Fabrication Data Manager

[madcad]

Any one know if this is usable yet?
Where exactly is the configuration stored? Is it alongside the Project on 360?
Not sure why we need another content editor instead of AutoCAD>CADmep. Mind you it might get it into peoples heads that Fab Parts are native to Revit and not an Add On. When I speak of using AutoCAD to create our content they think we are using AutoCAD content in Revit.......

https://help.autodesk.com/view/RVT/2025/ENU/?guid=GUID-72D357DB-1ACF-47F1-B541-F9EE862300A1

[Darren Young]

Yes... I know if it's usable.

No.... it's not usable for just about anyone aside from an engineering only firm and even then it has limitations.

Best things users can do is avoid this pile of excrement.

[madcad]

I do like an unbiased honest opinion.
Thought as much.
Not sure Autodesk understand the product or the customer needs enough to make a solid go of this.
I am not concerned about using it for configuration or content creation but hosting the config in the cloud could be good.
Can it do that reliably?

[DotNet]

They seem to be investing significant development time in the project at the very least. Something like this, depending on how well it is implemented, has the potential to be extremely useful. Game changing, even.

You can run VB Script inside of AutoCAD MEP parametric parts for absolute control. Revit MEP families enable calculated parameters driven by conditional statements. This is far more powerful than the generic calculated dimensions we get from CAD/CAM/EST. We rely heavily on COD scripting. I don't see why they couldn't easily add this functionality to fabrication parts.

Fabrication will need to evolve and integrate with modern software and workflows if it's going to stay relevant long term.

[WilliamNY]

For what it's worth, AU has a class for Fabrication Data Manager this year; if you're going, check out catalog number BES1081. The course description says that they'll be debuting new features as well as improvements to existing features. Interpret that how you will.

I have another course scheduled for that time so I won't be able to make it, but I'm curious if anyone else goes.

[Darren Young]

They had the same class last year. Not much has changed. Nothing significant anyway.

Even within Autodesk, there doesn't seem to be a consistent thought as to what or who FDM is for. Spent nearly a decade on the Sprint reviews for this. Finally gave up.

At a high-level, "it'll enable everything" they want to do with advanced workflows, etc. But it's been 8 years to get to this point...where it's not really usable.

We just on a call w/Autodesk today and the other opinion on what FDM is for, is for all those customers who want to move to Fabrication but don't know anything about CAD, CADM or EST so have no easy way to learn it.  Past comments to my inquiry as to "when will it be done enough that contracts like me can use it" resulted in the response "who'd to say it's for customer like you?"

Imagine if they implemented FDM as well as they did Fab parts in Revit. Good start but never really finished. There's no public API for DFM either. The work I need to do in my database is not faster w/FDM, it's much slower.

There's some major advantages to the FDM concept as Steven/DotNet said, but that's all predicated on them getting done.  Moving to FDM is a major decision point. If that's where you do your fab db work, there's really no going back once you get in too far. So if you run into any major roadblocks, you're pretty screwed.

Unless you're a piping only engineering company, it's not worth paying attention to until they get price/labor in there as well as Sheetmetal support. But even then, it doesn't talk to EST or CAM so... there's that.

I wouldn't have spent 8 years on Sprint calls every two weeks if I didn't believe in the idea. But I dropped off as I love faith and confidence in Autodesk actually accomplishing it.  A year into the project, I asked how long it would take to finish during a customer council meeting at AU. We were all told another year. I told them then it was at least 5 years. And unfortunately I was wrong....it's been longer.

[DotNet]

I'm not convinced unaltered integration with CAD/CAM/EST is Autodesk's plan, or even the correct way forward. The fabrication database leaves much to be desired, especially when compared to features like routing preferences and calculated parameters driven by conditional statements.

Autodesk acquired fabrication primarily for its extensive CID library and to strengthen its position as the dominant MEP software provider. See attached image. Much of what the fabdb has to offer, aside from CAM, can be done better and must evolve to integrate with modern software and workflows.

Subtleties like notches, seam/connector allowances, ancillaries, and even material/labor costs are manufacturer specific, and are not required at model time. All of that is just a function of geometry, quantity and project requirements. One of the main things keeping contractors from modeling in Revit was the monumental effort it would have taken to generate accurate (virtually endless) MEP content. Now, even with the out of the box ITM files we have access to things like drop cheek elbows, 6 inch throats/extensions, and OGEE offsets, which was almost unheard of before. The Fab CID library alone made switching to Revit a solid and viable option for MEP contractors.

As far as CAM/EST integration... I don't think it needs to be (and traditionally hasn't been) as complicated as it currently is from an MAJ file standpoint. The geometry and quantity data generated from modeling with accurate CID content is all that is required. Everything else is just a function of that and external variables. Wouldn't it be nice if we could easily apply our company standard notches, seam/connector allowances, ancillaries and material/labor costs to any MAJ/RVT file instantly? There is no reason why this isn't possible, especially if those files don't need to be purged and heavily scripted clean of bloated/irrelevant/outdated/incorrect CAM/EST data.

[craigjonnson]

I don't look at the FDM tool as a replacement, there's currently too much missing from FDM to even think about it.
I'm sure I, like many of you, am the only database manager within your firm. The training or documentation for database setup and configuration for ESTMEP & CADmep isn't great and heavily relies on senior staff passing down that knowledge.
FDM allows trusted users to get into some elements of database configuration, like adding a new service and I like that restriction. I like that it allows the next generation of users some understanding of building content without the risk of screwing the entire thing up.
There are some great features in FDM like changing a connector name, and it finds all content with that connector and updates that connector vs having to run scripts or manually find every itm to change it manually for one. It would be good if there was some bi-directional management in FDM, but definitely would require admin access before that becomes a thing.

