Jump to content
3DCoat Forums

V4.1 BETA (experimental 4.1.17D)


Recommended Posts

  • Advanced Member

3DCoat reminds me of a kid who wants to grow up too fast and be like the big boys without going through the pain of growing up, you need to be patient, grow with solid foundations, mature and gain respect, throwing as many tools loaded with bugs together is not the way to go, perfect the tools you have and then add new ones only after all the problems with your existing tools have been worked out to EVERY ONES satisfaction. Reputation is everything in business. If you want to make tools attractive to artists then stability, reliability and usability are all what matters, an artist is like a driver of a car, he/she doesn't care how the car is built or how shiny the tappit covers are on the engine, or how many pistons are in the engine, he/she just wants it to start, drive reliably to and from their destination and park the car for use the next day with the knowledge that tomorrow the car will perform as it did today.

T.

  • Like 2
Link to comment
Share on other sites

  • Reputable Contributor

3DCoat reminds me of a kid who wants to grow up too fast and be like the big boys without going through the pain of growing up, you need to be patient, grow with solid foundations, mature and gain respect, throwing as many tools loaded with bugs together is not the way to go, perfect the tools you have and then add new ones only after all the problems with your existing tools have been worked out to EVERY ONES satisfaction. Reputation is everything in business. If you want to make tools attractive to artists then stability, reliability and usability are all what matters, an artist is like a driver of a car, he/she doesn't care how the car is built or how shiny the tappit covers are on the engine, or how many pistons are in the engine, he/she just wants it to start, drive reliably to and from their destination and park the car for use the next day with the knowledge that tomorrow the car will perform as it did today.

T.

+1
Link to comment
Share on other sites

  • Reputable Contributor

+1

I also see the performance bottlenecks that have been part of the app for the longest, much like a bug. It's a big problem that has gone ignored for a few years now. The CPU multi-threading was a big step forward, but it didn't solve the long-standing problem of large brush radius' bogging down performance (Voxel/surface sculpting and Painting). This is why I have kept harping about CUDA. Users are NOT getting the most out of their expensive hardware. It desperately needs to be updated/recompiled (to take advantage of the newer features of CUDA 4-5) and put to use in ALL brushing and deformation operations, in 3D Coat.

Andrew, could you contract a specialist in GPU multi-threading (CUDA, OpenCL, OpenACC), to look into this? As I stated before, Mudbox utilizes the GPU to do all the heavy lifting and large brush sizes do not bog it down.

  • Like 1
Link to comment
Share on other sites

3DCoat reminds me of a kid who wants to grow up too fast and be like the big boys without going through the pain of growing up, you need to be patient, grow with solid foundations, mature and gain respect, throwing as many tools loaded with bugs together is not the way to go, perfect the tools you have and then add new ones only after all the problems with your existing tools have been worked out to EVERY ONES satisfaction. Reputation is everything in business. If you want to make tools attractive to artists then stability, reliability and usability are all what matters, an artist is like a driver of a car, he/she doesn't care how the car is built or how shiny the tappit covers are on the engine, or how many pistons are in the engine, he/she just wants it to start, drive reliably to and from their destination and park the car for use the next day with the knowledge that tomorrow the car will perform as it did today.

T.

I agree. It's getting terribly buggy. I would strongly prefer to see bug fixes and workflow improvements over new features for a while. The new features just keep introducing new bugs or causing existing bugs to be lost in the mix, or worse still, causing the bug to worsen.

Please fix bugs and improve workflows. Thanks.

  • Like 1
Link to comment
Share on other sites

  • Advanced Member

I've just built my first new workstation in 4 years... and 3dcoat is truly awesome on this new hardware. I hear a lot of angry complaints about bugs on this beta testing forum. (Beta testing!)

Shouldn't bugs be being reported, instead of complaints about how development is progressing?

I love the fact that 3dcoat is the best voxel sculpting software out there - that also has amazing UV and retopology tools and a full-featured paint room.

I say keep up the amazing work guys!

