How about Open Sourcing Silo?

#1

After what I can see it hasn’t received any updates the LAST TWO YEARS or so.
Is this application about to vaporize into oblivion?

If so I think it’s a shame that they don’t open-source it, so that others can take the responsibility of bringing this program up to date and fix eventual bugs etc.

This is a open suggestion:
I work as a programmer using C++, Python, C# and some other tools.
I’d be honored if I could make this project move forward by engaging a group of fellow programmers to take care of this extremely valuable little modeling application. :wavey:

There is a definite need for an application like this for people that are sick of all the never-ending bugs that keeps flowing in and out of Autodesk’s products and for those without the financial backing needed to use these programs legally and do not qualify for student software. :cry:

Now that The Foundry has snatched Luxology, the prices will only keep getting higher, and Modo which was once a plain polygonal modeler starts to take on a “cram it all into one package” appearance. :banghead:

Some people might like that, but I believe in the old UNIX philosophy, to have one tool do one thing extremely well, and then combine multiple such tools to create a production chain. :applause:

So, if anyone is with me on this way of thinking, let’s gather the forces and start by contacting the representatives over at Nevercenter.com.

I have been using both Maya, 3DS, Modo and Houdini legally in the past, but I just can’t keep up with the costly updates.
Therefore I’ve started to learn Blender, and it really is a gem in the rough. Some more years of development and this post would never been written, but still I feel that having a separate, fast and STABLE application for polygonal modeling somewhat of a must.

Personally I have a commercial license for Modo 401, and it does what I need, but this is a way of trying to start a community effort, for our future as 3D enthusiasts, not only for the professionals in the industry.

I’ll include a poll, to check out the interest for trying to buy the rights to the source code from Nevercenter and turn this into a Open Source project. :buttrock:

#2

Frankly, I think you would be better off moving this poll elsewhere since this place is pretty much dead. And it’s not like it is only Silo users that would be interested. So maybe post another poll in General Discussion or have the mods move this thread there… :slight_smile:

#3

I’m a developer as well, and cautiously like the idea. Blender is like someone taking a ball of clay and rolling it across a filthy carpet. It is a horror story of UX design. I fear that open source development of Silo might lead to uncontrolled overgrowth of the program.

Either way, I doubt nevercenter would ever open it up. I’m guessing they’re just trying to ride it out a bit longer while they still have any selling point or draw in the industry, trying to sell the IP rights to the highest bidder.

#4

They might not even have the source code anymore.

#5

That or some other tragedy could explain the situation.

If its gone, its gone.

Maybe we can make an opensource spiritual successor.

#6

I would love to see this program gain more support and features while still being a dedicated modeler program. Then maybe have an animation program develop along side it. I love working in this program but it’s limitations as far as file import and export and a general lack of support is irritating.

Hopefully this program doesn’t get sold off to some greedy shmucks like Autodesk who will bloatware it and jack the price up so they can extort large companies and down right shit on individuals.

#7

Don’t forget about Autodesk slamming on a “View Cube” and some stupid “Fly Through” tools as well.
Plus integrate the severely damaged edition of MentalRay they push onto their clients. (XSI Might be the only of them that has a relatively proper implementation of MentalRay integration.)

:banghead:

#8

Good point. That might explain why they haven’t done any updates for two years.
But they’re still milking the cow though.
Selling obsolete software. But it works, so I guess that makes it “business as usual”.

I work as a developer myself, so…

#9

I believe Silo is built on top of a cross-platform UI library which is commercially licensed (can’t remember which one, but it came up as the reason for the longstanding problems with Silo and some version of Mac OS X) and this dependency might make open-sourcing Silo problematic.

Sadly, I think the best way for a programmer with good UX sensibilities to contribute to the 3D world would be to get on the Blender bandwagon and try to help out – the major UI push with Blender 2.5-2.6 has paid dividends – it’s far from perfect, but much better than it was.

#10

An absolutely understandable and probable but tragic scenario. Let’s hope not…

If I where to start maintaining and developing features for this app, I’d use a C++ GUI framework like Digia’s QT for the GUI part and have very strict UI and design Guideline for fellow programmers, following common User Interface Guidelines and tried and tested approaches from other applications with the ability to customize as much as possible on a user level.

On a conceptual and architectural programming / source-code level it will be GIT and obligatory peer code review, utilizing the best of the OOP paradigm, as it fits best for this kind of application.

C++ for the meat and potatoes and Python for the API (maybe also the GUI) layer is a nice combination I think.
:thumbsup:

#11

Maybe that will attract some more people to the post.

Chris

#12

All the code rights issues aside, why would they?

If they sell 1000 licenses on steam it´s an easy 40K. And Steam is a gigantic market. Still lots of good revenue to be had there with Silo.

#13

One option might be to encourage them to start a kickstarter campaign to generate investment to fund the development of a Silo V3.

