i was really interested to see what they add to Mudbox 2015, but so far i dont see anything significant. looks like another release that can be ignored.
no new sculpting tools, no new mirror functionality, still no way to manage presets(alphas, stencils etc.) so many things that can be added to improve core sculpting but this release looks barely any different from 2014. what is going on ?
is autodesk planning to kill off Mudbox ?
The Maya 2015 UV editor improvements make me super happy. They've finally caught it up with the 21st century with the improved unfold algorithm. Plus, one of my favorite features of Softimage's UV tools, the colored UV boundaries, is included. Built in checkerboard and distortion shaders, plus the improvements to the move UV shell tool... SO many little things fixed or improved that will speed up workflow!
In regards to them taking so long to add some of these changes, I'll just move on with the "better late than never" mindset. Some of the changes, especially the placement thing, seem like they'll be a great help to the things I model so I'm looking forward to the next version of Max.
Still, I'll probably pass on the upgrade, the changes are nice and all but they are too small to justify the cost.
Echoing this. Wouldn't mind better viewport perfomance, but that (plus the quad chamfer and shaderfx) is the result of a year's development?
nope... a lot of features are taking more time than a year and therefore they are not showing up in this release... there is a lot more goin on behind the scenes... for example ICE (moondust) in softimage took about 3 years to get into a release...
Thanks for that info, oglu. For a forum full of game developers, there's seems to be a lack of knowledge about the realities of software development.
except when a single guy can write a plugin in < 3 months (quad chamfer) and it still takes how many years for them to put it in as a standard feature, AND push it as a major feature worthy of an upgrade fee?
im not talking about those minor features... those are more a management decision... you all know how slow this is moving in a big company...
go and listen to the latest modo modcast... brad is talking there how fast it is to create a prototype software and how long its taking to create a final product ready for prime time...
except when a single guy can write a plugin in < 3 months (quad chamfer) and it still takes how many years for them to put it in as a standard feature, AND push it as a major feature worthy of an upgrade fee?
come on man.
All that stuff is fluff to me, all I care about is performance improvement. What's wrong with the quad chamfer plugin that shipping it with Max 2015 fixes? Does the creator of the Quad Chamfer plugin get any money from this?
if you think changing the code of all tools working with subD in an app isnt major change to support openSubD in maya than you should think about it a second time... its affecting nearly every tool in the app... its not just throwing in some new code... you have to make every tool working with this new code... its a major core change that took about two years...
if you like to do the same with max its even more complicated cause of the history stack and all deformers...
then why is everyone so angry that Autodesk "took so long" to incorporate it into Max!?
people are complaining that they're advertising it as a major feature worthy of upgrading (it should have been a service pack addon), as well as it taking so long. it shouldn't have needed a plugin in the first place.
The main problem is how buggy autodesk software is, I know bugs happen in everything, but a big company like autodesk putting out yearly releases, barely adding anything new, then charging 3k for it, its just not worth it.
Gonna get even worse over time, there focusing most of there attention on Maya with the oldest code. Pretty much why BiFrost is a separate application, especially with the multi-threading in Maya.
They killed of XSI with the newest core and are now porting some of XSI's features over to Maya.
They killed of XSI with the newest core and are now porting some of XSI's features over to Maya.
newest core is not equal best or most efficient core... xsi is in big troubbles cause of two reasons... to thight mental ray integration and ice is trapped as operator in a stack based core...
to update mental ray to get close to the stand alone version it would take to much work each release.. to fix this issue you have to decouple mental ray from the xsi core like they did in maya in version 2012 and 13... and the troubles with ice showed up with the first try to integrate rigging to ice... it was to slow... and to make ice what it should be it cant live in a stack... again major core changes...
but i feel with all xsi user thats no excuse to let xsi die...
Autodesk might be saving the actual GOOD feature additions for future extension releases which will only be available for subscribers so that people pay more for the subscription.
monster: Have you done much python scripting using the 2014 extension? I was wanting to use it for a big re-write of internal tools to help integrate with our contract partner's tools but it ended up being a waste of time. I was really expecting a lot of common maxscript functions with the same names or at least some sort familiarity between the 2 but it was pretty different. Maybe they are wanting the community to write those libraries or wrappers for them.
