Jump to content
3DCoat Forums

V4.1 BETA (experimental 4.1.17D)


Recommended Posts

  • Reputable Contributor

I have had to move to ZBrush to get my jobs done because I cannot rely on 3DCoat any more, I think the turning point was when I pressed the save button and my scene file was destroyed as the program crashed and I lost many hours of work, what is amazing is that I am running ZBrush in Linux under wine and its very stable and Reliable, this is what Andrew has to do, drop all the development and make 3DCoat rock solid, stable and reliable because nobody wants to loose their work or waist their time. 3DCoat has such a great potential but nobody will ever take it seriously while it is in a constant state of perpetual Alpha/Beta. Once 3DCoat is made stable and reliable on all platforms then people will take notice and place trust in the program to get the job done, the last thing an artist wants is distractions whilst in the creative frame of mind, distractions kill creativity. New features should only be added then released to the general public after they have been thoroughly tested under every possible scenario that the user may use the feature in. As I go through all my various versions of 3DCoat I find the same pattern of some old bugs being fixed, but far from all, and also a swag of new features which are full of bugs which you hope will be fixed in the next version, which then repeats the cycle over again, or even worse, a feature which has been stable for quite a while suddenly becomes unstable in the latest version. I think this just has to stop if you want 3DCoat to be respected as a serious player.

T.

See...this is the downside to allowing the public to use beta builds. Folks will encounter bugs in beta versions and fuss about stability, forgetting that there are stable builds for them to use in a production setting. I agree in principle about stability trumping new features, but we want to have our cake and eat it too.
Link to comment
Share on other sites

  • Advanced Member

AbnRanger, with all due respect, 3DCoat is always in beta, the so called "Stable Builds" are just builds with some of the bugs fixed, there are still way too many, I have used all the builds for many years now and I can tell you that there was never a Stable build, I always ran into issues which I had to find work arounds for, maybe its a Linux thing, I don't know, but I do know that its killing my enthusiasm.

Link to comment
Share on other sites

  • Contributor

Contrary to many here I only use a small subset of 3dcoat tools, and I've to agree with Tser, even with using a small fraction of the app tools I always encountered bugs in 3dcoat versions even stable ones. Now at the beginning I was reporting bug, but I've to admit I've just been watching for over a year now because I felt we weren't much to actually do it and most of it went unheard.

So I won't start another whine chain but I just want to add my support to Tser's comment.

Link to comment
Share on other sites

Status info for mantis users:

I moved all issues with the following status to the v4 project.

- feature request

- feedback

- new

Reminder:

Please check the resolved issues inside v3 and set them to "closed".

Please confirm/check the resolved issue before closing them. If an issue is still not working, move the issue to v4 as new or feedback. Leave a comment for reason.

If you can't move any issue, drop me an info by mail or by a forum pm.

Best wishes and thank you

Chris

Link to comment
Share on other sites

Well... sad you need to take this decision

yes... randomly V4 corrupt the files when save its... i loose some works too.

For my day by day work im using V3

----------

Separate the stages is needed

- V4: new features.

If users need its, they can buy V4 or upgrade from V3.

V4 is a new beginning of development until the release of v5.

New features add bugs, so must be very clear what release is stable or unstable

- V3: Stable version

Users needs a truly stable version, ready for a day by day work.

V3 was an stage of development from V2, and will be great to try to build an almost bug free version.

--------------

some software add a comment: stable / unstable to their releases... why not 3DCoat ?

and last...

Will any moderator add the label HIGHLY EXPERIMENTAL to this V4 thread please ?

regards !

Link to comment
Share on other sites

  • Advanced Member

I'm agree with you Tser, i really wish the 3DCoat team to focus on stability for the next release.I wish they stopped any new development and make it rock solid.

In the same time, i never encounter the pain that you get.I mean, i get weird bug, even when i saved my file.

I use to increment every 10-20 mins, in 3DCoat,Maya, Nuke and even ZBrush (Zbrush crash a lot too, especially for >8 millions poly).

I never relied to any software even if they are stable,network can break,electrical issue, unplug electrics plug, push the power button accidentally, etc.

