Jump to content
3DCoat Forums

Feature request: No more features.


Ghostdog
 Share

Recommended Posts

  • Advanced Member

I'm stealing this idea as it says something simply, and so well it's better than any so-called 'roadmap' a team of suits could come up with.

Performance and stability should be PRIORITY No.1 for the next 6-12 months, in my opinion...here's my vote for no more wiz-bang features until those areas have been fully addressed.

I stole it from here

AbnRanger, cheers for putting it so well.

Link to comment
Share on other sites

  • Contributor

I also wondered why PTEX is getting months of development while the voxel and surface sculpting tools are still awkward.

It's been a long time since i spoke here, but i have to say i totally agree with that, i'm actually not using 3dc (although i bought a license) and still working on zbrush because of the feeling of the brushes in voxel sculpt...

Link to comment
Share on other sites

  • Advanced Member

It's been a long time since i spoke here, but i have to say i totally agree with that, i'm actually not using 3dc (although i bought a license) and still working on zbrush because of the feeling of the brushes in voxel sculpt...

I am in the same boat, I had to move my projects all back to ZBrush, something that wasn't my first choice, but hours of merg-bars(merge progress bars running on only 1 of my 8 cores) in the voxel room from switching between surface and volume modes(and the many vox tools that trigger a merge process) forced me to so I could at least try to make my deadlines on these projects. I'm not doing much of anything with my license now, considering selling it if I dont see any performance improvements in the coming months. Which all in all "sucks", as I enjoy using 3DC for sculpting far more than any other tool I've used thus far.

Link to comment
Share on other sites

  • Contributor

Yeah, even sculptris feels more confortable (in multithreaded version), with its alien controls and alpha stage, kinda sad actually. I asked like many other to have the brush reworked, still waiting to be heard.

Right now 3dc is feature packed, but looks aimed at hobbyist not people meaning "serious business" (if gameart/high poly sculpting can be qualified as serious that is).

(awaiting a philnolan nag in..

Link to comment
Share on other sites

  • Member

It's been a long time since i spoke here, but i have to say i totally agree with that, i'm actually not using 3dc (although i bought a license) and still working on zbrush because of the feeling of the brushes in voxel sculpt...

same here ... i only use 3d coat for quick concepts and do the real work in mudbox, because of performance issues ... in a real world working environment where there are tight deadlines you simply cant wait hours for some merges and quadrangulations and whatever :/ sad

Link to comment
Share on other sites

  • Contributor

same here ... i only use 3d coat for quick concepts and do the real work in mudbox, because of performance issues ... in a real world working environment where there are tight deadlines you simply cant wait hours for some merges and quadrangulations and whatever :/ sad

Yep, the only reason i still launch it now is for its retopo tools, imho its the only part that looks coherent and solid. The others are full of functionnality none of which are really working together more like side to side not to their full potential if they were though out in a cohesive environnement (who said roadmap ?)

I don't know what's Andrew's view on this, but i don't judge the quality of a software based on the number of its features, mudbox 1 was fairly simple and basic, yet it gave a simple and effective alternative to zbrush, it found its users with its brush engine (which was awesome compared to zbrush 2).

3dcoat is still searching its way, nothing is solid (except retopo), everything is WIP, i bought it, feeling that Andrew's dedication to add features would fill the gaps, it seems its now starting to be a curse.

I will stop there, cause i don't want to turn that thread into a 3dc bash, because I don't have much time to spend on these boards and because I still hope this software can become something really great if some planning is made.

Link to comment
Share on other sites

  • Advanced Member

I feel bad for posting... can't explain why though. Generally not a complainer... I live with near constant chronic pain which keeps me pretty balanced actually.

Anyway couldn't sleep (not because of post hehe!) but felt I needed to get that off my chest.

Cheers,

Robbie

Link to comment
Share on other sites

  • Contributor

I feel bad for posting... can't explain why though. Generally not a complainer... I live with near constant chronic pain which keeps me pretty balanced actually.

Anyway couldn't sleep (not because of post hehe!) but felt I needed to get that off my chest.

Cheers,

Robbie

100% Agreement from me if it counts for any.

Nice to see Andrew has made a good start on existing problem stuff like Shader/Lights consistency. Voxels could do with attention and I find many, Many little inconsistency in UI that need to be looked at, which spoil things somewhat. Some little things that should be working are sort of do, but not as you would expect. Overall things could do with an overhaul toward usabillity. 3d coat is beginning to feel like a collection of many seperate programs again, Like back in V 2 before UI improvements.

