What is the risk of working in projects in unsupported versions of Revit?

[skyeg321]

Hi All,

We are trying to develop a policy at my company for working in older versions of Revit. We know Revit 2020 will be unsupported by Autodesk after the new year and we have a lot of big projects in that version. We are advising that project managers keep their projects in supported versions of Revit but we know there will be people hesitant to upgrade in fear of warnings and wasting project fee on it. So I'd like to fully understand the risks of not upgrading and working in older versions. We know that if there is a problem with a project and its running in an older version of Revit that Autodesk support will make you upgrade in order to work on it. But does upgrading before an issue happens cause these issues to be easier to fix? Does being proactive in upgrading actually result in less risk of file corruption or something like that? We are just trying to figure out what to tell people to do. Because saying all projects must be upgraded seems heavy handed and advising that its ok to stay in unsupported Revit versions also seems bad. 

[syman2000]

That depends on your IT and how the software works with Microsoft patches. I run an old project as far as 2017. However I find older software may experience random crashes or random software quit without warning. It can get annoying, but it is up to your firm if they are willing to take the risk of losing work or having software just rage quit on you.

[HVAC-Novice]

I assume your firm will have projects in current, and in older versions. Here a few disadvantages:

- potential issues with newer Windows and other IT versions

- bugs not being fixed in older versions

- may not be able to get Autodesk support for a specific problem if the problem is based on the old version. 

- user has to learn workarounds needed in older versions, when newer versions may have a native solution

- any new families or schedules you develop in new projects can't be used in older projects. 

- no gain from new features. A huge one is 4K support that  doesn't work in old versions. Being bale to use large monitors is a huge productivity and pleasure gain IMHO. 

- user having to work in multiple versions and keeping everything apart must be frustrating. Especially if multiple projects need attention at the same time. 

- anyone getting a new PC will need to re-install all the old versions instead of just the current one.

IMHO, the multi-designer projects should have a clear upgrade path decided on at the beginning of the project. It is recommended to upgrade year by year and not skip years. and everyone should have the the very same patches installed. 

[skyeg321]

@syman2000 We definitely don't want any rage quitting.

Ooh @HVAC-Novice that upgrade on a 1 year basis statement is scary. We only do new versions on even years. So we have a lot of projects looking at going from 2020 to 2024. would you say go 2020 to 2022 to 2024? or just try going for the 4 year jump and fixing warnings as they arise? 

[RSomppi]

The rule of thumb that I've always gone by is that a project stays in the same version as it's inception unless there is a need to upgrade. IMHO, an arbitrary decision to upgrade mid-project is risky, especially in a collaborative environment. The upgrade can create a lot of work if everyone is not ready for it.

What if any of the consultants involved do not have plans to use a newer version? What if it has problems like '24? Upgrading for a single project could be detrimental to their bottom line.

I've always advised that upgrading a project just for the sake of upgrading should be avoided.

[HVAC-Novice]

How to upgrade projects with other players should be communicated BEFORE the project starts. It also depends on  how long a project lingers around inc. construction time. I have a few projects I started many years ago and (due to budget reasons) slept for a year or two. Now I'm not even sure if they get bid out 2024, or later. and I'm not going to keep an 2018 or other version around. YMMV. Our contracts with sub-consultants include the requirement to upgrade when we do. Never heard that being an issue. My projects are relatively small, though. 

Due to the recent "schemas" issue I changed my mind a bit from "being always up to date" to maybe upgrading to the new version 6-9 months after its release. By that time it should be known if there are problems. Service packs and patches should be applied shortly after release. If I could go back in time, I would have switched to R2024 after it is confirmed the new patches fixed the schema issues (they say it may take more patches). 

I now try a "zero warnings" design to begin with (will see how long it lasts to stay at zero). This may or may not help in upgrading. I also will be more diligent upgrading all projects annually instead of digging out an old project and then jumping 4 versions.  I have a script to batch-upgrade all my families. I never had an issue with upgrade and skipping versions. But it is the official recommendation to do it annually. I bet most of the things I had to fix after an upgrade where items that were wrong to begin with. Better and cleaner design = less upgrade hassle (I hope).

It is a balance between having the newest features and bug fixes, and a mature version. Like you don't buy a car in its first model year. Old versions have bugs that were fixed in newer versions. So, having an old version doesn't mean it is bug free. At this point I would not want to upgrade later than a year. 

[curtisridenour]

I agree with everything you said @HVAC-Novice. Most importantly, waiting a good 6-9 months before using the latest version. I want to wait at least until the first service pack. We had a several issues with crashes on 2023 even after the first service pack. It seems stable now though.

We have several older projects in 2018 and 2019 that have come up again because of RFI's and construction phase services. I don't always see a need to upgrade a project for a minor amount of work. But, if we get a contract mod, I will advocate for upgrading to our current Boiler Plate version which is 2023 at the moment.

Historically, 2016-2017 was a terrible upgrade. The text changes made the amount of work required to fix the model beyond tedious. I have not seen a similar issue since that upgrade though. But you never know.

[HVAC-Novice]

I think waiting for the Service Pack 1, AND the first patch to that might make sense as long as you monitor the forums if there are issues. Probably wait 1-2 months after SP 1 to see if anyone has problems. 

Also consider if those files will be used in the future (for future projects, or reference). If they get used later, it may be worth to just upgrade them now instead of in 5 years. I bet realistically no one upgrades a 10 year old project and does it 10 times for each version (if anyone even has the old versions installed).

I work for an owner of many buildings. So, my old project files may be useful for later projects and keeping them up to date is worth it. If your relationship with the client is finished at completion of a project (like they build a strip mall and sell it after construction) , you may not see that value. YMMV.