Memory effect, also known as battery effect, lazy battery effect, or battery memory, is an effect observed in nickel-cadmium and nickelβmetal hydride rechargeable batteries that causes them to hold less charge. It describes the situation in which nickel-cadmium batteries gradually lose their maximum energy capacity if they are repeatedly recharged after being only partially discharged. The battery appears to "remember" the smaller capacity.
#virtual-reality
1 messages Β· Page 96 of 1
well, nothing else going on, so we are filling the vacuum π
Li-Ion / Li-Po still degrade over time and use, though
See your smartphone for reference
yeh, except it's going to be a few years instead of a few month π
There's no denying the tech is vastly better π
Enerloops are the go to AA rechargeables for a reason π reliable and can last for a few years.
@fleet veldt if you get Li-Ion or Li-Poly AA batteries, please let us know how it works for you
Thats cool and all...but you literally can't bend over facing away from the tracker
but looks like directly underneath it, should be pretty good
yeah, but it's better than nothing π
Li-Ion require surge protection circuit. But Li-ION 14500 were always there.
they have relatively low storage capacity. Like 1300mAH at most.
plug charger into power strip and avoid power surge issues ?
@mighty carbon huh? see no connection
All about the power packs that propel your projects!
what do you need surge protection circuit for ?
@mighty carbon so they're not explode
@mighty carbon the link i've posted have all the info
yeah, and that article specifies that all of that taken care of by on-cell or on-charger protection
@mighty carbon you could overheat the cell by fast discharge too
you can, but how will that happen in Touch ?
@mighty carbon "probably" wont happen
if that was a risk, there would be no li-ion or li-poly batteries in AA format
oh, just figured out, 14500 are not AA...those are 3.7V
good to watch https://www.youtube.com/watch?v=mzOrPxDsIkU
meh...hoped 80% on gear
lol, they already dropped $200
nah, it's not available in my country anyway...
π¦ Not even through Amazon ?
"This item is not eligible for international shipping. "
what country are you in ?
Russia
Vive - yep, not sure about Gear VR, but probably too
I wonder if Rift couldn't pass Russian FCC (at some point it didn't) and that's why it's not being sold there π€
does anyone know if Onward use Unity Asset store assets for characters and anims ? (sounds like entire game is made of assets from the Asset Store)
pretty sure it does
o.O
its kind of the point with unity
its a different through process from UE4
where the market isn't as good, the engine is less modular, and people are more independant
though, lets be honest, for being c++, UE4 is amazingly modular
it sure is
im having no trouble at all with psvr stuff
the actual platform code, like ps4 or xbox one, is minimal
the absolute mayority of what i have to touch is on the morpheus plugin, mostly on MorpheusHMD.cpp
and thats about it
Alot of features that Unreal has built in is only available as a plugin / marketplace asset for unity too, so I guess that counts
Kind of have to use some to attain parity
I am just building a case for myself for getting assets from Marketplace and spending time on animation / effects and BPs (and building story/lore). So far the case is in favor of getting assets instead of building it all from scratch.
π
not VR, but would be handy to have
@mighty carbon look at the hub level in robo recall, probably over half of it is marketplace content
I see
I think bullet train used some too (could be why it couldn't be released as a free example)
there was a talk at Oculus connect about one of the early vive zombie shooter games that was written in UE4, I think they said they used a lot of marketplace content too
but that they did some modifications to a lot of it
in the robo recall office you can see they took the same fan from the saigon office but added some other details to it
(clinton crumpler was in their credits so he might have done some of the work for them with the original unbaked source assets too, but the majority of it is unchanged from that marketplace pack)
yeah
I am not sure if I'd modify too many of the assets I am after, but I'd certainly make more if needed in the same style
We just released our ue4 (forward renderer) VR experience
Check it out, I double dare you.
Looks really good! congrats π
thanks @fleet veldt
Are any of my Unreal Slackers homies going to be at SVVR this week?
I was the only programmer, but there are 4 artists worshipping me nearby π
And it was full time for about 12 weeks
So if you're a solo dev, don't fret
So did you do much work under the hood?
c++? The whole thing is BP
I was looking at the SteamRenderer to change aspect ratio... but didn't get it working in time
(aspect ratio of the display mirror I mean)
Or was that an american car joke?
Yeah
nice
woo, just put a hyper evo 212 fan on my cpu because my tempertures were high and wow what difference it made. I was getting 98 deg when building lighting etc and it was causing cpu to throttle speed, now it doesn't go over 45 deg and runs in turbo mode when building lighting. woo hoo
@deft badge Nice work! Was actually looking forward to that release... at least the trailer made it look like it'd be a nice exploration demo π
Thx man, were pretty proud of it.
Is it Vive only or Rift too? π
Just vive. It was released at a motor show. And we knew the hardware it would run on. Hence vive only.
Fair call π
It wouldn't be too hard to port. It's just motion controller location and trigger really
Yeah. Also menu button
I did not f much today, just watching downloading stats like an idiot
Haha π I still check the stats!
Actually I mostly check the reviews. I've gotten a half dozen in non-english languages π
Nice! We got a German one today
Hehe π Feels good to have an 'international audience'.
So yeah, between that and my reply from John Carmack, it's been a pretty cool week π
Sweet. How do you contact guys like Carmack? Just all casual and downlow?
#jcarmack Yo johnny, I got a sweet locomotion system, whatcha think?
#jjams sounds good. Make a game out of it.
Yep. Googled his email at oculus. Sent him an email with a juicy VR click bait subject and voila
Sweet. Yeah, it's nice to remember that these guys are actually contactable sometimes π
I did try to justify my existence in my email to him. I.e. Jjjams != nut job
Anyway he got back to me, which was super cool.
Nice.
So what kinda game you gonna make with it? FPS? Adventure? At liberty to say? π
FPS fo sho
Sweet.
It's both what ue4 is good at and it'll show off the movement system as well
Yeah. Movement systems are most fragile under high action stress..
So a good system would robustly allow for that.
I've been cruising around one of the infinity blade levels tonight
The grasslands one?
The Elven ruins
Is that the snow one?
Actually that is the grasslands one βοΈ
Nice.
Here's what I wrote to Lord Carmack:
Hi John,
I discovered a novel solution to nausea free locomotion in VR.
I'm a professional VR dev in Sydney Australia, and I know the issues and current attempts at solving this issue. I.e. Not a wack job
I have tested the solution on my VR colleagues to confirm my non wack jobiness. π
I'm almost certain that I hold the key that opens the door that leads to VR sickness free FPS games, and other unsolved locomotion questions.
Here's the catch (or benefit), the idea is easily replicated. So I don't know how to proceed.
So here I am, as a long time Gfx programmer and fan of your work reaching out for advice on how I might present this solution to the VR community and to hopefully set myself up to continue to explore the VR frontier.
I look forward to hearing back from you.
and his response:
It is almost impossible to personally profit from an innovative idea in isolation. In theory, patent protection is supposed to offer that path, but it just doesnβt work that way in practice. I donβt know anyone that has made money from a patent other than lawyers. Iβm sure they exist somewhere, but not in my sphere.
