renato Posted November 23, 2004 Share Posted November 23, 2004 I'm considering the possibility to buy Maxwellrender, but I have some questions that maybe someone can shed some light on it: Can I run Maxwellrender in two machines? (I'm looking the possibility to share the cost with a friend and split the license in 2 pc's). Maxwell guys say that it is a "float" type license, but I don't understand the term very well. Can maxwell do net rendering?, for large still images or animations it would be very cool. Does it work with max standard materials? or which materials are not compatible with maxwell? does it work with standard lights? does it accept photometric data? What about stability. Does it crash? does it come with good documentation? Thanks in advance Link to comment Share on other sites More sharing options...
kunstraum.tv Posted November 23, 2004 Share Posted November 23, 2004 to all the points (except the two machine license and standard material) i have to say: actually no ! maxwell is really early alpha. all the things you are able to do now are very basicly. this is okay, but from the functions you mentioned maxwell is very far away...give them some time Link to comment Share on other sites More sharing options...
abicalho Posted November 23, 2004 Share Posted November 23, 2004 It isn't. It could have been, but they chose to go in another direction. Physical accuracy requires real-world units, which both Max and Cinema lack, but I guess Maxwell will have some way of dealing with that. Sorry, Ernest, but MAX's Radiosity engine is accurate. And 3dsmax does support real world units, has always supported since MAX 2.5. I've tested all 3 renderers in MAX 7: Lightscape, Radiosity and Mental Ray against the real Math formulas that compute the result (not taken from code) and found out that all 3 generate very similar results in Lux, with less than 5% difference between them. The only difference came from Area Lights, since the Radiosity engine's algorithm deals with them differently. MAX Radiosity is a different engine, it does have different meshing, but if you're using Point Lights, IES Lights or Spots, it's not innacurate when compared to Lightscape. I put my hand on fire on this topic and will challenge anyone who says it's incorrect compared to Lightscape to prove so. And if you're inclined to test different renderers, you can also use mental ray's GI in MAX 7. It's as accurate (or more) than Lighscape. Alexander Discreet Quality Engineer for Radiosity Link to comment Share on other sites More sharing options...
nisus Posted November 23, 2004 Share Posted November 23, 2004 Can anyone use the Maxwell on the Unchallenge models and give us some specs for that? Tnx! rgds (half-face) nisus Link to comment Share on other sites More sharing options...
Ricardo Eloy Posted November 23, 2004 Share Posted November 23, 2004 This is getting hot... Alex is right when he says max can handle real world units, Ernest. And Max's radiosity IS accurate, even though it works with statistics. That said, maybe we should focus on some other issues here. For example, the questions Renato made about photometric data. I know it's early alpha and stuff (as marc mentioned above), but one has to wonder how Maxwell will handle this kind of data. If it doens't support it (yet), how does one get real world results if not using real world lights? Nisus: Good idea! Then, we can have some real benchmarking. Link to comment Share on other sites More sharing options...
Ernest Burden III Posted November 23, 2004 Share Posted November 23, 2004 This is getting hot... Alex is right when he says max can handle real world units, Ernest. And Max's radiosity IS accurate, even though it works with statistics. Hot's good, Rick. OK, that's news to me. I had thought that in Max you cannot switch between unit types without scaling the model. Is that wrong? Certainly Cinema treats size as a point of notation only, it doesn't care if they're right or not. The larger point is how nice is is to be able to just put a light (including IES) with a certain strength into a space and have it behave in a predictable way--no falloffs to calc (guess), no faked ambients, it just works. It sounds like that is the news on Maxwell. Lightscape is a fast, accurate raytracer. But when you include the time to do light calcs, that first image can take a day (less so with newer PC). The next 2500+ images are really fast, though. There is an animation on the Maxwell site of the sun orbiting an architectural whiteboard model. I would love to find out how long that took. Link to comment Share on other sites More sharing options...
renato Posted November 23, 2004 Share Posted November 23, 2004 I was very attracted to the fact that maxwell handles very accurate light distribution, based on real world physics. But I can see now that maxwell is very far from this,or maybe it's not, but now it cannot be used in a production workflow, of course is an early stage, but not many people are happy to spend $395 in a beta renderer that is very limited. Even if I spend my money on this, I don't know if the final release is gonna cover my expectations, so, I think the guys from nextlimit made a mistake in selling their product at this very basic stage. They should have released a free beta, (watermarked or whatever) and then the customer would have more arguments to decide if they buy the renderer or not. Link to comment Share on other sites More sharing options...
