The silica beads with moisture indication is an often misunderstood product which leads to false assumptions. They only indicate the moisture level of the silica itself and when it need drying or replacement; they provide no information about the moisture content of the filament. They really should put a label on these containers informing customers of this.
Heat is the most effective way to remove moisture. While some may argue that storing filament in a dry container—such as an AMS reading 0% humidity—will dry it, this process takes weeks to months. Applying heat releases moisture much faster.
The only reliable way to confirm moisture is the issue is to weigh the filament, dry it, and weigh it again. With Flashforge filament, which typically comes on a plastic spool, you won’t need to account for moisture in a cardboard core. Use the X1 bed drying mode documented in this wiki. Although 65°C for ten hours is the recommendation, it likely won’t take that long. Start the process, weigh it after an hour—if it’s losing weight, that indicates moisture is being removed. If there’s no change, there’s no need to continue. Keep drying until the weight stabilizes; at that point, the removable moisture has likely been purged. Try to refrain from going over 70°C or you may wind up with a mess that is well documented here in this post.
Suggested Diagnostic Step
Saves a lot of filament and time
Here’s a diagnostic step that uses much less filament: print a cone primitive with roughly the same height and diameter as your model, and print it in vase mode. This produces a quick 31-minute test instead of a 5-hour print, allowing for faster and more filament-efficient testing. It’s also a demanding test, as the single-wall print will reveal even minor calibration issues.
Click on image to zoom
- Right-click and select cone.
- Scale the cone to the 116 diameter that your current model has.
- Cut the cone at 27mm to mimick the height of your current model
- Select Spiral Vase mode
- Set bottom laters to 0 which will mimmick you current model.
Suggested changes to your settings
While this may not solve your issue, it may make the print perform better overall.
-
Switch to Arachne which will leave few gaps in the corners of the filament internal walls. In your model, it may only be seen to have an effect at the base but it doesn’t hurt to turn it on.
-
Change the wall order to outer/inner. This should have a noticable effect because the outer walls will set up first. It may make it worse if the filament is not properly calibrated but it’s a good diagnostic step.
-
Set the following parameters:
- Top Shell, Bottome Shell and infill density to zero.
- Wall loops to some value larger than your models total wall thickness, 21 is the example here.
This sill give you a very nice smooth series of walls that will have no infill pattern and aid in setting up a good track for the filament to cool properly. 
- Try tree support. I realize that tree supports can be tricky and they can take a very long time(minutes for this kind of mode) but if you can dial-in the exact places you want them using the support paint tool, they can be a real life saver for a model like this.
Here’s what it looks like without painting in blockers. The supports want to crawl through the grid.
Paint blockers on(red) and Green for yes
And you get a supports that are easier to remove that won’t mess with your top surface.
- Last suggestion - Try 0.08 layer height to see if it will help. If your filament is not calibrated it will make it worse but if it is dialed-in, it will produce a much better looking model for this type of model.