Wishlist SU 8... Let's not !
-
I honestly cannot think of one 'large scale enhancement' that would not require SU to perform better for it to work.
My point here is that Google needs to go back to basics and build a better foundation for SU, we want our favorite app to improve and enable us to do basic things that other apps do but we are restricted to an aged foundation that cannot carry the load of these new tools. So as much as we dream of cool new features we have to be realistic that SU cannot handle them.What bothers me more than the poly problem is Googles non communication policy, they have never discussed this burning issue ever and probably will never as they have no solutions.
-
Give it back to @last and let them run it. Google can lease it or something.
-
DacaD, you are demonstrating my point rather nicely. Aside from the game engine none of those ideas would bring any long term improvement. SU will not last any length of time if all they do is add 'features' like this. These are the sort of features i think should be implemented through ruby.
I am not arguing against better support for large models, it is entirely necessary. I'm just trying to point out that it is not the only thing we should be thinking about. SU needs big new features to keep it going, it needs stuff that is going to differentiate it from the other packages out there after theyve all finished copying the inference system and push/pull. this is the stuff i think the development team should be working on.
-
@solo said:
What bothers me more than the poly problem is Googles non communication policy, they have never discussed this burning issue ever and probably will never as they have no solutions.
Too true, i think improving communication between the developers and the community would ease matters a lot.
shakes fist at google policy
-
I suppose my question is - Why should Google care about Sketchup beyond the basic tie-ins to things like Google Earth. I can't imagine that the revenues from Pro are that substantial and they certainly take a lot of negative flack from the 'pros' who use it.
-
@solo said:
My point here is that Google needs to go back to basics and build a better foundation for SU, we want our favorite app to improve and enable us to do basic things that other apps do but we are restricted to an aged foundation that cannot carry the load of these new tools. So as much as we dream of cool new features we have to be realistic that SU cannot handle them.
What bothers me more than the poly problem is Googles non communication policy, they have never discussed this burning issue ever and probably will never as they have no solutions.
Quoted for agreement. The way development is going now, saying SU is starting to fall behind is something of an understatement.
sighs
@double espresso said:
I suppose my question is - Why should Google care about Sketchup beyond the basic tie-ins to things like Google Earth. I can't imagine that the revenues from Pro are that substantial and they certainly take a lot of negative flack from the 'pros' who use it.
I think you are right. They don't care - SU's just another marketing scheme to them. I mean, they certainly have the resources to be able to compete with the one or two man dev teams out there, now don't they? Anyone remember the vids they used to announce the release of 7? QED.
-
@jmmad said:
3.- Concentrate efforts to take the 3D to the web once and for all.
I absolutely agree; here is my take on that:
Rather than starting with a wishlist, I suggest we try to persuade Google that Sketchup should be a major element in linking human activities via the web; and that both the design and promotion of SU 8 should reflect this.
My summary conclusion is that producers should provide accurate Sketchup models of their products and present them, not in a warehouse, but directly in the user interface. Models are promoted in a file sharing type network, each with its own data formatted for comparative analyses. Users call up previews, download alternative products and assemble selections in their own models.
I believe this plays to three key elements of the web - searching, advertising and, not yet really exploited, data sharing - and, technically and commercially, works as well for water taps as for major developments.
Chris
-
Google can fix that stupid bug when zooming in very close to object.
Better dwg export. Offten lines are missing and some unneeded lines are showing up.And one more thing: when a plane is created and under this plane you place some object (really close but not connected to plane) then when zooming out that object beyond that plane shows up.
Later I will add some screenshot to better understand about what I'm talking.
Thist ugly bug was presented in SU6 and still presenting in SU7 like all other bugs!
Nothing was fixed!PS. Remus I'm not talking about new tools I'm talking about fixing.
-
I hear some of you guys wishing for a good 3D web implementation.
I am not against that, but I am afraid that if it would become the full focus of the SU dev team, we can forget about high poly support and a smoother core all together.
Also, I like Lineas suggestion that Google needs to be more on its tows and watch its own market instead of staring at the belly button and just do what they think would be cool to impress their bosses.
Having a 'market watcher' as that Ubuntu guy would be a great staff member. At least than we can expect things people really care for in upcoming versions.As it is now, no one can really be happy with how things are going.
-
Solo said
@unknownuser said:I say fix the core so that it supports higher polys then open source it and let the scripters loose.
I agree. I think the problem is Google don't seem to understand who uses SU and where to develop. If their only reference point is 3d warehouse then we're all screwed. Open source SU as soon as possible.