Ricardo Eloy Posted November 23, 2004 Share Posted November 23, 2004 You got yourself a point there, Renato. Unless they have something expectacular under the sleeves, they should have done like it was done with Vray. Release public, watermarked betas to see what people want and how it behaves. But I'm sure they know what they're doing. Link to comment Share on other sites More sharing options...
DavidR Posted November 23, 2004 Share Posted November 23, 2004 Yeah, like Brazil/Ghost did, too, and Rhino. It helps spread the word fast. How about a Sibenik-type challenge? Let's propose a big scene with some fine detail and everyone do their best. That wills ettle things and/or show progress, new features, etc. Sibenik was great for that, and also showed that scanline guys can out-do mental ray guys sometimes. Link to comment Share on other sites More sharing options...
Jeff Mottle Posted November 23, 2004 Share Posted November 23, 2004 I too was thinking about why they opted to charge for the early alpha (or give an early discount on the final version) and I came to a few conclusions: 1. They automatically ensure that the "alpha" testers are serious professionals and not just some hack who wants to render monsters and make unproductive comments about the product. Also if you buy the product, you are much more likely to actually do some serious tests with it. 2. If they get the serious people involved upfront, they will more likely be future customers becuase of all the time they spend upfront learning the application. Which is good for them, becuase they get qualified users making high end images and that only lends itself to a more professional looking product. 3. Cashflow, Cashflow, Cashflow. It does not matter what you are selling the sooner you can start a cashflow the better. Maybe they took a risk taking this approach, but I think it was a wise one. Just take a look at how many views this thread has been getting. Granted it's been posted on the home pages for 2 days, but this thread is the most popular one in CGA history for the time period. No other product release has generated this much excitment. All this for a product that is not even a product yet...I'd say they are doing well. The final test will be the product release, but look at where vray started. I'm sure the developers of the product will take note of what is important in production and make the neccesary adjustments in the application. Link to comment Share on other sites More sharing options...
Ernest Burden III Posted November 23, 2004 Share Posted November 23, 2004 I too was thinking about why they opted to charge for the early alpha Overall, I agree with your points, Jeff. Well said. The only addition I can make is to point out that I was part of the original Rhino public (free) beta and can say that it attracted top pros doing real work, and the discussion was highly professional and productive. It was a model for how to develop a product. But this way is fine, too. Link to comment Share on other sites More sharing options...
Tim Nelson Posted November 23, 2004 Share Posted November 23, 2004 Yea thats fine if they want to charge us to test their product for them. But I'll be holding onto my $400 for a while until I can see some more positive results - especially if I have until June '05 to decide. Link to comment Share on other sites More sharing options...
Jeff Mottle Posted November 23, 2004 Share Posted November 23, 2004 The only addition I can make is to point out that I was part of the original Rhino public (free) beta and can say that it attracted top pros doing real work, and the discussion was highly professional and productive. It was a model for how to develop a product. But this way is fine, too. Yeah McNeel set the trend for this type of public testing, but I think a product like Rhino is automatically condusive to professionals. Not too many non-professionals would want to use Rhino. Kinda like you will never find a Catia for dummies on the shelves. Renderers seem to attract a much broader audience. I do know what you mean though. Link to comment Share on other sites More sharing options...
jucaro Posted November 23, 2004 Share Posted November 23, 2004 Until it has proven itself in a real production pipeline, I'd see it as a well thought of hype. Just my 2 cents. US$400 is not easy here. Link to comment Share on other sites More sharing options...
Tim Nelson Posted November 23, 2004 Share Posted November 23, 2004 US$400 is not easy here. Hehe, it's not always easy here either! I don't think its hype though. I love the look of things so far, although I am leary about the render times. The first renderer I ever learned to use was Accurender. To me it was a brilliant program - mainly because of its ease of use and fairly decent results (for its time). Unfortunately, it hasn't had a new version in about 5 years and its technology is long outdated. But its learning curve was pretty short and it was really fun to work with. My hopes are that Maxwell will be my new solution. In the meantime I will waste no time practicing with Vray! Link to comment Share on other sites More sharing options...