Creating the breakthrough killer app VR game is a way to make a few million dollars now, and possibly start a franchise that is the foundation of a powerful company in the future, but that requires doing hundreds of things right, regardless of if there is a fundamental key insight in one aspect of the gameplay. I profited not because I invented the FPS, but because I wrote Wolf, Doom, and Quake.
If your idea is good, it will certainly be copied, and you almost certainly wonβt be credited for it. Thatβs just the way things are.
I would suggest that you ship a product demonstrating the idea and write about it. Include a developer mode in your game that lets people switch between conventional controls and your idea so they can be A/B tested. Even if you canβt deliver Robo Recall level of polish to a game, if you really do have the ability to deliver free locomotion and exploration, there is a lot of powerful, low hanging fruit to grab. If it works, and you have been publicly advocating the technique, it may open some doors for you to deploy it with more resources.
Yeah, like I said, super stoked to hear back at all.
Very measured, intelligent response (all Carmack like)
But also, great to hear that his thought process largely mirrors my own direction! π
(i.e. Welp.... all I can do is make something awesome with it!)
So yeah... that's most likely the path I'll take too
Gonna have a chat with Lisa from HTC as well.
Also have a chat to the boss about maybe investing in a game with the new mechanic
Good luck!
I think at the end of the day, it'll turn out VR locomotion is a problem that is solved from multiple vectors. Just need to make sure the details are right.
Yeah, I agree. I have def used aspects of other locomotion systems to improve mine.
Bit by bit, piece by piece, we'll solve it.
Anyway, I better shut up about it. Otherwise it'll be a big ass'd disappointment when it flops π
Hehe π
@deft badge what do automotive brands think of VR? Worked with any other automotive brands?
This was my first gig as a pro
btw downloading it as we we speak
At the motorshow there were quite a few VR experiences though
It was the Geneva motorshow
S8 + new Gear VR are going to be unveiled today
Enjoyed it @deft badge !! Great job! What was the general feedback you got? Did first tiime users of VR love it? The environments look spectacular!
Lots of positive feedback. They went nuts over the animals.
@deft badge Just tried it. Really pretty. Kinda wish HMDs had better resolution π
But the opening scene had hella loading issues for me.
Constantly dropping into the steam loading screen until it got to the second scene.
Also, wish there was teleportation... couldn't get into the driver seat properly in my playspace π
Yeah it takes a while to load all the assets
We should have done more, but we were all sick of it by the time the motor show was done.
Yeah fair call. Guess it's not as bad if installed on a SSD.
Cheers to the team. Hard to tell where the polys ended and the skysphere started.
@mighty carbon gearvr, cheap audio, tell me all about it!
what should I be using?
oculus audio plugin?
FMOD
it's the only solution right now
and it's free if your budget is under $100k
(or check licensing terms to be precise)
what is the cheapest in terms of performance do you know?
in one of our maps, if we tun audio OFF it takes 1 hour to overheat
which is awesome
as soon as audio is on it takes 30 minutes
that's very cool!
@mighty carbon so you are saying there is no way for me to get the default audio in Unreal to "work"?
why is that?
I mean, it is working, it's just seems to be very expensive
should I turn off spatialization? should I switch to HRTF?
how about audio quality max channels?
there is no HRTF with UE4 on Gear VR
that's why I said FMOD is the only way to go
or wait for 4.16-4.17 and then you don't need anything but UE4 + Oculus plugin to get HRTF, occlusions, reflections and other goodies
if I don't want any of those goodies, what are my options?
should the default audio "just work"?
or are there performance problems?
like I said, if you want mono - use UE4. If you want proper sound - use FMOD with Oculus plugins
for 1 project both options are free, and I am sure FMOD + Oculus is more power efficient
(even if you don't care about HRTF, which you should as 50% of good VR is in audio)
nice stuff
this one is cool too:
@raven halo have you seen by chance nice ocean material that would work exceptionally well on Gear VR ?
funny you should ask
short answer is: maybe?
the proper answer is way longer
remember how I told you I'm hitting these last minute thermal problems?
my game is like a platformer type game
with different levels
in almost every level I'll have like a "bottomless pit"
if you fall there, you die.
we replaced that with a "black liquid" instead
so if you fall on it, you'll die.
I also have liquid platforms, they work similarly to the black liquid in the vertex shader
so anyway...
I've done some thermal tests and these are my results:
red means it's turned off or made invisible
green means it's turned on
the numbers are the minutes it took for the overheat warning to pop up
the only cost of the black liquid and platform liquid is in the vertex shader
I have no idea what to make out of that data though
So going back to your question: I don't know until I have figured out what is causing the problems
watching videos on Gear VR kills battery as fast
my charge went from 99% to 70% in like 10 min last night
sometimes it's a bug in Oculus software
were the videos already downloaded or being streamed in?
downloaded into phone's storage
yeah, but I also had tracking bugs, which I reported already
so it's not usual battery drain
I could watch entire 90 min movie and still have enough juice left in the phone
yeah, that's the bug.. Not occasional for me
@mighty carbon you seem to be the man for this question. I am trying to figure out how I would go about and design a POI(Point of Interest) interface to direct peoples attention toward a certain point(using arrows or something, not sure yet). Would you do this using stereolayers?(Rift, not gear vr)
Its important that it doesnt get hidden behind other objects in the world
if it's for Rift, why not to do object's highlight with an outline or something ?
Because, if you're standing with your back against it, it doesnt really get your attention
cant rely on audio
then yeah, use Stereo Layers (or render UMG into stereo layer)
personally I am more for not using HUD much in VR
doesn't help with immersion
Normally I would agree with you, but the experience we are designing this around(step-by-step guides) its important to have a clear consise system to keep leading people the right direction
aye, make sense
wow, stereo layers seeeeems very unstable
keeps crashing my 4.15
interesting π
Maybe its smarter to come up with a 3d widget system instead
stable on Gear VR π
you can easily make a 3d widget that is invisibile but renders to a layer
for super high quality vr interfaces
that is what I am doing at the moment... but it is crashtastic π
I am wondering if I should just do something like EPIC did in Robo recall
make a basic cylinder mesh with a texture that fades on if you're facing the wrong direction from the POI
that one is a layer i think
robo recall abuses the hell out of the layers
for all the hud
Hmm, as far as I remember its a mesh inside the VR pawn
that is not locked to the camera
I really like the POI system Robo Recall has, gotta try and see if I can pick apart how they did that ingame interface
I remember us testing those older Razer Blades and having massive issues, mainly due to the Optimus chipset and not having direct access to the NV chip
ah nevermind, only noticed the external gfx card bit now π sorry
@tawdry dragon We recently implemented a POI system where an actor in the world gets your attention by floating in front of you (NOT head locked) and pointing a line towards what you should pay attention to. It works quite well π
think of it like an annoying fairy floating around try to re-get your attention if you look away π "hey hey, no no look at me then this thing"
there is a "gaze threshold" involved to when it moves, so it doesnt just follow the center of your view constantly
haha that is pretty funny π
@heady parrot - the reason why I am going for an UI based POI system is because the objects we are gonna be handling are man sized, so if they start floating infront of you... well you know π
Some people might get an heart attack
Luckily I found the stuff they are using for offscreen warning in Robo recall and remade it in my own system π
no no, I dont mean the objects themselves π I mean things like an arrow
with stereo layers ---- thanks @mighty carbon π
I just hope those areant head locked layers π
the small arrows are π its not very big in your field of view
its 100% the same method EPIC used inside robo recall, I think it works quite well
great if it works for you, I just got a pet peeve for anything head locked.. so many issues
Agreed... Personally I would love to do some fucky stuff. But for this prototype readabillity and ease of use is above anything
So having UI, that cannot be occluded by the world, that show precise steps are very important to us
our target is not gamers or people who even know what VR is π
sure, but thats no excuse to give them eye fatigue or make them feel ill π
Well, if they are following my steps and do as they are told, they will never see the small arrow π
but its a valid point. Will have to test on my co-worker who gets easily motion sickness π
Anyways, I'm off! Have a good one
you too π
@tawdry dragon I made a widget component that renders its output directly to stereo layers with no in between, haven't had any crashing issues with it.