Link to comment
Share on other sites

well... i think that this build is a step forward

The only way to bug kill is to post every bug founded here, in the 3D-Coat V4 Beta testing thread

the V3 development is dead... so what can we do ?

help and wait... help and wait... i cant see another way

Link to comment
Share on other sites

  • Contributor

Am I really the only one getting:

-holes/tears/ disconnected patches (inside or at the surface of the mesh) in surface mode with liveclay

-cubes stretches floating around the model at times in voxel mode

Without any way of replicating something that happen totally randomly or at least with a pattern so convoluted you can't spot what you did to get those bugs ?

Really ? Like since v 3.0 for the voxel bugs ?

I'm ok with reporting bugs, but if there's no way to report it properly: what can I do except say they exists and they've been a major annoyance for years now ?

Other thing, everytime an user gets an unusual issue the way of dealing with it is directing him to beta versions (ok bugs get fixed I get that) and then "delete options.xml" ???

Why this file gets corrupt so often in the first place, why write on it during sensitive operations ? I see you did create temp save file to avoid corruption, nice, what about every file ?

What is beyond me is that you don't realize you DON'T NEED to add new features to be at the same level (or above) as the competition, all you need is stability and predictability. I don't know if you noticed but you got top grade industry artist on this board, or at least you had, because most of the profiles of those I know have been inactive for quite some times now... you should ask yourself why and I'm sure you've an idea by now.

Link to comment
Share on other sites

Hmm I'm now confused to no end ... buggy new features? please be more specific, because latest features introduced are not buggy, are stable, other thing is some tools in LC that still have bugs like ANY development process, I don't think there's a perfect software out there without bugs and that's why we get release cycles (along with new features) Max, Maya, Mudbox, Zbrush,... name any, ALL have bugs, pay a visit to their artist forum and you'll see a similar situation.

I think many things related to bugs can be attributed also to lack of knowledge on way to use the tool. Well, the documentation is being updated and rewritten for V4 and this is important too.

So I think is good to spot and reports bugs (preferably in Mantis) but a bug free application is something non existent, what we will have is an asymptotically stable application and we have made certainly progress toward that. Just compare 3DCoat one year ago ad now. Development is not a closed circle, is an spiral process.

Link to comment
Share on other sites

And those random holes very difficult to replicate are related probably to memory leaks, which we are working to solve , they are particularly difficult to spot but will squeezed them down.

I really don't understand all the rant about new features and it saddens me a lot, perhaps because of that I don't feel comfortable visiting this forum.

Remember that a drop of honey attract more flies than a gallon of gal.

What is beyond me is that you don't realize you DON'T NEED to add new features to be at the same level (or above) as the competition, all you need is stability and predictability

I think an application is dead the moment it stops innovating, we are not aiming at a clone. Stability and predictability? we will get there...

Link to comment
Share on other sites

  • Reputable Contributor

@ Farsthary, Glean out the good, spit out the bad... That is what I do with bugs. If I let bugs get the best of me and ruin my day, like having to redo my preset brushes (still in beta) 3 times in one day I would not feel comfortable visiting 3DCoat but how could I then post bug reports at Mantis by staying away. I just keep my spitting bucket by my side for them nasty old bugs whether in beta 4 or version 3 and I know you guys are working hard to give us a stable 3DCoat.

Like you said bugs will always be in software, even with the best intentions.

Humans are buggy too even with the best intentions, just glean out the good and spit out the bad

Don't let the bugs of human nature keep you from the forums.... :D

Thanks for all your hard work...

Edit: I been using 3DCoat for 5 years, I ain't gonna go nowhere,,, 3DCoat she's my honeydew...

Link to comment
Share on other sites

  • Reputable Contributor

To lighten the mood, I will repost my "Poem of the Mantis bug hunter"

Poem of the Mantis bug hunter

Wake up in morning, bug huntin' on my mind

No time to be creative,wash my socks or dine

Them bugs are crawling all over the walls

Must do something or I will lose my mind

Grab my shotgun, bug huntin' I go

Wasting them bugs till the midnight glow

Wake up in the mornng, bug huntin' on my mind... :beta:

Link to comment
Share on other sites

Thanks Digman, I understand is not artist fault too, is very frustrating having a bug destroying your hard-work or wasting time, and the first things that comes to your mind in that moment is the entire genealogy of the makers. In that sense I need to be more prudent with feedback and reply.