But once again you right they should focus only on 3DC stability for the next version.

kind,

Pixo

Link to comment
Share on other sites

  • Applink Developer

I think that Andrew has done great job fixing bugs lately. 3d-coat is much more stable than year ago. Well at least in my opinion.

I understand that there is some bugs that are really hard to see. That's why I would like to suggest that Andrew would add some kind of

undo log window. When users finds a bug, he would easily see what actions has been taken and maybe find a pattern for that bug.

And I have to add that LC looks really promising. Still some issues but already working quite nicely.

Link to comment
Share on other sites

  • Reputable Contributor

AbnRanger, with all due respect, 3DCoat is always in beta, the so called "Stable Builds" are just builds with some of the bugs fixed, there are still way too many, I have used all the builds for many years now and I can tell you that there was never a Stable build, I always ran into issues which I had to find work arounds for, maybe its a Linux thing, I don't know, but I do know that its killing my enthusiasm.

That may be somewhat true, but let's be honest here. You're saying that you're trying to use V4 BETA and complaining about IT'S stability issues. What's wrong with this picture? Lately, 80-90% of all work Andrew does, is squash bugs. Do you understand how hard it must be for him to have a userbase that hounds him relentlessly for more and more features (he tries hard to please) and then get hammered from the other side, by those who clamor for the features to stop and squash bugs only? He's walking a tight wire, here. It's a difficult balancing act. No wonder he's taken a lot of vacation time in the past number of months. This must be wearing him thin.

If you want stability, go back to a stable build and don't try to use any of the (in-between) BETA builds. You say the stable builds aren't stable...well, then, how many of your issues got posted to Mantis and went ignored, when it was released? That's all Andrew can do. Squash the bugs that get reported. At some point he has to move on and start developing the application, again.

I understand the frustration, but you have to REMEMBER not to be doing all your critical work with Beta builds. That goes for any software. 3ds Max, even after official releases, has Service Packs that squash bugs that users have to deal with. Maya is the same way. CAT, 3ds Max's newest Character Rigging and Animation system, is still buggy...3-4yrs after it was released. I upgraded to finalRender R3 (third party render, like VRay) a few years ago, reported the constant crashing with the Interactive renderer (the main feature I upgraded for). They just pointed the finger at Autodesk (3ds Max's internal limitations), and I get burned for $400 on a tool I can't even use. Only solution...upgrade to V3.5, and it will crash less. I say all of that to say this....no software is without it's own degree of stability issues.

Andrew could squash every single bug reported to date, and people would still fuss about stability. I don't know how one man can solve this dilemma, except to publicly tell everyone to quit asking for features. He probably gets dozens and dozens in his e-mail each day, all while squashing bugs.

If it keeps going like this, I fear that he'd throw in the towel and sell it off to another company (many here know of one in particular who already tried). To which it would either get included with it's other software and essentially rot on the vine (going undeveloped), or increase the price and see all the frequent feature addons disappear.

Link to comment
Share on other sites

  • Contributor

As an amateur programmer I know this:

Coding feature is fun, bug fixing isn't. I'm sure Andrew feel the same way.

Question is: when the only thing your software needs to be a new standard (or at least a real competition to the top tier) is predictability and stability, efficient interface and usability polish, do you keep adding feature because it's fun or you bug fix the hell of your software to make it go to the top ?

It's easy from a customer perspective (and to be honest I'm not even saying this because I consider 3dcoat as an investment, I got plenty of tool to fallback if 3dcoat fails me, I just want to see better tool* I'm not expecting anything as a customer would), I agree, but you've got to chose between fun or running a business.

Harsh way to put it but well...someone got to say it.

*I'm thinking about things like liveclay here, things that could shape the industry for years to come IF 3dcoat gets reliable enough to be a valid alternative to quad/subdiv.

Link to comment
Share on other sites

  • Reputable Contributor

As an amateur programmer I know this:

Coding feature is fun, bug fixing isn't. I'm sure Andrew feel the same way.

Question is: when the only thing your software needs to be a new standard (or at least a real competition to the top tier) is predictability and stability, efficient interface and usability polish. Do you keep adding feature because it's fun or you bug fix the hell of your software to make it go to the top ?

It's easy from a customer perspective, I agree, but you've got to chose between fun or running a business.

Harsh way to put it but well...someone got to say it.

