To embed or not to embed...

[rusty]

Hi,

 

I want to know more about the pros and cons of embedding resources in a project (and therefore embedding in the chor).

 

For myself, I have since my own 'dawn of time' with A:M embedded everything, always. Why?

 

Dependability and minimizing loss is always my first priority. And, probably around 10 years ago I started embedding everything for that reason alone--I seemed to have less problems doing it this way. Part of this was that separating the current ongoing work from the resources I had on disk isolated and protected those resources. I ran into situations where A:M crashed, my PC crashed, there was a power outage, I screwed things up and could not undo and this as often as not messed up resources on disk I'd spent hours perfecting. Oh, I backup, backup, backup everything incessantly still...I lost more work if I had resource files simply linked into a project.

 

Model development mirrored the above--saving the model then embedding it while working on it provided automatic protection of what I'd accomplished so far.

 

Then came the need to consolidate. If I moved a project file or sent it to someone else...everything (except images) was all in the project file nice and neat.

 

Then came my resource pool. Reuse became central to everything I did but resources in the 'pool' could easily become messed with unwanted changes when I was just looking for something and I forgot to embed (once I decide on using a resource it is copied to the project folder...again consolidation...everything used for a project is in the project's folder). At first I made everything in the pool read only but... this was a hassle when I wanted to add a new model to the pool or update the library files. So, my personal workflow became open and immediately embed it.

 

But best practices change and regardless of everything above, if it makes more sense to not embed, or, even more important, if it makes things more reliable and reduces loss...I'll do a 180 on how I do things.

 

Please let me know your thoughts, experiences and your own rational on this topic.

 

Cheers,

Rusty

[higginsdj]

Its a work practise. If you embed, you have one copy that only you can work with of any element used in that project - and only one version - that projects version. So if you want versions of ANYTHING then your have to create versions of EVERYTHING - ie the project. Not a very efficient way of doing things and how on earth do you keep track of versions of individual elements within the project?

 

What is your resource pool? What tool do you use? What is your file system/method? What are your naming standards/file structures etc? If your library/pool is properly structured I can't see how it could get messed up - especially if you are using the right tools. It's a bit like software libraries. You check stuff in and out of the library and those tools typically control your versioning. This is how it worked in TWO and SO and I dare say these libraries may be somewhat bigger that yours and they were managed across a lot of users over the internet! (My copy of the TWO data pool is 10Gb while my copy of part of the SO data Pool - just the parts I worked on - is 5.5Gb)

 

Cheers

[rusty]

Its a work practise. If you embed, you have one copy that only you can work with of any element used in that project - and only one version - that projects version. So if you want versions of ANYTHING then your have to create versions of EVERYTHING - ie the project. Not a very efficient way of doing things and how on earth do you keep track of versions of individual elements within the project?

 

What is your resource pool? What tool do you use? What is your file system/method? What are your naming standards/file structures etc? If your library/pool is properly structured I can't see how it could get messed up - especially if you are using the right tools. It's a bit like software libraries. You check stuff in and out of the library and those tools typically control your versioning. This is how it worked in TWO and SO and I dare say these libraries may be somewhat bigger that yours and they were managed across a lot of users over the internet! (My copy of the TWO data pool is 10Gb while my copy of part of the SO data Pool - just the parts I worked on - is 5.5Gb)

 

Cheers

 

Hi David!

 

Thanks for replying. Good to hear from you again and I hope all is going well. I get the impression you don't like embedding but you had more questions than thoughts or suggestions. I wish I had the time to answer all of them but I don't, sorry. However, if you think back, I'm sure you were introduced to many of these things in 2005/2006 or there abouts. If not, I've discussed all of them in previous posts here.

 

Cheers,

r