The Global ARCHICAD Community

Stay informed. Get help. Share your knowledge.

Discussions closely related to ARCHICAD. (Example: Do we need a Linux version of ARCHICAD?)

Moderators: Barry Kelly, Karl Ottenstein, LaszloNagy, ejrolon, gkmethy, Csaba Kézér, mtron

#306944
bouhmidage wrote:
Thu Feb 13, 2020 10:00 pm
Another point, Free objects and plugins will streghthen archicad position in the market, revit plugins and objects are almost free, in the other hand, archicad has usefull and good plugins, but with extra cost...
Software development services are like architectural services - they require training, experience and professionalism. There is no such thing as free software, just like there are no free architectural/engineering services. It might be invisible to the end user, but someone always pays.

The cost to an end user is often determined by the market size. A large paying market means lower cost. The market for architectural software is already very small (as a whole) and the ARCHICAD-using segment is much smaller again. That means there are fewer customers to cover the cost, hence the cost will be higher. Again, someone foots the bill – end users, GRAPHISOFT, manufacturers etc.
#306946
jl_lt wrote:If this was an excel spreadsheet, there would be many circular references alerts by now :lol:
And here is one more. :wink:

jl_lt wrote:-As i read in an article the other day (cant find it again), a lot, if not most, of Revits manufaturer content we see isnt even made in Revit. Its done in inventor or other modeling software for industrial design, which is where the objects are originally designed and manufactured via CAAD/CAM. As Autodesk owns most of these software, the translation to Revit is easier if not automatic i might guess. Trying to compete with this process would seem nonsensical from a Graphisoft point of view.

Sure, but Archicad can import almost any 3d model format. It is not a problem of having the geometry. The problem is to make it parametric inside Archicad and there is when, for example, offset modifiers and other similar options could make the difference. To make the process easy and automatic, Archicad could have an option to link a morph object to an external file (with all 3d formats). If the manufacturer modify the product in their native CAD/CAM software, they would just have to resave the morph linked file. Then inside Archicad we would add parameters (offset modifiers, BM, Surfaces, etc...) to set, for example, commercial sizes, color, finish options. All product iterations would be automatically update in the Archicad file.
#306947
Braza wrote:
Fri Feb 14, 2020 12:27 pm
The problem is to make it parametric inside Archicad and there is when, for example, offset modifiers and other similar options could make the difference. To make the process easy and automatic, Archicad could have an option to link a morph object to an external file (with all 3d formats). If the manufacturer modify the product in their native CAD/CAM software, they would just have to resave the morph linked file. Then inside Archicad we would add parameters (offset modifiers, BM, Surfaces, etc...) to set, for example, commercial sizes, color, finish options. All product iterations would be automatically update in the Archicad file.
Aren't we trying to over complicate things here. Ok parametric is great for core components like windows & doors, which GS cover quite well and can be manufactured to suit. BUT if I want a manufacturer's specific bathtub shouldn't it be a fixed object? I mean I would not expect to ring up a bath manufacturer and say "Your Aqua-tub is 1700mm long but I need a 1735mm one to fit my drawing can you make one?". If a manufacturer offers CAD components in a range of sizes & finishes then each one should be unique inline with their product codes. Having third parties redefine a component's parameters can only serve to introduce errors. I do think there is an over expectation on provision of components, you only have to consider the LOD (Level Of Detail) issue to appreciate how quickly the geometric data of an object becomes irrelevant and the BIM data becomes inversely valuable. The only point at which truly accurate geometric component data has any value is in client detailed presentation & rendering. In many cases generic objects will readily fulfil that need. You need to ask what you are trying to deliver, if a simple morph with appropriate BIM data can represent the required component then why worry that My Component Co, doesn't have a 3D parametric CAD file for their widget?
#306951
And hence the proposition of giving Archicad GDL objects the ability to absorb the parameters from a global database of manufacuters dimensions and specifications input directly from their catalog database. The object then adotps the specifications and nominal dimensions of any given real world object. Its not represented realistically as in a super detailed model, but it should be enough for many projects and scopes. But me sez, lets move on! as the absence of a GDL GUI is far from being the only problem that needs to be addressed to achieve bigger market share
Last edited by jl_lt on Fri Feb 14, 2020 7:06 pm, edited 1 time in total.
#306965
@DGSketcher

When I suggest a morph object linked to an external 3d file, I was talking more to that kind of product that has a basic geometry that can vary in standard dimensions. i.e.: A fiberglass shower base that has a series of standard sizes... If you use an offset modifier over a linked morph, you don't have to create a 3d geometry/object for each size. And if the manufacturer, lets say, changed an edge from sharp to round, all he (or whomever were working to him) would have to do was update the 3d linked file. It is just an isolated raw idea to improve market share.