This doesn't necessarily mean Google are totally out of the loop. Ubuntu (linux) employ a "community manager" (he's called Jono Bacon http://www.jonobacon.org/) whose sole job is to report back to Ubuntu about scripts and plug-ins and needs in the community and recommend that they are integrated in the core product. Because of this Ubuntu are able to release a new version every 6 months.Chris said
@unknownuser said:My summary conclusion is that producers should provide accurate Sketchup models of their products and present them, not in a warehouse, but directly in the user interface. Models are promoted in a file sharing type network, each with its own data formatted for comparative analyses. Users call up previews, download alternative products and assemble selections in their own models
Maybe I'm shortsighted but is all this virtual product stuff really necessary? SU is going to need very high end rendering capabilities to satisfy the marketing reps. For the architecture industry I think SU have already missed the boat, or probably were never on the same boat as apps like Revit that now have repositries of building products.
But for consumer products I can't see SU being instrumental in this way. If people want something, they buy it. We like everything in 3d, but I just can't see most people orbiting vr models of everything or building a 3d room to place that lamp before buying. When 3d printing really takes off and you can download and print a product at home, then maybe SU has a place, but high poly support will be essential. Google could waste alot of time on this and it could amount to about as much use as Lively.Espresso wrote
@unknownuser said:
I can't imagine that the revenues from Pro are that substantial and they certainly take a lot of negative flack from the 'pros' who use it.
It's worth remembering that SU wasn't always free and would never have got anywhere if people didn't buy it. I don't imagine Google are exempt from the current financial crisis. If it's a question of whether or not Google need the money, then I think they should let us know. I'm still happy to pay for the tool that I use most at work but I won't bother if there is no point.
This comes back to the open source suggestion. Alot of open-source projects give their product away but make money by selling educational materials, providing training and hosting events. So Google could still turn a profit. Donations are also an important revenue for o.s. ventures, if people care enough, maybe this approach is better than Pro licensing. -
@linea said:
Maybe I'm shortsighted but is all this virtual product stuff really necessary?
Well maybe not short sighted but please consider this:
As far as I know designers have always made "virtual products", it is just that they call them drawings, sketches, models and so forth. All I'm suggesting is that instead of the designer making a model of a door, manufacturers' make their own versions and each offers it directly in the designer's interface, most importantly when requested (designer clicks Find). For this to work the versions have to be compatible. Sketchup provides a simple method for them to be so with very few rules.
@unknownuser said:
Advertising and media agencies will be all over this straight away, then SU is going to need very high end rendering capabilities to satisfy the marketing reps selling these products.
No I don't think this is true. More likely rather than use traditional agencies, manufacturers will look directly to expert modellers to provide models suitable for intended purpose. For general design and construction purposes they would likely be raw and light with emphasis on supporting data; for presentations, yes, probably high end rendering.
@unknownuser said:
For the architecture industry I think SU have already missed the boat, probably were never on the same boat as apps like Revit that let you integrate building products.
Absolutely not - there are very few manufacturers doing this; there has never yet been a comprehensive move for manufacturers to be brought in to the design process as early as possible, yet they contribute an ever increasing amount of design input. I think the aim is to persuade building owners to require their designers use accurate models of real products. Then the building industry will then have a chance to act like a proper industry.(Egon-esque!)
@unknownuser said:
But for consumer products I can't see SU being instrumental in this way.
I think, with respect, separating out consumer products is irrelevant, simply because we live, work and shop in buildings, and whatever we consume should be accommodated by their design. (e.g. how many Barbie dolls fit on a supermarket shelf?)
What I think is better than having wishlists for improving SU is to suggest other ways for SU to fit in with what seems to be Google philosophy. (i.e aiming higher than the SU team)
Chris
-
@chrisglasier said:
What I think is better than having wishlists for improving SU is to suggest other ways for SU to fit in with what seems to be Google philosophy. (i.e aiming higher than the SU team)
ChrisI tried to avoid having a 'classic' wishlist here, hence my first post.
Maybe I need to change the thread title. -
@kwistenbiebel said:
Did you read the first post in the thread?
Maybe I need to change the thread title.I did...
@kwistenbiebel said:
A wishlist for SU8? Nah.....
I don't think people want to get into a wishlist for SU 8 as making one for SU 7 turned out to be a useless endeavour...
Well, parametric modeling was one of the biggest feature requests for SU. And it_'s here. Note that I could live without it - it will probably be useful for manufacturers, not me, but you cannot say ths was not listened to.
As for the poll - yes, go ahead and change the topic title as it doesn't have too much to do with the poll itself.
