Fool me, this is real good shit ! [the VR thread]

[seibu]

You know what? I'm starting to get a bit worried about the Oculus SDK.

Firstly, the new one (0.4) crashes. A lot. Check out the reddit threads. It also exhibits some pretty awful programming / design decisions, such as the way it tries to implement being a service without actually using Windows' service system.

Secondly, there's no Mac or Linux version, and no ETA on when they'll arrive. The email announcing the SDK didn't even mention the lack of Mac and Linux support. Previous release emails have.

Finally, the drivers are no longer open source. The demos are open source, but the service which interfaces with the hardware is not. This is the most worrying development for me. Are they permanently closing the source for the drivers? If so, we wont be able to port to other platforms without reverse engineering. For example, when the first SDK shipped without Linux support, it only took one enterprising coder a few days to port it to Linux. This isn't possible for SDK 0.4 because we don't have the driver source. I work in academia and do some very heavy to the metal work with the DK1. I can't work with closed source drivers.

I worry that, software wise, the company is heading in the wrong direction. I know saying this is not cool, but I wonder if they've hired the right people. The SDK has always been pretty poorly coded right from the first version, but if it also becomes closed source, we can't fix it. The Rift is one step close to becoming a paperweight unless you want to use it in exactly the way Oculus want, on the platforms they want, complete with all the bugs and inconveniences they can't be bothered to fix.

[Mat]

That's weird, because, well, Carmack.
Not usually a name I'd associate with shoddy programming and closed source. And he's always been very happy coding for different platforms too.
Maybe he's not that involved with the SDK?

[gilesgoat]

You know what? I'm starting to get a bit worried about the Oculus SDK.

Firstly, the new one (0.4) crashes. A lot. Check out the reddit threads. It also exhibits some pretty awful programming / design decisions, such as the way it tries to implement being a service without actually using Windows' service system.

Secondly, there's no Mac or Linux version, and no ETA on when they'll arrive. The email announcing the SDK didn't even mention the lack of Mac and Linux support. Previous release emails have.

Finally, the drivers are no longer open source. The demos are open source, but the service which interfaces with the hardware is not. This is the most worrying development for me. Are they permanently closing the source for the drivers? If so, we wont be able to port to other platforms without reverse engineering. For example, when the first SDK shipped without Linux support, it only took one enterprising coder a few days to port it to Linux. This isn't possible for SDK 0.4 because we don't have the driver source. I work in academia and do some very heavy to the metal work with the DK1. I can't work with closed source drivers.

I worry that, software wise, the company is heading in the wrong direction. I know saying this is not cool, but I wonder if they've hired the right people. The SDK has always been pretty poorly coded right from the first version, but if it also becomes closed source, we can't fix it. The Rift is one step close to becoming a paperweight unless you want to use it in exactly the way Oculus want, on the platforms they want, complete with all the bugs and inconveniences they can't be bothered to fix.

Gosh ... all that worries me ...

I haven't seen all that yet but it should be soon, as a developer "I want things that work and/or that I can make it work nice".

One thing that worries me .. but I am not sure if I can talk about it here and/or it's into the realm of NDA .. is that "if it will ever be possible to implement properly a certain thing on PC" .. not because of the Oculus HW .. because of the PC OS ..

Now I think "the real test" of all this will come when they really will start to be "more commercial" and focus in selling the HW/stuff around .. that's .. when "the fun" starts I think ..

Besides .. if it "continues this way" I think you'll be able to count on your fingers the "number of developers that done some VR WITHOUT using the Unity engine" .. you can count me as "one" .. probably you can add 3 or 4 more people ..

At this rate Oculus is going to be "an add on for Unity" or such

[gilesgoat]

That's weird, because, well, Carmack.
Not usually a name I'd associate with shoddy programming and closed source. And he's always been very happy coding for different platforms too.
Maybe he's not that involved with the SDK?

No I think he's quite a lot .. and I think at times that maybe "he's the one desperately trying to make things go really well" .. see my other comment about Unity ...

[gilesgoat]

Now .. said that .. I MAY BE WRONG - I REALLY DON'T KNOW but ..

I am getting a bit "this by nose impression" .. "the big industry with big guns" seems to me "still a bit reluctant/unsure" about this whole VR thing in general .. I get the feeling that "there's not that investment that should/could be" and they are still "waiting and looking to see what happens".

Despite some incredible enthusiasm, as the one we have , and the incredible potential of VR I have the feeling that quite a few fear "could be all a big flop" ..

Said that .. gives me also the feeling that "a lot of people into VR" are NOT coders at all and I can see A LOT - "too much" - of this Unity all over the places, fair play, still people wanting to create VR content but "not the people that would sit down to create optimized code/VR app".

