mos6507 - all messages by user

2015/7/9 17:01:58
It's all happening here! ziggy72 wrote:

The top one is without AO, the bottom with.

This doesn't look like real AO to me. All it appears to be doing is creating a fuzzy drop-shadow around every 3D object's edge. It is not finding the actual nooks and crannies in 3D space the way AO is supposed to work. Notice what it's doing with the legs of the chairs for instance, AO would not be generated there under normal circumstances. It's like a floating black cloud rather than something enhancing shadows. It SHOULD be pooling at the right angle between the edge of the rug and the walls, and it's not. This really doesn't look very useful.
edited by mos6507 on 09/07/2015
2015/5/28 1:19:21
Unreal Engine, going forward A while back Digimania came in here and all but said that Muvizu was at "end of life". In other words, just bugfixes but no new features. They are focusing on other priorities now.
2014/8/30 6:18:49
Digimania Show reel This is their RenderDigimania product. I took a look at it. All it does is take FBX scene files you export from something like Maya and render them with a realtime engine. There is NO capability to actually script your animation in the program. It's of no use to Muvizu users.
2014/5/27 17:19:46
Death by Jelly - scene 1 WIP All signs indicate that we are at the natural endpoint of a generational cycle with machinima software.

Even Plotagon, which isn't even a year out of the starting gate, is refactoring to mobile in an effort that smells of "Plan B".

Making truly cutting edge and useful machinima software is not easy to do on a shoestring. And all you need to do is compare the existing offerings to today's commercial games and you can see how far off the mark they are in almost every category.

(For example, go to youtube and search for Vindictus Character Creation if you want to really despair over the sorry state of machinima programs compared to today's games, and Vindictus came out 4 years ago)

Games are getting more and more cinematic while simultaneously not letting you drive them as machinima software. Fluidity of movement, physics, lipsync, destructible environments, it's all coming together. Why? Because that's where the money is. It's just that all of the R&D money is going into games and nothing filters over into machinima.

So the engines (Unreal or whatever) aren't the problem. It's having the resources to build a tool on top of it.
2014/4/18 15:00:48
The Copyright people are out of control! "Why does it have to be "preexisting" music in the first place? Plenty of free music out there for anyone to use."

Because the free music has no pop-cultural resonance. If there's a free song that is in a similar style of, let's say, Beatles Yesterday, all it can do is rely on the intrinsic qualities of the song as a first-listen. If you use Yesterday itself, it evokes all of the thoughts and images surrounding the song and all of the ways it's been used throughout the decades, and you can create connections through that in through the themes of your story. It just has weight to it as a cultural artifact that free music lacks.

If you actually look at how Youtube is used, it is very much a device in which people remix and comment on pop-culture itself, usually through parody (think of all the Gangnam style parodies). But it can also be used in a more serious context. When Youtube flags stuff, it is telling people what they're doing is theft and not freedom of expression. This is a form of collage just as much as anyone who uses a compiled score in their work, like 2001 or Kill Bill. Regardless of whether permission is given or not, the creative impulse is the same and it should not be punished when nobody's making money off it.

edited by mos6507 on 18/04/2014
2014/4/18 13:44:09
The Copyright people are out of control! This problem with Youtube ContentID is reaching a critical stage and people all over are talking about moving to Vimeo.

Sites that integrate with Youtube in some fashion should really look into supporting Vimeo.

As far as dogs on skateboards go, Youtube's not going anywhere, but for hobbyists who need to rely on preexisting music, Youtube is all but unusable now.
2013/9/20 0:00:33
Is there any way to use 3D assets from Xtranormal? OK. I see it. I'm not used to the user-interface here.
2013/9/19 21:14:27
Is there any way to use 3D assets from Xtranormal? I see that Muvizu doesn't have private messaging. best way to reach me if you don't join our forum is via youtube PM.
edited by mos6507 on 19/09/2013
edited by mos6507 on 19/09/2013
2013/9/19 20:45:48
Is there any way to use 3D assets from Xtranormal? "TTS is not very popular."