Most of the time I found myself wanting to write the tool mostly in maxscript and then send commands to python scripts from time to time to get a result assigned to a mxs variable. It doesn't seem to work that way but works the other way around if you run maxscript from within python with their returned FPValues. Then again I'm a noob with python so I'm wondering if you had better luck than me.
Also there's not a python interpreter in the Listener and anything requiring a ui was troublesome. I tried pyqt which worked ok but that was beast in itself to learn on top of python. Ideally I would of wrote the ui in maxscript but again you cant easily get returned values from python that I know of...
Thanks for that info, oglu. For a forum full of game developers, there's seems to be a lack of knowledge about the realities of software development.
No. We just see all around there is software for game developers (!), that is progressing much faster, where company listen to feedback, and is much cheaper.
It boggles my mind how anyone can defend autodesk 0_o.
It boggles my mind how anyone can defend autodesk 0_o.
... because some people are *gasp* actually happy with the updates? The maya tools look great and I'm actually looking forward to the new release.
If you don't like it there are many alternative apps, or perhaps learn to script the tools you feel you really need? Just feels like a lot of unnecessary bitching and gets kinda boring.
Do you pay your own software are are you employed? Because most of the bitching comes from the price policy 4485 euro including taxes, a yearly subscription or an update for 2500 (and thats 20% off atm) thats a lot of money for a single seat. Put 1000 on top if you want it floating. So if I set up one seat for a new artist this money has to flow back somehow, thankfully you can rent it now, but the price policy is again horrible here
it costs 245 Euro a month here, while in the states it costs 195 Dollar, thats 140 euro, or 100 euro difference in price. For the same digitally downloaded product, no shipping from the states to here, no translation needed, nothing it's the same.
Do you pay your own software are are you employed? Because most of the bitching comes from the price policy 4485 euro including taxes
I think it's more down to the fact that people feel AD isn't making enough innovation with their updates and that the software is archaic. I was just saying that there are other apps to try like Blender and Modo that may accommodate to people's tastes more, but some would rather stick to the first app they used and complain about how crappy it is with lame updates. Obviously the artist has to use whichever software the studio paid for, there's not much that can changed about that, like you said licenses are expensive.
I'm honestly surprised (or actually maybe not surprised) that in max you still cannot setup shader networks like in maya or softimage.
i.e. connect a noise map to the uv coords of another map. It's a simple enough thing in theory.
All the slate editor and such have just reshuffled a patchy feature set, and there is even more glue and fluff added to a creaky base in this release. Oh well.
All these new things I doubt I will use in anger, but we will see.
In the studio on my team we are still using 2011 for production and currently evaluating 2014. It's just a shame we had to pay the subscription for the time inbetween!
Anyway, I'll carry on picking up modo with a dream of a software agnostic future.
In the studio on my team we are still using 2011 for production and currently evaluating 2014. It's just a shame we had to pay the subscription for the time inbetween!
same here, have fun with the tangentbase crap, not sure if it was fixed in 2014, we have full licenses just for data conversion collecting dust here :X
I'd be happy if they fixed bugs. I worked with large scenes, with a lot of active modifiers, materials and imported models and there are at least 20 crashes a day or constantly encountering things that don't work properly.
same here, have fun with the tangentbase crap, not sure if it was fixed in 2014, we have full licenses just for data conversion collecting dust here :X
TBN is working in 2014. You can also switch it in the options (as long as you include a bit of voodoo hlsl in the viewport shader)
you also have to either make a dx11 shader or revert to dx9- switch off gamma, plus they fixed a long standing bug which means one or other of the T or the B is inverted now (or switch in the texture, your choice)
Once all that is done it works well.
DX11 shader performance is much better then DX9 (as you can imagine) but I didn;t dig in much yet. Just getting the basics up and running.
I would just be happy if they could make the Unwrap UV modifier in Max remember the UV window placement when I put it on my second screen, and remember that I want my edges to be 'thin' and not 'thick' and remember that I don't want normalize ticked on! Why can't it just do simple things like that? I don't really care about innovation when there are tiny little things that could use some thought.
I'd be happy if they fixed bugs. I worked with large scenes, with a lot of active modifiers, materials and imported models and there are at least 20 crashes a day or constantly encountering things that don't work properly.