So I don't know sometime it's a bit of "strange feeling" as in "you get the feeling there's still lot of uncertainess, confusion, indecision about what to do and how to do" in all this.

[seibu]

Besides .. if it "continues this way" I think you'll be able to count on your fingers the "number of developers that done some VR WITHOUT using the Unity engine" .. you can count me as "one" .. probably you can add 3 or 4 more people ..

At this rate Oculus is going to be "an add on for Unity" or such

I completely agree. Unity is very nice I'm sure, but Rift content I've seen created with it, personally, I've not found impressive. Unity also has pretty bad Rift integration, given their devs have advance access to the SDK.

For me, the beauty of the Rift was its simplicity. It's a display and a USB HID device. The only thing which really needs to improve from the DK1 is the display (resolution, refresh rate and especially low persistence). This stuff with a webcam to track headset position looks like a horrible, limiting hack to me. The 0.4 SDK's client / service architecture is overcomplicating the Rift further. Give me a good display to work with and easy access to the headset gyro data. That's all I need as a developer.

[gilesgoat]

Besides .. if it "continues this way" I think you'll be able to count on your fingers the "number of developers that done some VR WITHOUT using the Unity engine" .. you can count me as "one" .. probably you can add 3 or 4 more people ..

At this rate Oculus is going to be "an add on for Unity" or such

I completely agree. Unity is very nice I'm sure, but Rift content I've seen created with it, personally, I've not found impressive. Unity also has pretty bad Rift integration, given their devs have advance access to the SDK.

For me, the beauty of the Rift was its simplicity. It's a display and a USB HID device. The only thing which really needs to improve from the DK1 is the display (resolution, refresh rate and especially low persistence). This stuff with a webcam to track headset position looks like a horrible, limiting hack to me. The 0.4 SDK's client / service architecture is overcomplicating the Rift further. Give me a good display to work with and easy access to the headset gyro data. That's all I need as a developer.

I fear here we are "colliding" with an "OS issue" .. where I'd dare to say, nor Windows, nor Linux (?), nor else (*) can do ..

Camera tracking .. "works for Morpheus" .. which means to me "you can do it, if the conditions are right" ..

Now my fear is that "VR" as in "very real time processing of sensor data and images" really "does not fit" nor with Windows, nor with Linux, this is the field where ATI or NVIDIA should step in saying "I do some really special drivers for you that allows you to" ..

Ok at that point drivers won't be OS as you like BUT "if they would works and make things works as they should/CAN it could be worth the price".

What lightly pisses me off is that "there are lot of docs about do and don't " on the OR SDK telling you "things to avoid" and "how to make the VR experience better" .. can't say I've seen that been followed that much.

I see a huge list of "complaints" about "what Unity does or does not or does bad or should do better" with OR .. and almost no one writing " the better code " but "waiting for it to someway appear" .. ending up in "if Unity does not do a very good job in improving the VR plugin you'll never see really good VR" ..

I've been a bit to "the metal" with OR as well yes the SDK is in various parts like "for now like that should be improved next version maybe" ..

But I think there are at least quite a few people struggling to make the SDK good and willing to get good VR content.

All I can say is that LS is really doing "its best part" to create "really good VR content", on "all the VR that we can put hands on".

But we'll see .. I yet haven't had time to put hands on the latest OR SDK .. I am too busy with "things" .. but trust me I've been quite vocal on the OR forums about "tech issues" and "if there's something to say you can bet I am going to say it"

[Dio]

I am getting a bit "this by nose impression" .. "the big industry with big guns" seems to me "still a bit reluctant/unsure" about this whole VR thing in general .. I get the feeling that "there's not that investment that should/could be" and they are still "waiting and looking to see what happens".

It's what I've said before in this thread: because VR games do not in the general case overlap with non-VR games - a few types like simulations excepted - it means that VR isn't about selling a peripheral, it's about building a platform.

Given the additional cost of development as well, the big boys won't, indeed can't jump until Oculus or someone else can demonstrate a big market - millions of units.

So the early days of VR will be A. in those limited game types and B. indie studios, a few of whom will make an absolute shitpile of money if the platform does actually take off, because for a couple of years (until the big boys really have spun up) they'll have a platform of a few million units to target and the ability to do it with a sub-£1m budget.

Everyone in the industry thinks Oculus will trivially ship 200k units, maybe as many as half a million. But to build a platform they have to take it beyond the game geeks and the techs and sell to the real world, and there isn't one person in the world who is sure they can do that. The spectrum of opinion I've heard varies from not a chance through 50/50. I've not heard anyone bullishly positive about the market. I've not heard anyone negative about the device, mind. Everyone loves it. Personally I think that's why Zuckerberg bought the thing; he wants one, and the games to play on it.