Held from May 8-11th in Silicon Valley at the San Jose Convention Center, NVIDIAβs GTC 2017 session schedule is chock full of deep tech talks that weβre looking forward to. Among them, Senior NVIDIA Research Scientist Anjul Patney will overview the companyβs latest learnings from their study of the βperceptually-basedβ approach to foveated rendering. Simply β¦
can someone please tell me why single player games can't be turned into MP games, unless entire gamecode is rewritten ?
coming from id Tech it sounds bizarre and wrong :/
@mighty carbon I think quake 1 was the last id tech game with co-op in the campaign (or maybe quake 2)
You get automatic support for some aspects. Some stuff just requires a human to decide (if a bunch of stuff like particle effects trigger as a result of some action do you just replicate the action and do the triggering logic on the clients, or do you replicate each effect individually)
my thing is that I'd like to start off with single player (since I have no clue about MP anyway), and then bring someone, who knows what they are doing, on board to expand whatever I have into MP and take it from there
I'd hate to spend all this time and effort on BP if it has to be rewritten anyway
what kind of thing are you making? some things are much easier to convert than others
Isn't a lot of it deciding what to replicate on the server and what's handled by the client? Don't know if it's a total re write but probably a pain in the ass to do later
They just did a stream on it.
why is it a pain to do later ?
for example, in vanilla Quake it was easy-peasy to make SP games support coop
in Darkplaces engine port with introduction of client-side code, it became pain in the ass if you wanted to use new client side code (it was based off Quake 3's code) - the code structure was entirely different from default Quake code
even in Doom 3 BFG (id Tech 5) it wasn't that hard to network entities. No code rewrite was necessary. Just had to tweak exiting code and add objects to replication functions.
so, when people say "rewrite is needed" or "it will be really hard to turn SP into MP" about UE4 projects I puzzles me greatly π¦
If you write your initial code using a lot of SP shortcuts or an architecture with a general lack of respect for keeping things straight, good luck porting that to MP easily π
If you write it well though, I don't think it would be too bad. Plus, you can always write things for MP, in an MP way, even if it's just SP at the moment
Though in these instances, you're talking about editors in games, not engines. I'm guaranteeing there's a lot more than you think to think about
well, I am better at art / design than in code, so π¦
The devs most likely did a lot of it upfront, to make it easy for the end user
Writing it from scratch is a lot different
I see
I wish there was guide "How to structure your SP BP code to ensure easy upgrade to MP in the future" π
Just a pain in the ass if you have a bunch of stuff a different programmer has to sift through who isn't familiar with your setup. Quake was pretty simple compared to what games are doing now. I.e. Physics takes a lot of special care so if you have a lot of that in your game you need to make sure it's something the server can manage.
@mighty carbon actually you should hit up vblanco about this. He is working on a Mp fps right now. He has probably learned some good practices on setup. At least it worked good when I tried it.
I might just go for SP with some sort of async MP that doesn't involve players being together in-game on a server and if there is a demand, next installment will be with MP from the start (or massive update to original with MP support)
What are you going to make? I am starting a new game in like 2 months when I'm done with this current one.
We could team up maybe if you like my idea
For multiplayer coop. I'm thinking a small scale dead island ripoff with some unique mechanics as coop, or just a quiddich type game using my locomotion mechanic. Just looking for wide audience appeal.
yay more vive trackers on the way! can order on the UK and US sites now
@mighty carbon if the majority of your stuff can run on the server, it's easy to convert it to MP
what do you guys think, how much tutorial does a VR game need? are players generally fine with figuring stuff out themselve or is it too frustrating when it's not immediately clear what you can do and what you have to press for doing it?
depends on your target audience for it and how complex your interactions are really. have seen a lot of different people use VR and everybody tends to do things differently, especially if they aren't familiar with the capabilities of motion controllers and whatnot
well my target audience are those people that buy VR games on steam
I'm quite sure they are completely familiar with the motion controllers by now
I think some sort of tutorial would definitely be needed, but I also know that I will probably change how many things work later so it will quickly become outdated and I'll have to spend quite some time with keeping that tutorial up to date, which I would prefer not to do
Not sure what's up with this bundle price
And if this means Vives are getting a price cut soon
But, if you assume not, and Vives are still $800. You can get yourself a 1080 for $324
But..
a 1060 is $200+
So, I just going to assume we will be seeing Vive price cuts soon
Vive+ 1060 for only $40 more...
Yeah, my bad
I am doubting that ASUS is selling their GPUs that cheap
Seeing how Rift got their pricecut, this may be HTC's way of saying, well we can do the same, but we add a little something
maybe the rift price cut already hurt the vive sales too much
The Vive has been out for some time, so the production is done, paid for months ago
Lowering the price is common after some time
but HTC likes to make money from it, since the rest of their business isnt running too great
but thats not the case here since they were outselling the rift 2:1 or something like that
Outselling the Rift doesn't mean it's selling well, just better
"Well" being a function of their development, production cost, and financial situation
and psvr is selling twice more than both pc headsets combined
you also forget that vive launched from the start with motion controls
so every enthusiast that wanted motion controls had only the vive for a considerable long time
its exactly what i did in fact
i saw that i wasnt going to get the hand controllers, so i sold my rift, that i preordered by min 10, and got a vive with that money
my wife accidentally ordered a PSVR last night. I had to regretfully tell her to cancel it as I haven't been paid for a bit π¦
its nowhere near as good as pc headsets
but if you crank the resolution up and make it work at 90 or 120 fps, it looks great
good luck getting 90 fps at 140 screen percentage on a base ps4 tho
im doing that on a pro, where the game looks the same as the PC version
the Pro has a technique more or less the same as MultiRes on ue4 nvidia card
but its not implemented on ue4
too bad, that thing would let me push 180 screen percentage for a great image quality
I dont want her to get one tbh. She will play Resident Evil and that's it. I keep telling her that the next gen are the ones she wants to get into, this gen is a temporary measure to get into the market
it sure feels like it
for the psvr
after developing for it, its clear that the ps4 has the power
the base ps4 not that much, but the pro can
i know the psvr is the cheapest (well, actually oculus is now similar price ) but its clearly lower quality
its made of cheap plastic, cheap rubber, and basic lenses
and tracking is terrible
I thought OSVR was
hdk 2 is still worse than the DK2
no way
We have one also :p
and yeah the SDK is a bitch
I've only used the HDK2 a couple of times but I was super impressed with the upgrade. The HDK1 was utter garbage. Kind of what I imagine the PSVR on a standard PS is like
and compared to dk2?
in fact for ages the only had extended mode and you had to flip the viewport as it rendered sideways π
compered to DK2 its slightly better visually
somewhere between DK2 and CV1
HDK1 or HDK2?