But of course much more has to be done previously.
#306987
A bigger market share? Here's my short list:

1. Get the people at Maxon to recode the 3D interface and make it smooth and seamless like Cinema 4D, even at high poly counts. The fact that if you create a complex profile wall for a more realistic look, the interface slows to a grinding halt, is inexcusable. I am considering purchasing a $1500 video card just to get the redraw speed faster. I have 0 issues in Cinema 4D with very complex scenes.

2. Purchase (or at least partner) with the people at Enscape and create a Mac version of the software as well, so that we can have dual screens setup and see our projects in semi-photo realistic 3D. Sure, Twinmotion sync is great, but the software is severely lacking in shadows and realism. Plus, you have to re-sync every time you modify anything. Enscape is real-time and works brilliantly. Showing clients their projects in the regular 3D window is.... well.... embarrassing.

3. Partner with ChaosGroup and give us the ability to use Vray inside Archicad. Further, make the materials seamless between the regular 3D window and the Vray rendering engine (i.e. you create a Vray material and it automatically knows what to show in the regular 3D window instead of having to manually create two different materials and hope for the best).

4. Include a much stronger and realistic set of surfaces/materials in the library.

5. Allow much more in-depth editing/creating in sections/elevations. The limited access we currently deal with is a bit silly, and constantly having to edit in floor plan or 3D simply because the options aren't available to us in section/elevation is cumbersome.

6. Create a full featured take-off system where we can accurately calculate the materials necessary to build the project (i.e. lumber, steel, etc.).

7. Seriously improve both the stair and rail tools (especially the railing, as it's a bit of a nightmare to get it exactly how you want it right now). A great example would be creating a custom railing panel, but the software isn't smart enough to know how to scale and stretch the profile to accommodate different angles.

8. Vastly increase the library to include truly usable objects such as interior window shades, curtains, trees/plants that look real, the list goes on.

9. Allow slabs to be rotated. Why on earth do I have to turn a slab into a morph to stand it up?

10. Solve the issue of the pet pallet jumping around constantly where you have to chase it with your mouse in order for it to snap to a stable position on screen. This has been an issue since I started using Archicad at version 17, and I'm guessing long before that too.

No one wants to purchase software that has so many issues. That's where you start. If you are going to spend a bunch of money on a piece of software, especially such a professional one as Archicad, it needs to be rock solid stable and feature-filled. Graphisoft keeps putting out a new release each year but they never seem to understand what needs to be done to actually impress their user base and get them to stick around.

Though I have no interest in switching to Revit, it always amazes me how fast and stable it is when I see others working in it. Archicad still blows Revit out of the water when it comes to its feature-set, but if Graphisoft really wants to show the world that Archicad is the best software for architects, it needs to be world's beyond where it is today, and at a more affordable price point ($1495?). Get the masses hooked and Autodesk will have a real problem on its hands.
#306989
kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>..."3. Partner with ChaosGroup and give us the ability to use Vray inside Archicad. Further, make the materials seamless between the regular 3D window and the Vray rendering engine (i.e. you create a Vray material and it automatically knows what to show in the regular 3D window instead of having to manually create two different materials and hope for the best)."[


This is probably never going to happen.
Why, you ask?

The makers of Corona render Engine - Render Legion - are currectly developing a render plugin for ArchiCAD (it's actually in open alpha testing stage now and you can download it from them to test it for free).
Which sounds great and all, but you see, Render Legion is now owned by,......yep,...you guessed it.....Chaosgroup.

It's doubtful they'd ever create a Vray plugin for ArchiCAD which would only end up competing with another one of their subsidiary products.

Personally I feel it's for the best (especially if they see it through to maturity) as Corona is vastly superior to Vray (for what WE do).


