When is a software package "Dead and "gone"??

#10

I think Shake will be used on features for years more. The studios who bought the source code from Apple are free to update and extend it, build it into their own software in modified forms, etc. and will probably keep using it until they decide there’s something else that’s worth the time and money (mostly the time) to switch.

I’ve seen studios still using Elastic Reality many years after Avid discontinued it. They were using it for roto, not morphing, but if a tool works well and you already own it and know how to use it, the fact that it’s no longer getting developed isn’t by itself enough of a reason not to keep using it on more films.

-jeremy

#11

A lot of people still use lightwave and silo. :stuck_out_tongue:
runs away

seriously ??
it is necessary ?
how about that one …?
amazing lightwave rendering

http://forums.cgsociety.org/showthread.php?f=154&t=789958

it was still on the front page yesterday

now show us some of the amazing work that you are doing with those 3d software that are still ““alive””

then it would help you to validate your opinion

thank you

#12

Xtrm3d, Bucket was just joking (and I got a laugh out of it) about Lightwave’s (perceived) popularity and Silo’s lack of development. Everyone knows they’re solid programs.

The one that gets me is XSI7, when people go on about how it’s dead now Autodesk own it. That program is going to have legs for a loooong time.

I’ve seen studios still using Elastic Reality many years after Avid discontinued it.

Via XSI, or…?

#13

Its a slow death with autodesk, not like their going to discontinue the program or absorb it into maya. (I’m saying this half jokingly.)

#14

Perhaps, but the point is that even if you never upgrade it, it’ll be good for a lot, for a long time.

#15

ho yeah, companies have been keeping around an SGI box just to run an old copy of Elastic Reality for the morphing.

However, in the last four years many clones of the tools have apeared in all compositors, and the fact that ER doesn't support floating point images (like OpenEXR) makes it now totally obsolete.

I have indeed heard that people have been using it for roto as well, but I can't see why.  It only has bezier curves (bsplines would be a better option), there is no tracker, and the performance is not great.  Except for its nice shape grouping feature, you might as well use an old copy of Commotion if your going to go for old.  Or go for Combustion, whcih does have the bezier/bspline, dopsheet, etc.

p.s.: I 'm the person that put Elastic Reality in Softimage|XSI

#16

Not much point in me showing rigs. A bunch of bones or joints rarely makes me people go wow, no matter what it is demonstrating.

Anyways, sorry I offended thee. I promise to mail your goat back to you as soon as I acquire enough stamps to cover postage. :thumbsup: (another joke, I do not have a real goat in my possession, only the imaginary type and I have plenty of postage stamps)

#17

I’m still using Animation:Master! :slight_smile:

#18

heh, you would be surprised. It will be a while.

#19

The only real problem with using software that isnt supported anymore, is that you’re stuck if you run into a bug or voodoo of some sort. Typicly when you’re deep into the most advanced features and really pushing the envelope of what you can do, then you might discover that this causes the program to crash or do the wrong thing. You cant go shouting at the backoffice support to fix it, and you cant fix it yourself because you dont have the source code… and if you did, its not properly documented. And so on.

#20

The problem may occur when your PC is gone :cry: The new PC will require new OS, because the old one (if it is really old) will be unable to run or to run properly - no drivers, some kind of noncompatibilty. This new OS will not accept old software as you can not run a bunch of software in Vista. So this is the magic formula for kicking users to spend a lot of money for upgrading.
Did you saw masters (mechanics for example or painters) and their tools - they are old, rusty, ragged, skewed, nasty looking but they know them from many years and can make miracles. They know them and all they need is time to use them.
I am still using RealDraw 4, which is about 4 or more years old (recently was an upgrade to v. 5) but I know it from 2001 when it was v. 2.2 so it works. And I am using XP 32 :slight_smile:

#21

Part of the reason everyone’s saying Shake is still in wide use is that a lot of the big studios bought the source code. So a bug could be fixed, or it could be re-compiled with a newer compiler, etc.

The home computer I’m typing this on is XP Professional Version 2002. I wonder if Microsoft will make another operating system after Windows XP? Did they get bought by Apple, just like Nothing Real? :slight_smile:

-jeremy

#22

Multi-year productions don’t upgrade mid-production. They still use older versions of things. One house that I checked out a few weeks ago was still using shake 3.

#23

I’d say the biggest thing would be the bug/ feature side of a program, i mean if you look at programs today, they aren’t out of date, even after a few months/ years, you can still do amazing work with them. But what you can’t do is get support, be it in the form of bug/ software support or development and feature support.