Yes it is. Because it's not practical to record all live voices. And anybody who attempts to do a whole show with just their own voice sounds like a guy telling a bedtime story and "putting on voices" with falsettos and what not. Lame. I'd rather use the variety of TTS and compensate for the lack of dynamics through other means.

BTW, it's actually a skill to write scripts for TTS engines. To a novice, they will mispronounce everything and it will be a monotone. But if you get good at it, the pronunciations will be spot on and they will actually seem to emote a bit. Don't just write it all off as Stephen Hawking.

"I wish they could have been absorbed by someone instead of disappearing forever."

The holding company (Fairhaven Capital) still owns it, but I doubt they'd let it go for what it's really worth on the open market.
edited by Jamie on 20/09/2013
2013/4/10 17:48:47
To the community at large With software, people WILL pay into it (even a monthly subscription model) if they feel like the money is actively going into R&D and the software is evolving. However, they won't be paying $100-200 for Muvizu to pretty much remain as it is today just so the company doesn't shutter its doors. Muvizu doesn't have to phone-home on the internet to be used (with the watermark). If Muvizu goes out of business, people will be able to keep using it in perpetuity. The main reason end-users will want Muvizu to stay in business is to improve the product. I know that sounds harsh, but that's the the reality of it.
Software MUST improve and evolve. I don't deny the team have worked hard, but the improvements to Muvizu since inception have not been very dramatic as far as what it enables you to do with the software. The prop interaction, better facial expressions, 3rd party asset marketplace (with characters and animations), all these things are not outlandish requests. They are standard in other popular systems. The best way to compensate for not having enough R&D is to crack it open for 3rd party plugins and let that process snowball.
I actually am not aware of a single machinima project that has made a lot of money. I wish there were a central database that kept track of such things. Popularity, yes, many youtube clips have gone viral. But I have not heard of anybody breaking out and making an indie film with machinima that has generated a healthy return. I do think that day will come, but it will only happen when the machinima companies position themselves so that they offer a richer feature-set and don't overly restrict animators with EULAs and TOSs.
There is only one high-profile project in the works I know of, which is someone trying to do a feature-film using Source Filmmaker.
Technically speaking, Valve is hands-off people's projects as long they don't use Valve assets, and Deep is all custom-built assets. Nevertheless, I find it hard to believe that Valve doesn't stand to make some money off of the supposed $19 million budget. I haven't heard news about this since then, so he may not even have a go-ahead. However, if you buy something like a seat for Maya, Autodesk doesn't hold their hand out to James Cameron for a percentage cut from Avatar. You offer a tool, and you charge a fixed amount of money, you process the transaction, and you get out of the way. That's how I think most people would prefer it to be.
I do think there are opportunities for machinima companies to find novel revenue streams besides the software, however. If your characters are highly distinctive, then you trademark them, and try to popularize them Hello Kitty or Mickey Mouse style. I haven't seen a lot of that attempted. Even the TF2 characters used in SFM haven't seen the sort of spinoffs that I would have expected. It's a missed opportunity. I think ultimately the most successful machinima company will be one that recognizes that prebuilt assets are exploitable intellectual property and the best way to exploit it is to crossover and be more of a direct content-provider.
2013/4/10 16:50:37
Watermark: new terms of use There is a phrase, at least in the US. "The customer is always right".
What that means, unfortunately, is that businesses need to have thick skins and tolerate being "called names". Corporations may be considered persons, but they don't have feelings that need to be protected. Yes, it's a double-standard, but that's how it's supposed to be done. You don't respond to customer complaints (which can get very nasty and personal) by doing the equivalent of the proverbial "hanging up the phone". That is really terrible PR. Negative feedback is a symptom. You have to disregard the ad-hominem and look for the substance of the complaint.