Is just that is very frustrating to see how I sacrifice my entire nights extending my work time for 14+ hours a day / 7 days a week, in front of my computer with no social life, and zero leisure time, to get something useful and interesting and if possible a breakthrough contribution just to find out a negative atmosphere of comments that really makes me think is not worth the effort. So yes, I think I should start thinking more on myself, work will no go anywhere and there will be work even after our death, I can't please everybody: some users want more features, some want stable ones, and some want even less tools.

When I finished Decompose tool yesterday in the afternoon I was so happy saying to myself: -Artist will love this! that I reward myself taking a walk in the city. That's when I allow to myself to take a short rest or recognize myself or make me a present for what I have achieved because when some tool is not finished or is not working I literally put myself in jail till I finish it, and I know is not healthy, but the pleasure to get public credit for an achievement is something I got addicted since Blender times. How fool I was!

What really makes me keep going is that I enjoy, just for myself, my job and I see that is the only motivation I have left, the day I loose that, don't know what would be of my life, but definitely will be a real life.

Have a nice day everyone.

Link to comment
Share on other sites

  • Advanced Member

Hey everyone,

just wanted to chime in. I for one enjoy new features and I think 3D Coat's development cycle is pretty cool in the sense that it is so fluid and organic. I use the program as my only sculpting software and have been using it quite a lot lately at work and on personal projects. I don't really get upset when I encounter bugs because every program I use has them. 3D Coat probably has a few more, but then I'm always using the latest build (at my own risk) and I also realize how small the development team is, so I understand. This is a pretty awesome and bleeding edge innovative app. There are bound to be problems. I think this community does a great job of steering the development of the program and reporting bugs. I'm sure it's one of the main reasons why Pilgways team is able to be so small yet still be able to compete with the big boys. So overall I'm very happy about 3D Coat and excited for the future.

All that being said, I think it would be great if they are able to get 2 more people on board, 1 coder to work on stability and performance and the other a talented artist to work very closely with them on a full time basis.

My 2 cents

-G

Link to comment
Share on other sites

  • Reputable Contributor

I appreciate the new work...and the new De-Compose tool will be handy for more than just troubleshooting the issues found in LiveClay. But I think it's hard to see one another's perspective. We can't see how much work you and Andrew have to put into doing both features and bug fixes. And you can't see just how frustrating it is to lose work, fight with the application to work around problems, then find time to screen record, and post bug reports...often times with no reply back.

It literally is a full time job on our end. I mean I spent almost 2 full days recently trying to merge that glove you saw, into the Paint Room and get a decent result. I uploaded the workfile, typed in a good description.....and a week later I haven't heard a peep about it. I just wanted Andrew to try it himself, and see how problematic it is for us users sometimes. And then maybe communicate back what steps he took to get a clean result. In fact, I downright dare Andrew to take that file, and get a clean bake from it. I tried about 20 different workarounds. When a user encounters this level of frustration on a continual basis, it really wears on them, mentally. Especially if there is a tight deadline involved.

I didn't just learn 3D Coat yesterday. I tried every trick in the book to try and get a decent baking result. Never did. How, then could I reasonably work a 10-12hr/day paid job while fighting with 3D Coat for 2 days just to do what should only take 1-2 minutes? I've lost a few hours worth of work just in the past 2-3 days, due to bugs. Do you see now, where the negativity comes from? It comes from repeated and continuous encounters with bugs. Every app has them, but 3D Coat has about 10 times more than any other I've ever seen. Is it the application architecture, speed coding that leaves behind problems?

