Posts made by brodie

I'd start by downloading something like a fence DC which copies the fence posts as you scale the fence up. That'll show you how to fix the size of the fence posts and how to determine how far apart the posts are from each other. You should be able to then translate that into what you're doing where the 'fence posts' are 1cmx1cm squares and the number of copies is the total length divided by 1cm minus 1.

So if the total length is 10 cm the equation would be calculated 10/1 - 1 = 9 giving you your 1 original square and 9 copies.

-Brodie

Great job Whaat. I've only used it a bit but the Loop tool alone is worth the cost of admission. I've tried some alternatives with no success.

And the Tube mapping works brilliantly. Absolutely on my 'must have' list. As far as I can tell this pretty much solves SU's lack of UV functionality. And what's more, it seems to work in a very SU like way, very intuitive, just a couple buttons and pretty easy to figure out after skimming through the youtube tutorials.

I can't recommend it highly enough. Let me know if you ever need a kidney.

-Brodie

In that case you may be able to accomplish your goal with geometry rather than textures. Instead of a 1cm x 1cm texture maybe you could use 1cm x 1cm components. You could fix their scale in the DC options to 1cm x 1cm and then have add a COPY option to the component so that when you scale the overall object to, say, 50cm long, it creates 50 of those 1cm components. The only drawback would be that you'd have to limit the scaling options to 1cm increments. Does that make sense?

-Brodie

I, for one, welcome our new overlords.

From an ArchViz perspective I think many of us here seem to have forgotten the frustrations of Google's ownership. We'll never really know how much control Google had over SU's development but since Google's acquisition, has not each release been fairly underwhelming? Some nice tools have been added: solids, dynamic components,...I'm sure there must be others, but over the last several years I don't remember ever seeing an update and being really excited to try out a new feature.

The Warehouse has been nice and I hope it continues in some form, but it too has hardly developed since it's inception although it has a lot of room for improvement.

Google used SU to build there Google Earth empire and pushed its development to that end. But that was THEIR goal, and it rarely aligned with many of ours.

I know little of Trimble but the speculation I'm seeing suggests they purchased SU for somewhere between $50-$90 million. That's a pretty significant investment so you have to assume they have some plans to develop it further. I also hear they are somewhat involved in BIM which is something I see a lot of requests for around here.

-Brodie

I've run into the same issue and I don't believe there's a good fix for this. When you scale an object, the texture scales too and that's all there is to it. If your end goal is rendering some rendering software may be able to help you out, but nothing within SU itself.

I'd like to see some sort of plugin fix for this perhaps that could go into all groups and reapply the materials to fix the stretching but I don't know of such a plugin

-Brodie

As others have mentioned, Alienware is overpriced. If you're a gamer without much of a budget, you don't want to build your own PC, and you like your PC to have lots of lights, it's the way to go. Otherwise, choose another option. You can build your own, spec out a dell, etc. If you don't see the options you like you might also look into CyberPowerPC. I ordered a machine from them for someone and it seemed to go well and they provide lots of options at a reasonable price. Like Alienware they focus on the gamer market but you have a lot more options. It's much more like building your own PC but you don't have to fret so much over compatibility, how to install a motherboard, and you get a warranty. I built a ShuttleBox PC awhile back. Also a good experience but the small case sort of limits your options a bit and can get kindof warm if you're doing renderings or intensive CPU processing.

-Brodie

Fraid so. It's not really SU's fault exactly. It's an OpenGL limitation. You run into the same issues with 3ds Max's viewport for example, it's just that SU tries to do a lot more in it's viewport than Max so the limitations become more apparent. However, the i7's do have a good turbo feature which boosts power to single processor programs so you are getting some benefit. Vray, should take full advantage of your processors so you might try looking for a Vray benchmark scene just to make sure your render times are in line with other folks with your CPU.

-Brodie

This test used to work quite well as a comparison but was completely broken after an update some time ago. It's no longer good for any sort of comparison.

