Wishes
Post your wishes about Graphisoft products: Archicad, BIMx, BIMcloud, and DDScad.

Please Fix Bugs/Glitches

4dProof
Advisor

Not sure if this is a "wishlist" item or not... but Archicad bugs have been compounding for a few versions now. I have not had a single file since AC22 or 23 that DIDN'T have model view options go missing in the view map. This is not related to any changes to the MVO, or any other trigger I can pinpoint. I just know, at least 3-5x on every project, I will have to reissue because my floor plan showed up with "simplified windows" or "door/window markers off" all related to MVO going missing.

 

I have honestly lost all confidence in Archicad as a tool for design and documentation. Very basic things like dimensions going missing or showing incorrectly (yes, even in AC25), view settings randomly falling apart/disappearing, zones not showing up in the door/window schedules for no apparent reason.

 

This is not to mention the countless crashes that I have to deal with every day, if not every hour.

 

A big part of what I do for a living is trouble shoot these problems for architects; and I feel like a real **beep** when all I can tell my clients "sorry, the software kind of sucks some times". Many times, it does turn out that assumed bugs are due to user error; but I have personally witnessed the missing MVO and missing/incorrect dimension bugs with no obvious user caused trigger.

 

At the very least, we need to know that dimensions will not disappear and view settings are going to remain stable and not show "missing". If an architect has to review and redline every drawing for both drawing issues AND software issues, you have to ask, 'why are we even bothering with this tool?' Archicad used to be the most flexible, versatile, and reliable design and documentation tool on the market, but faith is dwindling in the product, if not completely lost on my end.

 

MVO Missing.png

BIM solutions and trouble shooting (self proclaimed) expert. Using Archicad 26 5002 US on Mac OS 11.5.2
66 REPLIES 66

“Missing” MVOs (which are not missing, nor have they been modified; they are just unfound by the View; creating new MVOs may trigger it; have never been able to reproduce it by design, but it does happen all the time) has been a standard feature for a few versions now. I do most of my work on BIMcloud and I don't remember now if I've seen it also in solo.

No. I can't find a way to reproduce the issue of finding missing portions of dimension strings, but taking a quick glance at my current project, I see three dimension strings which have portions of dimension strings missing, which had previously been fully dimensioned. You SHOULD be able to place a dimension and not have to keep obsessing about whether or not the last command blew it away. This is bread and butter stuff. I just installed the preview version of build 5000 for AC25 earlier this afternoon, so I'm hopeful that this issue may have been resolved. This was one of the improvements listed.

"DEF-6576 DOCUMENT: Dimensions associated with wall corners sometimes went missing."

iMac 27 (2020) 3.8 GHz Core i7; 128 GB RAM; Radeon Pro 5700 XT 16 GB GPU (macOS Sonoma)
MacBook Pro M1 Max; 64 GB RAM; 32-Core GPU (macOS Ventura)
ArchiCAD 27/26/25 Full (latest builds). User since AC12 (2008).

To be fair Karl, just because the problem isn't consistently reproducible doesn't mean it doesn't exist. You are basically victim shaming users for not being able to point a finger at a piece of code to say it is broken. The problem is buried in the software and arrived following updates in 23/24. If Graphisoft want to hang on to their customers then the duty to fix such volatility rests with Graphisoft, not the end user. For me the dimension problem was sufficiently bad that I now use 3D documents to maintain dimension stability in my plan drawings. I was sick of apologising to a VIP client as to why random dimension strings were incomplete or displaced after they changed during publishing. I can get away with using 3D documents, for most users I doubt that is an option. 

Apple iMac Intel i9 / macOS Sonoma / AC27UKI (most recent builds.. if they work)

I completely disagree with this it's not shaming. 

If it looks like a duck and quacks like a duck it's a duck.

If some users never experienced the issue but some have them there is certainly a way to reproduce the issue and if you can't they it's not going to be found. 

My assumption is that is is user error is some way or another, a specific procedure that produces the bug. So it could potentially be both. That's what karl is after.

Find the process that causes the bug so it can be fixed.

Instead of complaining on the forum use the info the moderator suggested and solve the problem as it's clearly only some of people's problems and not all.

Problems are always able to be consistently reproduced when dealing with any coding language, it all comes down to procedure.

So crunch the numbers find the process and solve the problem?

Which users where? What machines? What versions? What files? What were the last ten commands executed?

I'm not saying that there aren't bugs in the code but I can say that every time I have had an issue I can always reproduce it at some point every time to be able to get the problem solved.

@SenecaDesignLLC I agree with you, BUT if the code is written such that the results are dependent on a diverse set of variables and not properly stress tested by the author then these seemingly random "bugs" will manifest. That stress testing should be done by GS as your average user cannot be expected to spend all their time trying to trap such bugs when they have little understanding of the code, they simply want to reliably add dimensions to their drawing. Enough users are flagging this up as a random fault, so GS need to dig in and identify why it manifests that way and not use the failure to formally report the problem as an excuse to not fix it. Unfortunately GS are very good at using that excuse...

Apple iMac Intel i9 / macOS Sonoma / AC27UKI (most recent builds.. if they work)

If you can't reproduce the bug how in the hell could you ever expect them to find it and fix it let alone know about it in a sense unless you're using 2-year-old software you're still a beta tester

And again I've never seen such bugs on any versions that I've ever used so how do we explain the difference? Clearly something is going on between procedures and if your procedure generates a bug and it's exclusive to you and a few others versus the world then yes it really becomes up to you to pinpoint it so they can fix it the end.

Still looking?

Browse more topics

Back to forum

See latest solutions

Accepted solutions

Start a new discussion!