(* it's also worth pointing out that they're not doing this in collaboration or in conjunction with Graphisoft.
They're doing it of their own volition, with seemingly no support or input from the GS guys (who, understandeably have their attention with their Twinmotion partnership).
I guess it just turned out that the Render Legion guys had some ArchiCAD fans or supporters who thought a plugin would be great (if not profitable) since there's a lot of requests at their site from users of other more widely used software (Revit, Sketchup, Rhino come to mind) for them to develop a plugin for them. I guess we lucked out this time, because all those other software have a Vray render bridge.)

kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>>...."7. Seriously improve both the stair and rail tools (especially the railing, as it's a bit of a nightmare to get it exactly how you want it right now). A great example would be creating a custom railing panel, but the software isn't smart enough to know how to scale and stretch the profile to accommodate different angles."

Agree, 100%.
I think when it comes to the stair and railing tool - and ESPECIALLY the railing tool - Graphisoft developers were guilty of that age-old sin of over-thinking things too much and creating an overly (and unnecessarily) complicated monster when simpler options were already at hand at their disposal.
By this I mean, why can't adjusting the panel alignment on the railing tool be as straightforward and intuitive as adjusting a Curtian Wall grid panel where you can almost literally go right in and move it physically and pull nodes back and forth or adjust the grid spacing in real time - and all in the 3D window - rather than having to navigate a dialog box and use guesswork with numbers and now which end to move or adjust the panels to get it just right.
I mean, it's RIGHT THERE (in the curtain wall tool), the simple basic logic you can use, and an element that's virtually the same kind of grid-based system.
(in-fact I can't tell you how many times I just gave up with the railing tool and just used a curtain wall tool, instead, with a custom-created accessory object as a handrail).
In comparison I like the Stair tool a fair bit more, but that's only because a lot of the editing functions tend to be in keeping with the standard editing logic of existing construction elements in the program (like walls, slabs, etc) - which means the learning curve overhead for us, the users, is cut down a little bit.

But still, both tools seem to have that hint or aroma of 'lack of end-user feedback during the development phase' about them that now shows in how they just seem like they over-thought far too many things rather than keeping it simple.

For the railing tool, I think they should just hire the guy who made the EXCELLENT 3DMD railing object and buy-out CadImage, and intergrate both, while paring down the complexity of the existing tool they themselves made, and give their users the best of all worlds.

kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>"....8. Vastly increase the library to include truly usable objects such as interior window shades, curtains, trees/plants that look real, the list goes on."


See the discussion above on why it might be better for them to improve the custom GDL-object creating environment and tools to allow people to populate and improve the library themselves with smart parametric library objects rather than having to wait for them to update a decades old library with objects so lacking in variety, versatility and usefulness.

kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>....9. Allow slabs to be rotated. Why on earth do I have to turn a slab into a morph to stand it up?"

But then wouldn't it then be a wall?
I'm just joking (but only half so).
I think there's a fair question to be asked as to what degree you can alter an element until it essentially becomes another different construction element in everything but name (and certainly in looks and geometry).
Like how far can you canter or slop a wall before you consider it a roof? And if you can't slope you slab far enough, why isn't it then not a roof instead or a mesh?

This might be the underlying reason why Graphisoft probably refuses to allow certain transformations on certain elements and still have then read in the program as they are or originally started (without going the 'Morph' route - in which case they then lose their native abilities and behaviour. Like a wall converted to a morph to allow a certain transformation, obviously won't be able to host windows or doors anymore and can't count as the boundary of your zone or room edge. Or likewise a slab which then probably reads differently in plan and also loses all slab properties).
I've always thought that if they were going to persist with the route of forcing people to have to go the 'Morph' route to get the transformation they want, then why not allow for the re-classification (not just in the object ID sense) of that same morph back into it's original element category - and at least read that way (like for scheduling or other purposes)? And for certain Element-> Morph -> original element class, which allow it, they would even allow some of the original behaviour.
For example if I turned a wall into a Morph to allow me to create a double curved (on two axes) sweep on one side, then assuming I still had a straight edge that I could denote as a wall reference line, the program could allow be to reclassify that Morph as a wall (or a 'special' wall with one of those warning yellow triangle) and which for Wall purposes would adopt those functions (like inserting a window or window hole or door relative to, and hosted at this reference line with the hole or opening expressed as an extrusion straight through your (original) morph as is the native behaviour of windows and doors on walls. And like normal Windows and doors, they would be paramatrically hosted and tied to it and positioned relative to the reference line.