You know, from many of my past postings, I agree, but Andrew DOES bugfix....constantly. He has said on many occasions, that if you e-mail him a critical bug, he usually drops what he is doing and tries to get it fixed. He's been pretty responsive to me, when I take the time to screen record the problem (and sometimes send the file). He's tried to balance the two (features and bugfixing). I think one of the problems, and what Chris Werner keeps trying to remind folks of...is that most just come here and drop their issues. It's MUCH more effective to report it on Mantis, so Andrew can approach it in an organized, systematic way. Here...it just gets lost in the clutter.

This is why I recorded a video and put it on the Youtube channel, covering "3D Coat Support Options." But very few views and no responses saying whether it helped or not. I think Andrew will need to place a sticky on the forums and announce a new feature freeze for a few months, and ask users to please withhold feature requests until that period is over. That should reduce much of the traffic he gets daily.

Link to comment
Share on other sites

  • Reputable Contributor

Status info for mantis users:

I moved all issues with the following status to the v4 project.

- feature request

- feedback

- new

Reminder:

Please check the resolved issues inside v3 and set them to "closed".

Please confirm/check the resolved issue before closing them. If an issue is still not working, move the issue to v4 as new or feedback. Leave a comment for reason.

If you can't move any issue, drop me an info by mail or by a forum pm.

Best wishes and thank you

Chris

EDIT:

First, I forgot to mention thanks for working on Mantis to help the bug reporting process...

The only problem is that some of those new issues were submitted a long time ago and just clutter up the new beta 4 mantis report section.... I moved and or deleted two of mine that got moved over but there are others that the users submitted but who are not even on the forums any longer. Other users might not be reporting bugs anymore. Also some had pictures or videos with them and they did not get moved. Again I would suggest moving them back to version 3. Any active mantis bug reporter can move his current issue over to the beta 4 section .I would hate for Andrew to have to go through each one of the very old new issues that in alot of cases are now fixed in the course of development.

The move or delete for your report is in the attached file section on the right, That threw me off for awhile because I thought it was for the attach images.

post-518-0-62584400-1351529052_thumb.png

Link to comment
Share on other sites

All new release V4 have bug fixes...

but there is not more development to V3

Next... back to the future report:

I am trying to make as stable as possible last 3.7 build before posting first V4 beta

Next builds with essential changes will be marked as V4 beta.

------------------------

If you feei it unfinished, just point me to the list of problems, may be I missed something important.

But 3DC can never be called "finished" as there is always big potential to grow or improve.

But I tried to fix what can be called bugs at least what is not related to LC. LC stabilisation will be performed in V4 testing cycle.

Here's a first draft, I'll add things as I work and recall other thing I didn't think about at the moment.

----------------------------------------------------------------------------

I think Andrew will need to place a sticky on the forums and announce a new feature freeze for a few months, and ask users to please withhold feature requests until that period is over. That should reduce much of the traffic he gets daily.

But but... the beta users are the bug reporters...

How can they report new features bugs, if team devs freeze the new features development ? :wacko:

Link to comment
Share on other sites

  • Contributor

At some point when you're mecanic and you need to inspect the wheels of a truck, it's better if it's still, or at least slowing down (a lot)...

Cryptic talk is cryptic. Anyway 3dcoat's developpment is the first I see where the customers are trying to convince the dev to take his time and polish the existing software before adding features. I think that speaks volumes about the potential of the software and the reality of the user.

Link to comment
Share on other sites

  • Reputable Contributor

At some point when you're mecanic and you need to inspect the wheels of a truck, it's better if it's still, or at least slowing down (a lot)...

Cryptic talk is cryptic. Anyway 3dcoat's developpment is the first I see where the customers are trying to convince the dev to take his time and polish the existing software before adding features. I think that speaks volumes about the potential of the software and the reality of the user.

That one had me smiling, Please Dev don't add any new stuff for a just alittle while, we got our hands full learning all the new tools you have given us... Polish what you have and let us catch our breath. Yes, Andrew has always out paced us and we are huffing and puffing to keep up, plus dodging some of the pot holes that are left in his wake... :rofl:

I'm sure Andrew has a road map for what he wants to include or add to version 4 before it is released.

A little story...