[DotNet]

It's not ready yet. They're still working on sheetmetal, calculated dimensions, and other things. They have however secured ownership of our databases and fabrications extensive CID library. That bought them some time for sure. They do not appear to be in any hurry. They have a lot of work to do.

I think the documentation for Fabrication is far better than any of its peers when you include the online community, XtraCAD, AU classes, and Darren Young's extensive COD script library. The APIs also extended Fabrications relevance for some time to come.

If we've worked with and customized other complex MEP software over the last 10-20 years, Fabrication is comparatively intuitive if you ask me. It does however leave much to be desired. Whether Autodesk is up to the task or not, it needs a fresh coat of paint and many core features need to be refactored or outright overhauled.

The Fabrication Database Pattern Options for example, are global. See attached. Because of this, at best much of them are mostly useless and at worst they are outright destructive. Patterns are heavily dependent on sheet size, material, connectors, and many other things. The "Machine If Length <" option for example is a great way to hose all of your existing projects. For galvanized TDC, the optimal value here is typically somewhere around 55". For Ductmate, ~59". For S&D, ~58". Simply setting this value to 55" globally will wreak havoc on all of your existing projects. If you catch it late, don't have recent backups, or talented COD scripters, you're pretty much screwed.

Irrelevant, outdated, and/or incorrect CAM/EST data has no business in our models. That's all just a function of geometry, quantity, and external variables.

The model drives this data, not the other way around. While we do need a "database" to drive services, systems, routing preferences, calculated parameters, conditional statements, and other things, it's important to know the difference. Everyone is so lost in the idea of BIM that it's easy to become complacent. No data is better than bad data. As with anything, putting the cart before the horse is a recipe for disaster.

[Darren Young]

8 years ago when I first was on the Sprint team for FDM, I was one of the few who actually supported it.

I actually do support the concept and idea. How it graphically shows you how things work together is a huge advancement for anyone trying to get into Fab for the first time.

The biggest problem w/FDM is they're just taking too long. Told them that 5 years ago... it's important... it's needed... but just stop.  They just don't have anough resources. Right now is the second time several in our industry rattled gages high up to get more funding for fabrication. Last time was circa 2016/17? (I forget). Near as I can tell, those resources added back then were quietly drained away. Now fighting to get them back.

Everything they're doing takes a huge amount of non-fabrication work because they're trying to put it on Forge (Platform Services). There were no services for hosting content. So everytime they wanted to buy a blender or can opener for the house, they needed to build a whole new foundation and house.

The other part of my issue w/FDM, is I just don't trust them for several reasons?

1) What happens when FDM is far enough people use it and there's no going back... but they stop development like they did for Fab Parts in Revit? Most of the core functionality in eVolve, Naviate, etc. really should be part of Revit. There;s some core things missing.

2) They have no plan or timelike for EST/CAM. Simply can't move forward without those being addressed.

3) It's too imature. Given the recent flurry of calls/emails I recieved sincde 2025 came out when they highlighted FDM, I loaded it up again. However I was told they needed to move their stuff to a new environment for performance reasons and that I needed to re-upload my database. Say what? What if that _was_ my only database? This seemingly lack of care with what I've done is a huge red flag for me.

4) It's in Technical Preview.... still. This simply means it's not suitable for production and any such use is at your own risk.

5) There's no API or COD script support. And much of the work I do uses a lot of techniques and scripts to do bulk managment efficiently. Using FDM will increase the time I need to do my work.

6) Things like RFA based stiffeners seem to be walking back from Fabrication and headed more to a RFA based workflow without any logical explination. Until they explain this in a way that makes sense to me, it gives me some distrust. When I asked who asked for this, I was pointed to a Wishlist item from a user we all know who wanted external angle rings to display and quantify. Except that same user, listed that as a request last month in our discussions with Autodesk. In other words, the person they added the feature for doesn't want what they did. Autodesk just "does" and doesn't consider that we havbe a vested interest in "how".  Additionally, they publically sought feedback for how to automate placement of these RFA based stiffeners. Excuse me? Isn't there something called SMACNA? Are not Specifications already in Fabrication? Yes, there's room for improvement but they didn't ask what's wrong with it or how it could be approved... they seemed to be asking like they knew nothing.

7) I've had Autodesk staff themselves suggest FDM isn't for customers like us. That it's for those customers who are new to Fabrication who don't know how everything works. Think Engineering only.  So if it's not for us... I'm in no damn hurry to use it.

  It's not robust enough for 3rd parties to use it. That's why GTP/Strauts developed Lightning. They just couldn't wait any more.

Conceptually however... totally on board. What I'm not on board with is Autodesk and their approach. Andy's put the database in the cloud twice now. He already had a cloud API for Fabrication when at Autodesk. They were forced to scrap it because it wasn't built on Forge. Now he's done it again with GTP/Stratus' Lightning product.

This tells you getting it in the cloud does not need to be hard or take a lot of resources. Al lthe things getting in the way of getting the database in the cloud and making FDM a reality are NOT technical. They're artificial limitations based on Autodesk decisions (so we're back to the trust issue).