PDA

View Full Version : Opinions – LW rolling upgrade strategy ?



Yog
08-21-2006, 05:53 AM
I was wondering about the average users opinion on Newtek’s decision to make Lightwave-9 a rolling upgrade for it’s possible 12-18 month cycle ?

Certainly it keeps more users happier, or at least more patient, for longer. For example, if LW-9.0 had been released as a traditional upgrade in it’s current state, i.e. with very little changes to Modeler, little in the way of character animation tools, no leaps in GI rendering, and practically no tool integration/streamlining (the opposite in some cases), many users would be quite critical, however, as NT have stated that these area will be looked at during the rest of the LW-9.x cycle, most users will keep their peace to see what evolves, even if it takes another 12-18 months to achieve.

So there are obvious benefits to Newtek, as well as users not having to wait another full upgrade cycle before (hopefully) their area of interest is addressed, but is it the right course for every one ?

For my personal copy I feel this is a good thing, but in all honesty I don’t feel I could recommend LW-9 to the company I work for because although there are signs to where the program is heading, I don’t really know where the program is going to be in 12 months time. With planned changes to fundamental areas of the program who’s to say what areas are going to be stable at any given month ?
How do users/managers from other companies feel ? Am I just being old fashioned or snobbish in thinking that this is a tactic aimed more at keeping students/hobbiests happy, whilst at the same time not providing to sustained period of stability that companies require ?

I'd be interested in what others think.

kopperdrake
08-21-2006, 06:57 AM
Doesn't Max have a similar kind of maintenance strategy? It doesn't seem to hurt them, or does it?

I for one don't mind - if it's bought Newtek some time with this iteration and kept my favourite 3D app up and running then I'm happy. I use it professionally, I know other studios using it professionally, though none of us are large by any means, but they've also updated it. This release bought enough changes to keep people happy and it's proving stable so far. In all reality, if you're using GI at this time then you'll be using a 3rd party plugin for the project and comping effects you can't render afterwards, though the choice would be nice.

I think it's going to depend on studio to studio, but I personally don't have a problem with it.

Wonderpup
08-21-2006, 07:04 AM
Clearly this strategy is a compromise between the need to rewrite and the need to stay alive in the market. Personaly I am realtively happy with what Newtek have done and where they seem to be going. I take your point about stability, but it's also true to say that a 'rolling' development allows bugs to be addressed more quickly- so I guess it depends on Newtek squashing bugs faster than they introduce new ones.

I do think some of the expectations for 9x are over optimistic- for example I seriously doubt we'll see a new character animation system added before 10 or later.

Yog
08-21-2006, 07:48 AM
I take your point about stability, but it's also true to say that a 'rolling' development allows bugs to be addressed more quickly- so I guess it depends on Newtek squashing bugs faster than they introduce new ones.But there lies the rub. In every major upgrade of LW since at least LW-5, there has been a bug or two that has been introduced by the patches that was serious enough to warrent a patch of it's own. Those were just effecting very localised portions of the program, however what is proposed potentially effects large areas of the program in one go.
Take modelling tools in Layout. If a bug slips through here it potentially effects, the mesh deformation tools, sub-D interpolation, animation tools in general, and most likely rendering.
Add to that the other areas that Newtek staff have to a larger or lesser extent indicated will be looked at in the LW-9x cycle (reorganising of Modeler tools, enhanced edge tools, updated animation system, revamped GI lighting, etc), and the program begins to look as fluid as quicksand.

The tool junky in me likes the new direction as it means I get drip fed new tools throughout the year, but as the person who has to maintain the section and sign his name to the justification reports I would prefer Newtek put one feature frozen product on the market, then spend the next portion of the year refining and bug squashing, without adding to the very real possibility of introducing a new show stopper.

It's amazing how conservative you get when it's your name on official reports ;)

kopperdrake
08-21-2006, 05:12 PM
As a side note, you don't *have* to install the patches. You can always wait for the first iteration of each release if you like? Or choose to upgrade half-way through the release only, or wait until others have found a small amount of bugs to make you happy to take that plunge.

GregMalick
08-21-2006, 05:56 PM
As a side note, you don't *have* to install the patches. You can always wait for the first iteration of each release if you like? Or choose to upgrade half-way through the release only, or wait until others have found a small amount of bugs to make you happy to take that plunge.Or you can do what I do - save off the point releases into it's own directory and then upgrade. Disk space is cheap and you can easily fall back to a safe point release. Mission critical work can be safely done in the version you trust most. I have actually kept versions back to 7.5 I think this is a common tactic.

kopperdrake
08-22-2006, 06:17 AM
Heh - I kind of do a similar thing, I take an older PC and install there first, play around with it and if it works then all's well with the world :)