Friday, October 9, 2009

Out of the Blue: Why I care about Blue Mars

With barely any content to speak of, no currency, audio, media, or voice, and a User Interface that allows for limited communication and camera movements, it’s not hard to understand why Blue Mars has its detractors. Despite all of its early Beta deficiencies, however, there are compelling reasons to believe that it holds great promise for current and future aficionados of 3D immersive environments.

Most Second Life residents I’ve spoken with don’t know quite what to make of Blue Mars,
but none appear to believe that it will supplant or even displace their virtual world, though I do sense a modicum of concerned solidarity and defensiveness. There’s this Capulet-Montague thingy going on… like, if you so much as praise one aspect of Blue Mars it’s a form of treason or something. Personally, I subscribe to the Romeo and Juliet canon: “Can’t we all just get along?” Or am I misquoting?

Conventional wisdom would indicate that Second Life, which – ironically – could be called a mature virtual world by comparison, with its entrenched communities, tons of content and hundreds (thousands?) of events a week, is here to stay… but Blue Mars, in its infancy, has not yet begun to demonstrate its gravitas. It will take time, money, and a growing community of dedicated developers and residents.

Would you feed this little bitty baby vitamins or Kryptonite?

In development

Every time I have a chance to speak with Jim Sink, recently named CEO of Avatar Reality, the company that owns Blue Mars, I come away feeling energized and hopeful for this budding platform. In a recent conversation, I asked him what they were concentrating on at the moment. “In order to create a vibrant virtual world community, we need to have outstanding content and we need to attract extraordinary developers. In a coming release, we’ll be offering the ability to create clothing, and we’re providing new uploading tools, too,” he explained.

Meet Blue Mars’ Ruth. Yes, these are default avatars created by unfashionistas at Avatar Reality, with a little fiddling I did around the eyes within the Face Customization editor. Please hurry up fashion designers, and make me something I want to wear

“At present, most of our energy is focused on laying the groundwork to prepare Blue Mars as a development platform. Of course, things like the UI are exceptionally important – we know how important this is to the community – and we’ve already begun to implement changes based on users’ feedback. There’s no shortage of things we need to do.”

Money makes the world go around (especially if you can cash in/cash out)

And one of the things that Blue Mars plans to do soon is launch the Blue Mars currency and, as Jim puts it, “the fundamentals of the economy.” I wondered out loud how musicians who depend on tips would make out in Blue Mars since only developers can cash out. To this, Jim had some very good news: “All developers will be able to cash out and anyone can become a developer, but Blue Mars requires more information and needs to know a lot more about you if you are going to be moving money in and out of the platform.“

Location, Location, Location (and putting a price on things)

Another impending feature will reside on the web-based Developer MyPages: Soon developers will have the ability to upload blocks of shops, set prices to items – including ready-to-use residences, clothing and stuff, and also real estate. Land! This is cool, but here I am going to dig my heels in the ground hard… there ain’t NO way I’m going to have a cookie-cutter home. Not here, not there, not anywhere. I once lived in a development and had the bad habit of driving right by my home because it was indistinguishable from all the others on the block. This just isn’t my thing. I don’t do it in real life, and I sure as heck am not going to do it in my virtual life. Yes, I’m that neighbor that annoys everyone by painting her house purple.

Blue Mars Betti in the kitchen? No, no, no! This isn't the kind of cooking I plan to do

Web to Blue Mars and back

Blue Mars has every expectation that they will be unveiling new MyPage web pages for end-users, too, from which they can edit their in-world profile, and purchase Blues, among other things.

Repeat after me: It’s a PLATFORM

Many of us, myself included, have been calling Blue Mars “a virtual world,” when it is actually a platform for many virtual worlds, and a staging area for all kinds of new – to me - talented people. For developers, entrepreneurs, artists it is a next-gen technological solution from which you can build your environments. For end users (the jury is still out on what we are going to be called… Residents? Martians? Colonists?) it is intended to be a destination offering rich diversity and choices.

These gorgeous photos of coming new content courtesy of Richard H. Childers, president of Virtual Space Entertainment (VSE), one of the first City developers on Blue Mars

“Scurvy-inducing 3D”

Jim Sink recently tweeted, and of course, I looked, “CryEngine isn't just for realism. Check out this amazing work that would look right at home in Blue Mars.”

Games… yes! Blue Mars is flinging its doors open to members of the thriving Crytek modders community, inviting them to create their own games and content. A recent scrumptious example is Monkey Island 2, in which free-lancing 3D artist Hannes Appell first demonstrated how, with a little “voodoo,” he could enhance the gorgeous hand painted 2D concept art with a basic camera projection. Those pirates sure are mean to little froggies.