PTex implimentation was always going to be long winded So I get it has taken up lots of Hours, But I would now concentrate on major usabillity. Voxel improvements maybe not a priority here so much as Cleaning up the clutter and making the tools more usable. I say keep New Features minimum and work on what we have to make this tool great, Before re inventing the wheel with PTex like new developements.

We Spoke little while back about Removing the cursor at draw time too cos it gets in the way, Which never materialsed. Its these little things of which there are many that make the program more usable day by day, More comfortable to work in, which is what it needs to be. I love the program to death but currently it has its quirks, and its great in many areas, But master of none.

I hope thats does not come across too negative. Its meant in quite the reverse.

Link to comment
Share on other sites

  • Advanced Member

I agree that performance/stability is crucial, but at the same time, I sympathize with Andrew being pulled in so many directions.

Software development is a tricky business. The more you develop, the more people want. Focus is hard. I have no use for the shader stuff, but others find it very useful. Others have no use for Ptex, to me it's essential. We're all "correct" in what we want.

Ptex will hopefully allow Andrew to break into high-end VFX work, where licenses will be sold in 10's or even 100's, and possibly allow him to hire more staff which will allow him to do more in a shorter amount of time.

Performance/stability keeps people using the software and prevents "straying" to other software when they get frustrated, but it doesn't natively "sell" software (it's not good for marketing).

Again, I feel Andrew's dilemma on all this stuff. 3DC has become too big and successful :) So, I think we all have to trust him to identify priorities, based on this type of feedback, but not beat him up too much :)

Cheers,

Peter B

Link to comment
Share on other sites

  • Advanced Member

I dont know about the performance it seems ok to me, it runs rubbish on my pc but then I know and use it within the limits of the spec of my pc, it works fine.

What with pcs becoming more powerful, and cheaper now, how much more performance can be gained and at what effort for only say a five percent gain, and in which it could later be redundant.

I think refinement of the tools is needed more than peformance, so that your better able to manipulate voxels or paint.

Link to comment
Share on other sites

  • Advanced Member

I dont know about the performance it seems ok to me, it runs rubbish on my pc but then I know and use it within the limits of the spec of my pc, it works fine.

What with pcs becoming more powerful, and cheaper now, how much more performance can be gained and at what effort for only say a five percent gain, and in which it could later be redundant.

I think refinement of the tools is needed more than peformance, so that your better able to manipulate voxels or paint.

I disagree completely with regard to work on performance not speeding the app up by 5%. Have you ever used a similar app like zbrush with multithreading turned off? It's VERY SLUGGISH on large meshes. Then turn on multithreading and watch the speed increase WELL over 100% great in frame rates and other geometry based processes in ZBrush. The benefits that come with a well-implemented multithreading backbone to an app like 3DC is usually well over 100% speed in crease when you go from 1 core to many cores. In some of the 3D apps I used in my pipelines, multithreading speeds up processes by over 500%+. This is not an if, but a definite truth. 3DC lacks multithreading in MANY MANY aspects of its toolset, especially showing itself to be true in the vox room.

Or perhaps adding CUDA and/or OpenCL to an app for performance, does this not produce FAR MORE than 5% performance boosts. It's VERY EVIDENT as to what can be done to gain a minimum of 100% speed increases in 3DC's vox and paint rooms. The speed increase that a well-implemented CUDA or OpenCL upgrade to a 3D App is already providing well over 500% increases in speed for the apps using these GPGPU additions to their apps. Consider an OpenCL/CUDA Quadrangulation process that is also multithreaded, would be lighting fast on systems that are 4+ cores with a medium level GPU.

IMO the voxel room toolset is VERY powerful and pretty-well implemented(though better naming would be nice). Refine the tools AFTER performance is addressed. Performances in the vox(and paint) room is abysmal now. The app uses only 1 of my 8 cores due to its lack of multithreading. This is NOT what a modern professional 3D app is expected to present to its users when its price keeps increasing in multiples. Even Sculptris uses multithreading to let me make models MANY times higher in poly count due to its usage of more than 1 core. I'm not going to waste time stating the benefits of multithreading, just read up on it if you arent already familiar with it

http://en.wikipedia.org/wiki/Multithreading

If anyone wants to see a sloppy wasteful use of tri's when compared to every other 3D app on in this field, just create a cuve in the vox room and hold the W-Key down, for me i see MILLIONS OF WASTED POLYS. yes I understand that the voxels are displayed using a polyskin that is much like ZBrush's unified skin, but Zbrush has the benefit of multithreading and a super fast display algorithm to compensate for such high poly counts Just one example of room for optimization in 3DC, perhaps even using a free core to realtime decimate voxel sculpt regions with low incidence-angle-difference averages per-region, or any of a million other speed-ups 3DC needs to warrant a place in a pipeline with TIMELINES(and not hours of merge progress bars).