2
i have a fove headset sitting here too, at least it adds something to the game, the osvr is just a few steps behind and offers no benefits to anything really
Did you get Gear VR controller by chance?
@pearl tangle ^^
@full junco that was the thing with Quake - everything was server-side by default. I am guessing in UE4 everything is client-side by default?
Either way, I am stuck with particles at this moment :(
What's a good template for Unreal Engine to start with? I'm looking for something with touchpad walking, not teleportation, and every template I find seems to have the teleport on by default, and I'm simply too new to change it (so I'm probably missing the really obvious things)!
Guys, need a tip from you. I have Vive on hands and mainly developed a game for my client based on how it does look and feel in Vive. BUT. Time came to the point where I need to do oculus support. I already found that it has different HMD offset, but client keep telling me that controllers are very off the positions, they're very low and sometimes are under the ground (While vive controllers are on their place and work fine with vive)
So here's the question - what's going on? Did someone faced such issue?
@cursive bolt you need to have character-based pawn then, I did it before. So mainly your input will be just similar as for monitor, something usual, you know, but then comes VR in place, you'll need to implement for this type of movement 'degrees step' for easier navigation around and also make sure you'll be at the center of the characteer. I guess there is no templates around for such things for now, but you can easily experiment with nodes and stuff, ask community for assistance and you'll get what you want :)
Have anyone tried to read in VR? How bad it is?
Not optimal.
it's hard to do, resolusion is still low
2k is not enough for small text and feels like 640x480 at usual screen mostly :)
I see. -_-
Size of the text has to be real big, like in Pulsar: Lost Colony
^ yes
just wondering how far are VR UI desktops from now...
Imo, when we get better screens in the HMD's
vr desktops are something more a game than usual tool for every day, believe me
for now
until lwe have really high resolutions
also HMD is not so lightweight to have on the head all day long
owwwkaaay, but we don't have a hardware which can surely handle such resolutions easily
or.. 3way sli?
no way
pimax working on 8k already
but what about the hardware?
:)
we will not be able to make anything to let it run smoothly
until we have no powerful enough hardware
i KNOWHOW to render 8k fast enough...but...it's me
Β―_(γ)_/Β―
scene should not be too complex for even 2k/90fps
not telling about 4k and even 8k
and forward shading also doesn't help much with that
it doesn't matter how compex scene is (well, not matter much)
@wraith sky okay then, but what matters?
you can easily run doom at 100 fps in 4k with that gpu if you lower a couple settings
@toxic wedge latency
lenses is not the issue as well, probably would cost more but meh...
latency comes from the how fast will GPU process the scene, shaders, logic and all such stuff
the PSVR screen is RGB wich means similar amount of subpixels, but its lower velocity and has oled black smear
the vive lenses are fresnel with a higer width than the oculus, so instead of a godray, it does light circles
so we're limited on GPUs, not some kind of abstract latency. or you know something that could possibly give a kick for a whole VR industry
and CPUs as well, for sure
hardware
its a balance of plenty of stuff
that pimax thing is 4k, but its a worse headset overall compared to vive and rift
does it upscale? I've seen such things in some others HMDs
the idea is to track the pupil and render full res only in the area of interest (human eye actually see good only within angle of 5-10 degrees https://upload.wikimedia.org/wikipedia/commons/thumb/2/27/AcuityHumanEye.svg/500px-AcuityHumanEye.svg.png
oh, you mean that. yes, as far as I know Nvidia is working on it and it's kinda dynamic scale of rendered frame resolution, but it's not the latency
yeah, they work but w/o pupil tracking, they're doing that for peripheral vision
and it still doesn't change the fact that we don't have sufficient hardware to easily handle extreme resolutions at very high framerates
hm. as an a concept way of thinking - is it possible to do kind of downscale resolution at the borders of the frame? or downscaling just a parts of the rendered frame? at the moment, on unreal engine 4
20 degrees is really small spot on full screen, about 3%
downscale made by rendering in passes, i.e. 110 degree angle rendered say at 5% of resolution, then half of that at 10% then half of that on 20%...sum are will be still 100% of the screen but performance wise...probably 20%
@toxic wedge Basically it's incompatible with how everything works from the game to the hardware
It's not impossible, but it's not straightforward either
@toxic wedge not possible for current way of rendering, only hardware support could compensate VERY-VERY low latency...eye movements are 1000Β°/s
It's a typical example of optimization that might be more costly than the gain itself
@cobalt relic you probably won't be able to render 8k that fast for....like 10 years
I don't think you wanted to reply this to me π
@cobalt relic "It's a typical example of optimization that might be more costly than the gain itself" thought You were talking about eye tracking
Nope
@cobalt relic ok, sorry then π
I barely can imagine how incredible should look 8k at 144hz
like.. life?
just imagine it in full-resolution on the complex level
daaaaamn
Movie resolution at 24fps can look lifelike already
@toxic wedge probably...or even better, in reality we're limited by real physics optical effects...in VR we're not
yyyyeep
@wraith sky I've seens some illusions that exploit the eyes blind spot and it always is so freaky how you have a blind spot but dont' realize because your brain does a Photoshop content-aware fill in heh.
@fleet veldt my field in uni was optics, so i know a lot of stuff about it, but unfortunately not experienced much myself
Thanks Pirog. Didn't realize I should've just started with a normal FPS template! π It works.
@mighty carbon no, theres not really something like a "default" in UE4, you can run stuff wherever you want to run it
I'm going nuts. I'm trying to capture ingame footage through the Rift with Geforce Experience and it's recording all my desktop sounds but not the ingame audio
geforce experience has no options at all to let you pick which "playback device" to record
:C :C :C
Also GFX is shit for editing, since it records directly as mp4
@mighty carbon havent played around with the release version yet. its essentially identical to the daydream but there is an extra button on the bottom. they pretty much just shrunk the vive controller
nvidia had a great paper about foveated rendering a while ago https://research.nvidia.com/sites/default/files/publications/foveated-siga-16-v1-for-web.pdf
@pearl tangle Well, yeah, but I am wondering about vibration feedback in it
(if there is any)
@clever sky surprisingly there is no word anywhere about S8 supporting Daydream :/
Yeah I know. D:
But the specs are there, and they are a Daydream partner or something like that.
yeah, but they don't have to have S8 Daydream compatible (especially that Daydream started off poorly and still doing poorly afaik)
I wonder if some of the cheaper phones Samsung makes can be Daydream phones
damn, it's been a month since I got approved for Oculus hardware and haven't received tracking code yet π¦
Also, would anyone know if it's possible to make a VR game with the TrinusVR API without using the SteamVR API ?