I think what Beat is saying is that no one is expecting a complete stoppage of features, but there does need to be a lengthy and deliberate freeze of new additions, while the application goes through a substantial stabilization process. Andrew will need to announce this publicly and ask the userbase to temporarily withhold their feature requests while this is underway....so you both can focus without so much traffic and noise.

Link to comment
Share on other sites

  • Reputable Contributor

Hey everyone,

just wanted to chime in. I for one enjoy new features and I think 3D Coat's development cycle is pretty cool in the sense that it is so fluid and organic. I use the program as my only sculpting software and have been using it quite a lot lately at work and on personal projects. I don't really get upset when I encounter bugs because every program I use has them. 3D Coat probably has a few more, but then I'm always using the latest build (at my own risk) and I also realize how small the development team is, so I understand. This is a pretty awesome and bleeding edge innovative app. There are bound to be problems. I think this community does a great job of steering the development of the program and reporting bugs. I'm sure it's one of the main reasons why Pilgways team is able to be so small yet still be able to compete with the big boys. So overall I'm very happy about 3D Coat and excited for the future.

All that being said, I think it would be great if they are able to get 2 more people on board, 1 coder to work on stability and performance and the other a talented artist to work very closely with them on a full time basis.

My 2 cents

-G

+1. One for stability like you said, and one who specializes in GPU parrallel threading. CPU Multi-threading is only good up to a certain point and then it totally bogs the application down. I'm not even talking about uber-ridiculous poly counts or texture map sizes. Performance hits a wall fairly often, and I upgraded hardware the last two times just to try and squeeze more out of 3D Coat. It hasn't proved to be financially feasible, when 3D Coat really doesn't take full advantage of what I already have. CUDA? Let's be honest. It's so seldom utilized in 3D Coat, it's practically not worth mentioning, on the website or promotional materials. Andrew implemented it to some small degree when he released V3 over 4yrs ago. It was at CUDA 1. Here we are at CUDA 5 and not a single thing has been done to bring it up to date, nor to use it where it's needed the most.

I've tried to be patient, but after 2+yrs of being ignored, it gets very frustrating.

Link to comment
Share on other sites

All that being said, I think it would be great if they are able to get 2 more people on board, 1 coder to work on stability and performance and the other a talented artist to work very closely with them on a full time basis

+1.

It's very right words, gbball.

.

Link to comment
Share on other sites

  • Advanced Member

Thanks PalSan,

Obviously, it all depends on whether or not that fits with Pilgway's plans and budget, growing a company is a major decision. Keeping things lean makes a lot of sense. I just think that Andrew's strength is creating new features and solving the really tricky problems. Raul, same thing.

Also, I think it is unfair to compare against other applications too much, I think 3D Coat has it's own thing going on.

Link to comment
Share on other sites

  • Advanced Member

And those random holes very difficult to replicate are related probably to memory leaks, which we are working to solve , they are particularly difficult to spot but will squeezed them down.

I really don't understand all the rant about new features and it saddens me a lot, perhaps because of that I don't feel comfortable visiting this forum.

Remember that a drop of honey attract more flies than a gallon of gal.

I think an application is dead the moment it stops innovating, we are not aiming at a clone. Stability and predictability? we will get there...

No disrespect intended, but have you ever actually tried to create a complex sculpt from the beginning to the end, meaning, starting from the curves tool, to sculpting the various components of your model with high details and definition, then retopo-ing the model components on to creating good UV's for them, merging the components to the paint room to texture and then exporting the lot to be picked up bu Maya or Softimage etc.

I can tell you its like taking a journey to hell and back.

You should try it, and then let us know how stable the tools are. The trouble is that you guys code too much and don't thoroughly test your creations in the same manner as an artist would who is working in a production pipe, playing with one tool all day is one thing, but using one tool after another as you refine your sculpt, your topology or UV sets is quite another.

Also, I wouldn't recommend leaving the honey out, you have more than enough bugs to play with.

T.

Link to comment
Share on other sites

  • Advanced Member

I think Tser is right.

It has been proposed numerous times...

You guys really should take a week off and do nothing but create a model. Everybody here would understand.

