History Bug

This is a common problem experienced by a number of the community and in serval community conversations. You don’t need my model! There is a common error in the implementation of history. When the history is large with complicated models something happens and errors are created where there was none, and items disappear as an effect of missing references.

I know I am talking to various people under your profile name and have to explain servals times. I find your customer service very poor and totally unfocused.

Forget backup - solve the problem. I no longer have interested in regaining the 2 months of lost work. I am interested in it not happen again. You have given @Jon1 assurance the bug is being looking into.

Again what is the bug resolution number and its name?

As I mentioned in the other thread we are aware of an issue that can cause a similar problem to what you described. The fix is targeted for version 5.660. I understand that you are not happy with our customer support.

When is v5.660 planned for push update?

We are releasing new versions every ~2 weeks.

5.650: July 1st
5.660: July 15th

+/- couple of days

1 Like

When people are back in the office, please provide a written statement of the scope of the related bug fit in v5.660.

I’m not sure if I understand what that means.

The bug fix relating to our collective problem needs a scope statement or description to inform the develops. Please provide this statement so that the community can rest assured the correct problem is being resolved.

The issue that can cause this problem is a bug in Siemens Parasolid, the underlying modeling engine of Shapr3D. The problem is that some direct modeling operations can be flaky (providing different outputs for the same inputs). To avoid this situation, we are always going to serialize the entire Parasolid session, and load it when you open the design, to guarantee that there is no feature tree evaluation is happening when you reload a design.

2 Likes

i.e. in Shapr3D words, what are you fixing.

Thank you. I’m not sure why it was so hard to get to that.

1 Like

I’d like to offer some advise. Do not use the profile of ‘Istvan’. Use ‘Support’ and introduce yourself in conversations. Poor customer support should not tarnish the CEO and founder’s image. And if you do this quality control can be gained in the various conversations.

People like to know who they are talking to, and good customer support should be rewarded.

I agree, this should have been the first or second response. As customers we are in the dark with what is happening.

It’s like feeling sick, going to a doctor, you explain your symptoms and the doctor knows what the problem probably is, knows they’re working on a vaccine but the only thing he tells you is: “explain your symptoms again”.

Glad to hear your team is working on a fix. Thank you for the update

Since I don’t know anything about @RevenueMonkey 's issue besides that there might be some data corruption, I can’t be sure that this fix would solve his problem. This is one of the many many bug fixes that we are working on (as we do for every single release), so I can only guess, but this is the only bug fix that we are working on that can cause data corruption. I also don’t know if it’d fix your issue, as I don’t know enough about it. Also please note that this will not resolve the issues in existing workspaces.

2 Likes

If I were you, I’d be posting a download link of the pre-history tree release to every single person having issues so they can revert back to a stable release. This is a major fubar exasperated by tone deaf support.

1 Like

Not possible, the designs are not backwards compatible. None of our stability metrics have changed, but I understand that for those who are impacted this can be very frustrating. These issues only impact a small fraction of our user base, unfortunately those who are impacted are experiencing very frustrating issues.

1 Like

I’m not confident Shapr3D appreciate the impact, a small fraction doesn’t ring true to me. All users are affected by this issue that create complicated models. If there aren’t many users in that class, I’d question why you need history for simple modeling. I believe it was grossly underestimated how many people enjoy the agility of the pre-history version. In my world if you want a fully fledged design tool you choose SolidWorks or the likes. Recognising an identify crisis is difficult. What is Shapr3D without agility?

By the way I’ve no idea how software can loose a reference causing a item to be deleted. It seems its still underestimated how massive a software error that is.

It’s <1% of our monthly active users. As we have hundreds of thousands of monthly active users, that number is relatively large in absolute numbers.

No. Those users are affected who are using certain direct modeling features in a way where the Parasolid bug occurs. While we have a way to reproduce the issue, Parasolid is a black box for us, so we can’t provide exact guidance on what to avoid.

No, it wasn’t. We’ve beta tested the new version for almost a year, ran hundreds of interviews, tests and more. Every major change in a product’s lifecycle will leave some of its users unsatisfied, but we need to move forward, we can’t stay at a local maxima.

1 Like

I remain unconvinced, this feels massively like an overconfident statement.

1 Like