Have you enabled hardware graphics acceleration? >Windows>Preferences>OpenGL>Use hardware acceleration

@ Alan - hmm, I do this all the time. works for me without problems. Even with rendering. I'm not sure what the problem is. Am I missing something obvious?

@ Remus - thanks!It's great when you learn little tips like this even after years of using an application almost every day!

Quite extraordinary! Stunning work!

I have a suggestion for your next jaw-dropping model here:- http://thefutureofthings.com/upload/items_icons/ATLAS-cern_large.jpg

Perhaps there is a Ruby script to select everything on a particular layer. This will allow you to assign materials to faces really quickly.

Assigning materials by layer in theory is a really efficient way of doing things. If you create geometry on a specific layer, you don't need to assign materials, it is done automatically. This sounds great but the drawback is that you have no flexibility with texture positioning.

@burkhard said:

about 10 hours watching videos against 1 min posting and a big thank you

...but you're right this is a very basic question

Oh Burkhard - shame on you You haven't seen these cool videos before? I know that most people are too lazy to read, but come on, it's just like watching TV!

I guess this means that some people haven't seen all the cool stuff on YouTube either. There are so many things to improve our knowledge of SketchUp there, no matter how good we think you are.

Actually I don't think that Google will drop the 'Pro' thing. They have the same sort of arrangement with Google Earth. Also, lots of large corporations don't like to rely on free software for core functionality. I may be wrong, but I believe that the principle was that professionals would want export options for use with other applications - I'm thinking renderers in particular. Hobbyists and enthusiasts would just use the core modelling features, but professionals need interoperability with other apps.

It's obvious that there isn't enough difference between the free and pro versions to justify the price difference. Like Alan, I think the Pro version for SU7 will have a lot more features for the money rather than abandon the principle altogether.

Personally, if I were Google, I would just have retained the previous version as the free application, but just added support for the latest file format.

However, we won't have too long to wait for SU7 5 guess.

Have you seen the video tutorials? They are excellent and explain most things about using SketchUp. It might save you time posting.

I think the subdivide & smooth plugin would really help you with this. It's probably the easiest way of building complex organic shapes.

I use 'pipe along path' which gives you an option to set the number of segments for your cylinders. I often use just 4 or 5 if they are for something which isn't close to the camera.

IES is a b@stard to use. I worked with some consultants who modelled the environmental performance of one of my buildings with it. There are big problems with creating the building model though. Hopefully this plugin gives them a better way of doing that.

IES is also very slow in its analysis. AFAIK it is a tool aimed at engineers rather than architects and requires entry of some quite technical data, as it takes account of the proposed heating systems and lots of HVAC stuff.

The consultants I worked with have stopped using IES and replaced it with a system called Hevacomp. Hevacomp has been bought by Bentley, and is based on the EnergyPlus analysis engine, albeit heavily modified.

I have tried the EnergyPlus SketchUp beta. It kind of works, but this sort of thing isn't easy. I don't think IES is going to be your thing if you want to do energy analysis with SU.

I would be really interested to hear anyone else's views though.

[Edit: I just noticed that there is a free application called VE-Ware on the IES site which works with both the SketchUp and Revit plugins. Something to try out next week I think...]

The latest version of ArtLantis is easy to create animation with. It is also very fast if fantastic quality isn't what you are after.
I used it for years before Podium.

Damn, Craig beat me to it, and it kind of makes what I was going to post a little redundant, but it's probably worth posting anyway.

@pbacot said:

I just wrote a long reply and lost it due to some login glitch. grrrh. I can't bear to do it again. So I spare you ... in brief.

Other programs cost much more,have their own problems,and are less responsive to user; often dying a slow death due to lack of development, while someone (I surmise) is mining the company for their retirement. You don't know what "lazy"is.

Er, I work in local government, trust me, I do.

@pbacot said:

Where else is there something like the SU developers and community? Some much more expensive Pro softwares do not have an open macro system, do not respond to forums, do not have training, etc. .

Lots do though. This forum isn't unique. If you want to see what really flaming the software developers on their own forum is like, head on over to the Art*Lantis site.

@pbacot said:

SU gives a decent foothold into 3d for those who can ill spare the cost in time (which is money, I hear) to take on other offerings.

I know. I love SU. But let's face it Google is a massive company with huge resources. They seem to be making all their money without trying, and developing software that they generally give away. You have to admire their altruism, but nothing hones a product like competition in the market. If you are totally reliant on one of Google's products, the warm feeling of satisfaction due to goodwill will wear off if their software isn't doing what you want. The competition is moving ahead fast. Other products already have an equivalent of push-pull for example. Google needs to take on the opposition head-on and it can't afford to dither and take its time over basic stuff. You are absolutely right that time is money and if some of the competition becomes more efficient, if changes the cost balance - possible away from Google.

We are complaining precisely because we really want SketchUp to be the best 3d modeller out there. We see all these great plugins by people who don't earn a vast amount of money out of them, and the obvious question it raises is,"Why the hell aren't Google doing this?"

Let's see what they could do for example. What about a monthly/quarterly/annual competition for the best free SU plugin? And what about maybe giving a cash prize? Perhaps a trip to Boulder to meet the team? What about initiatives for the SketchUp team to release ruby plugins which add functionality? We aren't talking about @Last here, we are talking about one of the most powerful computer companies in the world. Where are the benefits we hoped Google would bring us? If you think I'm speaking out of turn - great. Actually I don't care because stimulating discussion which may just make Google push development harder is what we want.