[Surfr]

Maybe I misunderstood the point but I didm;t realise that Elite Dangerous, Asseto Corsa etc were all Unity projects. Were you talking about the SDK developers when you said 'maybe 4 people..' or is literally everything a Unity development these days?

[gilesgoat]

Maybe I misunderstood the point but I didm;t realise that Elite Dangerous, Asseto Corsa etc were all Unity projects. Were you talking about the SDK developers when you said 'maybe 4 people..' or is literally everything a Unity development these days?

I am talking about "People that made a VR game/demo NOT using the Unity engine".

[gilesgoat]

I am getting a bit "this by nose impression" .. "the big industry with big guns" seems to me "still a bit reluctant/unsure" about this whole VR thing in general .. I get the feeling that "there's not that investment that should/could be" and they are still "waiting and looking to see what happens".

It's what I've said before in this thread: because VR games do not in the general case overlap with non-VR games - a few types like simulations excepted - it means that VR isn't about selling a peripheral, it's about building a platform.

Given the additional cost of development as well, the big boys won't, indeed can't jump until Oculus or someone else can demonstrate a big market - millions of units.

So the early days of VR will be A. in those limited game types and B. indie studios, a few of whom will make an absolute shitpile of money if the platform does actually take off, because for a couple of years (until the big boys really have spun up) they'll have a platform of a few million units to target and the ability to do it with a sub-£1m budget.

Everyone in the industry thinks Oculus will trivially ship 200k units, maybe as many as half a million. But to build a platform they have to take it beyond the game geeks and the techs and sell to the real world, and there isn't one person in the world who is sure they can do that. The spectrum of opinion I've heard varies from not a chance through 50/50. I've not heard anyone bullishly positive about the market. I've not heard anyone negative about the device, mind. Everyone loves it. Personally I think that's why Zuckerberg bought the thing; he wants one, and the games to play on it.

One thing that I think so far none approached "seriously" it's "the potential of use of OR/VR OUTSIDE the video games market" .. I am sure there are tons of possibilities.

The way I see it so far is :

1. VR *CAN* WORK and *CAN* "deliver very interested experiences"

2. VR NEEDS stuff MADE FOR VR with VR IN MIND ( so that's why "not immediate overlap" )

3. VR *CAN* give experiences that nothing else can

4. VR games or not IS interesting stuff by its own with lot of potential.

Now, I did not want to say it like that but I believe "the big industry" is just leaving "the small and indie take the bite and the risks" and - possibly - create that "big emerging market" for them .. we are basically "the cannon fooder" ..

Once "something big enough for them" will be established/proven to be possible bet "millions will start to come into" and "the big 4" will come with big guns shooting all else out "dominating" once more again, 90% of the profits will be theirs and the remaining people will be left begging for the remaining 10% that probably will be 1% ..

Despite all this - possible shit scenario - "I don't give a shit" .. if it's bound to happen it will happen in a way if it happens it's going to prove "that VR IS succesful"
what I care is that "whatever, the device and VR is really COOL and it's really cool shit that can do wonderful shit".

I want to make "wonderful shit" for me and others and hopefully survive into all that will follow that will be.

I mean "man, I've been walking INSIDE a Klein bottle that *I* made and other shit crap" ... "a dream I tought never possible" .. and I made it my own for me .. so "I made it" I've seen what I wanted to see thanks to this new tech.

Does it sell, will it sell, will it not ? .. I don't know .. but what I know is "that the tech is super cool and CAN do things"

But "Booh to the big industry" , we are the pioneers, NOT them .. they just come "after" .. I am sorry guys, we were, you won't .. you just came when the road was already open by someone else

[gilesgoat]

Eh seibu about "the company in wrong direction" .. I don't know .. "seeing how things are going" one could think/say only one thing they have to do "give a fuckload of money to Unity and push/help them to make a mega working ultra optimised plugin" .. and that would fix 90% of the stuff, people will stop complaining and they'll have quite a lot of "fast made" SW ..

Actually I am kinda "surprised" it's not already like that and/or if it is I am not aware of ( too busy working on my own stuff ).

[OX mkII]

Will all the kickstarter Oculus funders automatically become shareholders in facefuck for being able to buy out a company before it even has a product to market? A bit bloody naughty if you ask me as it potentially changes the original kickstarter vision.

[Baggypants]

Will all the kickstarter Oculus funders automatically become shareholders in facefuck for being able to buy out a company before it even has a product to market? A bit bloody naughty if you ask me as it potentially changes the original kickstarter vision.

Tough.

[Bod]

Tracker info says my Occy is curretly at Feltham, but expected delivery is on Moday !!!
It's only 150 mile away & UPS won't deliver on Saturday. FFS