Maxwell Render 1.5

[IC]

I really don't like the stock comparison, because stocks are bought and sold.

 

I can't sell my Maxwell licenses. And they really aren't worth much to me otherwise. Oh, and I bought them under the assumption that NL's promises were worth something, which they weren't.

 

So, closest investing comparison I can make is to Enron.

 

It is a tenuous comparison but I think the point is that an investment can pay off or not.

With software I don't think you would ever really be hoping to ever sell it on again, particularly for profit, but in terms of productivity, you will feel a loss or a gain.

[3dworks]

Yeah, and fireflies that are attracted to glass... are you really able to use 1.5 like that?

 

I ask with genuine interest, because the OSX fireflies and 'fleas' are the problem that is keeping me from trying 1.5 out.

 

yeah, that's an unpleasant release bug. at least, with the latest update (1.5.0.1) the problems seem to appear only in scenes with dielectric materials in combination with emitters - sky with sun should work OK. a final fix will be out very soon - they are working hard to find the compiling problem that broke it in the very last builds just before delivery.

 

markus

[AJLynn]

I'm going to sound like a broken record here, but a full-time or at least half-time release engineer would make their jobs soooo much easier...

[Devin Johnston]

I've been using Maxwell for production work since the Beat came out, it was limited at first but with the release of V1.1 I've been able to make Maxwell my primary render engine. The quality it is capable of producing is amazing, clients love it and even though there have been more setbacks with it than successes it is a usable and valuable tool. I had the same hopes for V1.5 and against better judgment I installed it on my network the day it came out. Unfortunately there are several things wrong with network rendering as well as the physical sky which have caused me to go back to V1.1. In my opinion this release simply wasn't tested thoroughly enough and as a result many of the bugs which could have easily been discovered were missed. You may say that you've heard all this before, and unfortunately you have because this is what happens in almost every one of their releases. So the fact that we keep talking about the same things really has nothing to do with wanting to bash NL, it a symptom of a bigger problem that has everything to do with NL's practices as a company.

[Paul Griger]

I have used M~R for some portfolio work and was happy with it once I figured my way around [plugin for ArchiCAD]. Some workflow bugs need to be worked out that I think could present challenges for a professional, but for a student portfolio, with me as [a more forgiving] client, it did a great job.

 

I'll try 1.5 [i used the previous version, 1.1] once others tame 1.5

 

Cheers

[leoA4D]

Updated MWR plug-in for Cinema (07.05.31) is out and the MWR/MAC app update is available (07.06.01).

 

Re MWR/Mac app update:

Hi everyone,

 

An update for Maxwell for Mac OSX is now available from the usual download site. This update solves various specific OSX bugs in the interface and rendering:

 

- Issue with numeric inputs in Maxwell Studio and Material Editor fixed.

- Dots ("fireflies") in the render have been fixed in OSX Intel version.

- Hang when rendering some specific materials has been fixed in OSX Intel version.

 

Thanks,

 

Nicole

[superbad]

In their defense, I do have to say they have become better at chasing down all the bugs that pop up after a release. We've had two OSX patches, and several plugin patches since the release. I still think they could do a better job of pre-release testing though.

[dave davidson]

I still think they could do a better job of pre-release testing though.

quoted for agreement.

[leoA4D]

Also, there are at least four NL people (Juan, beatriz, nachob, nicole) actively answering questions in the threads. A help question to NL got two exchanges within 24 hours, too.

[Adehus]

Also, there are at least four NL people (Juan, beatriz, nachob, nicole) actively answering questions in the threads. A help question to NL got two exchanges within 24 hours, too.

 

That's a refreshing change. If it continues even *after* the user scrutiny of 1.5 wears off, I'll even concede that it's an improvement.

[leoA4D]

That's a refreshing change. If it continues even *after* the user scrutiny of 1.5 wears off, I'll even concede that it's an improvement.

 

Quoted for agreement.

[RyanSpaulding]

Today was the 1st day I went into 1.5...

 

What in the holy hell happened to Maxwell? Seriously. The interface is worse, quality is worse, ect.

[Devin Johnston]

How long has it been since you used it?

[3dworks]

Today was the 1st day I went into 1.5...

 

What in the holy hell happened to Maxwell? Seriously. The interface is worse, quality is worse, ect.

 

maybe could you be a bit more specific about what exactly in the interface is worse and show us an example which demonstrates that quality (do you mean render quality?) is worse in your opinion?

 

markus

[RyanSpaulding]

How long has it been since you used it?