Yes we all love SketchUp, and I guess most of us (me included) love Google. But let me illustrate with a metaphor. If you have a child who is performing well below the level of their capabilities, and you have done the softly-softly approach, what do you do? TO use a UK colloquialism - you stick a rocket up their ar*e.

Using colour by layer isn't the best way to assign textures. It isn't really compatible with most renderers for a start. Although it is very efficient for modelling, it doesn't allow any flexibility for positioning textures on surfaces at all.

It would be useful to have a Ruby plugin which reads these textures from layers and assigns them directly to faces.

@gaieus said:

@Least Google didn't drop them but developed them to their current stage.

Actually I'm not sure many tears would be shed if they dropped LayOut.

Hmm, interesting. The developer of another global illumination render engine using similar algorithms told me that Vray wasn't physically accurate.

From my understanding, any biased engine using FG sampling can't be 100% correct because it relies on interpolation between samples. The further apart the samples, the less accurate the solution will be. These solutions also rely on other algorithms to increase sample density in the corners and other high contrast areas. Although with lots of samples interpolation should give accurate results, unbiased algorithms don't (AFAIK) rely on clever maths to arrive at a solution which makes an educated guess at what is happening between sample areas.

However, I don't have a really detailed knowledge of this stuff, so may well be wrong.

I looked up 'Carmack's reverse' before I posted. Wikipedia has some information here:- http://en.wikipedia.org/wiki/Shadow_volume

There are a number of different, but apparently similar approaches. Proper shadows really must be made to work. Such a basic issue still in Version 7 (I'm assuming that it's not going to get fixed by then) of an application by one of the largest and richest companies in the world really isn't good. And they have a truly awesome amount of brainpower at their disposal.

I'm sorry if it seems like I am having a go at Google. I certainly wasn't sniping at Craig who was honest enough to front up about it.

I rely on lots of Google's products, and they are almost all free, which is fantastic. However as a company they are rich and powerful. When they bought @Last, we all expected big progress. What did we get? LayOut, Sandbox and support for Google Earth. Layout (which I believe was formerly known as Grizzly) was one of the killer features we were all looking forward to. Something that worked like SketchUp which would add 2d drawing and labelling capabilities to SketchUp so we could throw out our CAD packages. That's what we wanted, but what we got was something that fell way short of that. I don't know how much it has been improved, I stopped using it when my family thought I had developed Tourette's syndrome. Google Earth support is nice, but not something we were exactly clamouring for. The sandbox tools? Again nice, but not as impressive as Subdivide and Smooth or LightUp IMHO.

On the downside we still have the limitations with complex models and shadows.

To quote Spiderman ( ), "With great power comes great responsibility" and I'm not sure Google is taking enough.

That's obviously correct, but unbiased engines should always give you a greater degree of accuracy with less fiddling around. Biased renderers rely on optimising much less accurate rendering algorithms to balance speed and accuracy.

@craigd said:

but I can tell you it's not as simple as you think. We're not happy about it either, but it is not in our control.

I'll admit that obviously I don't know remotely as much about it as you do, but I suspect I have heard more lame excuses. From my (admittedly ignorant) standpoint it is quite easy. Google pays whoever the money to licence the relevant algorithm, or some of the mega-brains at the Googleplex develop an alternative or a slight modification/improvement which doesn't violate the patent.

Later this year you are scheduled to release SketchUp version 7, and Google is one of the richest and most influential companies in the whole world. You have a different set of rules - or at least you ought to. I thought this was supposed to be one of the benefits of Google buying out @Last.

Looking at some of the brilliant plugins here (UV mapping, freeform deformation, Subdivide & smooth) it makes me think that much better versions of these ought to be integrated into the next release. No offence to the guys who coded these - they are brilliant, but you develop the application, you really ought to be able to do it better. There is also the persistent problem with SketchUp's bility to handle large complex models well. To be honest it looks to me like you Google guys spend too much time on your slides, Segways and table football to do enough work. Yes, I know I shouldn't believe everything I read in the newspapers, but I would be delighted if you prove me wrong.

I am a massive SketchUp fan, but if I'm totally honest it looks to me like Google is getting corporately fat & lazy. I think it's fair to say that all users are expecting some really big advances for V7.

So no pressure then.

There are some good video tutorials on YouTube dealing with texturing that might help you. This (http://www.youtube.com/watch?v=NV0mLPQDcJ4&feature=user) is one. This (http://www.youtube.com/watch?v=HHsRRLCAOm8&feature=related)is another.

There are 2 fundamental issues here, speed and complexity/ease of use. Biased renderers like Vray are fast. Vray is very fast, and is the render engine of choice for lots of professionals. It's not quite 100% accurate, but it produces some great images anyway. Fry and Maxwell (and Luxrender) all use the Metropolis Light Transport algorithm to create physically accurate simulations of lighting, but this takes a lot of time - at least with the current level of hardware sophistication. In 5 years or so, we'll most likely all be using this technology for day to day rendering.

Vray is a great render engine, and can produce results as good as unbiased renderers. The drawback is that it's complicated. Very fast, but very complicated. If you want your images to be truly stunning, you have a steep learning curve. The unbiased renderers are easier and quicker to set up.

I don't think Kerkythea is in the same league as these other engines. I have tried it and looked at the gallery - a lot, and that remains my own personal opinion.