He then went on to demonstrate – just for fun - how original Monkey Island 2 backgrounds can look and feel in a modern 3d game engine, by building the sets with Maya and then exporting them into Cryengine's Sandbox editor.

Hannes is a Bavarian and now lives in Germany where he will soon graduate from the Institute for Animation & Visual Effects of the Filmakademie Baden-Württemberg. If LucasArts doesn’t hire him after this little “experiment,” someone sure will. This video has only been posted on YouTube for 19 days and already it’s had 329,000 views and the comments beg him to “MAKE A REAL MOD OUT OF IT.” A little Blue Mars, Hannes?

Role Playing

I’m also like a stuttering deer caught in the headlights when it comes to Role Playing, but were I an elf, Trekkie, furry, or inclined to hang out in locations that attempt to accurately portray a historical era, such as Second Life’s Roman Forum or the 1920’s Berlin Project, Blue Mars’ built-in ability to control incoming content such as how an avatar looks or dresses would make suspension of belief considerably more intense.

Photo courtesy of VSE, a Blue Mars City developer

Architecture and design

I like the way h3oworldz, a CryEngine modder group, used the platform to demonstrate how valuable a "3D realtime walkthrough" a structure or set can be to demo environmental effects and scale. They would likely be interested in hearing about DB Bailey from Second Life and his success at using a virtual world to demonstrate a real life structure to ultimately sell a real life project.

Vinyl"Club from h3oconceptz on Vimeo.


VAM United is a German group working on a science fiction movie project using the CryEngine2 game engine. They’ve created this proof-of-concept as they prepare to produce a feature-length machinima based on a “loved and acclaimed” book. I highly recommend that you watch it here for better quality, but here is the YouTube version for you non-clickers:

What will the work of the likes of Lyric Lundquist or Lainy Voom or Colemarie Soleil look like, shot in Blue Mars?

Look at this one. Tell me, with a straight face, that you can't make use of these graphics and physics to create your vids...

See also:


Spence said...

Fantastic. I tried Blue Mars but it didn't do anything for me. Their user inter is horrible. I hate not being able to zoom back or send IMs. Not being able to fly over their cities is also a nuisance. I didn't realize they were going to make so many changes so soon. This gives me hope. I am looking for a new world. Maybe Blue Mars is it after all. I will have to brush up on my Blender skills.

PingT said...

Bettina, when do you think they will have more content and voice?

Can we make animations there yet?

Especially liked the last two tasty videos.

sororNishi said...

I think you are absolutely right, Bettina, but the future of a Virtual World like this is so determined by the philosophy of the controlling body.
If they are predominantly Realists then will NPIRLer's have a role to play in its development?
This is what I am waiting to see ... Flora Virtua Exotica or roses and oaks?

Pay said...

I was disappointed at how slow it seemed and I am using a fairly decently powered PC ... if the complaint against SL is that it is requiring more and more computer power and creating a gap between power users and the rest of us, what does that say of Blue Mars, which hasn't even begun to populate itself. Movies of SL look better than SL, too ...

Bettina Tizzy said...

@Spence: It seems that Blue Mars is concentrating on the platform's stability and on the development tools first, and will then target the UI.

@Ping: In a previous interview, Jim indicated that they want to get a few things sorted out before they introduce voice, and I know you can't make animations there yet. From what I gather, they are working hard and will be introducing many new features over the coming months.

I have heard that there's lots of content in development.

@soror: Since it is a platform for many virtual worlds, it will all depend on the City owners.

As Jim pointed out when he tweeted about Monkey Island 2, “CryEngine isn't just for realism. Check out this amazing work that would look right at home in Blue Mars.” In other words, it's about fidelity. I think your trees would be spectacular there. In so far as NPIRL's role in Blue Mars' development, I can tell you here and now that I am going to celebrate and foster great NPIRL content there, just as we do in Second Life.

*It's not about excluding Second Life artists' content, it's about taking pleasure in the work of all great virtual art, no matter what the platform.* It's bloody exciting, really!

@Pay: I'm so glad you mentioned this. From what I know so far, your computer was likely slow because of your graphic card. About 3 months ago, my two year old state-of-the-art $1,000 Nvidia Ultra died, so I had to replace it with a newer, more powerful card. I was SHOCKED to discover that the logical replacement - also by Nvidia - cost me $135! See where I'm headed with this?

Masami Kuramoto said...

If I remember correctly, EA's "Battlefield" franchise does not use CryEngine but Frostbite.

By the way, the trick demonstrated in the Monkey Island video is not special at all. The technique is called camera mapping and has been used for quite a while. It is not specific to CryEngine but works with any other engine that allows importing custom meshes and UV maps.

