topic Re: Flaw in stair logic in Modeling

!Restored: Flaw in stair logic

Anonymous — Thu, 25 May 2023 14:57:34 GMT

The problems with the stair tool stem from a flaw in the logic. The stairs are being defined by where the riser is rather than where the front edge of the stair tread is.

In this post I've attached an image showing a stair that does not have a nosing overhang:
- the 3D matches the 2D; and
- the handrail is the correct height above the nosing of the stair (as stipulated by regulatory requirements).

In the subsequent post I will attach an image of the same stair with a 30 nosing overhang:
- the 3D no longer matches the 2D; and
- the handrail is no longer the correct height above the stair nosing.

These problems occur because the 3D of the stair is being generated by the outside face of the riser rather than the front edge of the tread. The nosing overhang is being added to the front edge of the tread instead of at the back where it should be, with the riser face moving back by the distance of the nosing overhang.

In the real world stairs are set out by their treads, not the risers, and the locations of all of the other elements of the stair is determined relative to where the treads are located.

I'm really surprised that a program as mature as Archicad 11 contains such a fundamental flaw in its logic. It is not reassuring.

Re: Flaw in stair logic

Anonymous — Sun, 15 Jun 2008 05:38:24 GMT

In this image of the same stair with a nosing overhang. Note how the additional tread width has been added in front of the riser rather than at the rear of the tread with the riser moved backwards.

This behaviour must create havoc on winder stairs with the consequence being that they are not modelled correctly.

Re: Flaw in stair logic

Anonymous — Sun, 15 Jun 2008 05:44:35 GMT

To demonstrate how absurd this logic is I have attached an image with the nosing overhang increased to the maximum possible (127 mm / 5"):
- the 2D is now nothing like the 3D; and
- the handrail height is a long way short of the required height above the stair nosing

This logic flaw could quite easily lead to the built stairs being non compliant with the result being the Architect being sued to fix the problem.

Please fix it.

Re: Flaw in stair logic

TomWaltz — Sun, 15 Jun 2008 11:55:56 GMT

mikem wrote:
I'm really surprised that a program as mature as Archicad 11 contains such a fundamental flaw in its logic. It is not reassuring.

The Stair Tool is one of the least useful and most highly flawed tools in Archicad. It has seen no real improvement in years and is something Revit resellers love to proclaim.

You're right, it's absurd that the Stair Tool is still this bad.

Re: Flaw in stair logic

Erich — Sun, 15 Jun 2008 14:43:14 GMT

mikem wrote:
This logic flaw could quite easily lead to the built stairs being non compliant with the result being the Architect being sued to fix the problem.

IMHO, this is hyperbole, any architect that fails to check their work and makes such an error should deal with the consequences.

While I too would love to see an improved stair tool, we should not begin to imagine that the software is meant to take the place of thought and consideration on our part. Thankfully we are the architects not the software.

Re: Flaw in stair logic

Bricklyne Clarence — Sun, 15 Jun 2008 21:33:56 GMT

Erich wrote:
IMHO, this is hyperbole, any architect that fails to check their work and makes such an error should deal with the consequences.

While I too would love to see an improved stair tool, we should not begin to imagine that the software is meant to take the place of thought and consideration on our part. Thankfully we are the architects not the software.

IMHO, this (what you said above, specifically) is absurd.

It's somewhat akin to implying that if I buy a car whose braking system wiring keeps failing because of poor or loose connections, then in the event of an accident caused by brakes failing, I should be the one to blame for not checking that everything was in working order before I drove off - despite the fact that the manufacturer assured me that his product works fine. That's ridiculous!

When car manufacturers discover errors like this in their products, they recall every single one of them and fix the problem as soon as it becomes apparent. It's called taking professional responsibility.
Thankfully GS don't have to recall every license that they sell, but only have to issue a patch or hotfix to resolve the issue. Which begs the question why such an error would still be there after so many versions and so many hotfixes over the years with all the complaints on this particular tool.

And nobody is seriously suggesting "that the software is meant to take the place of thought and consideration on our part" as you said. But the logical and reasonable expectation is that the software will work well and as expected.

Of course the architect is still responsible for checking the accuracy and veracity of the drawings and documents he produces, but if GS's pitch is that ArchiCAD will save him the time of having to do all this himself, along with coordination errors, through an automated and parametrically driven software process, then what's the point if the architect still has to go through all the drawings again as if a human had drawn them just to make sure that the oh-so-accurate computer software didn't make a logical error?

If there's a feature that Graphisoft can't get to work properly then it shouldn't be there to begin with.
You don't place the blame or the responsibility for this kind of error on the customer especially if the error is not so apparent to all, without considerable testing and especially when you sold them on the fact that your software works properly and most especially if the error is part of a buggy tool for which the users have been clamoring an immediate upgrade and fix going almost 5 versions back.

Re: Flaw in stair logic

Anonymous — Mon, 16 Jun 2008 00:47:43 GMT

The fact remains that the logic is flawed and because of this even with a small nosing overhang:
- the handrails are not at the height the dialog box says they are.
- the 3D is showing the treads in a different location to the 2D.

That is not correct and will never be correct.