I turn my head to ask a question about which road he is to be taking, Andrew? Andrew?, Where did he go?, I sigh, hitch up my pants and run out the door after him...

Been using 3DCoat for years and will continue to use it for years to come... :paint2:

Link to comment
Share on other sites

but there is a point of no return here

Pay attention that versions from 3.7.14 it are actually V4 betas and many features that you see there are actually V4 preview.

But actually it hard to distinguish what belongs to V3 and what to V4 because V4 is just logical evolution of V3.

To polish V4 means polish new features

so... first needs to finish the developing, fixing the new bugs during the process.

or...

Stop Beta tools development, remove beta tools, and bug fix V4 ?

How will be possible if V4 = V3 + Beta Tools ?

Link to comment
Share on other sites

  • Contributor

No you don't get it the point of no return is reached when you have a base so full of bug that adding anything just create cascading bugs.

If you don't stop from adding from time to time to evaluate the robustness of your wall, you will build a roof over crappy foundations and crumbling walls. NOT GOOD.

If you don't fix, "sanitize" a codebase from time to time you end up with so much bugs at all level you don't know what to fix and how to fix it anymore as everything can have an impact on the surface: the end user experience.

What I'm saying is not news for any programmer, but it's a tedious task that needs to be done before going to the next step of adding more features that require more complex changes to the code (deeper levels).

Keyword:

REFACTORING ! (yeah the "Ow no, not that again" part of every dev)

Link to comment
Share on other sites

  • Reputable Contributor

@ Carlosa, I think what most of us mean, and I have very clearly stated, once all new features have been added for the final release, there is a time allotted for putting 3DCoat version 4 through some hard testing to revel any bugs.

Development betas always break things. A release candidate is where all new features have been included. Development has stopped for that cycle. Next is to test the RC to uncover any bugs. Fix those and then release the final version.

We can not truly test for final bugs because the development betas at times break things that have been working or compound a problem as Beatkitano stated.

I just posted a bug report about an minor e-panel bug that was never broken till now. Only when a RC is released can we test to fix all the remaining bugs. This is not a opinion, that is a fact...

Once the RC has passed inspection and is released, then you restart development. Version 4 has been almost 2 years in development. Taking one month to test a release candidate is not asking much and puts the icing on the cake for the final release.

The above is a SOP in number of software development cycles. Does mean that your release version will be free of all bugs, of course not but there will be a lot less of them.

Link to comment
Share on other sites

  • Contributor

Problem is with that logic (and I understand the intent) is that it won't work, because I'm sure Andrew as any programmer is eager to release his baby to the world when it's (for him) done.

This means we need to find every (virtually cause that can't be achieved) bug before final, except the time required to be able to do that is just huge (think a QA team which doesn't actively try to find bug but encounter them by using the software, not so efficient ! It takes times to find peculiar bugs), something that may not be possible for a small company as pilgway that may need the cash from v4 update.

Imho this is doomed to fail as it failed for the last 3.x builds...

If Andrew take a break for bug fixing even if it's only temporary as new features integrated later on will introduce bugs (unavoidable) to audit his code, we can expect a much better chance of finding bugs by the users and by Andrew himself (even bugs not visible to the user I bet !) during the RC stage.

I also understand Andrew will have a hard time to admit this time is needed as I'm sure he's thinking about Raul presence and what he can bring to the table. If Andrew freeze the trunk it means raul is somehow slowed down on his work which isn't good since Raul time in Kiev is flying.

Link to comment
Share on other sites

  • Reputable Contributor

Andrew, plows ahead but I will still stand in front of the steam roller at times, My chances of being flatten to the pavement are pretty good but I since I don't die from it, I will at times will hold up the logic sign in front of the steam roller and carry an air pump with me... :D

Link to comment
Share on other sites

  • Advanced Member

If Andrew take a break for bug fixing even if it's only temporary as new features integrated later on will introduce bugs (unavoidable) to audit his code

I'm agree but let release a stable version for the V4 , like this nobody will complain and tell Andrew that there's not a stable release.

There's something like 1 month before the next release.Imho it would be good to deliver a very robust one , and then go back to dev new features.

Nothing prevent users to request features on Mantis, cause i'm pretty sure that Andrew know how to manage his priorities.