There's a bunch of these things, VRidge, OSVR for cardboard etc
@granite jacinth its awesome that they do it, but that headline is clickbait
So that would enable to use a smartphone to play VR pc games
having an easy way to integrate ads is good for developers, but since its will only be possible on viveport stuff, it wont be used that much
Failing at seeing the potential but maybe that can be cool
@granite jacinth thats just in vive port though, which no-one really uses thankfully
so far the OpenVR / SteamVR ecosystem itself has no such thing
@tired tree it would be nice though
devs could create a free version of their game where the world is filled with ads π
Guys if i packed a vr output form ue4 using the vr template that comes with the engine, when i open the .exe file will the monitor be splitted ?
cuz it looks like a normal windows packaged output
will it be compitable with oculus ?
on my desktop monitor ?
should it look like that ?
some one here is going to be very happy
it's for the best. He only did damage by wearing that shirt
What's everyone using to develop VR? Vive or occulous?
I use both
It's a big investment so I want to get the right one for dev
Hey gang I am a solo developer. I use the engine for art and enviornments more so than creating games.
here is my website: lathanmastellar.com
I need some help with something really specific within the engine
what's the issue? π
So I tend to ramble and not make total sense when talking about this stuff. But in short, I am trying to create a "Heaven" in VR heavily inspired by Dantes Inferno.
all development is expensive π
I have a large part of this scene completed and I can share some screne shots but I am trying to acheive the effect of all of those souls swiriling around the center.
Hardware
Do you all have any resources or links that could help me out?
well, there's https://realtimevfx.com/
I'm not aware of anything like that in particular
Vive is $1300nzd atm
Thanks brother I will check it out
I might pick one up...
Rift recently dropped in price
guyz, I was wondering if it's ok to package a VR project using the Vr ue4 template without connecting a headset then play the game on another PC which is connected to a headset?
You can package a game without having the headset and run it on a computer that had the headset without any problems. But would take you an insanely long time to test things like that
@pearl tangle oh, that's tough, thx too much
@opal quest http://www.gamesindustry.biz/articles/2017-03-29-ccp-ceo-i-would-call-the-vr-installed-base-huge
Oh man, so Lucky got fired? Damn!
well I don't think he was doing much with things after the facebook buyout anyways
yeah, probably
reminded me of Todd Hollenshead from ID - got walked out after some time after acquisition
probably to offset $500M he has to pay to Zeni
well Luckey was getting sued directly, not facebook. So if he is no longer at Facebook then he will probably have to cover his own legal fees
alright, got a go for Rift. Should join your ranks in ~6 weeks
get your extra camera(s) with it too?
you will struggle to go back to mobile once you get into desktop vr though π
Abandoning previous projects once you've gotten a taste of full VR development is a non-zero possibility π
its also easier
ha, well, hopefully I find it in me now to drop the ball on mobile
you get a hell of a lot more freedom with development on desktop vr. you now can just create, rather than worrying about the hardware. although annoyingly with rift you need to focus on the front facing side of things rather than room scale so it's more restrictive depending on your idea
well, I'll be stuck figuring out controllers first (I see many people here come and ask for solution for motion controls)
so technically it's no different than mobile VR.. Just will be stuck with different issues
I should probably play a few games just to get a feel for it and see what locomotion and visual cues look like
the motion controls are super easy when you just use the vr template
hmm.. I wonder why people make it out to be hard to implement
idno. rarely had problems with them. There are a couple of little bits here and there but there is usually a solution around nowdays. Problems usually come down to attachment + collision or using physics objects and late updates but think those issues are fairly well updated on the forums with different workarounds
Hi, can someone help me with the camera for oculus? it's stuck on the floor googled for an hour tried a few stuff no success
is the headset working before you start unreal?
have you tried it in an empty build using the vr template?
is the yaw pitch and roll picking up but just not the translation?
actually it's lock to hmd and oculus default location is ground level
so i basically need to change hmd position
so you mean it's not rotating the view at all in the VR preview on a new VR template map? So it's like it's not detecting the headset at all. Is the image even appearing in the headset?
the headset is working fine only problem is it's on the floor level
yeah thats not working fine then. does it work with stuff on steam and oculus store correctly, as in it's only unreal thats not getting the positional tracking?
Weird.
Because it should work out of the box with the RIft on VRTemplate.
Try packaging the template and running that.
15.1
I mean, Oculus driver
dont see any driver version for oculus but i got my m.2 this morning so... just reinstalled everything today
ya app version 1.130362051
should tell you in the oculus home software
hey guys, I have an oculus vr and when i compile a project from ue4 i can look around but i can't use the game pad to move around,
any ideas?
-using vr template locomotion map in ue4
thanks
@analog topaz I dont see a movement component in that pawn.. you'd probably need to have a movement component as well as some input mappings to axis. I dont see that in their template., unless im missing it. Not sure what your level in BP is, but you can try reparenting the VR BP from pawn to "character" to get a movement componenet.. then open a FPS template next to it and copy the movement stuff out of there.
you mean to teleport or to walk around?
@pearl tangle good question. I assumed walk around since he said "move around". but maybe he meant teleport.
@deep willow there is a blueprint node that specifies if you want the VR tracking origin at floor height or eye height
Anybody have experience with using Nvidia Flex with VR in UE4?
@ripe bone I'm actually right in the process of building the 15.0 Flex branch and testing it out soonish
Building it takes ages. I assume everything should work fine since Nvidia Funhouse uses it.
@dusk vigil Funhouse uses it interesting
So, finally figured out smoke trail particles last night. Looks pretty decen, but kills performance to shit on Gear VR (or last night update for Oculus software screwed up performance)
You guys have any preferences on VR template or plugin setups? I've been between runeberg Vr, Mitch's Vr and the unreal template
You could do it from scratch (by learning from those or from Robo Recall)
Robo recall is rather limited
for good reason though, they didn't need a ton of interactions
Hello guys! How use the mitch's VR template?
@zinc rose thanks for the reply, thing is my 2B usb twingamepad is not working even if i add an input to project settings, the gamepad is entirely ignored " it works in othergames" using same machine.
and no i meant moving around not teleporting
@little nacelle follow the training videos?
@tired tree what kind of interactions Robo Recall input solution wouldn't be able to work with ?
and of the more advanced movement systems, more robust physics interactions, two handed interactions
its possible to add it to the framework
but not something you could "learn to do it" off of
as it doesn't exist in the project
I see
by 'two-handed" do you mean messing with physics objects with both hands simultaneously ? (since in RR you can use 2 weapons at once already)
yeah two handed manipulation
two weapons at once isn't thesame
all blueprint probably a mix of Mitches tutorials and RR nodes would be best to learn off of
I am guessing mitch's VR stuff is your work, right ?
no
oh, which one is yours then?
VRExpansion
ah, ok
I specifically note that my plugin is meant for more advanced interactions though, I suggest Mitches stuff for people starting out
does RR stuff allow you to open swinging doors naturally ?
I don't think there is anything supporting that in the project
can check real quick
because now i'm curious if they started in on that...don't think they did though
no worries.. Just wondering if I should use RR to get going with Rift/Touch and maybe later expand the BP code to do more (or not, if project isn't popular enough)
I wouldn't mind using your solution if it's just drag'n'drop thing - I'd probably spend more time digging someone else's code to understand hows and whys, instead of working on the game π
mitches templates would be better to work off of than RR generally if you want just BP
I wouldn't use my plugin just starting out unless you want some of the features it has
I see that all of his work is outdated and he stopped supporting his template
instead he runs new YouTube training channel and released a book about UE4 and VR
did he? He put out something fairly recently I thought
think he was just low of free time with the book and all
I guess my Google-foo isn't at its best today :/
do you have a link to most recent VR template ?