I know this full well from my experiences with Xtranormal, who, six months ago or so, when the going got rough, saw fit to completely shut down its forums and start aggressively censoring Facebook of anything too overtly critical. If you behave that way, it paints a portrait of a company that is more interested in sanitizing its public image than actually improving user-satisfaction. Since then, they've lost just about all their techies and are really nothing more than a zombie-company. That's what happens when companies have a top-down "my way or the highway" attitude. I'm NOT saying Muvizu is like this, but you never ever want to even come close to operating that way. You have to be willing to pay attention to criticism and not circle the wagons.
The fact of the matter is that Muvizu IS a business, and they've been developing the software for over 3 years now, which is a long time for a startup like this. There comes a time where the vested interests expect to see a return, otherwise they will fold. You have to begin to assess whether the expectations that existed at the project's inception have been met, and if not, why? Technology marches on. For instance, lots of attention has shifted towards the (free) Source Filmmaker recently. That didn't exist (at least not officially) when Muvizu started. You have to keep reassessing the landscape and make adjustments.
However, you wind up confusing users if your business model and TOS keeps changing too much. That's again, what happened with XN. It was too much of a moving target. Ultimately XN committed the final insult, which was to completely prohibit anyone from making money with XN videos, which is far more grievous than charging per-minute as Muvizu is now doing. For creative people to invest their mind-share over a long period of time in a technology, changes like this can feel like they are having the rug pulled out from under them. So the better a company is able to anticipate where things are headed and isn't constantly zigging and zagging, the happier customers are likely to be.
I don't know how many of you are aware of this, but crowdsourcing is peaking right now. If you have an ambitious project you want to do with machinima, there's no better time to try to get it funded than now. What's the worst that could happen? People turn their nose up at the minimalism of the graphics. But it's worth a shot. So how these companies deal with people trying to make the leap to getting their projects funded is extremely topical.
2013/4/10 14:07:24
Watermark: new terms of use Muvizu could pursue an asset-based business-model like iCLone and Daz3D if they really wanted to. They've been talking about allowing for custom characters for years now, but it never happened. Their architecture just isn't as extensible as the others. If they had tackled this problem, then we might have seen a much more varied set of assets available for Muvizu, which would have expanded its appeal.
Valve also just opened up a workshop for Source Filmmaker, but there are currently no monetization options. With a small team like Muvizu, they have to leverage 3rd parties. Being able to port over static props isn't enough.
2013/1/31 23:41:27
New character? "everything is glued down and non-moveable and you can't change anything"
Not for me it's not. I removed the chairs and laid down positionable props. That's not the stock captain's chair. I have a whole separate utility called Set Decorator that lets you reskin sets, bake props, and customize the lighting, just like Muvizu. I'm sorry State Plus wasn't officially released by Xtranormal so if you want to look at it as though it doesn't exist since it's a hack, so be it. I certainly tried to get them to fold it back into the application, but they were stupid.
"I can't walk down a hallways or open a door because state is very limited in that respect."
State characters can certainly walk. They can't open doors, no. Neither State nor Muvizu offer true character interactions. Yes, you have guitars and mics and things, but you can't pick up arbitrary props.
"my character t walk off the bridge with a camera pan opening a door and follow him down the hallways. State does what is called "talking heads" very well. One room at a time."