And that's where the actual problem is: While any other game engine supports mesh import, SL knows only parametric and sculpted prims. Linden Lab must implement mesh import soon to remain competitive.

Another thing that differentiates SL from other game engines is normal mapping. This is a technique used to make flat textures appear three-dimensional by modifying the surface normals and, as a result, the way in which the surface reflects light. It's a common way to fake detail on low-poly geometry meshes, but SL does not support it (i.e. it does not support uploading customized normal maps in addition to the predefined ones).

While CryEngine looks nice, it has its own set of problems. The most important one is the lack of OpenGL support. OpenGL is a platform-independent graphics API that is available on Windows, Linux, Mac and elsewhere. A 3D world wide web will never be based on an engine that does not support OpenGL.

However, I hope that Blue Mars will be a huge success. Strong competition may finally force Linden Lab to open up the grid for teleports between SL and OpenSim regions.

Aliasi said...

MUST Second Life? The entire advantage of prims is it made creating in Second Life possible for non-professionals. So of course, once it got big enough to attract professional attention, the pros wanted nothing more than to grab the reins out of the amateur's hands... :/

Ruina Kessel said...

Is there some reason SL couldn't have both Mesh Import and the base prim system it has now?

Masami Kuramoto said...

Aliasi, if you make a platform appeal to non-professionals, at some point only non-professionals will use the platform. Right now there are still some very talented content designers in SL, but they will be the first to admit that creating high-quality 3D models for SL is an extremely tedious process. That is the reason why SL looks ugly most of the time and why those designers crave for alternatives such as Blue Mars. Splitting up complex 3D topology into sculpties is time-consuming and redundant work that is more autistic than artistic.

Furthermore, as I pointed out here, drawing prim-based objects is extremely inefficient and the number one reason for low frame rates on the client.

SL's rendering engine is not as bad as many people believe. It is just wasting too much time drawing ugly and inefficient geometry.

It's OK to keep prim-based building tools, but mesh import is a badly needed extension.

Bettina Tizzy said...

@Aliasi - SL doesn't have to introduce mesh imports, but it needs to unless it wants to go the way of many bright young things that have since disappeared from our Internet landscape. Progress :) Remember Mosaic, Netscape, and Compuserve?

@Ruina - I don't think there's any reason why SL can't do both, no. In fact, LL has stated that they are working on it.

@Masami - Ouch! Lol. Yea, you are probably right about Frostbite. I will look into this.

You make a pretty strong case there. It was painful to read! And I do agree with you about SL's rendering engine. One has only to look at the Grand Odalisque to understand that.

For me, the greatest reason to celebrate Blue Mars' existence isn't the graphics, physics or creation opportunities. It's the foundation for protecting the IP of those creations/creators. In that regard, they've already had Linden Lab's lunch, and they'll eat LL's dinner, too.

Eladrienne Laval said...

My computer runs horribly in Blue Mars (it's an old piece of crap), but you know what? I keep going back because I keep having this nagging feeling about a little something called "potential."

Aliasi said...

Meh. People seem to forget prims were introduced into SL for a damn good reason - to keep bandwidth down. Sculpties are a relatively small increase thanks to sculpt maps, and it STILL greatly delays the time until everything is rezzed. Demanding to use tools designed for an entirely different task seems quite strange to me; yes, it takes time to learn how to build with prims, and how to build efficiently, but I believe you become a more skillful modeler for it. It is not impossible to build beautiful things in SL, very obviously.

Does this mean I'm wholly opposed to some kind of mesh import? Not at all, not if it keeps to sculptie-class bandwidth requirements. Otherwise, the problem we have now - amateurs using gigantic textures on twisted torii, high-resolution sculpties, etc, with no heed for performance or bandwidth - will only be further exaggerated. I feel the adaptability, the freeform creativity of SL - WITHOUT a board of censors/corporations to make sure your 'product' is profitable, unoffensive, and unlikely to provoke legal action from the Concerned Citizens of the world - is pretty fundamental. Those who use it wholly as a social venue, or some sort of collaborative 3D modeller, or speak in bizspeak terms about incentivising incomes may not see it, but I don't believe either of these things could have existed, and SL would not have gotten to where it is, without that underlying substrate.