 

The 1.1 days. So the interface was a lot different and I'm finding that the AA isn't as good as it was. Doesn't off-hand seem to be faster...but maybe it is. Also, when I use the physical sky...there's some pink in there.

 

*looks outside*

 

No pink to be found.

[Devin Johnston]

It's not faster especially if you’re using the 64 bit version; the pink sky is a big problem for me and many others. NL says it's physically correct but no one is buying that, apparently they are working on a new and better sky but God knows when it will be available. You must be using Studio; I haven't messed with it so I don't know if it's better or worse in 1.5.

[RyanSpaulding]

Yeah, MicroStation has an exporter...not a plugin. Most everything comes over fine...I just have to tweak in studio.

 

So they're saying that pink...is based in reality? REALLY?!? How can they deny error when all we have to do to verify is peek outside???

[Devin Johnston]

That's exactly what I've said to them, in fact there is an active thread right now discussing this very topic. There are several people including an official 'beta tester" who are saying right now that the pink sky is physically accurate even though it may not look like it. When asked for an explanation how this kind of a glaring error could have made it into a release they have no other explanation other than to say it's physically accurate. I've asked for them to re-implement the old sky to which they say if they do that then people who have been using the pink sky to do work will be inconvenienced and will have to rework their scenes. When I point out that everyone who has done a Maxwell scene with a physical sky up to V1.5 can't work on their projects without reworking them and once they fix the pink sky people who used it to do work will have to rework their projects, they just say to bad.

 

This says to me that they don't consider people doing Arch Viz as a priority, and they don't test any of their software before they release it.

[AJLynn]

Wow. That's pretty impressive even by the established standards.

[Sawyer]

I got this update a few weeks ago I thought it was a strange way to word this:

 

Sky:

 

- This patch does not include an update for the Maxwell Render sky model. The current sky is not just a matter of rightness and physical correctness, but subject to personal taste. Many clients need this patch to fix some critical issues while working on images, but they do not want to change their images and need the sky to stay the same. Thus we have refrained from a radical change in the look& feel of the software in this small update. The Maxwell Render development team is presently working on a radically improved physical sky.

 

Not just a matter of "rightness" and "correctness".

[Devin Johnston]

The argument is so ridiculous I don't see how they can make it without laughing. You'll notice that they don't even think about all the people who have been using Maxwell for years and are now inconvenienced because none of their old models look like they should. It's simply a matter of incompetence on their part and as soon as they saw that people didn't like the new sky they should have issued an immediate patch to fix the problem.

[AJLynn]

Technically, I think everybody is correct in this. On the one side, people wanting their renders to look good - and usually the pink does not. On the other, the NL people who have researched Siggraph papers and put this physical model in the software. (I think, but don't know, that they're using the Preetham model, which is simplified here.) Notice that in the graphics in the descriptions of the Preetham model, there's always that pink. I've got a bunch of thesis images rendered in fR2 for C4D that have the same pink...

 

which often is actually visible in photos taken on mountains. Do a search for mountain on hxc.hu and you'll see a lot of it. The problem is that the Preetham model exaggerates the effect relative to what we'd see with the eye or camera, and that usually we're not rendering scenes on top of mountains. Shots that demonstrate Preetham usually use a camera at high altitude looking out at an unobstructed horizon, like you'd see standing on a mountain, like what most environments in our models are actually simulating without obstructions on the horizon... Just a thought but I think there are a lot of steps in this process where anyone involved in the whole thing can make an error that results in a pink sky while thinking that what they've done is perfectly correct.

[Devin Johnston]

I'm not even saying that the physical model in V1.5 isn't "physically correct" as they clame, what I'm saying is it doesn’t look correct. Certainly the client who sees one of these pink sky renderings isn't going to understand or care about the Preetham model, all they know is when they go outside during most of the day the sky is blue not pink. I'm not trying to argue with you, I'm just saying what is the point of creating a physically accurate sky if it doesn’t look right.

 

You know what’s funny AJLynn, you just explained why the sky is pink and it’s perfectly logical and I can understand why it might have been done. We have been asking NL to explain it for weeks and no one has even come close to explaining it as well as you did. Maybe you should offer to consult for them.

[AJLynn]

I still think they need to change it and make it less pink

[Fran]

In the meantime, while they are "busy" re-inventing a revolutionary new Sky™, there are a lot of users who just need a simple patch to make productive use of their product. Clients who like the pink sky don't need to apply the patch. It's so simple. But's it's a well-known fact that simple things can't be any good.