I added camera moves to State. Can I do a steadicam shot like in Goodfellas, no. A straight dolly down a corridor yes. But not a slinky path. But there's more to telling a story than how fancy the camera moves can get.
"my plate is full but if this is your desire give me a couple of months and you can have that exact bridge in muvizu. "
I can give you the bridge in OBJ if you want to try to import it, plus the Wrath of Khan chairs. I was just comparing it to the stock Muvizu offering with all of its off-kilter angles and solid colors.
I mean, if you're going to talk about importing objects into Muvizu, it's only fair that I show you what can be imported into State, officially or not. I imported a ton of props from Star Trek: Elite Force. Anything that can get spit out in OBJ that has a single diffuse map per mesh should work. It doesn't require the weird ASE format that Muvizu does.
You know, I'm trying to be objective in this thread and I don't think it's fair for me to be accused of trying to touch off a holy war. If there's something Muvizu can't do well, then that's a fact. But whatever I do with State, whether it required hacks or not, is still what this thing can do, otherwise I would not be able to demonstrate it in clip after clip. The capabilities are there.
Just to get this back on topic, I have yet to see a 100% straight (no tongue in cheek) Muvizu clip. If someone has ever successfully accomplished this, I'd like to see it.
My belief is if I finished duplicating the clip in question, I would wind up with something that, emotionally speaking, falls far short of what I'm doing in XN. Is my work Pixar grade? No. Is it "Meet the Heavy" SFM grade? No. But I did it in like 3 hours.
2013/1/31 23:22:38
New character? "in other programs we would basically have to spend hours setting up everything and animating everything before we could get to the final product."
That clip I did with the robot took me, I dunno, 3 hours start to finish maybe? It wasn't mean to be that much of a big deal, and I even forgot to turn the Orion's eye blinks back on. It's just not a big time-investment to work at this level. Most of it spent on getting the body language right, none of it requiring the sort of keyframing you see in SFM, for instance. I'm not going to argue that Muvizu can't be fast doing the things it is good at, or that other software can be time-consuming, but XN is really fast. And XN is designed specifically for dialogue scenes. I am not enjoying the way you're supposed to enter dialogue wav files in through Muvizu. It is a slow process with tons of clicks, and no TTS supports. Plus, they apparently changed the UI in the last update and I don't see a new tutorial on how it's currently supposed to be done. You can not click to import the WAV directly in the dialogue section. You have to make an audio track and then link to it through the dialogue, which just adds an extra layer of complexity. The tutorials only feature a single speaker so it's not really showing me how to do a typical one on one scene like this. If anyone has a good dialogue tutorial for the most recent build, let me know. I do want to see how far I can go with it.
This is my first attempt to duplicate the look. I know it won't come close, but I am more interested in the virtual-acting capabilities than the look right now.
2013/1/31 23:05:32
New character? I don't want to be disrespectful in this thread, since I definitely am an interloper and naturally here is where Muvizu fans congregate. However, I think there's been a little too much uninformed Xtranormal-bashing here. I don't blame anyone for having a negative opinion about it, since the company really sabotaged itself, but the original desktop software (State) was capable of a lot, and I have spent over two years expanding it. Barely anyone besides myself uses the expanded system, and State has since been replaced by the inferior XD (not that I can't help anyone curious enough about this stuff get it on their system if asked.)
This is the kind of work I want to do in Muvizu.

I don't think it's capable of those sorts of nuances. I am taking some time to try to recreate this scene on my own, but since it's really the first true Muvizu clip I've done, it's slow going.
I'd be glad to supply the audio for this if anyone wanted to prove they could do this. I mean, it's only a few lines of dialogue and a ton of body language.
As far as it being closed, the current system is, but State isn't entirely closed. I know how to import props and sets. Characters are doable but you have to really know Maya to prepare a proper rig.