Run the application for hours, export back and forth to Modellers and Photoshop, work with large files, use all workspaces.

Unless you really do this you will always feel misunderstood and sad and question whether the bad resonance is worth the whole effort...

You have it in your hand yourself to understand what all this whole noise is about.

Reporting bugs is all we can do but this will never give you a somewhat complete picture.

3DC at all times has been the most error-prone program I have on disk.

No other graphics application I own crashes comparably often and lets a model go bad irrecoverably with a single operation.

Link to comment
Share on other sites

Latest BETA4 was really essentially unstable in retopo room and I found it only after getting back from vacation... that is really not good, especially for Linux user who have no BETA3...

Sometimes I making some real models in 3DC and it always helps a lot. Maybe I will try to do something again, Raul will try to model something real too. In recent build all what is done is bugfixes + one little improvement of materials + Raul's tools. I plan to continue stabilizing. Generally Mantis helps a lot.

Don, you often tell about CUDA but I don't do this request for years because it will require much more efforts and time than will bring back efficiency. All I want now from CUDA is don't crash because it may just crash unpredictably without any visible reason. Maybe someday I will get idea how to do CUDA support better.

Link to comment
Share on other sites

  • Contributor

No disrespect intended, but have you ever actually tried to create a complex sculpt from the beginning to the end, meaning, starting from the curves tool, to sculpting the various components of your model with high details and definition, then retopo-ing the model components on to creating good UV's for them, merging the components to the paint room to texture and then exporting the lot to be picked up bu Maya or Softimage etc.

I can tell you its like taking a journey to hell and back.

You should try it, and then let us know how stable the tools are. The trouble is that you guys code too much and don't thoroughly test your creations in the same manner as an artist would who is working in a production pipe, playing with one tool all day is one thing, but using one tool after another as you refine your sculpt, your topology or UV sets is quite another.

Also, I wouldn't recommend leaving the honey out, you have more than enough bugs to play with.

T.

Ofer "PIxolator" Alon, is also an artist, you may like his software or not but you can't say it's not polished.

And because there's so much to say on every comment here (so good things said, some I disagree with but that's not an issue of perspective just experience I guess): Farsthary, with all due respect I admire your work, Andrew's work too, but saying 3dcoat would die because it wouldn't include new features for a while going into "bug hunt mode" is dead wrong.

I know tons of game (and movies for some) artists liking 3dcoat tools but not adding it to their workflow beyond retopo (and a bit of painting) because it's unrealiable/unstable/"buggy". Their words !

I know the UI changes we noticed in v4 was spearheaded by LJb, I don't want to drag him into this because he's not commented on the issue yet, but I know he wanted to do something because (and we talked about it on live demos) we were struggling finding ui items in the app. He admitted to me he was mainly doing it to be able to do clean video tuts on it for sale and he encountered bugs like me and wanted to change a bit about that by getting involved in the overhaul...

Now I don't know if he's doing it or canceled... but this shows this apps really needs polishing...

@Andrew beta 4 wasn't only buggy in retopo room. I got quite a few bsod which I NEVER experienced before on multiple workstations. And all those BSOD happened only when 3dc was open. So it *may* not only be 3dc's fault, but it's involved for sure.

Link to comment
Share on other sites

  • Advanced Member

Well, I've been working on V4 Beta 5 all day and haven't had a single crash. 5 mill poly in surface mode (which I know is high for what it is, but will decimate it at a later stage to bring down the file size and poly count).

http://3d-coat.com/forum/index.php?showtopic=6712&view=findpost&p=82592

Incremental saves seem to be working too. No artefacts inside V4, may try opening the sculpt up in V3 later on.

Ricky.

Link to comment
Share on other sites

  • Contributor

When I finished Decompose tool yesterday in the afternoon I was so happy saying to myself: -Artist will love this! that I reward myself taking a walk in the city. That's when I allow to myself to take a short rest or recognize myself or make me a present for what I have achieved because when some tool is not finished or is not working I literally put myself in jail till I finish it, and I know is not healthy, but the pleasure to get public credit for an achievement is something I got addicted since Blender times. How fool I was!