-
@kwistenbiebel said:
I tried to avoid having a 'classic' wishlist here, hence my first post.
Maybe I need to change the thread title.Sorry I was responding to: "The question that pops up is how you guys see the future for Sketchup?" that I read in the first post, and it seemed necessary to add the wishlists bit for balance.
Chris
-
@unknownuser said:
there has never yet been a comprehensive move for manufacturers to be brought in to the design process as early as possible,
Depends where you work, I know plenty of design practices (ourselves included) that get potential manufacturers in straight away.
@unknownuser said:
As far as I know designers have always made "virtual products", ... Sketchup provides a simple method for them to be so with very few rules
Good point, sorry I think I misinterpreted what you meant.
@unknownuser said:
I think, with respect, separating out consumer products is irrelevant, simply because we live, work and shop in buildings, and whatever we consume should be accommodated by their design
I agree with that too, but aren't there more tangible, immediate things that google could do with SU? -
@linea said:
@unknownuser said:
there has never yet been a comprehensive move for manufacturers to be brought in to the design process as early as possible,
Depends where you work, I know plenty of design practices (ourselves included) that get potential manufacturers in straight away.
There is a difference betweeen "comprehensive move" and "getting them in", and it's not the fault of architects. I have discussed this with many in various parts of the world, and it nearly always comes down to old-fashioned notions of accountability. Here is a couple of diagrams we made years ago to illustrate the difference between traditional competitive tendering and potential competitive collaboration.
@unknownuser said:
... aren't there more tangible, immediate things that google could do with SU?
Use it to promote models as the basis for data linking infrastructure perhaps ?
My regards
-
good points Chris, I concede. :
-
Hi Aidus, hi folks.
@aidus said:
And one more thing: when a plane is created and under this plane you place some object (really close but not connected to plane) then when zooming out that object beyond that plane shows up.
What you describe is probably not SU's fault but more a limitation of Open GL. When you zoom out of such a plane, the distance between the plane and the object behind becomes very small compared to the distance from these to the observer. In fact, the difference in distance from the object to the observer and from the plane to the observer becomes so close to zero that Open GL sees it as zero and thus, shows both.
Just ideas.
-
For me, Google simply needs to focus more on 'core' features. Fast, accurate modelling which will work smoothly with large and complex (or even tiny and complex) models, with a wide enough range of modelling tools to suit all but the most demanding users. Silver shadow, solo and others have shown us that you can model more or less anything with SU with the right plugins, determination and skill.
The only ways the core can really be improved are (IMO) :-
- Better manipulation of complex geometry
- Fix Shadows
- More options for working with curved surfaces - Loft tool please!
- More options for texture mapping
- More options for tighter integration with 3rd party plugins - Python support sounds cool, and not too difficult
- More import/export formats - the real key to broader acceptance and use
Things like Style Builder I think are a distraction. Not many people really wanted this. Yes it might be nice for some, but I never use them at all. They are the sort of things that really ought to be worked on when the core is really well-tuned.
I would quite like to see some extensions to DC as well, and for them to be coded in C not Ruby. I'm having difficulty using multiple nested arrays inside components, and would like the option to substitute components based on formula values.
With a lot of the recent criticism, it is easy to forget how good SU still is. Because it is so easy, many of us are pushing it beyond its current capabilities in terms of 3d manipulation. This is a good thing and an opportunity and I'm sure Google realises this - in its own way.
I do want this stuff to happen in the next 6 months though, or I'm going to start looking at MoI. That doesn't mean I'll necessarily switch, but that my requirements and abilities have progressed a little and that SU isn't perhaps the ideal tool for most of my 3d work any more.
-
I think we all secretly know that it is getting to the end of the line.
It is obvious that they just can't get SU to handle the mount of poly's that the future will demand.
Z-Brush can handle billions of poly's while SU can't handle 100,000. And that is now! What about 1 or 2 years from now.
With the absolutely disappointing release of ver7 (I have never had this many program crashes in any software ever!) it put the writing on the wall for me.
Handling poly's is what makes a 3d software. Period. That's what these softwares do. Handle poly's. SU can't and that will be the death knell.
I am forcing myself to use other softwares where I used to use SU. If it wasn't for the fantastic Ruby scripts that have come down the pipe, I think I would be done with SU for everything but rudimentary architectural work, which stinks, because, as we all know, there is no more intuitive and fun to use box modeler.
But, if they can't improve the core code, which seems apparent that they can't, then SU will be relegated to nothing more than a toy.
Advertisement