I think you would agree that this bridge is better than Muvizu's.
2013/1/31 16:33:45
New character? The Heroes and Villains stuff definitely is a step in the right direction. I should spend some time with that.
2013/1/31 16:08:18
New character? The bald one is a Deltan. Ilia was the character in Star Trek: The Motion Picture. My character's name is Ilsa and she's the ship's counselor, like Troi, rather than a navigator.
Sorry for the topic-drift.
I've added a bunch of extensions to Xtranormal State and I would have helped Muvizu be able to get their characters to emote better if they were interested in any of this sort of thing, but apparently they aren't as they didn't even see fit to reply to my emails.
Now Xtranormal is dying and there are more options available like Source Filmmaker so Muvizu doesn't seem as attractive to me to eventually migrate into the way it used to be.
2013/1/31 14:54:33
New character? "True. the one thing that Muv really has going for it is the fact that you can litterally customize the characters any way you want"
Customization is a broad term. I can't get them to look the way I want them. I want them to look like this:
2013/1/31 11:54:52
New character? Muvizu has never made a character with normal proportions. They are basically muppets. They have skinny floppy limbs, big cylindrical heads, and cucumber-type noses. The superhero characters are better, but still on the comedic end of the spectrum. And since they've never come up with a way for 3rd parties to make their own characters, it really limits animators to doing comedy and satire. I've seen a few brave attempts at a serious clip with muvizu, but they simply aren't credible, because the instant association of comedy that one has when one sees the character designs. So you always are thinking that it's tongue-in-cheek or for kids. I wish machinima software wasn't so closely linked to the style of the software company's internal assets and it could be judged purely on the merits of the engine itself. Moviestorm has a similar problem, because all of its characters are some variation of a store mannequin with cold, dead eyes. The reason I gravitated towards Xtranormal is they had more diverse assets. I think Muvizu hasn't really acknowledged that its art-style has limited how many people decide to use it. The software's been around long enough that they probably should start to ask some tough questions about what is holding it back from hitting critical-mass. I thought that thing they stuck on the homepage a while back was meant for gathering that sort of feedback. To me, the character designs are at the top of the list.
2011/6/15 20:48:05
Facial animation ziggy72 wrote:
, but...consider this - how many 3d packages dont have a manual? How many have courses which, realistically, you have to go on to learn them? How many dont mention vertices, polygons or UV maps? If you really want all that (and the control that it will give you) then Muvizu isn't the place to look for it.

Where in my post did I say I wanted that?
What I would like to do, for starters, is be able to bring the Xtranormal characters into Muvizu and have them act the way they do in State. Just continue where I left off. I bought them, and State is soon to be abandonware. If I have to port the gestures over bit by bit, I will. Once the data is in Muvizu, THEN I can use all that simplicity. (I have no desire to keyframe every knuckle.) But I will have the style of characters and animation to which I've become accustomed. Enabling this doesn't mean turning Muvizu into Maya. All it means is exposing and documenting the architecture. To date, Muvizu does not sell assets the way XN does, whereas XN wraps their assets in a very thin layer of DRM and obfuscation. So the money angle really isn't there. (I don't know how Muvizu intends to monetize its IP, BTW)
Anyone who cares to know what I intend to do can just look at Nothing that I've done so far turns State into Maya. It's merely opening the system up so that you can make it more flexible. If you don't want that flexibility, don't use it. But at least it's there.
What I don't like is that animation systems are too tightly woven with the assets that the creators ship with them. So Muvizu can not be evaluated as a general purpose animation system, even though it is one. It has to be evaluated as an animation system that has a limited number of assets of a particular style which behaves in only one way. Once the system is more open, then some (albeit a select few) people will go in there and expand the assets out.
Isn't this what people were talking about earlier in this thread? It's not like I'm asking for anything that different.
We already know that they are working on character imports, which I intend to utilize. But what I don't want to have happen is get the characters ported over and suddenly they start angling their hips and swinging their arms around like the Veggie-tales guys. It will be like they are being possessed by some crazy spirit. I need to be able to port over the entire personality, if you will, or do my best to simulate it.
Muvizu shouldn't just be hardcoded to utilize one set of animations, one set of expressions, one set of idles. It should be either bound at the character or theme level.
This is a very niche segment right now and there aren't a lot of options available. I've put all this effort into State because there aren't any greener pastures that I can see. You can name-drop all you want about Moviestorm, iClone, or Daz3D. I've checked them all out. I know what they do well and what their downsides are. I'm just saying that I fail to see the downside in accommodating my request.
I think some people are interpreting this as a malicious attack against the company and I wish people would just listen to the content of my argument before blindly defending the status quo.
pages: 1 2