still very expensive on their own too for good ones
Yeah. Niche market.
Will stay expensive.
But at least they're widely supported by driving games.
So they're worthwhile if you enjoy the genre.
in Spain you can findd PSVR in any store
they have more stock than what they need
after all, spanish salaries arent that high, and we are a bunch of casuals
Need to get me on that PSVR money train.
DWVR not for sure. but im looking to release VRMultigames ultimate edition in asia too
but gotta have to localize it of course
I think Sony will be quite interested if I can get this freedom locomotion stuff working well for PSMove ๐
it wont
I suppose I could get a PSVR since I got a PS4 now, not sure if there will be enough games
And it does!
Its gotta be better than the barren Oculus store ๐
@wicked oak why do you say that?
I can get it working with forward.
but with such an small tracking space?
forward facing only. Hell, I can use it seated.
im designing PSVR expecting people to be seated
uhm, if it works well seated, might do the trick
Yeah it does ๐ Most people were only interested in CAOTS
?
dash step = dashing teleportation steps.
Oh yeah.
That is a thing
Meanwhile PCVR has honeyselect.
And others.
Q: VR + Particles + Physics : Is it possible to make particles ( currently using a vector field system to make some nice effects thanks to @upbeat kestrel 's python script ) react to physics objects more than just collision, like being able to swipe them with VR tracked controllers to really get them to fly about from the hand waving at them?
hey simrak
the only way to affect particles with something other than collision that could work for that is to use vector fields attached to limbs
and enable global vector field influence in the particle system
we do that a lot on this project http://www.magicdancemirror.com/ to have your movement affect existing particles
use vector field components with a fitting vector field, attach them to your motion controllers
Thank you! I will investigate
@dusk vigil one important note: only GPU particles can be affected by vector fields
All righty
Some sneak peeks of what I'm up to : )
Very early prototype still, hoping to hit the stage where I can start showing it to potential investors in about a month
We are using Perception Neuron mocap suit and Hunter's Gesture VR plugin as the essential core of the game
cool ๐
So far I have been warming up to Unreal quite nicely, just started learning it about 3 months ago
lol, did you read comments to that Kotaku article ?
so many VR naysayers
still comparing it to 3D TV and 3D glasses
you mean VR isnt 3D glasses ๐
VR suit, I wonder when you'll be able to get full wirework rigs for VR
https://media.licdn.com/mpr/mpr/AAEAAQAAAAAAAAOZAAAAJDc0NjI1MTgyLTZlZDctNDA2Mi05YmY3LWU5ODI0ZmU0NzgxNg.jpg lol wow they wernt far off with this one
Thats Lawnmower man 1992
does anyone know payout schedule for Oculus ?
(monthly like Steam or quarterly as consoles)
Over the course of the last 50 years, capital markets have significantly codified a generational model approach consumer technology products. New โgenerationsโ of products are expected to come at a regular pace and provide increased features and reliability at the same prices. But depending upon the product type, the typical life cycle from one generation to โฆ
lol, this chat is getting more dead every day :/
it would be nice if racing wheels and flight sticks work on Android. Gear VR racing/flight game with proper controls would be awesome.
something like this https://community.infinite-flight.com/t/joystick-support-in-infinite-flight/13715
We recently added support for Joystick and other peripherals on Android and iOS in our latest 15.10 update. Here's what it looks like with the ideal setup: [image] Android What you need to reproduce this setup: Android Device Phone, tablet, game console... most should work. USB/OTG Cable This is a special cable that acts as a converter for Mini USB to Full Size USB. We recommend getting one that has multiple input. We have tested with this one but feel free to report the cables yo...
Yeah, why is this so dead, where are the devs! hah
I just happened to see this article: https://community.infinite-flight.com/t/joystick-support-in-infinite-flight/13715
Just imagine not only doing the same for UE4 games on Android, but also on Gear VR or Google Daydream!!!
Is it something that is already supported out of the box? If not, can Epic implement such support for flight sticks and racing wheels in UE4 ?
Thanks beforehand
missed opportunity btw
just imagine how cool it would be to make racing or flying game for Gear VR and use actual steering wheel or flight stick ?!?!
@wicked oak ^^^^
sweet
some documentation at last
so it is compatible with instanced stereo too
neat
too bad its 10X0 gpus only
becouse its nvidia only
ue4 really dislikes platform specific stuff
if its like Ps4 only, or Xbox only, its ok
but on PC, it has to be consistent no matter if you are AMD or Nvidia
that's not possible
why not to have a render path that kicks in on Nvidia, but not working on AMD. When AMD steps up the game and releases something similar, Epic would just add render path for AMD.
There will never be a unified solution until there are several vendors on the market
sooo, apparently NOte 8 will have 4k screen and some enhancements for VR
I hope it doesn't mean yet another Gear VR model :/
I think we found a really hacky way of doing Mixed Reality
it's untested but in principle it should work
Get your green screen and everything set up, get your camera in a fixed position.
Then using networking spawn a Spectator pawn that is locked in place relative to your player
match the spectators view to your camera view
green screen in the spectator pawn's view
boom
mixed reality
wow, Epic isn't going to update Oculus Mobile SDK in UE 4.15 to ver. 1.0.4 ๐ฆ
WTF?!?!?!?!
๐ฆ
Wow its almost like Vive > Oculus??
@odd garnet why?
I don't think this place is 'dead' just not as busy as it was when it first started
FYI I have 79 messages in here in the last 24 hours.
@odd garnet thats how most people and how epic have actually recommended to do mixed reality. they don't think it's a good idea to run a single machine for it and networked is the best way
Epic has a better relationship with HTC. They put me in touch with the head of tracking at HTC and within a few hours the guy had setup a meeting with me. Takes months to get anything out of Oculus support
Just PM Zuck on FB ๐
@mighty carbon that note 8 thing sounds interesting. not for gear vr, mobile 4k rendering won't work for a long time, but it means there are 4k screens available for desktop VR!
@pearl tangle is there anyone else who does it?
I'd love to be able to follow footsteps rather then re invent the wheel
That said; when I do it I'll show everyone and help if I can
@odd garnet there is a bunch of stuff up on the udn on that and there are a couple a PRs on GitHub that you can grab and merge in your build that will help