What would that end up doing?
It would add uv distortion, the same way you would do so in hlsl shader.
Using procedural values as inputs for other procedural maps is the key part of procedural shading and max doesn't expose everything or supply utility nodes to perform basic maths or other operations.
It's mostly irrelevant for realtime use since you can do all that in the shader in engine, but it's just 'one of those things' which 3dsmax doesn't do quite right. (and softimage does, but is being shut down!)
I would just be happy if they could make the Unwrap UV modifier in Max remember the UV window placement when I put it on my second screen, and remember that I want my edges to be 'thin' and not 'thick' and remember that I don't want normalize ticked on! Why can't it just do simple things like that? I don't really care about innovation when there are tiny little things that could use some thought.
For the thin seam display by default, just find your "unrwapuvw.ini" in your "<Maxroots>\plugcfg_ln" folder. Open in a texteditor, search for "ThickOpenEdges" and set it to 0. Save & restart Max and you are done. Did'nt find the normalize option in there though
You can save the windows positions there too by using the following entries ( here's my example ) WindposX2 means right window border etc..
Maya updates are looking good. The UV tools might actually be usable now. Also the new booleans look far more reliable and interactive.
The modeling toolkit is very powerful but it hasn't quite become fully integrated with the app yet. I find I am constantly having to activate/deactivate it, pretty much any standard modeling tool will cause it to shut off, it's very "all or nothing". That being said it is still worth dancing between it and the standard tools because they are very useful.
So have they actually done anything at all to improve NEX inside of Maya? It was a pain like you say activating/deactivating NEX all the time, not really a smooth workflow.
Guess we have to wait for that in 2019, meanwhile Modo continues to knock out awesome stuff every release.
Replies
http://www.autodesk.com/products/autodesk-3ds-max/features/new/gallery-view
just a shame all my clients are like "max or maya plz".
Now about 3ds max updates.... And yeah I agree lee.
edit: I didn't notice it on the Maya LT product page, but apparently they did add Geodesic Binding, as well.
no new sculpting tools, no new mirror functionality, still no way to manage presets(alphas, stencils etc.) so many things that can be added to improve core sculpting but this release looks barely any different from 2014. what is going on ?
is autodesk planning to kill off Mudbox ?
My thoughts exactly.
I didn't even see the video for that, this and the new DDO suite.... I think my head is going to explode :poly127:
model in Modo, export to Max?
agreed..
+1
I really want that max placement tool feature in maya though. So annoying that we probably wont get it for awhile.
Still, I'll probably pass on the upgrade, the changes are nice and all but they are too small to justify the cost.
Echoing this. Wouldn't mind better viewport perfomance, but that (plus the quad chamfer and shaderfx) is the result of a year's development?
That may have won me over
nope... a lot of features are taking more time than a year and therefore they are not showing up in this release... there is a lot more goin on behind the scenes... for example ICE (moondust) in softimage took about 3 years to get into a release...
except when a single guy can write a plugin in < 3 months (quad chamfer) and it still takes how many years for them to put it in as a standard feature, AND push it as a major feature worthy of an upgrade fee?
come on man.
go and listen to the latest modo modcast... brad is talking there how fast it is to create a prototype software and how long its taking to create a final product ready for prime time...
All that stuff is fluff to me, all I care about is performance improvement. What's wrong with the quad chamfer plugin that shipping it with Max 2015 fixes? Does the creator of the Quad Chamfer plugin get any money from this?
then why is everyone so angry that Autodesk "took so long" to incorporate it into Max!?
if you like to do the same with max its even more complicated cause of the history stack and all deformers...
people are complaining that they're advertising it as a major feature worthy of upgrading (it should have been a service pack addon), as well as it taking so long. it shouldn't have needed a plugin in the first place.
Gonna get even worse over time, there focusing most of there attention on Maya with the oldest code. Pretty much why BiFrost is a separate application, especially with the multi-threading in Maya.