I don't know how much of Blue Mars' content is streamed as Second Life's is, but I'd suspect the answer is 'very little'. It sounds more like a turnkey solution to making your own virtual amusement parks - be they World of Warcraft-style games or entirely social critters like Metaplace - than anything I'd identify with Second Life. This is cool and I can see how many people would find it more useful - businesses who don't want rogue elements to come in with penis missiles, artists who want only the interpretation they set down - but I don't want an amusement park, with the world handily pre-created and transferred to my hard drive. The whole mesh vs. prim argument seems to be a reflection of this. Any but the most simplistic mesh (i.e., no more detail than a sculptie) is impractical to stream with current technology, so far as I know. Which means, like every other virtual world out there, you're just moving an avatar through someone else's playground.

HeadBurro Antfarm said...

Great post Bett - I've started in other MMOs but never stuck with them because SL has spoilt me. So far BM is the only one to keep bringing me back because I have a feeling it will be equally as great as SL.

For me, as a non-content creator type, I look for a world where I can weave my own tales and roleplay - to this end I've already crossed over into BM with my tale Far From Home:

I look forward to many more tales to come thanks to BM and many other worlds yet to be born :)

Masami Kuramoto said...

Prims were introduced because subdivision modelling requires a much more complex set of editing tools which would have been overwhelming to the casual user and too difficult to build into the SL client.

However, the low-bandwidth argument doesn't hold. When you approximate non-primitive shapes using clusters of prims, the result takes more time to rez because each prim and each individual texture is a separate object in the database.

Sculpted prims are even worse because every single one of them is the equivalent of a 32x32 mesh. That's horribly inefficient if all you need is e.g. a bevelled box. And you still have to cluster them if the rectangular topology of a single sculpt map fails to describe the target shape you want to build. SL's long loading times are a result of its fragmented lego-style structure where every single brick has a unique name (i.e. a UUID) and a personal profile in a database, even if it's just a pixel on your screen.

By the way, I would like to add something to the professional vs. non-professional argument discussed earlier. Aliasi said:

"So of course, once it got big enough to attract professional attention, the pros wanted nothing more than to grab the reins out of the amateur's hands."

This sounds like there is some kind of barrier around the realm of professionals which the amateurs are unable to cross. Of course this is not true. I know quite a few builders who saw the introduction of sculpties in SL as an incentive to learn Blender, a (free) tool they had never touched before. In case anyone is still looking for a real-world example of SL's educational qualities, there it is.

Dale Innis said...

Yeah, I've been having my ups and downs about Blue Mars. Currently I'm very much "wait and see, and someone tap me on the shoulder if anything interesting happens". There's lots of wonderful stuff that they *could* do, but so far very little has materialized. They are going to do this, and going to do that, and all sorts of really amazing stuff is going to appear Real Soon Now, and in the meantime we should look at these nice videos from vaguely-related systems with completely different usage and business models.

The reason that people coming upon it for the first time think of it as a virtual world rather than a platform is that that's how they've been pitching themselves at the retail level; their website starts "Blue Mars is a free to play massively multiplayer virtual world...". So it's an obvious mistake to make. :)

Blue Mars could be a really amazing platform for a diverse set of virtual worlds, enabling people with good ideas for new worlds to start them up without having to do all the groundwork of user onboarding and client design and billing and authentication and all that. I think that would be very cool! But they've got a long way to go to convince me that they're actually going to accomplish that. So far the evidence is sparse imho.

aliasi.stonebender said...

Kinda silly to have a running commentary in comments, but...

You're mistaking bandwidth for end rendering, sai Kuramoto. A prim takes almost no BANDWIDTH - it's a terse description of a mathmatically-determined shape caused by manipulating a circle, square, or triangle through a region of space just so(roughly speaking). The majority of data transferred is in other items attached to that prim - textures and so forth. A mesh must have FAR more data attached - information on every vertice, a UV map for skinning, etc, etc. A sculptie is essentially a compromise.

Now, it is entirely true this approach leads to rendering more polys than a smartly designed mesh would, for an object that isn't a direct derivative of one of those simple forms. That's why occulsion culling was such a big deal when it was added back in the day... but to be perfectly honest, at the time SL was first designed I believe network performance was considered a more significant bottleneck than graphical performance, and this is still true (if not always for strictly technical reasons - consider the metered bandwidth plans that are popular with many ISPs; you can always buy a bigger GPU).

tldr: don't confuse networks with graphics cards.

Masami Kuramoto said...

"You're mistaking bandwidth for end rendering"

I'm not mistaking anything. I was talking about database congestion, caused by fragmented storage of geometry. This has nothing to do with network performance.

Fragmentation in databases has the same effect as fragmentation on your harddisk. You have to understand that retrieving a single record of 1 MB size is much faster than retrieving 1024 records of 1 KB size, although the total amount of data is the same. This is because of the overhead caused by multiple database queries.

Imagine a simulator that is filled to the max with prims. Downloading the prims alone results in 15,000 individual queries to the asset database. Each of these prims has...

