Porting NE1 to win x64, weird OpenGL bug

[-]

Hi guys!

First post here.

I'm student interested in molecular nanotechnology.

I find NE1 very interesting software. I want to use it to design and simulate simple nanosystems.

Currently largest problem with NE1 seems usage of obsolete libraries (e.g. Numeric) and incompatibility with modern OS versions, so my patches focus on these issues.

So, my "work" so far has been porting NE1 to Ubuntu 12.04 and windows 7, both x64.

I have some success with Ubuntu and Win7 port (now I work on Win7 port), but I got weird OpenGL clipping-related bug, described here:

https://github.com/kanzure/nanoengineer/issues/5

I can do a screenshot if you wish.

Intuitively, it seems that that's related to displaylists and lists of indices that are loaded into OpenGL API.

As I'm not that familiar with NE1 code, maybe someone here will propose hypotheses of possible error source(s)?

Also, I'd like to ask about this module http://www.nanoengineer-1.com/snapshots/NE1_Documentation/utilities.debug-module.html

Maybe it'll be useful in inspecting model tree (atoms) at runtime?

[Eugen Leitl]

On Fri, Dec 07, 2012 at 09:49:58PM +0400, - wrote:
> Hi guys!
> First post here.
>
> I'm student interested in molecular nanotechnology.
> I find NE1 very interesting software. I want to use it to design and
> simulate simple nanosystems.
> Currently largest problem with NE1 seems usage of obsolete libraries (e.g.
> Numeric) and incompatibility with modern OS versions, so my patches focus
> on these issues.
> So, my "work" so far has been porting NE1 to Ubuntu 12.04 and windows 7,
> both x64.

Thank you for your work. Maybe NE1 can be cleaned from bit rot yet,
after all.

[Bruce Allen]

Hi!

Please check out my project at https://sourceforge.net/projects/moleculardynami/

If you want to update NE-1 to stop using old packages that is great! I want
to as well. There are resources I point to on my page that can help you build a development environment for Windows. I don't think one exists for Linux. Windows SP1 breaks NE-1. There is a DLL in the NE-1 distribution that must be rebuilt for SP1.

I'd like to find others interested in working on Molecular Dynamics Studio to it becomes a collective effort and perhaps become successful.

-Bruce

[Bruce Allen]

What graphics card are you using? I seen your issue with the gto 7950.

[Bryan Bishop]

On Fri, Dec 7, 2012 at 11:23 PM, Bruce Allen <clau...@gmail.com> wrote:

I'd like to find others interested in working on Molecular Dynamics Studio to it becomes a collective effort and perhaps become successful.

I still don't understand why you're forking. I suppose I don't have to, but it's still peculiar. Are the other branches up on github not working? Have you considered filing bug reports?

- Bryan
http://heybryan.org/
1 512 203 0507

[Bruce Allen]

Bryan,

I'm forking because I'm an old programmer and don't understand that github thing. Plus, if everyone using it doesn't sync their changes with each other will anyone's changes work? As far as I know NE-1 is dead and provides a platform for building other applications? Correct me when I'm wrong.

My changes were made for my graduate thesis. Does anyone have a master plan/schedule for NE-1? I wish someone does. I want to add another 1000 features to it! If you check out my project you can see that all the feature functionality could be added to NE-1 including integrating LAMMPS.

I would like to find other developers, like yourself,  to update and enhance NE-1. If using github is a better idea I'd like to learn more about it. Ideally, it would be better to have a single source tree for NE-1. Do you know how many projects are using NE-1 as a root for their projects?

Thank you for your comments,

[Bruce Allen]

Check out this resource.

http://www.nanoengineer-1.net

On Fri, Dec 7, 2012 at 12:49 PM, - <kirag...@gmail.com> wrote:

[Bryan Bishop]

On Sat, Dec 8, 2012 at 12:03 AM, Bruce Allen <clau...@gmail.com> wrote:

Bryan,

I'm forking because I'm an old programmer and don't understand that github thing. Plus, if everyone using it doesn't sync their

Git is just another version control system, except branching isn't terrible.

 

changes with each other will anyone's changes work? As far as I know NE-1 is dead and provides a platform for building other

Like any version control system, you have to merge changes when you merge branches together.

 

applications? Correct me when I'm wrong.

I don't know how you define dead, but it's certainly not a platform that anything should rely on.

 

My changes were made for my graduate thesis. Does anyone have a master plan/schedule for NE-1? I wish someone does. I want

There's a few notes in the README file about the chroot upgrade path. There's also the non-gui branch that focuses solely on core functionality.

 

I would like to find other developers, like yourself,  to update and enhance NE-1. If using github is a better idea I'd like to learn more about it.

If you happen to hate github, there's a non-github hosted git repository for nanoengineer over here:

http://diyhpl.us/cgit/nanoengineer

 

The remaining active developers hang out in ##hplusroadmap on irc.freenode.net,

http://diyhpl.us/wiki/hplusroadmap

Ideally, it would be better to have a single source tree for NE-1.

I disagree. Older version control systems penalized branching and working on experiments, but that doesn't seem to be the case with modern development tools. However, I think the branch called "master" should be treated as the destination for merging things in.

 

Do you know how many projects are using NE-1 as a root for their projects?

No.

[Bryan Bishop]

On Sat, Dec 8, 2012 at 1:19 AM, Bruce Allen <clau...@gmail.com> wrote:

Branching has been a pain since I started writing software in 1981.

Branching shouldn't require terabytes of hard disk space, however. Try it out with git, and let me know what you think. Instead of copying the entire repository over into a subfolder, you can just reference the previous constructed delta and commit on top of those.

As for actually merging, it's nice to be able to pick and choose which commits to include because you can accelerate your timeline for certain features as much as you personally want.

 

I don't hate github.

I assumed you did because you were using sourceforge... my mistake.

https://github.com/kanzure/nanoengineer#readme

[Kirill]

суббота, 8 декабря 2012 г., 9:28:48 UTC+4 пользователь Godzilla1138 написал:

What graphics card are you using? I seen your issue with the gto 7950.

I'm using ATI Radeon HD 4250 (intergrated into motherboard).