Chad

#14

I know they are still selling Silo, as a long time user of the program, because they sell it on Steam and Smith-Micro is selling it for them along with their Camera Bag software.

#15

A fund was suggested at Nevercenter many times, however, the developers simply never responded to anything after a while.

Silo is small, fast and intuitive, the only problem I see with open source development is that people would try to add thousands of abilities, ala 3DS, and ruin a very good program. What Silo needs, and has needed, is about a half dozen additional commands like the ability to chamfer, bend, etc and then refining what is already there. Just my thoughts.

#16

This is excactily what I had in mind. A Silo 3.0 “Polygons Reloaded Edition” with the same tools, plus additional tools for doing effective Polygonal / Subdivision modeling.
I’d strip away the whole sculpting part of the application as everybody is using specialized applications for that anyway. I’ve just started to learn how to use the sculpting tools in Modo 701, and I rarely need to use the GoZ button anymore.
It’s actually that good. Combine that with Modo’s unmatched UV / Texturing tools and you have a really nice package there.

I actually went on and upgraded my 401 license (previously updated from 301 EDU) and the licensing model for Modo is something many companies could learn from. They give you the ability to upgrade for the same price from ANY old version to the latest at the same price, without any stupid subscription that costs you like $2K a year just to get some add-ons you don’t really need, just so you can upgrade to the latest Main Release when it arrives without paying for a new license. :deal:

Take Maxon as an example. If you skip the C4D subscription, updates are much more expensive, and if you’re more than two releases behind, you’ll have to buy a brand NEW LICENSE AT FULL PRICE! Now that’s customer loyalty for you! :wip:

I paid around $180,- for the Modo 401 -> Modo 701 upgrade during a Christmas special.
That’s 40% off, so the price isn’t all that bad to begin with.
$999 for a single commercial license with unlimited (and extremely easy to set up) network render slaves. Plus, the license is bound to YOU as a USER, wether you use it at home, at your friends house or on a laptop on the beach. You just can’t have the same license open on two machines on the same network without entering slave-mode on one of them.

Now that’s LICENSING with STYLE! :buttrock:

Wow, that was quite a digression!

But anyway, I was thinking in the same terms as the one I quoted.

Priorities:

  1. Stabilize / clean up existing functionality and remove unneeded bloat.

  2. Add the missing functionality in terms of modeling tools.

  3. Update the UI with a more user friendly (that means Icons are Optional) and modern look and feel, following common UIG’s (UserInterfaceGuidelines) for 3D applications, with a high degree of customizable parameters. For example, assignment of hotkeys can be done from a list of action history, and stored as user config files along with the other configuration of the UI etc. For the actual workflow, I’m thinking in lines of a central “Hotbox” kinda thing (for those who know Maya), but implemented in a more Modo-like pop-up Pie-Menu style, so that the user can have the whole viewport for modeling or choose a more traditional “Button layout” with or without icons and hoover help as configurable options.

  4. Implement new UV tools based on the existing ones and a texturing module that supports both OpenGL, and DirectX based render pipelines (GLSL/HLSL/CG), including a node based material editor.

  5. Introduce a render preview window that plugs into your renderer of choice, to quickly see if the textures / materials are as they should. Also a possibility to use game engine renderer to check out the looks after that kind of tessellation. These could be selectable via a drop-down in the render window and produce different versions of the same scene, with a “history row” in the bottom part of the window, showing the cached renders from your current session, with the ability to store all or export each .EXR to a permanent disk on file. Conversion options included off course.

  6. Create “state-of-the-art” documentation for the product and tools with both text and videos describing how to utilize each tool most efficiently, in addition to the standard PDF based documentation.

I believe there is a need for such an application, that can be used as the first and second stages of a production pipeline or as a complete tool for product visualization artists, without any form of animation, rigging or sculpting tools, but “just” a basic but Professional subdivision modeler that only renders through existing established renderers, be it Open Source solutions such as Lux Render and Yafaray or commercial suites such as Mantra, MentalRay, Arnold, V-Ray and Maxwell to name a few.

Chris

#17

Wow, Modo sounds really good. No wonder its the standard. I especially like how the licensing is handled. I seriously hate having a license semi-permanently tied to a computer.

#18

What do you mean by “standard”?

#19

If I’m wrong, correct me. I’m not in a position to argue about it, as I can only base the statement on what I’ve read, rather than first hand experience.

I was under the impression that Modo was as widely used as Zbrush when it comes to modeling tools.

#20

I’m one of those individuals that doesn’t believe that the source code of Silo will be released any
time soon, if ever. On the other hand, if some good programmer has the enthusiasm and free time to
do something about an abandoned 3D program, there is Misfit Model 3D. I’ve nothing to do with this
software…except for testing it for a couple of months on Windows 7 (and custom. all the hotkeys).
It’s very basic compared to Silo, no sub-D, more like MilkShape3D. OpenGL, very stable and simple.
Sorry for bad english.