@dusk vigil yes, but some of them became old now, it doesn't work anymore with 4.15
JC explains magic behind new hi-res / high performance update for Gear VR
Robo Recall update!
so basically well made stereo layers
i found exactly what he is talking about. I made my UI be a layer for Oculus in PC. But found it buggy on Vive and unusable on psvr
it does improve quality massively
you can very easily create a 3d widget, and attach its texture to a stereo layer to do that stuff
is it really tricky to have a UI made from UMG in VR? Im talking about something flat, not with stereo layers
yeah, just remember 3D widgets can get hidden behind other objects
@tired tree so you're the auther of VRExpansion? Would you recommend using your plugin when you start designing more interaction heavy stuff? So far, like natural door opening, I have used math to calculate the angle between the startpoint and current controller position etc. Is that made easier with your plugin?
Yeah I choose to do it with physics in my plugin
one of the grip types creates a non rotationally driven constraint from your hand to the interacted object
ie: any lever, door, wheel, w/e you can think of and mock up with simulating objects works with it without custom work
Interesting! How safe is it to use? Like, are you gonna stop updating suddenly or will it ease out? If I chose to use a plugin, my employer wants me to make sure we suddenly cant deliver stuff anymore because our fundament is based on dead code
I know it might be a rude question π
for simple things like drawers or linearly / rotationally limited items I also have an interaction struct that defines limits, so physics aren't use
Thats just what I need!
pretty sure I can't stop updating it at this point, too many games coming out using it now
plus its a fun hobby :p
cool!
Right now I am using a BP interface and some BP function library to do all the interaction stuff... But that means I have to "write" all the code my self
yeah, the default epic teamplate setup creates a rather manual node setup per object
My guess its also fully BP enabled? Or do I need to touch C++ to get it working?
nope, its intended to be BP
sweet
Well, the manual node setup I bypass by having a 35+ function BP lib right now with all kind of stuff I havd made for interaction
people were complaining thinking that they couldn't link to it in c++, had to remind them that bp functions are c++ functions as well
Thats what I really love about BP.. in the end, more experienced programmers can easily convert or reference the nodes without a problem
or use them as a learning platform
Anyways, I will definitly check it out! Seems like it might solve many of my problems and also save me from a headache or two π
its got a template project
that implements things in a rather generic way as examples
It you only need object interactions though, could get away with just using the GripMotionControllers
the full character is for movement / replication / things you might not need
well, might not be necassary now, but Replication is defenitly something people has been requesting from me
never got the normal motioncontroller component to work proerply
properly*
damn... 4 beers later and I spell like a 5 year old
it works now, in BP as of 4.14
but its not very efficient to replicate in BP
but I meant character movement and grip replication
hmm.. I never got it working in 4.14... Had some very weird offset
Well, that is just as important
Cool, thanks for the insight @tired tree - Think I will play with a bit later tonight when I sober up a bit π
Man, the Vive Pucks are fun to play with, but its like the first few months of the HMD all over again, SteamVR is throwing fits trying to keep them paired all the time
does HTC hire people to write these kind of articles ?
This kind of article is written by non sense people who only care about numbers. Anyway that won't stop game makers to be passionated about VR, and won't stop VR to evolve in the best way.
- close that page and go to work on my VR project -
lol, same thing for Gear VR is rated as "comfortable"
Founded in 1993, Immersion Corp. designs and licenses haptic technology thatβs come to be used in gamepads youβre well familiar with. Now the company is developing a new haptic programming system which aims to help game developers make better haptics effects for their games, faster. The latest generation of VR controllers use more advanced haptics β¦
can this be used for Rift / Vive ?
"According to GSM Arena, the Mate 9 Pro boasts a AMOLED 5.5 inch display with a resolution of 1080 x 1920"
Google is poisonous for mobile VR
@pearl tangle http://vuze.camera/product/he_camera/ Have you pre-ordered this one yet ?
but with this "360 cam" you'll probably have a hole on the top and at the bttom.
I doubt about the stereo since it's a square and not a circle shape
full sphere rather, not full dome
what the hell are you smoking there?! o.O
there is a VR cam in triangle prism case.. So what? Shape of the cam has nothing to do with stereo.
ah ok
Samsung's Gear 360 cam is spherical, but not stereo.. Mind blowing, right?! :/
Anyone here using Oculus Spatializer with Fmod ?
I have not but I would like to see people's comparison between that and the built in oculus spatial sound plugin
@glossy agate until 4.16 is out, you can't have HRTF sound in VR using UE4's native audio. So for desktop it's either Wwise or FMOD + Oculus Audio SDK plugins for either and for Gear VR it's just FMOD + Oculus plugins.
however, I was told that once UE4 has new audio engine, 100% (out of experimental), you won't need any 3rd party audio solutions (for Rift/GearVR anyway).
I still like FMOD though - can edit your sounds in non-destructive / non-linear manner and immediately have it in UE4
but $2k license for indies is a piss poor choice on FMOD's behalf
actually it's free for 1 game per year
but if you make more than 1 game per year ....
Then they will execute you!
every time i exploit a free software license in a commercial setting I'm just picturing that axe over my neck
lol
speaking of Audio SDK, Oculus released an update yesterday
did I mention that GPU particles butchered performance on Gear VR / S6 ?
(probably due to blended transparent material and not due to particles)
I wonder how I can get clever with it and not lose performance π
Or is it 1 license per team member. Each take a turn for multiple games
well, if you are gonna let your team members release your game under their name, then yeah, sure
Just new LLC per game. It's $75 in CA
lol
Cheaper than 2k haha
$75 + a truck full of headaches
(taxes, paperwork, trying to register with Steam, which will also going to charge you, etc.)
at the end of the day it's better just to wait for 4.17 and not do anything of that π
Nah, I have made a bunch for real estate it's not bad. And I'm close to the capital so I just drive it down to secretary of state
True
Steam isn't charging yet for vr games at least.
I wonder how it would work if I want to retain some of the FMOD functionality (effects, non-destructive editing, etc.), but not use FMOD :/ I see Oculus offers VST plugins for DAW, but how to set the whole thing up and what DAW to use
Would be cool if they did some sort of native implementation like they did with simplygone. That tool is amazing
And 3D sound really makes the vr experience amazing
hmm... nm, the plugin is just to be able to visualize spatialization.
Onward did a great job with bullets snapping by your head and the ambient dog barks in an ally. Bugs in the garbage ect.
well, 3D sounds isn't what I am talking about
Oh. I was looking for the spatialization. I thought fmod helped with that?
for example, FMOD has scatterer sound. Meaning I can have bunch of sounds as 1 sounds source and randomly play them while randomly tweaking each sound's pitch and volume and whatnot.
So, if I have a sound of cricket, I can make 100s of crickets by using scatterer sound.
