Everything related to Maxwell Render and general stuff that doesn't fit in other categories.
#397066
it really bugs me seeing that all other engines are pushing forward development- and my favorite (and best in my option) engine lags. you can see the steep decline of forum posts and facebook likes... why not just let people use maxwell totally free for 4 months or so, I don't care I paid money since maxwell 1. I'd like to see big interest and uses to maxwell in a way that will make those people paying users and that way speed up dev....
#397071
I don't know if that solves the problem. Maxwell is one of the more affordable render engines out there (especially if you are purchasing additional render nodes). The 'problem' I see for Maxwell is a lack of integration as compared to other render engines (especially with Maya & 3DS Max, which, at least here in the states, are still the ones that matter), and a lack of ease of use. The learning channels for V-Ray are everywhere. You can youtube your way to doing a lot of pretty complex stuff in V-Ray. Arnold as well, these days. Maxwell requires you to try and understand some pretty dense documentation. While there is a Maxwell material library, offerings like Vray Scans or Substance Source are grabbing the attention of a LOT of people. I think a deep integration with Autodesk and substance, along with better learning materials would go a LONG way to helping NL's market position, but I don't really know at this point. It seems many folks have left Maxwell and I doubt they are returning.

Also...

I know this is a personal issue, but it's stuff like this that drives people away....

I purchased to Maya licenses when I got v4. I only ever used one. I bought a new machine, downloaded the second license for that machine and asked if I could delete the old Maya license and get a license for Studio instead (which used to be included and, honestly, still should be since it does things the plugins cannot). I was told no, that it had been too long and that I would have to purchase Studio. Seriously? That seems insane to me (not to mention bad customer service).
#397073
Oh :) here is my 5 cents.
I love Maxwell. It gives me that beautiful time to tinker around.
But let's be honest - in order for Maxwell to climb back to top of the industry - it has to do a few very important things.

1. First of all - Priorities. Maxwell can't afford itself to be a generic render engine that is "good for everything". Nope, it never was and will probably never be - but that is more of a benefit than a problem. Maxwell should have a very clear vector to where it wants to grow. Say, Object Visualization (product / automotive / arch).
If that is the main priority then development should be tunnelled to that. Meaning stripping down everything that slowdown the the development of the best tools for that target.

2. Licensing burrito. It is vital for Next Limit to make money in order to produce great products - but it is as equally important for consumers to feel that they are supported. And if for next limit 500 euros is not a big thing - for independent freelancers it is. Big studios can easily swallow those sometimes irrational licensing policy but every software's core community is freelancers. Especially these days. So if someone want to swap one license on another - what's the big deal - tell him you'll make an exception for him and make him feel special. In return you'll get a boost of loyalty!

3. Integration. And Mathew is absolutely right here. Maxwell can't afford itself to ignore what is going on in the industry these days. Maxwell studio SHOULD NOT compete with proper DCC packages (read Maya, 3D Max ... even Cinema4D). Like it or not but Maxwell at it's best when used as a plugin. So deep integration is extremely important here. Ideally NODE based system (as the main app use that in their core) will make Maxwell a kick-ass blockbuster with unmatched image quality.

4. Yep - accessibility. You guys have a very extensive documentation and that was enough ... back in the days - not anymore. Do you know why other's software grow rate is higher ? they record videos (sometimes you feel like reading - but in most cases you don't .. kids these days .. right ? :) ) and if your users are not recording at least a few video tutorials every couple of weeks - you're vanishing from the web. So what other (relatively new developers) do, who don't have a huge fan base - they appoint community managers and (even engineers / technicians) to record video tutorials, with explanation on how things works and what cool thing you could do and so on.
Check out what allegorithmic does they have massive video playlist with various video tutorials, conferences, tips, tricks, project based stuff and so on. Basically everything for people to jump right in and start making cool stuff from the day one. Look what autodesk does ( Autodesk! Karl! ) beside the "experts" and some community managers you can easily find videos recorded by an autodesk technician who describes some new technique or a feature. And - if you think that Autodesk is a massive company and can afford themselves to sit back and not do this - well Autodesk thinks differently here.

But .. I love Maxwell. I do.
And I believe in it.
To me ... Maxwell render gives something very unique. That other render engines use don't. Something very organic and there is a character in pretty much every render. That separates it from other engines.

And if there is anything I can help with - I'm always open to participate in everything that will bring Maxwell back to the top.