But it is, you don't get my meaning, I'm just saying that since some previous bugs which potentialy affect everytool are not fixed it will eventualy create situations where the new users will not be able to understand with crashes/freeze/loss of work/frustration etc.

I'm not saying you did wrong, I actually see a quite cool use, and again even though 3dc has flaws this new feature for instance is simpler/more logical than lets says, zbrush counterpart (auto group>group split).

It's just that it's another stone set too soon which will affect the experience of unfortunate user.

Now I understand Andrew got you onboard for features like that, you're gifted for that you demonstrated it with blender, so maybe it's not the time for saying this. Maybe I should wait your work is done before, the thing is I feel a push for v4 release, and to be honest I won't buy a tool that may not reach his goal (dare I say promise ?), I could but that would only push you guys for more features which is definitely not what 3dc needs at this time.

Link to comment
Share on other sites

@ BeatKitano

but saying 3dcoat would die because it wouldn't include new features for a while going into "bug hunt mode" is dead wrong.

Yes, my bad, I was at the end my my coding nigth so can't think clearly and I realize we where arguing about different stuffs: Of course entering a freeze mode where only bugs are solved is a necessity (everyone's do) and I'm starting it too. Is just that things need its time, LC needed to complete most of its feature base before entering in bug and polishing phase, because that part is uncertain how long can take, it can be fast, it can be slow, so the feature level needed to be finished first.

At least this is my development model, it has the advantage that you can directly benefit from gradual increment of stability and require less slow down in development because coding a tool, switching to full quality/testing control,then coding a new tool, while less painful for artist, delays the whole process and is also painful for programers.

Some artist finish a painting in many steps, draft, drawing, mattes, color layers, shading others finish each part at once (very risky because you can easily loose the sight of the whole idea and goal, but valid anyway) my approach is the first. My father is a painter, and he told me that the consequence of letting others to see the picture without finishing is that they can't see the beauty yet, so critics will rain and incredibly, advices on how you should make your artwork....well coding is the same, that's why big guys in the industry have a closed development cycle and artist just see the polished product at the end of the year or some nice prepared demos. But I don't like that model either.

Link to comment
Share on other sites

  • Contributor

@ BeatKitano

Yes, my bad, I was at the end my my coding nigth so can't think clearly and I realize we where arguing about different stuffs: Of course entering a freeze mode where only bugs are solved is a necessity (everyone's do) and I'm starting it too. Is just that things need its time, LC needed to complete most of its feature base before entering in bug and polishing phase, because that part is uncertain how long can take, it can be fast, it can be slow, so the feature level needed to be finished first.

At least this is my development model, it has the advantage that you can directly benefit from gradual increment of stability and require less slow down in development because coding a tool, switching to full quality/testing control,then coding a new tool, while less painful for artist, delays the whole process and is also painful for programers.

Some artist finish a painting in many steps, draft, drawing, mattes, color layers, shading others finish each part at once (very risky because you can easily loose the sight of the whole idea and goal, but valid anyway) my approach is the first. My father is a painter, and he told me that the consequence of letting others to see the picture without finishing is that they can't see the beauty yet, so critics will rain and incredibly, advices on how you should make your artwork....well coding is the same, that's why big guys in the industry have a closed development cycle and artist just see the polished product at the end of the year or some nice prepared demos. But I don't like that model either.

Ok, I know about the last 20% of coding meaning 80% of time ^^

We agree I guess, let's just say I started freaking out because I know you guys are on a schedule, and so far I fear the deadline will be a two part word...

Now I don't want you both to take this personaly, I know how coding works (you comparison to a work in progress is true, I can see it on my side) I just wanted to be sure you were aware of it.

This all started because I know (unless you put a real tight case switch in there) your new tool will create huge issue with the old standing issue of voxels floating bits (artefacts) or liveclay patching/tearing/holes. Tnx.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...