And seriously, they have that little stamp for the backs of all the checks they receive. How could you stop using such a VALUABLE asset?
6 Likes
Considering I have found 2 critical errors with 2024.2 on vanilla Epicor screens (adding items to PCIDS and processing payments) I have doubts that pushing the new interface is going to come bug free.
Making dashboards is significantly harder and there is a loss of functionality there. You have to redo trackers and based on other threads on this very forum add events to get them to work right, something I never had to do with the classic tool. I don’t customize dashboards. I just add a BAQ and a few filters. No code. I can do a simple dashboard in the classic tool in minutes. We have nearly 100 dashboards. It will take forever.
And whoever got the idea to move trackers off the main screen and force them into a slide-out panel. Thanks for that. That is manual rework for each dashboard.
Simple things like disabling a button has been made so complicated that there are multiple threads and tutorials made just to walk someone through it. In 30 years of development I have never seen an interface work so hard to make something so easy to be so difficult. Clue → Make a widget to disable a dang button 
If there is now a widget to do that - great. But if you have to add events, widgets and rules just to change the state of a button it is far too complicated.
I am not even super deep in yet and I am shaking my head. Not sure how forcing companies to pay to redo all their IT work is a step ahead but I am a tactical level small picture guy.
5 Likes
I do agree with what you’ve said here. Even with the new wizards, the time to roll out a dashboard is drastically increased. Part of that is that we’re still learning how to do things the most efficient way, but I also think part of that is just the new reality. My biggest, single complaint with Application Studio is the inability to copy something that already exists. Yes, you can hack your way around this, but this is something that should, and NEEDS to exist, out of the box.
One option you could consider is that the classic dashboards can build an auto-translated, Kineticized version of your dashboards for you. They aren’t perfect, but they seem to mostly work.
I recently did a project where we did fully re-design the top 3, critical dashboards in Application Studio but the remaining non-critical ones we deployed via the classic Dashboard tool. As time permits, we’ll give them the full uplift treatment, but it gives a work around until then. As always, your mileage may vary.
You could also use the Edge Agent to give access to the Classic dashboards and still maintain the browser first approach. This would also allow you time to uplift the remaining dashboards. It’s not a perfect option, but it is an option.
4 Likes
I guess my complaint would be why it has become ok and normal for paying customers to be relegated to being alpha & beta testers of the software we pay quite a lot for? I can understand with a large product have bugs, but it seems their software R&D has been pushed onto the customer base.
10 Likes
Its just as fast to create a BASIC kinetic dashboard as it is to create one in classic, at least in 2024+ where the wizard works pretty well.
For the complex ones I think the toolset is improving (slowly) but its definitely a difficult learning curve especially with all the bugs.
5 Likes
I’m no expert but based on my experience I have to agree with your statement.
<rant>
The software companies’ development cycle today has to include FULL QA on all sprints. People from Epicor and some of their apologists on this site seem to allude to its “hard” or “impossible” to catch everything. I find it laughable that releases completely and totally break the functionality of Epicor in so many places.
I feel so sorry for people on the cloud that have no options past “flexing” their dates. Even that is insulting that you have to pay a premium to try to wait as long as you can for the major bugs to be worked out.
Epicor continually pushes out new features (YAY!), but they keep breaking other features. In addition, they never fix older features that are broken and have been for years (ICPO process with Credit Memos…). I get so tired of being told by support to put in an IDEA to FIX A BUG IN FUNCTIONALITY ALREADY RELEASED.
I think it comes down to maximizing profitability. Why employ good QA departments when they are moving forward just fine with allowing their customer base to be QA? It’s not like we can just dump Epicor at any point. Conversion is costly.
Is Epicor ERP a good system? Yes. I love what developers can do within the system, and the ease to integrate with Epicor from the outside is what drew me to their EPR in the first place. Is Epicor a good software development company? I’d say no. Their development and support process seem to be quite broken.
</rant>
Honestly, I doubt Epicor as a company would be doing as well as they are without this community.
12 Likes
I’ve got a half dozen BPMs just for this purpose - fix what they refuse to fix. Even if they do fix some bugs, it takes far too long.
3 Likes
One of my biggest frustrations…find a bug, report it and wait and wait and wait. Going through one now where it shouldn’t even be POSSIBLE for the bug to occur…an outright lack of data integrity.
That being said, I won’t dogpile on all of their support people. We’ve had some real good help on a lot of different issues. Some are aces, kings and queens…and unfortunately others are treys and deuces. The consistency just isn’t there. It’s like playing the lottery, you never know when you’re going to catch a winner.
4 Likes
My memory may be faulty but I thought the automation tools customers now have access to were a side benefit of tools Epicor built internally to do automated testing. I would have expected that a full testing suite of scripts would be used to beat the daylights out of a release.
