So question for the experts, does anyone actually use Mari in there work-process. I'm just thinking of if I should invest myself in to this program. I mean I know it has a multi-layered system, which might be more efficient then zbrush since its a bit crashy in zbrush. So does anyone use it or any good benefits of using it with zbrush?
Replies
I think you might have a better time in Mudbox rather than Mari right now, Mari is really designed for a film pipeline and not Games. Mudbox has a great layer system and whilst it might not have the procedural or masking controls of Mari Mudbox also allows you to sculpt and the baking of textures is great.I Teach both Mari and Mudbox ,coming from a games background I would go to mudbox first, you arebeing able to paint in PTEX and then bake that down into UV space at the resolution you require.
cheers
http://www.polycount.com/forum/showthread.php?t=123839&highlight=Mari+games
... and Ready at Dawn and quite a few others.
http://blog.selfshadow.com/publications/s2013-shading-course/rad/s2013_pbs_rad_slides.pdf
Games is a big push for us this year (I'm the product manager for Mari) and I'd really like to know the short-comings in Mari so we can address them.
We've got a new version out fairly soon that should have some VERY relevant updates it in.
On the list to look at next is better FBX support and Multiple-UVs. What else should we be looking at?
Please keep the discussion to technical rather than price or licensing though. We are very aware of those issues already.
well to answer your question you guys have some tough competition to be more specific, after doing some research
1. Zbrush vs Mari
a. Pros: the best and most popular clay sculpting program in the business
b. Cons: Does layers like Mari but it is very limited and often crashes
2. Mudbox vs Mari
a. Pros: Can clay sculpt almost as well as zbrush, the layered painting system here offers most the same features as Mari, with the exception of ptex texturing. Since it can paint on the lowest sub-d level, it makes it very hard completion for Mari, especially since it sculpts as well.
3. Substance painter vs Mari
a. Pros: offers a particle brush which can do painting simulations unlike any other product in the market
b. Cons: still beta and not as feature packed as Mari
4. Ddo plugin for Photoshop
a. Pros: Using the power of Photoshop it actually offers a lot more painting power then Mari, also gives you real material previews
Suggestions for Mari, really whats important to understand is that before even putting a model in to Mari most artists have already worked on that model in zbrush or mudbox, and since painting features are offered in that you will have to offer a lot to motivate them to use another program in their process. What I would suggest is letting artists do HDRI renders in Mari so that they could quickly demo their art to bosses. Also giving Mari material asset control which integrates with major game engine. Substance painter will mostly likely have this feature once they exit beta since they have it on their other products. But essentially it is about editing materials in Mari, then through a plugin or the likes that material would transfer over perfectly to game engines.
sorry for the flip "Designed for film comment" I guess I just meant that its initial focus was working with huge data sets and in a linear 16/32 bit pipeline allowing for large multi tile texturing at 4k which is beyond the remit of games
I suggested Mudbox because of the baking tools and the simple but well implemented painting tools
Mari is fantastic,the sharing of textures between layers makes it very efficient
The triplaner projection shader and the non destructive procedurals make texuring a thing of joy.
the layer stack workflow allows for a very deep non destructive workflow
what it lacks at the moment are processing tools for UVs and baking. it would be great to use ptex and bake to UV space when ready
I would like to bake between 2 models, if UVs need to be changed and keep all stack information
It would be great to paint opacity and view it in Mari
if you are used to using a 3D painting package that samples the surface you may find the paint buffer and baking a little difficult at first but you quickly get used to it
The learning curve is a little higher, but once you are comfortable with it is a joy to use, I would use it in production but one cant deny the cost is high compared to Mudbox for example and so a manager might be scared off at the cost
I hope that helps
thanks
Mari is a dedicated texture app in imo should stay that way, plenty of other sculpting programs that are good at that other stuff
I imagine this is already being implemented, but some added shaders for mimicking UDK, Cryengine and Marmoset materials for getting the final textures closer to what they would look like in-game would be pretty cool. Will update this if I can think of anything else, only recently started using Mari but love it so far!
http://vimeo.com/65408722
Yeah, Jgreasley already posted that.
I definitely agree though, used to texture in Mudbox before and don't wanna go back now, it's truly revolutionary... Hope a lot more studios are going to start using it. Shared channels ftw
I see the price as being a major signifier of their target market...