-Brodie

It's not uncommon given SU's limitations that you might not notice a lot of difference. SU doesn't use much of your graphics card and only 1 processor so there's a lot of untapped potential there. However you should be seeing an improvement with rendering software. What software are you using and what processor did you have before?

-Brodie

As I recall the intermediate 3dats book has a good template.

http://www.thecgschool.com/cgbooks/

-Brodie

Very cool, thanks for sharing. 2nd place is certainly noteworthy and the renderings look very good. Did you have a chance to see who got first place? Any idea what they did better?

-Brodie

Groups should work just fine. I think I (and perhaps most people) just get used to making everything a component. I've never found much use for groups personally.

-Brodie

I've got a table that I need to be able to stretch in 30" increments. However, the wood texture applied can't get deformed with it so there's a problem.

What are my options? Can I have the DC contain a table with proper UV's which gets popped in depending on how large the user scales it? Is there a plugin that will go back through and reapply UV's to components that have been scaled? Any other options?

-Brodie

@cheneymax said:

On another note...what would be the best forum to post a Maxwell for Sketchup question (It's another error )

Thanks again,
Max

I think Jason and myself both scan the extensions forum on a regular basis for maxwell questions

-Brodie

Aww, gotcha. Yes they do seem to be very similar so I think you're on the right track.

-Brodie

Oh, sorry, I forgot you were using the standalone plugin. The script only works with the full version. Sorry about that.

For an inexpensive solution you might look into Podium. You'll sacrifice quality for sure but it's dirt cheap and super easy to use. Make sure to see how it's working with animations though. A year or two back it had some bugs but those are probably worked out. You could also look into kerkythea if it's still around or twilight. Both good renderers that might help you out on the animation front. Kerky was free last I checked and I think twilight was very reasonably priced and had a lot of the same features as maxwell (which may mean it has the same animation limitations but it's worth looking into).

-Brodie

There is a workaround to create animations from Sketchup. There are plugins which will allow you to create a scene for each frame of a SU animation. When you export that to Maxwell you'll end up with a camera for each of those scenes. JD, the writer of the Maxwell plugin created a script that would make a rendering from each of those cameras essentially allowing SU users to do animations. This, I believe is the correct script if you're interested. http://min.us/mF91PT5wN

However, just because you CAN do something doesn't mean you SHOULD. In practice Maxwell isn't a good software for animations at this point. Even generating low resolution renders can be too slow when you need to render 1000+ of them. I've done it for an exterior animation before, but it's not a great solution.

I never found an ideal solution. I'd planned on going with 3ds Max and Vray but never got a chance to try out many other options and both of those programs have nasty learning curves. However they're the industry standard for a reason (flexibility, speed, customization, capability, etc.). If I were spending my own money, I'd take a good hard look into Modo. I've seen some very nice renderings from it, it seems fairly speedy, and I've heard it fun to model in.

-Brodie

Maxwell's motto is something like "It's as easy as taking a photo." It's quite true, but the unspoken caveat is that, as you probably know, taking a GOOD photo isn't easy at all. And if you gave an amateur photographer the ability to completely adjust not only every camera parameter but also every ENVIRONMENTAL and lighting parameter? Things could get crazy fast.

Well I've got a few suggestions.

1. Put the environment setting back to default. I almost never change them and when I do my changes are quite minor. You always want to start with the most realistic parameters you can. By adjusting your sun power to 10 times it's normal brightness and the intensity to 7 times it's normal intensity you're creating a world with a sun which is WAY brighter than it would ever be on earth. That's going to screw up all of your other lighting parameters (you'll have to adjust your camera to be way less sensitive to light - a higher exposure value EV - and your interior lights to be way brighter than normal to balance everything out). Also, Maxwell tends to give more weight to brighter lights. So if you have a super bright light and a bunch of regular lights the regular lights won't become very noticeable until higher SL's. By making your Sun so bright you're insuring that you won't see what it is your interior lights are doing for a longer period of time.