3DC is my favorite 3D sculpting toolset, that now, because of timelines, is only a retopo/UV tool in my pipelines. The vox room is now just a glorified ZSpheres implementation, only MUCH SLOWER. I'll hold onto my 3DC license for a little longer in hopes that I see a 64bit OSX release and multithreading added to the app, fingers crossed, or perhaps its Topogun time. I need projects to get finished, as close to the deadline as possible, regardless of toolset, but if going so far as to restart my CPU into 64bit Win7 with my rather powerful system isnt going to make 3DC perform much better, I have to be pragmatic. I dont like MANY aspects of other 3D app sculpting toolsets I use, but at least I can get more than 1 FPS and save my models without out of memory errors or merge-progress-bars that dont finish for hours. No disrespect, 3DC is a great tool, I really enjoy using it, but IMO at this point, it should be known that the vox room(and some aspects of the paint room) can barely be considered useable in many pipelines with deadlines.

on a side note:

I'm on OSX 3DC, and it's still 32-bit, this is UNUSABLE!!! Another example of a performance upgrade that would allow for much faster processes in the app. I have models I cant even open and work on in OSX, and must load up the 3DC demo in windows just to save. In January, before I bought my 3DC license, I was told the following by Andrew regarding the 64-bit OSX release: "I think it will be available during month or two." It's April, and no 64-bit OSX release. Yes I know he said "think", but still I see little-no attention to performance in the needed areas and we are in April now. I still have to boot up into Win7 just to get my models to open and save.

And a robust sculpt room to the tune of AbnRanger's proposals would be VERY useful AFTER performance is addressed.

Oh and, I can run most other 3D sculpting apps in a virtualized domain, but 3DC wont boot in those virtual boxes, has anyone else had this experience too?

Link to comment
Share on other sites

  • Advanced Member

The vox room is now just a glorified ZSpheres implementation, only MUCH SLOWER.

I did not try polyspheres, but I think the ability to melt voxels together in realtime and to cut through them etc. like mercury, is uncomparable to other techniques.

I was speaking specifically for my workflow. I love the "ultra-booleen" nature of 3DC voxels. I in fact prefer them to every other sculpting workflow I've used.. but I cant use them for anything greater than a medium level sketch due to the speed issues on even VERY high-end systems.

Link to comment
Share on other sites

  • Advanced Member

+1 to the OP. Speed and workflow improvements over new features. I wouldn't feel bad about being constructively critical, Andrews a programmer not an artist he needs good feedback to make the tool work for artists not just do the job on paper. Any genuinely good program is a group effort of the users and the programmers.

Link to comment
Share on other sites

  • Contributor

+1. He needs to know where the bottlenecks are (although I'm sure he is aware of many of them), and how it affects an artists ability to get his work done. This would be no different than if he were making proprietary tools for a large studio. If there are aspects of a given toolset that is holding up progress on a production, you can better believe the studio heads will be voicing there angst.

I realize that I do repeat myself here on certain requests, but the old saying does hold true..."The Squeaky Wheel gets the Grease." If users don't repeatedly hold up important issues they are having, it can easily be taken as something trivial and unimportant. I spent a lot of cash at the first of the year, that I could desperately use for other things, on building a new system that wouldn't be brought to it's knees by 3DC (and at times in 3ds Max). Although it helped a good deal, it still isn't enough...because the single biggest show-stopping issue in 3DC is that it doesn't use all the hardware available for many crucial functions. For that reason alone, I stand with ifxs on the Multi-threading issue. It seems to continue being ignored, but in my opinion, it's absolutely critical to get that resolved. The use of just one core on the biggest time-sucking, workflow-killing calculation hog in the whole program....and it remains ignored, while Ptex gets all the love. I don't understand it either. I mean all the little improvements in the world don't add up to a hill of beans when you have the massive black hole of sorts sitting right there in the middle of the Voxel Room?

Multi-level capability in Surfacing mode was on the slate before Ptex came along, and got leapfrogged. This is why I think we need to be measured with our feature requests, and focus on what can be done to refine what is already in the application and resolve longstanding issues. Just my 2 cents, again..

I too would like the full use of my 8 cores, but I read recently that some companies (Not Pilgway, of course) are waiting for faster cores because multiple cores are harder to program for. I think they will have a long wait to get a single core to perform as well as 6 with 12 threads currently available.

Link to comment
Share on other sites

  • Advanced Member
I too would like the full use of my 8 cores, but I read recently that some companies (Not Pilgway, of course) are waiting for faster cores because multiple cores are harder to program for. I think they will have a long wait to get a single core to perform as well as 6 with 12 threads currently available.