Re: Flaw in stair logic

Erich — Mon, 16 Jun 2008 00:55:59 GMT

Bricklyne,

I am not saying that the software issues are not a problem. Indeed, I agree that stairs are laid out from the nose of the tread. The failure of this tool to respect such practice IS problematic. Often in our designs we have stairs that cause us to have to go to great lengths to make work as the codes require and that the tools provided make that harder is an issue.

My comment was referring to the idea that such a flaw could lead to the stairs being built wrong and result in the architect being sued. Regardless of the tool used be it pencil, CAD or Virtual Building or even the point and shout method, it remains the responsibility of the architect to be certain that what they draw works. If I ever caught anyone working for me not checking something critical like this, I would be furious. Presumably, there are stair details to go with the pretty pictures and in those details such a discrepancy would be caught, at least that is the hope.

While I think the flaws of the program should be pointed out and, with any luck, be fixed. Engaging in hyperbole is not productive. And yes, if there's a feature that Graphisoft can't get to work properly then it shouldn't be there to begin with. That comment should be followed with the idea that stairs are a critical part of architectural practice and need to be in the program and need to work.

As for the car analogy, if it was your job to make sure the brakes worked and they subsequently fail then, yes, you are responsible.

Re: Flaw in stair logic

Anonymous — Mon, 16 Jun 2008 03:27:03 GMT

The important thing is to get this fixed. And it has to be fixed in Archicad 12.

Erich, just as we have a responsibility to check things, it is not unreasonable to expect that Graphisoft will check the different program parts to ensure that they are correct. It is basic Quality Assurance.

I find it hard to believe that this basic logic flaw hasn't been picked up before now, and therefore fixed. Why hasn't it been? Are there other isues like this lurking within Archicad?

Re: Flaw in stair logic

Erich — Mon, 16 Jun 2008 04:51:07 GMT

Mike,

Are there other issues...you betcha! It all depends on what you want to do and how you want to show it. The stairs have long been a sore point for users. You have picked up on one of many complaints. If you haven't done so already, vote here on that some of these issues are fixed. (Just one of many stair wish list items)

http://archicad-talk.graphisoft.com/viewtopic.php?t=1656&highlight=stairs

But don't hold your breath of it being fixed in 12.

Re: Flaw in stair logic

Bricklyne Clarence — Mon, 16 Jun 2008 09:06:45 GMT

Erich,

While I agree with the basic point you are driving at, I still have to take issue with your dismissal of Mikem's point as being "Hyperbole".

Yes, there have been numerous complaints about the stair tool and yes, some of those complaints have been raised in less than savory or even civilized or acceptable ways. I don't think Mikem's posts fall into any of those categories.

Furthermore, I'm of the belief that during the few times that GS bothers to venture into these forums to see what user concerns are, when they see posts like your dismissing other users' serious concerns as being hyperbole, unnecessary, or insignificant, that that basically gives them the unspoken license to ignore said concerns because they can justify it that an equal number of users are not bothered by that problem. I can't find any other way of explaining or figuring out how else they would ignore such a problem as this or even the stair tool as a whole, unless they hold this attitude or worse yet, unless they are or were at some level unaware of it. In which case how can a thread like this be "hyperbole", in any event? when at the very least, it's alerting GS to this problem, or if they are then it's also letting other users be aware of this problem beforehand.

Again, I'm not absolving any architect of the responsibility of checking their drawings but if I have to spend that extra time looking out for an error that I wouldn't even have to check for were the drawings done the traditional way and which ultimately defeats the entire logic of a parametric model with design integrity on all drawings, then that $4000 license fee is better spent elsewhere: either with another software or with human drafters who don't need to be told more than once to repeat this kind of error. It goes beyond the professionalism and diligence of the architect and speaks instead, to the efficacy and the inherent value of the software as well as the commitment of its developers to producing a viable product and serving their customers well.

And as for the car analogy, I don't know about Oregon, but most places I know, its not the customer's responsibility to ensure that the product he or she buys, works as its supposed to work, or, as it's advertised to work. One wants to be sure that it works well to be certain, but to expect a soccer mom, a teenage driver or just plain Joe average to be capable of taking apart of the car system to make certain that all the nuts and bolts are in place as they are supposed to be, so that the brakes will not fail, as you seem to be implying is to the very definition of 'hyperbole' IMO and to completely miss the boat. Some firms have entire divisions, - plants even- just to test and ensure against these kinds of things.

Quality Assurance!

.......one can't say enough about it and just how much other companies regard it along with its impact on consumer confidence.

Re: Flaw in stair logic

Anonymous — Mon, 16 Jun 2008 14:38:40 GMT

Looks like 12 is upon us. If it hasn't already been fixed don't hold your breath waiting.

Re: Flaw in stair logic

Jere — Mon, 16 Jun 2008 14:45:58 GMT

Bricklyne wrote:
Erich,

While I agree with the basic point you are driving at, I still have to take issue with your dismissal of Mikem's point as being "Hyperbole"..