2. Your camera settings aren't typical for an interior scene. Your EV is determined by a combo of your shutter speed, ISO, and f-stop. 9 times out of 10 I don't worry about what the specific shutter, ISO and f-stop values as Maxwell lets you input the EV directly and will adjust the other values to match. Based on my earlier rule you want to find a realistic EV for your scene. Currently you're at 14. If you look at the chart about half way down here http://en.wikipedia.org/wiki/Exposure_value you'll see that this is a common value for an exterior scene. For Home Interiors it should be more like 5-7. This will compensate for bringing your exterior environment to a reasonable level and as long as your interior lights are also realistic.

3. Materials - your glossy materials can take longer to clear up that noise. I think your floor looks good but your walls and ceiling should have a roughness of about 97 probably which will help with the noise. Also make sure that white ceiling isn't whiter than about 230,230,230. White materials take longer to clear up and in the real world no object is really whiter than about 230ish so going higher will mean more noise and less realism. You're better off adjusting the levels a bit in photoshop if you want a more pure white look - same as a photographer would do.

4. Make sure you either don't use glass for the glass panes or use AGS glass. When light passes through real glass the light rays are refracted and the calculations become much more difficult. Lighting a whole room with refracted light means super slow render times.

5. An underlit space will take longer to render. To fill your room with light Maxwell has to bounce the light all over the room for a long time and each bounce adds complexity to the calculations (ie. more noise). If you add some interior lights which will provide a lot more direct light to the scene you should see faster renders.

That should give you a good starting point. Keep in mind that the EV of 5-7 assumes you've got some interior lights on. If you have no interior lights it may be realistic to have a value more in the 4 neighborhood. If you start with that and tweak it a bit you'll be in good shape.

Getting back to what's right for you you'll notice that shaderlight was faster and easier and looks nice and clean. But if you look through the noise you currently have you'll notice that the Maxwell rendering is more realistic. If you're happy enough with shaderlight's output then it or a similar software may serve you quite well. But if you crave the realism Maxwell provides you may find it impossible to achieve with shaderlight.

-Brodie

Is there a question in there somewhere?

-Brodie

In general Jason's tutorials are the defacto learning resource but I'm not sure there's any one video in which he completely tackles your specific question.

http://www.vtc.com/modules/products/titleDetails.php?sku=34170&affiliate=halflifecomics

There's also the Maxwell resources tutorials which again deal a bit more generally with the issues.

http://support.nextlimit.com/display/tuts/Home/

The key to understanding Maxwell is understanding that it does not cheat - at all. The upside is the reality you can achieve. However, it can be frustrating at times. You've probably seen plenty of beautiful Maxwell interiors so obviously it's not a limitation in the software. So as you're solving issues like this you want to think of it as if you've just taken a photo of a room which turned out to be too dark. What could it be that caused this? In broad strokes it's either going to be a lack of light in the room or camera settings. If you're taking a photo of a room with the lights off it'll be really difficult to get a good picture, likewise if there's only 1 light source or all of your lights are really dim. Alternatively if your camera is set with an EV of 13 or 14 you'll end up with similar issues as it should be closer to 7ish.

There are more technical rendering quarks as well. If your scene is not to scale it could cause issues. If all of your lighting is being bounced off of reflective surfaces it could take forever for the lighting calculations to resolve, etc.

If you post some images of what you've got it would be easier to tell what may be going on - or even posting a MXS file.

As to your other question, it's hard to say whether or not Maxwell is right for you. I really like the software but admittedly it's not for everyone. I think the most fundamental questions come down to where you fall on the scales of needing photorealistic results, how much time you have to learn a software, and how much time you have for individual projects. Maxwell can produce highly photorealistic images with a fairly small learning curve but renderings take longer to resolve. Vray (or similar software) CAN produce photorealistic results but only after a steep learning curve and the renderings won't take as long.

-Brodie