- a UUID
- a name
- a description
- a creator UUID
- an owner UUID
- a set of permission flags
- a price tag
- a default click action
- a set of parameters controlling type, shape, color, shiny, bump map, transparency, glow, light, flexi...
- up to six (?) texture UUIDs (which result in additional database queries immediately)
- an inventory (i.e. a table of asset UUIDs)

It is this sort of geometry micromanagement which makes things rez slowly in SL. The network is not the bottleneck at all. Maybe it was in 2003, but now it isn't. Whenever a prim does not rez or a texture does not load, it's because the asset database is taking too long to respond to a specific query.

Custom meshes will reduce the number of database queries because they allow more visual detail to be stored per asset, reducing the total number of assets required per object. Unlike the hidden (and useless) overhead of a linked set of prims, every bit of mesh data will contribute to what you see on your screen.

Do you really think the creators of Blue Mars have not considered this?

Bettina Tizzy said...

@Masami, thank you for catching my mistake re: Frostbite. Blogger doesn't allow for scratching out so I just deleted that section.

@Eladrienne - Great running into you in Blue Mars yesterday!

@Aliasi - I understand your allegiance to Second Life and the opportunities for creativity that it affords amateur creators. Second Life is in a tricky situation right now. On one hand, it needs to keep up with the times and introduce mesh imports, but when it does, most of the better content creators will move towards making the majority of their content off-line. That content will look considerably better than prim-based creations, too, and the divide between amateurs and advanced/pros will grow.

I don't want to worry about that, and I'll tell you why. There is no reason why creators of any level can't learn to make their beauties with off-line programs. In a few days I will blog about one of the most famous creators in SL who took a few days off and learned to work with sculpties, and how *happy* that person is about it.

@Dale - You make an excellent point regarding the positioning. I'm with you regarding the potential, and hope to see it fulfilled.

@HeadBurro - I hope you will take me on one of your adventures when you do!

@Masami - What an extraordinary explanation! It is worthy of a blogpost in and of itself. Thank you for dropping that jewel here. I learned a lot from it.

Icon said...
This comment has been removed by the author.
Icon said...

@Masami OOOOMFG I've been crying your argument, I've been breathing it, sneezing it, and what ever else. SL needs to get on the ball if they a) want the place to be more appealing and b) want to speed things up. As a content creator myself, I can tell you how much of a PAAAAAIN it is making statues out of 50 or more sculpts when it could easily just be ONE OBJECT!! (and look 700x better). and considering that each of those sculpts is a 32x32 image translated to a prim.that's 50 images for 50 prims mmmmmm lovley. >=| forget about loading the rest of the place LOL. I know it's not a short throw to say LL's database will empty out like a clogged drain when they allow mesh import.

@Aliasi granted yes the gap between pros and non pros will grow but in the end everyone will be happy. Creatores will be happy for the freedom and the detail they can put in objects, consumers will be happy that they can have a full Victorian furniture set and not devour allocated object use for their tiny 128 prim parcel, and LL will be happy their servers have less "stuff" to shuttle around. I can't wait to see sims with some serious forestry since every tree is only one object AND only one tree texture for the whole place =D

Oh yea this is supposed to be about BM right, Masami got me excited lol.

Blue mars is looking to have tons of promise and is honestly second in my line of site ESPECIALLY considering it's free. I'm hoping the platform also allows for programing to allow for *gasp* an Aion, WOW, War Hammer - esque game experience (should you visit that city or cities). Graphics, killer; can't argue there. I love how MASSIVE each place is! no jumpy sim crossings and enough space for some awesome boating, flying etc. It would however be amazing if they could some how link areas side by side. I can tolerate a "crossing" if it's not so frequent.
I haven't seen what's the latest and greatest but I do hope they're thinking about attach points before the community outgrows their avatars (learn from SL.) Because I knooooooowwww there will be people who don't want to be people but furies, robots, and what ever else. Chubaka will still want to hold his cup of coffee with out loosing his hand. And skin/shape creation/modification needs to come soon to appeal to noobs. I mean not to sound bad but, well .............................. *cough* homogeneous a touch much? ^ ^;

either way I'm keeping my eye on it, mega potential to be had. more than is even being thought of right now. It's just a matter of will they go forward full speed or hold back and tease.

Bettina Tizzy said...

Icon... just to get you even more stirred up: Aion was created with CryEngine 1. Blue Mars uses CryEngine 2.

Icon said...

OH! =O could BM be the universal answer to MMO of the decade? I love SL and have a ton invested, but I will admit when social interaction is low, an adventure or two would help. But again.... would they do it.