If you’re not customizing dashboards, then it sounds like yours may be simple dashboards and shouldn’t really take all that long to convert them. Just a few months ago I converted about 85 dashboards in a matter of just a couple of weeks as we transitioned from 10.2.700 to 2024.1. I will say that it wasn’t my first time doing that, as I’d done conversions for another company a year earlier on 2023.1. However, I had forgot a lot that I was familiar with at that time and the improvements between 2023.1 and 2024.1 did make it easier.
If I could provide a few pieces of advice after having done this twice:
- As others have suggested, use the classic Dashboard deployment to generate the Kinetic app from the classic app. There is a conversion program to do this, too, but it is all or nothing, so I found that converting a smaller subset that it gave me more control to do them one-by-one in Dashboard.
- If you’re up on 2024.1 or higher, utilize the new Activity tracker to determine which of your custom dashboards aren’t being used and ignore converting them. This may cut down the number of dashboards that you have to convert
- While I know there is a way to use trackers as a slide out panel in Kinetic, it wasn’t easy for me to accomplish. What I did instead was either a) remove the tracker all together–regenerating the Kinetic app in classic Dashboard–and letting grid filters suffice for smaller datasets or b) when the dataset was too large to return all results and rely on filters, then I’d add BAQ parameters to the queries, regenerate in Classic, and then add the finishing touches in Application Studio .
Once you’ve gone through the exercise of doing your conversions, you’ll find that you’re more comfortable using App Studio, it will be easier to expand on your knowledge of how to use it, and then creating new applications from scratch in there won’t feel as daunting.
3 Likes
Brings up a question…do the automated testing tools cover the base and their “integrated bolt-ons” (for lack of a more appropriate description)? Products like Quickship, EDIHQ, EDA, Smart/IP&O…
1 Like
I rebuilt 89 dashboards and it did take sometime to learn App Studio but it didn’t take that long. I reviewed them and picked the most easy ones to do first; Single BAQ dashboards with a tracker.
It would be nice if there was a way to copy an existing dashboard. @mbayley’s template method comes in handy here, I’ve used it to “copy” a dashboard I wanted to use as a base layout for a new dashboard.
2 Likes
Apologist here. I would dare say that most of us could not write a test suite just for our own company that covers everything we do. That’s just one combination of settings and modules.
The combination of all the modules and settings for Epicor would be just astronomical. I agree that at least the individual components should be unit-tested, and some popular integrations, as well as any bug that was ever fixed, should have a test. End to end testing, especially with BPMs, Field Security, extra UI layers with verticalization and country specific rules is just not possible. You can try to get test coverage, but you’d end up wasting time writing tests for combinations that nobody even uses.
Otherwise, I agree with your sentiment completely. It’s very frustrating.
3 Likes
Agreed - that goes back to what I said earlier about Quickship, EDIHQ, EDA, Smart/IP&O…they’re all Epicor(-adjacent) products…so the automated testing needs to get at least as far as the handoff point to Product X.
2 Likes
No, that’s a difficult ask. My point was there are some major bugs getting through that are base functionality, not obscure instances where only certain companies would ever come across it.
If they are in the middle of their development of their internal testing scripts, then I can understand it. But to @CSmith’s comment, it feels like they need to spend more on resources to get this off of our backlog since it is part of the support services that we pay for.
2 Likes
Will DMT still have a command line mode in these Power Tools?
4 Likes
@Mark_Wonsil as far as i know (
Safe Harbor ) we are not doing significant changes to how DMT works (other than possible/normal improvements from Epicor Ideas )
4 Likes
I can convert a dashboard but it puts the tracker on a separate page (which is a pain to get to).
Moving the tracker to the main page is a pain and you have to override some settings to do it. A big thank you to Hannah who showed me how I can do it.
Also if you have 2 filters on the same field (like a date range) the Kinetic dashboard absolutely does not understand that. It cant have 2 filters on the same column. In classic this was trivial. With Kinetic I now have to redesign every query that uses date ranges (nearly all of them) and test them from the ground up. Adding extra columns to a BAQ to duplicate the date just seems wrong but its the only way that I got it to work.
But that is the easy part.
Creating an application from scratch. Horrendous.
The irony is that the hard part (kinetic functions) I learned in a few days. I’d never done them before but with some posts on this site I figured it out.
Figuring out why a text box is read only when it hasn’t been marked as read only or telling it to be x wide or a specific colour. Massive amounts of work for the part of the system that is supposed to be easy.
Or no option to delete an application that was broken out of the box.
I created an application with the wizard, made no modifications. Selected the object from a drop down. Preview. It just errors.
Can’t get a detailed error message, can’t delete it. Joy.
Update I think I can delete it but the option was confusing. I take back that statement.
I’d say this is a promising start but has a long way to go.
11 Likes
(post deleted by author)