Same thing with slabs turned into morph to allow slanting (or indenting for the purposes of drainage ridges and holes) and re-classed back as a slab and behave as such as well.

It seems (to me anyways) like this should have been the logical evolution of Morphs and morphs created from other construction elements in the program, but... I guess... 7 versions since it's introduction, Graphisoft devs don't feel the same way.


kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>>...."10. Solve the issue of the pet pallet jumping around constantly where you have to chase it with your mouse in order for it to snap to a stable position on screen. This has been an issue since I started using Archicad at version 17, and I'm guessing long before that too."


I thought you could already solve this issue currently.
(*not really an 'issue'.
More of a 'behaviour')

The program allows you to specify a fixed location on screen for the palette where it will always be when called up, so you know where to expect it, and under this setting, you can also manually move it to a new location if that location doesn't suit you.
That's been my default setting for as long as I Can remember (or at least since version 11 or 12 for me)....as I couldn't be bothered with constantly chasing that silly little pet around every time I needed to do something or change a command option.

Although now, there's another annoying behaviour where it comes up and pauses close to your pointer (rather than showing up at the default location) and then when you go to click on your option, almost like it suddenly remembers at the last minute that it's at the wrong location and quickly rushes to the default selected location, leaving you clicking on the wrong (And empty) spot and ruining your command. An 'Undo' obviously undoes it, but second time around it behaves correctly and shows up at the right location. This has only been happening for the last few versions so it seems to be like a recent bug that's somehow managed to escape the attentions of whomever is in charge of writing code for the dang thing.

(*Okay, I just checked,...
Check your settings under

Options > Work Environment > 'Dialog Boxes and Palettes' > 'Pet Palette movement'

I guess on my own end, I changed this setting way back, all the way going back to version 12 , I guess, and have never had to bother with it since I usually just carry over my old setting from older versions when I get a new version and just use the same 'User scheme' I saved going all that far back - with whatever new additions I have to adapt to it for the new features)
kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>....."Graphisoft keeps putting out a new release each year but they never seem to understand what needs to be done to actually impress their user base and get them to stick around."


Blame the yearly release cycle for that.
Ever since they switched to it, most of their new releases have been weaksauce diluted versions which only seem to exist to justify charging the Subscription group their subscription fees and keep that revenue going rather than putting out an actually useful improvement, and you'll be lucky to get any big new groundbreaking features every 3 or 4 versions.

And the really sad part is that I don't think anyone (other than the Graphisoft higher-ups and Numbers people) like it at all.
The actual developers CAN'T possibly be happy working under the deadlines they have to work to get these out yearly and catch all the bugs before release while ensuring a stable release.
And the strain is beginning to show - especially when you consider the debacle that was the massively delayed Version 23 release date and the subsequent problems it STILL came with.
kittonian wrote:
Sat Feb 15, 2020 5:41 am
>>>>"....Though I have no interest in switching to Revit, it always amazes me how fast and stable it is when I see others working in it."


I'm actually surprised they managed to stabilize it if it's "fast" and "Stable" since, the last time I used it steadily and professionally (around 2011/12), it was one of the most comically unstable and in some cases (read: BIG project files) super slow and buggy.

This is actually not good news since it means that Autodesk are getting their act together and actually closing the gap with Graphisoft in this regard (not that surprising given the vast resources they have), and not just relying on their dominant market presence to sell the product.
That should be a warning shot across the bow to GS, if they're actually bothering to pay attention.

Speed and stability were two things that they used to have the market cornered on (and in some situations, still do).

How long before Autodesk is actually releasing versions with actual useful (and requested) ground-breaking and innovative features - the kind we constantly keep begging GS to integrate, but which they never listen to us about?

Not good.
#306992
kittonian wrote:
Sat Feb 15, 2020 5:41 am
5. Allow much more in-depth editing/creating in sections/elevations. The limited access we currently deal with is a bit silly, and constantly having to edit in floor plan or 3D simply because the options aren't available to us in section/elevation is cumbersome.
This is one of my biggest frustrations in terms of usability / editing. If a core tool allows a node or edge to do something it should be consistent across views. Curtain walls is probably the worst case where you have to jump between views to be able to edit a subcomponent, even the baseline has these issues. What is irritating is these matters aren't considered bugs, so the usability never gets addressed even if you report them.
  • 1
  • 6
  • 7
  • 8
  • 9
  • 10