2023-11-15 07:01 PM
Can someone from Graphisoft please step up and own the current abject failure to deliver the renewed MEP Modeller?
Today I joined the many others who have had to return to using AC26 in order to deliver on our MEP commitments.
We all pay out for GS to deliver improvements to AC, yet here we are 7 months later with a key feature broken & software we can't use. Can someone at least tell us what we can expect in terms of delivery of fixing this MEP debacle and when?
It's bad enough struggling with the 3D screen "grey outs", and the disappointment of new features that suddenly became experimental, but finding a fundamental tool is no longer useable... 🙄
2025-11-10 04:34 PM
So i wasted 3 hours and checked if it is any better and the result are below.
– No oval ventilation ducts.
– Connection splitting problems. Everything is drawn as a continuous chain, so you can’t just delete a single segment easily — you have to cut it manually, which is imprecise and time-consuming.
– No parameter editing for tees. They are generated automatically and their parameters are locked, so you can’t create custom tees (e.g. 20/15/15). In Archicad 25, you could still make your own objects — maybe not perfectly, but at least there were some workarounds. Here, there are none (or I haven’t found any).
– No control over branch height in tees. The outlet position generates automatically, and all its parameters are locked. (It can only be changed in the route specification, but that affects all tees in the entire system.)
– Most MEP objects (ducts, pipes, tees, etc.) are closed-source (from MEP routing tool perspective). You can’t simply add or modify something, because it breaks functionality. Creating custom objects is very difficult and usually causes crashes or other issues.
– Element movement is unreliable. The “offset object” option (shown below) appears only in specific cases. Normally, you have to use “change position – move”, which later breaks connections, causing issues like the ones shown below.
– Strange glitches happen sometimes, e.g. a move point gets locked somewhere and you can’t add a node or move anything, because it automatically snaps to a random 2D/3D point.
– After switching to flanged connections, some elements stop responding. For example, I set the entire route to be insulated, yet one element ignores the parameter completely. Sometimes flanges don’t appear at all.
– You can’t change the flange width, only its diameter.
– When inserting two elbows using the new “double elbow” function and the connections are flanged, the result looks completely unrealistic - there should be 1 flanged and 1 not just like connection works in real life.
– Sometimes things just break — even with the same diameter, inserting a tee can produce something like this (see image).
– Valves can’t be rotated. The ones in the default library always insert in one plane only.
– The “two elbows from one” function doesn’t always work properly — sometimes strange things happen.
– Material reports are still made mainly for architects.
They’ve improved it slightly, and with some effort and templates you can make it work to a degree, but it’s still so complicated that it was easier for me to understand Archicad’s C++ API than to get precise material quantities here.
My solution written for Archicad 25 in 150h using the C++ API (with copilot and grok help) works far better.
– Drawing on crosssections and pipe connection suggestions sometimes work surprisingly well.
– 2D ducts (e.g. 140×51) maintain their height and width correctly when direction changes to 51×140 — I couldn’t break this with the new MEP tool even though I tried.
- they think about things like callculating air supply etc.. but i didnt have time or will to try it out (so i don know meabe its superb) becouse basic functions dont do what it should.
2025-11-11 01:27 AM
Hi @tomswons, very nice graphical description of where MEP could be more accurate in accord with real world components. Why not copy this post and edit it and put in the wishlist ?
2025-11-11 07:58 AM
Seems like a good idea 😉 probably will do it, thanks
2025-11-11 01:14 PM
Hi @tomswons ,
Thank you very much for taking the time to write this up! Very valid feedback, most points we are already aware and aiming to fix, but I also found some new issues along the ones you raised. Your 3 hours checking it for sure were not wasted, as it will help us a lot making the next release better.
2025-11-11 10:51 PM
Thank you very much for your time and for taking a close look at the issue. In the company I work for, for various reasons, we are committed to working in Archicad. Until now, due to competitive pricing, we were able to overlook certain limitations. However, recently, due to pressure from management, we need concrete results if we are to invest in licenses under a subscription model.
Could you advise when we can realistically expect at least the above-mentioned issues to be resolved? I do not expect miracles, as I understand how challenging it is to develop functionality in a complex environment like Archicad.
I realize that even 15–30 licenses may be marginal from your perspective, but for us, it is important that we can work in Archicad. I am curious about your roadmap regarding this matter. We cannot continue working at version 25 indefinitely, and our timeline is also becoming critical.
Of course, the architects in our company are satisfied, but what is missing is effective collaboration with the building services (MEP) teams as it is critical for creating sustainable fully modular building develoment. As we expand to various markets rights now time is not on our side.
Thank you again for your attention and support.
2025-11-12 05:03 PM
We are currently focusing on five key directions in our roadmap, with further details available here: Product Roadmap - Graphisoft Community. Several issues raised in your feedback will be resolved in the next release, while some, as you mentioned, are more challenging and will need additional development. If you have any questions about it, please let me know.
I believe that launching the separate MEP Designer product underlines our commitment to addressing the main pain point raised by you: “what is missing is effective collaboration with the building services (MEP) teams as it is critical for creating sustainable fully modular building development”.
3 weeks ago
Bailint, you seem to miss the fact that the main pain point is that we had a functional, slightly buggy, MEP toolset in v25 which your team replaced with a non-functional toolset as most of the tools are broken or have major issues. Please explain how that decision benefits your users. At minimum, the working toolset of v25 vs the current implementation should have been a choice for your users. We pay for a working product, not a half finished alpha version. It is your responsibility as developers to provide a product that works. Simply put, how can you justify giving a broken set of tools that previously worked to paying customers? Perhaps a more relevant question is: who do we speak to as paying customers to express our absolute disapproval of this implementation? To ignore this is grossly negligent on graphisoft’s part.
Monday
You are right to call out that we took away a toolset in v25 that, while imperfect, let many of you deliver projects, and that the new one might not yet been a reliable replacement for heavy MEP work. That has real cost for you, and that is on us.
The reason we did not just keep iterating the old MEP is that its internal structure had hit a wall: it could not realistically support the kind of documentation, calculation, interoperability and dedicated MEP product we are now building with MEP Designer. To get there, we had to rebuild the core, not just patch the old add‑on.
For existing projects, the practical workaround today is to keep finishing them in Archicad 25/26 with the old MEP toolset, since the new platform only starts from 27 onward. New projects that want to benefit from the new roadmap need the new core, but that does not change the fact that the transition should have been safer for you.
The current focus is on closing the most painful gaps you and others have listed in this thread and in the roadmap items tagged with MEP Designer, so that the new platform is stable and complete enough for day‑to‑day work. If you want to escalate as paying customers, the two paths that actually feed into decisions are: logging concrete issues and blocking cases here or via your local Graphisoft partner / support, and adding and voting for specific MEP Designer items on the public Product Roadmap.
Monday
That is as disaster of a reply.
Tuesday
So Balint, what I am hearing you suggest is that users who need functional tools that don't work in current versions should NOT upgrade to current versions and instead stick with the versions that DO work. There would be an argument to upgrade to newer versions "IF" ARCHICAD was backward compatible BUT it is not.
So, using a business case argument, if something isn't working the way it should then the VP of Operations should seriously consider getting rid of the people who created it and replacing them with people who can, don't you agree?