I think for the largest studios who own source code and/ or have custom built apps then it’s not really an issue, but for the smaller studios who can only afford the license, hence only renting the software, then the software becomes obsolete once it can’t do a certain feature which is a must for a project or several.

Just my opinion, could be wrong.

#24

Why programs like shake get abandoned:

They stop supporting modern hardware. No GPU accelleration. Unoptimized for multiple cores. Lack of OS and driver support. No x64 support. Inability to adjust to changing trends with new workflows such as 3D compositing. Attention from third party plugins shifts to other packages leaving no new features.

Also when you’re building a pipeline you want software that will grow and improve. It’s a lot of work to fit a new tool into a pipeline. If the software is going to continue evolving and improving over the years but still be largely compatible with your pipeline integration with a single investment. We have thousands of hours invested in software specific tools. If you are going to need to switch applications in 5 years because it’s so slow and lacking in features then you’ll have to also recode all your inhouse tools. That’s a lot of time and money.

That’s why programs like shake stick around so long in the first place. It’s hard to switch. So people try and divine a software road map which requires as little switching as is humanely possible. Staying with one package for more than a decade means less retraining, less development and less chaos. Buying into a product which has already reached EOL means a guaranteed switch. It might be 3 years before it really starts to feel outdated but 3 years isn’t very long in pipeline time.

#25

thought Steven Wright’s piece on this was intreresting.

http://www.vfxio.com//PDFs/Legacy%20Software%20Conondrum.pdf

#26

Anyone remember SEGA’s Animanium?

Im sure Robert Chang will but never knew many people who used it.

I recently talked about it and uploaded a video demo of it on the latest episode of cgRoundup (sorry for the shameless plug :))

Video Demo of SEGA’s Animanium (7.31mb)

I also wish Clay3d would have got developed further (and actually released). Think the latest version was 0.25? Always remember Martin Krol dropping that name quite frequently when talking about modelling apps, which of course made people insanely curious.

#27

Dude, I know you’re poking fun and trying to get a reaction, but your troll is just stupid. The original poster wanted to know about why people stop using software that has stopped being DEVELOPED. Well, whatever. Ha ha. XSI is dead too and who the hell uses Houdini anyway? :rolleyes:

Use tools that work for you, I think is what it boils down to. There’s a lot of software out there and not all of it is “better” - only “popular” because people run with the latest fad. As an example - Final Cut Pro is a great editing software, no doubt… but ask around and people are like “Premiere is dead” - yeah tell that to Adobe who just released a patch for CS4.

Its the OUTPUT that matters. Granted, knowing a piece of software can get you a job if that job uses that software… but if you are a freelancer with your own clients, all the clients care about is the output. And if they ask you “so, you use Nuke?” and you are using Shake… just say “yeah Nuke is awesome” and you get paid all the same.

#28

:open_mouth: I remember clay! I always wish I could have tried it out. MK seemed to think it was the bees knees back in the day.

#29

That’s a an interesting topic to me, having witness the demise of Elastic Reality, Matator, Media Illusions, and other applications. I don’t believe it is related to not supporting new hardware or OS (porting to 64-bit is really easy). Rather, it’s the more fundamental “what’s next?” problem.

I believe the reason Shake was abandoned is because it has nowhere go to given how it’s designed. I know users will have a hard time agreeing with this, because they’re completely satisfied with what it does and they have a few ideas about what could be added in it, but a software needs to grow its market. Shake is just a script interpreter, the UI is just a front to generate that script. Get a book like the Dragon Book of Compiler, and you’ll know how to write your own Shake. The script is parsed, the syntax tree is built in memory, and executed. If you want to go real-time, or put it in a 3D Space, all the sudden you’re limited by the fundamental approach the app is based on, and you’d end up re-witting it all and then it’s no longer the same product.

Beside all the corporate problems, what led to the cancellation of Elastic Reality, Matador, Illusion, etc was also the question What’s Next? That might be, Motion Graphics, go after Commotion, go after Discreet Flame. People look at how costly it would be, and just how big the market is, and the odds of making inroads, and the numbers simply do not add up. Generally, that next step is going to be two years or more of developement, and you’ll have to sell lower than the previous product, and the competitors will have had more two years.

No one really asks these hard questions early in life of a product but when the team comes in and say they need a year to re-write a large chunk of it, it’s the most dangerous part of a software’s life.