And I have to disagree with you and agree with Erich. I might even use the term "absurd" instead of hyperbole. If you get sued because of a flaw in your modelling, it's solely your responsibility. I fine your car/brake analogy strange. You're talking about the manufacturing of a built object. A better example would be a typographical error in a science textbook because spellcheck changed a word it shouldn't have. In that case it would be the author (i.e. the creator of the intellectual work) who was at fault, not the creator of the spellcheck software.

Personally, I would love a "perform building code review" function, but for now that responsibility lies with the designers.

Re: Flaw in stair logic

Anonymous — Mon, 16 Jun 2008 15:02:26 GMT

Its unfortunate that this topic has been sidetracked by a 'throw away line' I posted. It would be better if we all agree to disagree and the topic stays focused on what is important.

We deserve a definitive response from Graphisoft on this issue telling us whether the logic flaw still exists in Archicad 12, or not.

PS: Jere, regulatory checking of virtual building models already exists as a capability: http://www.solibri.com/

Re: Flaw in stair logic

Bricklyne Clarence — Mon, 16 Jun 2008 16:42:19 GMT

Jere wrote:
Bricklyne wrote:
Erich,

While I agree with the basic point you are driving at, I still have to take issue with your dismissal of Mikem's point as being "Hyperbole"..
And I have to disagree with you and agree with Erich. I might even use the term "absurd" instead of hyperbole. If you get sued because of a flaw in your modelling, it's solely your responsibility. I fine your car/brake analogy strange. You're talking about the manufacturing of a built object. A better example would be a typographical error in a science textbook because spellcheck changed a word it shouldn't have. In that case it would be the author (i.e. the creator of the intellectual work) who was at fault, not the creator of the spellcheck software.
...........

.....in your eagerness to disagree you must have missed the part where I stated (repeatedly, at that) that I don't absolve the architect of his professional responsibility to ascertain the integrity and accuracy of his drawings regardless of the shortcomings of the software.

The car analogy was and is valid - you're merely and needlessly splitting hairs here; we're talking about a manufacturer (or developer in this case) producing a faulty product while assuring his customers that it works fine, despite the fact that it impacts the customer's intention for buying the product negatively, and financially even if the customer is responsible on his own end using the product.

Which wasn't even the point in contention to begin with. The point was that the product manufacturer or developer - be it a car manufacturer, spell-checker developer or CAD/Design program developer - has just as much a professional responsibility on his own end, to his clients and customers ( just as the architect has his professional responsibility to his clients, regardless of whatever software or means he chooses to use) to ensure that his product will do what he advertises it to do and not leave flaws unresolved.

I don't see how that is so difficult to understand. But you're right; let's just agree to disagree.
I for one still agree with Mikem that Graphisoft still need to resolve this issue.

Re: Flaw in stair logic

Jere — Mon, 16 Jun 2008 17:28:54 GMT

Bricklyne wrote:
I don't see how that is so difficult to understand. But you're right; let's just agree to disagree.
I for one still agree with Mikem that Graphisoft still need to resolve this issue.

My eagerness to disagree? You're right about that. In fact, I've sitting around all morning just wishing to find something to disagree with you about. When I found something that I disagreed with, oh.....the joy. "FINALLY!" I exclaimed.

I don't believe it's difficult to understand either. I'm fairly confident I understand your point about professional responsibility (on the manufacturer/developer side) and professional liability (on the designer's side.) However, I still find your analogy odd. Comparing a handrail height on a drawing you're responsible for to a life-safety device in a motor vehicle seems a bit of a stretch.

But I'll talk no more about it. I think we can both eagerly agree this conversation is detracting from the point of the thread.

Re: Flaw in stair logic

Anonymous — Mon, 16 Jun 2008 17:39:07 GMT

I agree with MikeM as well. This tool always baffled me and I gave up trying to do stairs with StairMaker because I could never make the balusters and the nosing line up correctly.

Has this been fixed in 12?

Re: Flaw in stair logic

Laura Yanoviak — Mon, 16 Jun 2008 17:45:30 GMT

william235711 wrote:
I agree with MikeM as well. This tool always baffled me and I gave up trying to do stairs with StairMaker because I could never make the balusters and the nosing line up correctly.

Has this been fixed in 12?

From what I can tell, the original issue as posted by mikem oz has been fixed in AC12.

Re: Flaw in stair logic

Bricklyne Clarence — Mon, 16 Jun 2008 18:01:20 GMT

Jere wrote:
My eagerness to disagree? You're right about that. In fact, I've sitting around all morning just wishing to find something to disagree with you about. When I found something that I disagreed with, oh.....the joy. "FINALLY!" I exclaimed.......

....wow. Comedy is just not your thing it seems. Sarcasm neither, apparently.......but then again there's that thing they say about sarcasm being the last resort of something....

Good thing you're in architecture; no laughs required here......

Re: Flaw in stair logic

Anonymous — Mon, 16 Jun 2008 18:22:35 GMT

"Good thing you're in architecture; no laughs required here......"

I used to be next to an office like that with all of these grim-looking, young faces glued to their screens quietly tip-toeing around the older, bitter, angry, disrespected professional they worked for. 😉

I'd like to see the stair tool in action and then how easy or not it is to make custom objects for balusters.