To sum it up, I think that Maxwell have to define it's place, it's core industry and work it's way to the top.
If the primary focus is arch viz (beside the secondary ones) - them push into it. Develop tools that Architects can't live without. Same goes for object viz - give us something we so desperately need - and, I'll surprise you - this is not speed. It's not even in top 3. Everyone who uses Maxwell is fine with waiting extra time for an outstanding result - we just need specific tools that will simplify our process before we push that render button.

Partner with studios (and with individuals) - record video tutorials / organise conferences / do more case studies

Sorry for a long one .. I just take it personally. I'm with Maxwell since 2012 ... not since like beta or anything ..but I'm with you guys and want to stay with you guys and I want you to succeed.

Artem.
#397075
I agree with you guys, that's why I've started this tread, to let NL see what long time users think and feel.
btw I think integration with maya (what I use) is almost perfect but as I use MR only for arch-viz I may miss features that others require.
the maxwell TV was great but we do need more like that. I really think that making MR accessible to more people - even with long term trial periods - will help us all to get the feature we ask for in the "wish list" and bug fixes faster.
#397077
Well i can tell you the problem Maxwell render has from my point of view, no order of importance.


1) Integration. We live in a world where people want instant feedback, not instant crashes. FIRE is not remotely polished to give users a smooth and efficent workflow to light scenes.
The only thing it gives you is a very rough and noisy approximation of where the lightsource comes from. You cant really see anything else, from material issues, to geometry issues. And most of the time (which is extremely frustrating and hilarious too) it crashes when you change some parameter, like materials roughness, surface, color, or move lights. I mean its a tool made to preview and give interactive feedback and honestly it fails to do that with an average complex scene.

2) Voxelization needs seriously to be faster, everytime you start a render voxelization takes so long it defeats the purpose of trying to preview something. A modern scene now has a lot of geometry, not just a sphere on a plane.

3) Speed, this is a major issue, lets face it, Maxwell has a great look in terms of rendering quality, there are people that wants to wait all the time it takes to achieve that quality, but honestly other render engine are more than capable too, and im not sure the cooking time is justified anymore in 2018. Maybe the "super physically correct" path taken wasnt the best option. Maybe start approximating more things?

4) Development is extremely slow, only reason for this is because there might be too few people working on it. But the amount of features other engine spit out in one year (even months) makes Maxwell render pales. Maxwell 4 is the same exact as Maxwell 3.2, except you threw 600€ or more out of the window (sorry but im extremely frustrated by what Nextlimit pulled out last product update and price change).

5) Stability went really down a lot, i've never seen so many crashes on Maxwell studio than version 4. When i open 3.2 and work with it, EVERYTHING and i mean everything feels smoother, lighter, faster, more interactive and better designed. I dont know if its a matter of that UI interface change, if its a matter of license check crap, or whatever they have done.

Open 3.2 studio, open a scene click render > instantly sends to Maxwell render
Open 4.2 studio , open a scene click render > hangs for seconds (maybe checking license) then starts Maxwell Render.

Its just an example where is extremely noticeable. There are other areas where maxwell 4 is more clunky than 3. Which is again.. astonishing considering its a new product.
The fonts and buttons and interface in general feels way worse.

6) Plugin integration. that is somethin personal, i was using softimage plugin that of course is EOL, cant pretend to be supported, but yeah it got chopped off, Same with houdini. I dont use max or maya, when i upgraded from 3 to 4 i had to pick a plugin but since i use softimage, i couldnt care less what i picked so i went for maya (dont have maya). Now i asked if i could swap the maya license one time only , for 3ds max one (it costs the same) because the studio im going to work with has 3ds max, and they said they wont do it. Ok i respect that, but honestly considering the dark waters Nextlimit is navigating, i personally would have done that to a client. You litterally dont lose anything by allowing that swap, for one time only. Whatever.

Bottom line, sadly i dont think this product will live long, people who use it are very few, and a niche market. Im not criticizing the output quality at all, i just feel Next limit couldnt do worse on managing this product.