Ricardo Eloy Posted November 24, 2004 Share Posted November 24, 2004 1. They automatically ensure that the "alpha" testers are serious professionals and not just some hack who wants to render monsters and make unproductive comments about the product. Also if you buy the product, you are much more likely to actually do some serious tests with it. I'm afraid I don't completely agree on this, Jeff. I think they could have asked some serious pros to be beta or alpha testers if they wanted. The way I see, they are actually worried about making money since now so they can continue developing their product. You see, 3ds max has serious beta testers all around the world, and they don't exactly pay to be one (as far as I know). Of course, renderers are always more appealing to test than modeling softwares such as Rhino, for example, and more likely to have non-pro people lurking around to pointless play with. But that's also why it's getting so much attention from us, and I'm pretty sure Next Limit has got a smile from ear to ear right now just because of this thread. Link to comment Share on other sites More sharing options...
skogskalle Posted November 24, 2004 Share Posted November 24, 2004 all the maxwell rendertimes ive seen so far are insane.... (but pics look good) in Vray you can cut the time for GI calc by rendering the GI at a smaller resolution than the actual render... is this possible in maxwell? Link to comment Share on other sites More sharing options...
ceballos Posted November 24, 2004 Author Share Posted November 24, 2004 A test picture. The light is from the Maxwell Physical Sky with Sun light Rendertime : 150 min Link to comment Share on other sites More sharing options...
ceballos Posted November 24, 2004 Author Share Posted November 24, 2004 Remember this? Rendertime : 180min Link to comment Share on other sites More sharing options...
Fran Posted November 24, 2004 Share Posted November 24, 2004 What I'm thinking is that theoretically, you can create a light bulb or tube, apply a light emitter material to it, and stick it into your fixture. Maxwell should be able to simulate the distribution based on the actual construction of the luminaire. This is what happens in real life. Anyone want to try that? Link to comment Share on other sites More sharing options...
Andronikos Posted November 24, 2004 Share Posted November 24, 2004 all the maxwell rendertimes ive seen so far are insane.... (but pics look good) in Vray you can cut the time for GI calc by rendering the GI at a smaller resolution than the actual render... is this possible in maxwell? No! Does not use GI or Radiosity or ..etc. No fakes! that is the philosophy... Link to comment Share on other sites More sharing options...
Andronikos Posted November 24, 2004 Share Posted November 24, 2004 What I'm thinking is that theoretically, you can create a light bulb or tube, apply a light emitter material to it, and stick it into your fixture. Maxwell should be able to simulate the distribution based on the actual construction of the luminaire. This is what happens in real life. Anyone want to try that? I can try explain a bit more plz... what do you mean by stick it into your fixture".? Link to comment Share on other sites More sharing options...
rivoli Posted November 24, 2004 Share Posted November 24, 2004 No! Does not use GI or Radiosity or ..etc. No fakes! that is the philosophy... what do you mean does not use GI, of course maxwell calculates GI (which is global illumination). GI is what maxwell does, in a supposed physically accurate way. MAX's radiosity does not mean fake, it's a method to calculate indirect illumination. Link to comment Share on other sites More sharing options...
Andronikos Posted November 24, 2004 Share Posted November 24, 2004 what do you mean does not use GI, of course maxwell calculates GI (which is global illumination). GI is what maxwell does, in a supposed physical accurate way. MAX's radiosity does not mean fake, it's a method to calculate indirect illumination. I didn't want to t meant that, but there is no option of anable/disable global illumination or how many samples,photons etc... therefore it does not uses what all other renderes are using right now (however you want to call it). But you are right Global Illumination is Global Illumination so it was a expression mistake...sorry... and I am not sure if Max's radiosity is physically accurate...not much experience. Link to comment Share on other sites More sharing options...
Tim Nelson Posted November 24, 2004 Share Posted November 24, 2004 I was curious to see how one of the scenes posted at http://www.ixor.gr/maxwelltest/maxwell.htm would render with Vray. I had to import the .3ds file (the .max file wouldn't open in Viz 2005). All I did was add a IES sun (which I have never used before) and cranked up the environment light until it looked good. This one rendered at 8:40 (minutes/seconds) using DR with about 10.5ghz total. I would have to admit that the lighting looks slightly better in the Maxwell version, although its kind of hard to tell through the grain. But I did manage to get mine pretty close, with the exception of a couple weird artifacts. And the caustics look pretty good too. I don't know what happened with the glass ball & its shadow at the top though - doesn't look right to me. For those of you who know Vray, this is 1st & 2nd bounce set at 1, and using the default color mapping values. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now