We have issues placing holograms for Guides accurately. We know the Hololens 2 is capable of relatively fine precision (roughly 1cm of accuracy per 1meter of distance from target) but we can't seem to accurately place holograms in Guides. This is frustrating for a variety of reasons, not the least of which is the example I am linking below. In this example a user is attempting to accurately line up a hologram across one plane, which causes a shift in another plane, they then correct that shift and are back to where they started.
Can anyone please provide suggestions to get more accurate placement of Holograms in Guides during the authoring step?
Also, I am curious if there are there any plans on the current development roadmap to improve hologram placement methods? If not, I would like to humbly suggest a placement option that allows for translation along 2 axes only instead of always allowing translation across all 3. This option of locking out specific axes is already in Guides for rotations of Holograms and is extremely helpful. If we could get this same option for translation of holograms it would make accurate placement much easier.
[View:/cfs-file/__key/communityserver-discussions-components-files/785/20210121_5F00_102936_5F00_HoloLens.mp4:640:480]
Thank you Andrew. Please set me a message at any time and we can setup a call. Our team will appreciate your insight/feedback.
Thank you Andy and Dave.
I appreciate the insights. I understand the desire of making Guides easy to use and maintaining similar UX across both Generations of Hololens. Guides is a great program, and the ability for end users to pick it up and start using without needing training on the Hololens is a great selling point. But, this particular aspect of it is frustrating, especially when we are able to create custom dev applications using MRTK and our own tools to get the accuracy we are looking for. However, I'm very happy to hear that the team is aware of this need already and actively considering some way of improving hologram placement accuracy in the future.
My team and I may reach out to discuss further, including possible detailed suggestions on how to implement this suggestion. Among other things, we help clients implement Guides and this issue has come up occasionally as a potential blocker to adoption, so I wanted to see our options for being able to push through this.
Thank you both!
Howdy Andrew!
Thanks for the clear and logical suggestions. My name is Andy, and I'm a designer on the MR Apps team at Microsoft. When we created the hologram placement controls for Guides, we were really balancing effort VS accuracy, as the control mechanisms work on both HoloLens 1 and 2 in the same way. We don't want the user to have to press too many buttons or change too many modes when placing holograms, since that would place a tax on every interaction, and require more effort than we might want to put in, say, on HoloLens1 with air-tap. This shared UX across HL1 and HL2 has been a good solution for familiarity and customers with devices fleets spanning generations - once you learn the controls on HL1, you can do the same movements, but with direct touch on HL2.
You're right that HL2 users could definitely benefit from a per-axis or 2 axis modifier. We've done some awesome prototypes that take into account the surface reconstruction geometry of the scene to help snap and glide holograms against the real world, but it's not ready for primetime yet. One of our ambitions and principles with mixed reality UX, is to avoid overwhelming UI complexity with spatial controls. We don't want to add too many buttons to our movement controls, which could incur accidental button presses. However, there is certainly interest in our internal team to provide per-axis controls, but this work is not the highest priority item on our agenda today.
Our recommendation is to do exactly as you are doing for now, as shown in the video. Evaluate your placements from multiple perspectives to ensure the hologram is where you want it to be. When grabbing a hologram, if you only want to move up and down, then make sure your hand doesn't also move forward. The movement controls always use your current view angle to apply the motion, so it can be helpful to orient your view side-on, and only move left-right/up-down, or look top down when positioning, and only move left-right forward-back.
In the meantime, rest assured, that we are working hard to improve all of our mixed realty controls, and feedback like this is more motivation to bump up the priority. If this is a blocking matter, I'd be happy to schedule time to discuss in more detail and help you and your team be successful with both the current functionality, and any new functionality we may build in the future. If you have specific ideas on how you'd recommend a constrained axis control could be activated by the user, please do describe your thinking below.
Thanks for using Guides!
Hi Andrew,
I am reaching out to our engineers to review your video and offer some suggestions. In the meantime I love the idea of translating a hologram on a specific axis. We are working on a roadmap that includes an improved authoring experience and this is something under discussion. I suggest adding your idea to aka.ms/Dynamics365Guidesits as that helps show customer demand (and others can upvote your idea).
Stay up to date on forum activity by subscribing. You can also customize your in-app and email Notification settings across all subscriptions.
André Arnaud de Cal... 290,900 Super User 2024 Season 2
Martin Dráb 229,275 Most Valuable Professional
nmaenpaa 101,156