We'll see, but honestly im not happy at all with Maxwell at the moment, not been since 4 release.
#397081
Hey Max - you're right in a lot of things ... and it feels like it is personal - and it should be personal.
Can't relate crashing issues and voxelization waiting times ... everything is nice and smooth on my side (I'm on Mac Pro - 64 gb ram) so everything is pretty stable here.
Speed .. well .. as I said Maxwell is not for everything. Just like every professional product. Maxwell is designed for a very specific tasks and to achieve very specific results - even though it is a bit hard to see and Maybe NL should articulate that a bit more which might scary away people who would compare Maxwell to Vray or Arnold but will deifnielly attract those who seek for that unique quality in their images.

Alto I do agree that development roadmap feels a bit like "we're doing everything". While I think it is a perfect time for NL to stand out and say we're not for everything we're for this this and this - and we going to focus on this. That way Maxwell will get back it's industry .. as right now it does feels like it is for everything .. which is similar to majority other render engines .. but only they are faster .. and results .. well .they are acceptable.

Personally I don't mind slow development - as long as the tools in development are the ones that will be really helpful for a specific tasks. Rather than just trying to have what others have.
#397082
Nasok wrote: Alto I do agree that development roadmap feels a bit like "we're doing everything". While I think it is a perfect time for NL to stand out and say we're not for everything we're for this this and this - and we going to focus on this. That way Maxwell will get back it's industry .. as right now it does feels like it is for everything .. which is similar to majority other render engines .. but only they are faster .. and results .. well .they are acceptable.
I'm trying that, showing some things already working on "guess the improvement". And in general, we are re-writting several plugins, and on GPU-side multi-gpu is on testing and additives are the current big thing we are working on.
#397083
Luis - I believe there is no single person here who would think that you guys a lazy or don't do things ... absolutely no.
In fact - we are all here - for a very specific reason - and it is that we love Maxwell.

What I've meant to say is that it feels like that you guys have too much stuff in development at the moment. And it feels like everything that is in development is for different purposes and targeting different user groups, different workflows and even different industries. What I think would be beneficial for a long run is to clearly define the single direction and put all power into it. For instance for Archive - GPU is not really a necessity .. I would say that even for product viz it is not. We're here not for the speed. We're here for the quality.
I honestly think that you guys need to get together and decide primary goals and secondary ones - and this will dictate what exactly should be in development and what could wait. Otherwise you might do something that good (and even something that other engines have) but it will no add up to what maxwell is.
GPU for instance ... I really don't understand the craze with it. Especially in Maxwell's context when speed is not the priority but quality IS. Why would you sacrifice that ? If you need speed grab a keyshot ... or even Arnold (that is not bundled with Max and Maya) and you're good to go - fast and acceptable ..
#397084
GPU utilization seems to me as a low priority now, as the denoiser deliver very good results and cuts render times 10X faster. I don't think someone can really use the GPU engine that lacks core features of the CPU engine. I believe that adding the possibility to import other common materials formats is much more important to most of us, and getting the denoiser to use less RAM, than using GPU crippled engine...
#397087
That is right .. Denoiser is really cool especially when you're dealing with liquid (say leash from RealFlow) or with big open surfaces (i.e. architecture) And the way it work - for sure could be improved - but it is already a pretty solid solution. (Improved in terms of a user experience, rather than speed or ram usage)
#397095
or even Arnold (that is not bundled with Max and Maya)
Well, if you’re under maintenance or are a subscriber to one of the 3 collections, then you have access to Arnold. Check your Autodesk account. Rendernodes are NOT included. :wink:
#397097
luis.hijarrubia wrote:
Wed May 09, 2018 11:05 am
Nasok wrote: Alto I do agree that development roadmap feels a bit like "we're doing everything". While I think it is a perfect time for NL to stand out and say we're not for everything we're for this this and this - and we going to focus on this. That way Maxwell will get back it's industry .. as right now it does feels like it is for everything .. which is similar to majority other render engines .. but only they are faster .. and results .. well .they are acceptable.
I'm trying that, showing some things already working on "guess the improvement". And in general, we are re-writting several plugins, and on GPU-side multi-gpu is on testing and additives are the current big thing we are working on.

We have recently been told that future developemnts of plugins like MODO & Solidworks will be dropped. Will Next Limit reconsider to continue development? :(
  • 1
  • 2
  • 3
  • 4
  • 5
  • 8
Rhinoceros 6 support

I just installed the new maxwell_rhino_v4.2.6.3_WI[…]

is Maxwell Render abandoned?

"You can expect good news in a few weeks. Be[…]

Good day! We have just released a new version of […]

Hi there, after a texture tag has been assigned to[…]