It was about a year back that jgreasley said that:
"Developing for, and supporting, cutting edge vfx work is expensive. The budgets, schedules and technical demands are intense. VFX support includes everything up to flying developers on site to fix production issues at midnight. We've released 21 versions of Mari in 3 years in response to customer requests. At the moment the current price of Mari correlates pretty directly to the cost of meeting those demands."
Regarding price, it may work for a specific area of film and vfx, but its quite beyond what most consumers would expect for a painting application. If the cost is due to sending support and making changes asap, then perhaps a version without that is needed since I doubt Mari support will be flying out to individual user's homes or taking their request with the same weight as a studio. So as an artist you are paying for studio support and preference you will never actually be given. This probably needs to change to be more accessible...and competetive.
Perhaps some version for steam would be more appropriate. That said, knowing your target audience and working within their range, competitively, is whats going to make applications like Mari and Substance Painter a success in this regard. I would like to see more competition between the two, price and accessibility wise. Allegorithmic is winning out on this one (they are planning 16k support at a price point thats below $1000).
Whats ironic, at least at this point, is that Allegorithmic has tighter Modo integration with their texturing software and Mari, which is also by the foundry, does not. That probably shouldnt be happening.
Flying people on site isn't something that's included in the support price, but we have had to do it in the past
We do, however take everyone's input equally seriously. Our support is awesome, even if you're a company of 1.
We have individual users, literally working in their bedroom, who are as influential in the development process and companies like Pixar and Dreamworks.
Scott Metzger's history with Mari is a great example of this. He's a freelance guy and is doing awesome, cutting edge work with Mari and he has had features implemented in the last 5 or 6 versions.
https://www.fxguide.com/fxguidetv/fxguidetv-165-scott-metzger-on-mari-and-hdr/
We are certainly listening to everyone's input on price and features regarding the games market.
I can vouch for Jack on the Foundry's support, I teach Mari at Escape studios in London and Jack has come in many times to talk to our students about Mari.
He also implemented a environment shader module after I requested it and we are and educational establishment not a big VFX studio
I have just finished teaching a week course in Mari and the students love it,none of them had done any 3D painting before and they really took to it
I only mention price as I know it is a factor in games production, some studios pockets are not as deep as others
to reiterate
Mari is a best of class powerful texture painter
I would download a demo and give it a go
In my opinion Its missing
Baking functionality
Opacity in the viewport
Easy to load Game shaders
I hope that helps
Depending on the specifics, I'd be interested in that.
I have no doubt about Mari's capability, but I question the accessibility front and how its being presented for a market outside of vfx and film. The market has texturing applications, usually with dual functionality (such as sculpting), ranging between $300-800 tops. Then there's Mari, without that dual functionality going for $2000. Its like its not even trying to compete or it over-values itself, thus I freel strongly if it wants to hit the markets that include the average artist or game focused areas, its going to have to be more accessible and work within/for the other markets.
I can guarantee you there would be far more profit in quanity of sales if it was priced in the same range as current industry painting applications.
I'm going to be in the St Regis Hotel bar this evening between 6-7pm. I'll be wearing a blue shirt with a yellow paint splash on it.
Come say Hi and I'll grab you a beer.
You can drop me an email at greasley@thefoundry.co.uk.
We're running demos in a suite as well, so if you want to see some cool stuff let me know and we'll try and slot you in.
Sign up http://bit.ly/1pypYXg to see all the games goodies 2.6 has in store.
http://vimeo.com/90986118
1. An Opacity shader option
2. Having a GOZ type feature to quickly send files to Maya and 3ds max. Setting up a composite shader in 3ds max or maya can take a long time for artists that want to utilize Patches, there should be a function to automate this. Especially sending to Zbrush so that the programs can link versions in mari with subdivision levels in zbrush.
3. Also all game engines dont work with patches. It would be nice if when importing multiple OBj files in the same scene if it would automatically move their UVs in to different patches, Otherwise lets dsay your working on a character with armor and you want different uvs and textures for the character and the armor. You will either have to texture the various pieces individually, or merge, offset the Uv map, offset them back, unmerge.
1) Agreed, Opacity is always nice.
2)A goZ style feature would be quite nice.
3)You don't have to use the multi patch feature, just import multiple obj's into your mari scene and you're all set, each with their own individual 0 to 1 space.
For example "weapon_diffuse.UDIM#.exr" or "weapon_diffuse.<udim>.exr". But it depends on the renderer. Then the alembic support could be useful for you.
But I guess its not possible to send your files back from mari to maya, softimage or else. But you can´t create UVs or change the geometry in mari so you won´t need this export feature.