Link to comment
Share on other sites

but there is a point of no return here

To polish V4 means polish new features

so... first needs to finish the developing, fixing the new bugs during the process.

or...

Stop Beta tools development, remove beta tools, and bug fix V4 ?

How will be possible if V4 = V3 + Beta Tools ?

As I mentioned to you in PM, V4 isn't just V3+ beta tools.

V3 is feature locked at the last V3 build, it is not getting any more features. It makes no sense to develop two applications.

Link to comment
Share on other sites

  • Advanced Member

Is it a fact that WE are the testers of the program? Are there any people at Pilgway who thoroughly test each new added feature? if not, I would suggest setting up a group of testers whose job it is to find bugs, this will be the only way to once and for all, produce a stable and reliable application.

Link to comment
Share on other sites

  • Advanced Member

and who is to decide that an application is stable? Is Autodesk a bell weather for that? Maya crashes on me when I use particular functions, I just have to know that and save before trying. My point is this method works to keep this application growing. I agree that there are many areas of 3D Coat right now that are underdeveloped and in some cases don't appear to work. This program is pretty big, autodesk has beta with normal users of varied skill levels to test and find bugs.The only real benefit is they/we get the next version of the application to work with. They often pull features that were in the beta for the release because they are not "ready" yet. I'm happy with the level of service I get from 3D Coat on bug fixes and I don't get more crashes than I'm used to with 3d applications. Its rare I get the full surprise crash. This isn't a videogame with a room full of beta testers who's full time job is to hunt down bugs and report them in an orderly fashion. The best bet I'd say is to make people more official beta testers, and keep the beta more hidden from general use. Then there will be a stable release everyone uses and bug reports on, and a beta. This will give a better "front" for 3D Coat but then again, I kinda like the openness of the dev team to allow anyone to beta test.

Link to comment
Share on other sites

  • Reputable Contributor

and who is to decide that an application is stable? Is Autodesk a bell weather for that? Maya crashes on me when I use particular functions, I just have to know that and save before trying. My point is this method works to keep this application growing. I agree that there are many areas of 3D Coat right now that are underdeveloped and in some cases don't appear to work. This program is pretty big, autodesk has beta with normal users of varied skill levels to test and find bugs.The only real benefit is they/we get the next version of the application to work with. They often pull features that were in the beta for the release because they are not "ready" yet. I'm happy with the level of service I get from 3D Coat on bug fixes and I don't get more crashes than I'm used to with 3d applications. Its rare I get the full surprise crash. This isn't a videogame with a room full of beta testers who's full time job is to hunt down bugs and report them in an orderly fashion. The best bet I'd say is to make people more official beta testers, and keep the beta more hidden from general use. Then there will be a stable release everyone uses and bug reports on, and a beta. This will give a better "front" for 3D Coat but then again, I kinda like the openness of the dev team to allow anyone to beta test.

I agree that BETA versions should, in principle, be hidden from the userbase at large, and allow a group of regulars to test them. This way, the app doesn't get a reputation for being buggy and unusable. The public only sees relatively stable versions. However, Andrew doesn't want to do it this way. The more people who use the Beta versions, the more thorough the discovery of bugs are. I can see the advantages in either approach.

Is it a fact that WE are the testers of the program? Are there any people at Pilgway who thoroughly test each new added feature? if not, I would suggest setting up a group of testers whose job it is to find bugs, this will be the only way to once and for all, produce a stable and reliable application.

Yep...everyone who downloads the Beta builds is a tester, essentially. There are some of us who Andrew might consider internal testers, who send him thorough bug reports and so on. If we could steer more and more users toward using Mantis as the default means of bug reporting, it should make it easier for Andrew. He doesn't spend much time on these forums, so I don't even know if it's a good idea to keep the bug sections/threads on here, as it's redundant.
Link to comment
Share on other sites

  • Advanced Member

I would love to be a full time beta tester for 3DCoat, but financially I need to put food on the table, so I have to work, therefore time is very precious for me, wasting hours of creating time because of bugs in an application is not an option. Maya, Softimage, ZBrush and others do have bugs, but they are minor and don't cause you to loose significant amounts of time when they occur, so they are acceptable.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...