Oh that's really cool!
yeah
Huge time saver too
or what I did recently - I had a sound that sounds like a space ship, but I didn't really like it. So I used flanger and some other effects on it to make it sound better and in one click it was in UE4.
and I could tweak it while testing it in PIE
Does it also implement stereo sound?
yeah
you can do stereo, 4.1 and 5.1
or 3D
so it works for any platform
it used to be only $500 per additional indie title
I also wonder if $2k applies to DLC or if it's for main titles only
because if you can make money with core game + DLC, you can either stretch it until next year, or you can use earnings to invest into second license if you are releasing one in the same year
Q. Do I need a license for DLC?
A. No you donβt.
I wonder how they even track it? Does it use a key within the project like deploying to iOS?
ha, cool
they can track it I bet (if user is connected to the Net)
I guess FMOD license is fair
Free 1 core game + unlimited number of DLCs per year for all supported platforms is a good deal
(used to be each additional platform was not free)
Yeah. Sounds robust. Wonder how they came up with a model so different from algorithmic on the art side. I could go for a subscription model because it's pay as you go.
Like if you want to use it in a bunch of prototypes before deciding on a final game
if prototypes are free, it's still free
Oh nice. Yeah that's a really good model then.
yeah
Yep. For mixing and spatial sound
After playing a few games with really good sound I have to implement it. I have pulled off my hmd a couple times thing the sound was in my house and not the game.
Sound is crazy important
its april now in my timezone, I really wanted to get my game finished in march and damn, I still didn't succeed with that! I also wanted to have it finished in december. and I also wanted to have it finished before last summer. I'm not good with finishing stuff quickly... I'm the opposite of @wicked oak
im really not that good finishing stuff. it allways takes more than expected
DWVR should have been finished by mid February
but ps4 dev turned harder than expected, and i wanted to make some things better from the pc version
well but you have finished stuff, I've never finished any UE4 game yet π
the trick is to just release it
I'm always picking those games though that no one else wants to do because its too hard it seems
none of my games that ive released are really finished
but they are good enough so i released them
I think first impression is important, and I want to blow peoples mind on the first impression
thats hard with something that "isnt really finished"
then do vertical slice style, and do only the "first level" but REALLY well
check Vertigo devs, or Windlands, or Budget cuts
ha yeah, in games that have levels that works
what are you doing? i dont remeber
I have only 1 level
@mighty carbon that Business Insider article isnt too bad if you keep in mine that some people expected VR to take off immediately
I dont think facebook is super happy with oculus currently, its far from clear if oculus will be the dominating VR company that they would have thought it would be
@full junco First impression IS important. But really if you put your game out there.. "finished" (in your mind) or not. It will get a ton of hate. and a little bit of love. If gamers see 3d graphics they instantly expect a GTAV. If its not that, bad review, and hate. Just how it goes. Many dont seem to understand the "solo dev" concept. Even if what you do put out is super polished, runs very well. The hate will be in the form of "only 4 hours... expected at least 30 hours for the $19.99. " Just have to brush that aside. try and keep things simple and get your stuff out there in a decent state. If you wait for it to be perfect (in your ind) you'll never release it. lol
@zinc rose thanks. but don't VR gamers know by now that a VR game can't look like GTA V?
@full junco From my experience. Not really. Most gamers don't know much about that side of it.
I definitely think if I would get more hate than love then I did something wrong. more hate than love means reviews are not good on average, and thats bad
but by now people should have seen there there isnt a single VR game that looks like GTA V
youd think but thats not how it works. At least not on platforms like steam. You can sell 500 copies. End up with 40 returns. and only 8-10 reviews. The reviews arent a reflection of the game. Its true in life, that generally angrier people always post their views WAY more then happy people so of those numbers you're lucky to end up with a mixed review state
well but there are games where the average of reviews is good, right?
yes. either theyve hit masss. like a few thousand sales so theres talk / hype / word of mouth .etc... or they've sold very few and were lucky enough to get positive people posting positive reviews
one review whent heres lesss than 20 skews the entire lot to negative or positive.. each review is critical until you have so many that one doesnt matter much
when you first release you'll get people that play less than 5 mins on record (meaning they didnt go past the menu) and leave a negative review and refund the game.. for ....reasons.
lol why should they do that?
why indeed. A lot of time i feel like its other devs who are jealous to be honest.. but really who knows.
and then I would think no one takes a review after 5 min play time seriously
you are correct.. no one does.. but most people dont look at each review.. the look at the total at the front
therein lies the issues
look at any major studio game that is generally a rehash of 4 games prior.. like assassins creed.... just takes millions of dollars worth of marketing. eventually more positive then negative happens... though you can get lucky and the game catches on but.. it would have to be one of those special/simple concepts. A VR shooter isn't likely to "catch on" like Rocket league or angry birds did. so
if youre good at marketing you can get like youtube streamers and stuff to play your game and they influence ppl. if they say they love it. others will too.. whether they do or not or knwo if they do or not. but trying to get a youtube follower with a real followign to play your game is as hard as getting IGN or RPS to do a write up too
one thing I'll say, is that having good trailers and snapshots etc reallly is important... its one of the few ways a solo dev can boost the game to get more people on board which helps with that
well one problem with VR is that its very hard or impossible to show in a "lets play" that the game is great. people wont experience it from watching a video
yep thats a good point
which is why i find it kind of sad that the games that sell well, both in VR and not.. usually have promos that dont show the game at all
an old man freaking out with a headset on goes viral and the game does well.... or a live action video of a game(non vr) or CGI video that shows no gameplay.. but "wows" people. shrugs
hm, yeah
im not good at the marketing or publicity and its exremely hard as a solo dev with a full time job to make a game, keep it up to date AND handle the marketing, social media and PR lol
so i cant offer any advice on how better to market it. but good videos and screenshots make a difference
that is your REAL FIRST impression
yeah
just not sure its really possible to show in a video why something is fun in VR
so primarly video and screenshots are about showing the graphics
videos dont do it justice at all for sure. Its good to show the unique concepts of your game. Being able to smash a window and shoot out and down it. Or push a door open with your hand so it looks interactive. But seems like the most videos that make a game do well are the "reaction videos". Which i think is sad because most are probably staged
yeah
there havent been too many VR games that have been huge if you notice since the first release of the motion controllers. The ones that came in early did the best and stayed that way. Not that they arent good, but being first and "polished" is huge. Anything that comes from now forward will need a budget to make a real splash... or hit the lotto and get lucky somehow
same goes for non-vr games. You can have the best concept and idea in the world. Put it together wonderfully, but if no one knows about it or sees it... it will just sit there in the crowd
I remember when rocket leage first got huge.. it took awhile for it to catch on but when it hit 1million "buys" i read an article on it at the time and the devs were saying how they werent sure about putting it on PSN for free. I cant remember. But in hind sight its the best decision they ever made. PSN makes it difficult and expensive for indie devs which in turn makes releasing there harder so there is less there. They throttle their "free" releases each month. So most people grab the free two games this month and try it and keep it or delete it.
its a good way to get noticed and tried by many... theres probbably a gazillion free games on steam.....
i thinkt he best chance a solo dev has for success.. is simple games.. release several and hope one of them catches on a bit.
yea... same. π probably true of many solo devs that take it seriously
but even if no one is interested in my game, I would still update it every few days
I would just hope that people would be more interested once it has more features π
so I would give it half a year or so
if you have the time to do so.. thats good. I'd suggest you stay away from early access until most of the features are there
well "most of the features" would take forever
I dont have any fixed feature list, but I would always find things I want to add before I consider it perfect
so most of infinite is quite a lot
people WILL leave in depth reviews on things that you publicly say are not ready yet. Just be aware of that... while people that have good things to say will sometimes post, but most will "Wait til EA is done" so you'll get a skew there of reviews you may not like or that are "unfair".
either way you should start the steam process asap. Good or bad, you can get a VR game in the platform without greenlighting it. I suspect this will change as the VR space becomes more saturated.