There's the thing, instructions dont always scale perfectly across cores. Some things you cant have use all the cores. Most of maths are linear equations, how do you make them multi core?

I dont anything about it but am pretty sure its difficult, otherwise more programs would be doing it and we would see more substantial gains.

By rights though, Moores law and all that we should have 10gigahertz or so cpus by now, but that didnt quite pan out, voxels barring memory, would've been no problem with that.

Link to comment
Share on other sites

  • Advanced Member

I too would like the full use of my 8 cores, but I read recently that some companies (Not Pilgway, of course) are waiting for faster cores because multiple cores are harder to program for. I think they will have a long wait to get a single core to perform as well as 6 with 12 threads currently available.

this is not true at all in the 3D sculpting arena. Zbrush and Sculptris use ALL of my processors, and it shows. The difference between 1 core and 8 cores working is night and day for EVERY app that has implemented this BASIC aspect of a modern professional artist's tool. This is a basic addition to EVERY tool used in the pipelines I work on, with the exception of 3DC(which is why its now out of my pipelines for everything other than retopo, which is doesn't need those extra cores, but it could).

Multithreading support might have been a feature devs were waiting on a decade ago, but there are ZERO apps I use today for my non-hobby work that dont use multithreading.

Link to comment
Share on other sites

  • Advanced Member

this is not true at all in the 3D sculpting arena. Zbrush and Sculptris use ALL of my processors, and it shows. The difference between 1 core and 8 cores working is night and day for EVERY app that has implemented this BASIC aspect of a modern professional artist's tool. This is a basic addition to EVERY tool used in the pipelines I work on, with the exception of 3DC(which is why its now out of my pipelines for everything other than retopo, which is doesn't need those extra cores, but it could).

Multithreading support might have been a feature devs were waiting on a decade ago, but there are ZERO apps I use today for my non-hobby work that dont use multithreading.

Its an exceptional pain in the butt to thread a lot of procedures. to be respectful you might want to quit calling it basic. It sounds like sculptress and zbrush have done a very good job, It does them quite a discredit to assume it was a trivial task.

Link to comment
Share on other sites

  • Advanced Member

Its not as basic as people think. Multithreading is not supported by all appz as people say. There are functions in every app where multithread functions aren't supported yet. Lets not oversimplify things. I do agree we don't need anymore major new features. I'm worried that 3dc will not reach its potential in either painting or sculpting. Zbrush is still king and Mari is coming up which is to me super impressive.

I know Andrew must get bored to do one thing over and over that's why he works on different stuff for 3dc, but 2 major things need to be addressed, performance and multi-level sculpting.

Link to comment
Share on other sites

  • Advanced Member

pardon me, I meant to say "STANDARD" not basic. by no means do would I say that multithreading is "easy" or non-arduous. I've made posts saying just that. Sorry for my miscommunication. Multithreading doesn't need to be implemented in every aspect of an application, just the main areas that REALLY call for the necessary performance boost. It's not trivial, but VERY necessary for a professional tool that doesnt get in the way of the creative process, in my opinion.

P.S. Mari is reasonably priced and looks AMAZING. I've read that with Mari you can load in obj sequences to test my painting on for animation... if so, WOW

Link to comment
Share on other sites

  • Advanced Member

pardon me, I meant to say "STANDARD" not basic. by no means do would I say that multithreading is "easy" or non-arduous. I've made posts saying just that. Sorry for my miscommunication. Multithreading doesn't need to be implemented in every aspect of an application, just the main areas that REALLY call for the necessary performance boost. It's not trivial, but VERY necessary for a professional tool that doesnt get in the way of the creative process, in my opinion.

P.S. Mari is reasonably priced and looks AMAZING. I've read that with Mari you can load in obj sequences to test my painting on for animation... if so, WOW

My apologies, I misread what you said.

Link to comment
Share on other sites

  • 4 weeks later...
  • Member

Since I just started playing with the demo a few day ago, I might not be in the position to comment much, but after the first excitement the performance and stability quickly turned me down :(

Although I could imagine lots of cool feature I think the biggest feature is there already. No worry about topo during sculpting!

So my vote here is clearly to focus on perfomrance and stability too :-)

Link to comment
Share on other sites

  • Advanced Member

Performance can be dependent on what you're trying to do, how you're trying to do it, and your system specs

in relation to the 'what' and 'how'.

Like any software, it takes a bit of time discovering it's stronger uses vs. areas it doesn't perform well at...

(P.S. Schnupps, it might save you some time by putting your system specs in you sig as well) :)

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...