They killed of XSI with the newest core and are now porting some of XSI's features over to Maya.
newest core is not equal best or most efficient core... xsi is in big troubbles cause of two reasons... to thight mental ray integration and ice is trapped as operator in a stack based core...
to update mental ray to get close to the stand alone version it would take to much work each release.. to fix this issue you have to decouple mental ray from the xsi core like they did in maya in version 2012 and 13... and the troubles with ice showed up with the first try to integrate rigging to ice... it was to slow... and to make ice what it should be it cant live in a stack... again major core changes...
but i feel with all xsi user thats no excuse to let xsi die...
I gotta say I'm very happy with this 3ds Max update. Python scripting is awesome.
Most of the time I found myself wanting to write the tool mostly in maxscript and then send commands to python scripts from time to time to get a result assigned to a mxs variable. It doesn't seem to work that way but works the other way around if you run maxscript from within python with their returned FPValues. Then again I'm a noob with python so I'm wondering if you had better luck than me.
Also there's not a python interpreter in the Listener and anything requiring a ui was troublesome. I tried pyqt which worked ok but that was beast in itself to learn on top of python. Ideally I would of wrote the ui in maxscript but again you cant easily get returned values from python that I know of...
No. We just see all around there is software for game developers (!), that is progressing much faster, where company listen to feedback, and is much cheaper.
It boggles my mind how anyone can defend autodesk 0_o.
... because some people are *gasp* actually happy with the updates? The maya tools look great and I'm actually looking forward to the new release.
If you don't like it there are many alternative apps, or perhaps learn to script the tools you feel you really need? Just feels like a lot of unnecessary bitching and gets kinda boring.
it costs 245 Euro a month here, while in the states it costs 195 Dollar, thats 140 euro, or 100 euro difference in price. For the same digitally downloaded product, no shipping from the states to here, no translation needed, nothing it's the same.
I think it's more down to the fact that people feel AD isn't making enough innovation with their updates and that the software is archaic. I was just saying that there are other apps to try like Blender and Modo that may accommodate to people's tastes more, but some would rather stick to the first app they used and complain about how crappy it is with lame updates. Obviously the artist has to use whichever software the studio paid for, there's not much that can changed about that, like you said licenses are expensive.
i.e. connect a noise map to the uv coords of another map. It's a simple enough thing in theory.
All the slate editor and such have just reshuffled a patchy feature set, and there is even more glue and fluff added to a creaky base in this release. Oh well.
All these new things I doubt I will use in anger, but we will see.
In the studio on my team we are still using 2011 for production and currently evaluating 2014. It's just a shame we had to pay the subscription for the time inbetween!
Anyway, I'll carry on picking up modo with a dream of a software agnostic future.
same here, have fun with the tangentbase crap, not sure if it was fixed in 2014, we have full licenses just for data conversion collecting dust here :X
What would that end up doing?
TBN is working in 2014. You can also switch it in the options (as long as you include a bit of voodoo hlsl in the viewport shader)
you also have to either make a dx11 shader or revert to dx9- switch off gamma, plus they fixed a long standing bug which means one or other of the T or the B is inverted now (or switch in the texture, your choice)
Once all that is done it works well.
DX11 shader performance is much better then DX9 (as you can imagine) but I didn;t dig in much yet. Just getting the basics up and running.
Good luck!
It would add uv distortion, the same way you would do so in hlsl shader.
Using procedural values as inputs for other procedural maps is the key part of procedural shading and max doesn't expose everything or supply utility nodes to perform basic maths or other operations.
It's mostly irrelevant for realtime use since you can do all that in the shader in engine, but it's just 'one of those things' which 3dsmax doesn't do quite right. (and softimage does, but is being shut down!)
Oh well.
For the thin seam display by default, just find your "unrwapuvw.ini" in your "<Maxroots>\plugcfg_ln" folder. Open in a texteditor, search for "ThickOpenEdges" and set it to 0. Save & restart Max and you are done. Did'nt find the normalize option in there though
You can save the windows positions there too by using the following entries ( here's my example ) WindposX2 means right window border etc.. The position values though get overwritten when you choose "Options->Save Current Settings as default"
The modeling toolkit is very powerful but it hasn't quite become fully integrated with the app yet. I find I am constantly having to activate/deactivate it, pretty much any standard modeling tool will cause it to shut off, it's very "all or nothing". That being said it is still worth dancing between it and the standard tools because they are very useful.
Guess we have to wait for that in 2019, meanwhile Modo continues to knock out awesome stuff every release.