I hope it will change as soon as possible
you dont have to release it til your ready.. but you need to have the features described as playable so they can try it out
first you said you "suggest you stay away from early access until most of the features are there", and now you say "you should start the steam process asap"?
I didnt quite get that π
you can start the process and go in EA. but your game wotn be published or released or advertised at all until you hit the button to go live
so in effect, no one will know about your game, but you can release it when you want (give or take a few days for steam to try it and ok it) after you hit the button
whats the point of being in EA without anyone knowing about the game?
at least i dont think its advertised ... i cant remember if it showed up when i first put it up. the game wasnt ready yet for a few weeks while i was on there though
ah, you mean just setting it all up so that it could be released at any time
yes
yeah, now I get what you mean
I set up all the initial stuff with steam
just havent added the game π
good luck
thanks
I dont really want to setup the game in steam because it asks me about entering a name, I'm just not sure about that yet, and it doesn't say if the name can be changed later. So I just don't do it. I would hope setting that up doesn't take long, so that I can do that once I think my game is ready
@full junco I have never waited longer than 3 days for anything "needs approval from valve"
That includes playtesting the original build
Alcatraz Escape!
John "the rock" Alcatraz (youd need to get sean connery to do some voice work for you tho)
people always come up with funny names that are related to "john alcatraz" π
wow, Ghost in the Shell VR is not that good π¦
(on Gear VR at least)
except the date on the article is 31st of March π
can i use psvr on unreal engine or i need to ask sony for dev kit w/e?
Google-sensei is your friend https://forums.unrealengine.com/showthread.php?119966-How-does-one-get-to-develop-and-test-for-PSVR
particles, at last https://www.youtube.com/watch?v=6ZjtqHaKSlg
Apparently GPU particles either broken in mobile/mobile VR or just perform like sh#t
so I had to go back to CPU particles and have smoke trail to be so-so when flying in the ballistic trajectory
(and probably I should tweak smoke before the take off since the drone isn't running on coals π )
@mighty carbon interesting
indeed
it could be Mali GPU related, could be specifically related to Galaxy S6, could be broken UE4.. Who knows π¦
One of the reasons to move to desktop VR is that it seems like Epic and Oculus care a way more about desktop than mobile
well, Oculus supports Unity a way better than UE4 and ofc their own native stuff
night
This is pretty cool but it won't work with any UE4 game without some source fu
I haven't been able to disable the "limit fps when game not in focus"
is there a way to make symbol and have the player draw it as close as possible with motion control to get higher points?
i imagine you could find examples where people have done similar things for games using mouse
and convert it to 3d
@deep willow like this https://www.youtube.com/watch?v=0CNJ2fCj4xQ&ab_channel=GaneshTiwari
@zinc rose now releasing early access or not doesnt matter. You arent gonna get advertised in any place other than the "recently released" list
wich is spammed with 30+ games a day + dlcs
I just updated from Android 6 to 7 and now I get vertical tearing all over the place :/
Just came out
Basically the MetaVerse
@mighty carbon ok figured this one out.
Turns out Samsung is rolling out Android 7 on S7 devices
but Oculus doesn't support it!
proof:
and.... more people complaining about it here:
@raven halo no, I am still on 6.0.1 and I don't know when 7.1.1 comes to S6
I think it will be later. In any case, don't download it when it's released unless it's supported. My last 2 weeks have been wasted because of this. :/
I believe Oculus supports N. And if not now, it certainly will once S8 is out.
Read the comment in the link above. 2 weeks ago an oculus staff member claimed that as far as he knew, they didn't because of those exact reasons
Things might change with the release of S8
I hope
I also just realized what tough spot Oculus is in. Google doesn't really have an incentive to look after them or to even cooperate with them
Since they have their own thing going with Daydream
I was watching Ghost in the Shell VR last night and my phone got really hot really quick, @raven halo
And that's a video made by Oculus
So if they ever give you grief about overheating, throw this fact back at them ;)
lmao
they haven't yet
just someone at Oculus who was previewing our game commented that the performance seemed off on Samsung S7 with Android 7
Btw, it was a bad quality and no so good content either
ima need some testing in this few days
anyone up to test my game? for the next update
im trying to polish gameplay, new missions, and better graphics
I don't think that person is very technical, so I don't think he was even aware android 7 doesn't seem to be supported. Although I'm only basing this on a post by an oculus staff member, I'm trying to find more info to back it up
@mighty carbon I checked out the gits demo on rift and and gearvr. And the gearvr one seemed super blocky and lowres for a video :S
Yep
Not even stereo either
I would imagine if it's a video, it should be stereo and visuals should be like Final Fantasy CGI or Blizzard or Blur Studios
@raven halo do you see a spinner as soon as you launch your app? They used to have one, but with one of the system updates it went away for me.
@wicked oak how long is the stuff that needs to be tested? I wouldnt want to test a 20 hour campaign, but short stuff, sure π
wands?
Harry Potter wands
tfw you leave ONE SINGLE LIGHT from a particle spell and it costs 5 ms of render time
yeah, dont use light π
just tested instanced stereo again, it still makes everything slower
it depends on the number of drawcalls
its faster when you have lots and loooots of drawcalls
like i do
its faster for me
its mostly a tradeoff
more gpu time for less cpu time
if you are cpu bound like i am (damn drawcalls) then you get a net improvement
yeah, but most people are super GPU bound in VR
yup. Epic has it on for robo recall
due to their amoutn of drawcalls.
My game has very very simple materials, no lights. But it has fucktons of polygons and drawcalls
ive been polishing the game this last few days, it now plays much better.
huge improvement on balance and small stuff
and the ps4 performance improvements let me have enough headroom to run MSAA in pc
I'm still searching for a good way to prevent different LOD being used per eye
looks like the 3D is completely broken when the left eye seems a different mesh than the right eye
thats why I tested it again, but the fps loss is too much
why does it sound strange?
left eye can be further away from the mesh than right eye
or the other way
i find it strange that it checks distance on the 2 eyes
instead of the center of the camera
why would ue4 calculate LOD distances twice?
well the eyes are completely independent viewports I think
so it will just do everything it normally would
i think they did some optimizations. But i dont know exactly how much
the obvious stuff is to do culling once, for example. Instead of twice
or the lods
or the draw preparation
get your render list on one eye, then play it again with a shifted camera matrix
thats kind of what the instanced stereo does. It prepares a drawcall, renders it "at the center" and then the vertex shader just shifts it left and right
then they use some clipping to avoid one side rendering on the other
yeah, but since I'm not using IS...