Hi Greg,
The new Samsung Monitor I posted about does track eyes, and therefore, is suitable for a single user (unlike shutter-glasses). See Tom’s reply to my original message about the differences. But it does not change the 3D viewpoint, but only handles the flipping of L/R images as the head moves (at least, that’s my understanding).
I really appreciate that you hand-delivered a 3D monitor to Tom; kudos!
Cheers,
Gökhan
From: Greg Pintilie <gregdp@gmail.com>
Sent: Friday, 9 May 2025 12:49 PM
To: Tom Goddard <goddard@sonic.net>
Cc: tolung.bio@gmail.com; ChimeraX Users Help <chimerax-users@cgl.ucsf.edu>
Subject: Re: [chimerax-users] Re: Samsung 27" Odyssey 3D G90XF 4K 165Hz Monitor
I’d like to chime in here because I’ve been an avid user of many 3D displays over the years.
The Samsung monitor here sounds old school, in that it doesn’t track the user’s eyes, like the older technologies that were in retrospect uncomfortable by comparison. I have used SBS mode with these older techs, and it is a pain, I doubt it is worth doing more with it.
Tracking the eyes and rendering the image based on eye position and distance to the screen makes the 3D effect a lot more comfortable and natural. As Tom says this takes more computation and is more like VR. It is worth it.
I highly recommend the Sony or Acer displays that Tom mentioned, and small side note, I am the proud hand-deliverer of the Sony display to Tom. He has made ChimeraX interface to these displays work really well. I am more optimistic than Tom that these will be used a lot once people experience them and hopefully prices come down some more. They have decreased in price already substantially.
Thanks Gökhan! I missed that the price was in Australian dollars. $2000 is more reasonable but still pricey. I could contact Samsung and ask for a free display, but I give it about 5% chance of success. Also busy with features thousands of people will use (Boltz structure prediction) so it is hard to spend the time to make a display work that a handful of people might use. Sorry about that. Acer asked me to make their display work and brought the display to my desk, and my Sony display was also hand delivered from Stanford university. I think I was a sucker to accept those displays given the few people who will use them, but if someone delivers a display to my desk I am weak and curious and prone to making it work!
Since I am located in Australia (and the browser automatically detects location), that page shows the prices in AUD, not USD; looks like you missed that.
So, the USD price is $2K.
Can’t you contact Samsung directly and ask for a sample/unit for scientific software development?
The Samsung web page you referenced says the monitor is $3000, the same price as the Acer 27", and the Samsung also uses eye-tracking so it will probably only work for one person. The Sony and Acer displays both use OpenXR and eye tracking. This is good and bad. The good part is as you move your head to the side you see the side of the rendered objects. The bad is that most 3D applications do not support OpenXR unless they are VR applications. In contrast the Samsung uses side-by-side video input. That allows it to work with any 3D application that supports that common 3D format. Here's a description on Reddit talking about side-by-side input with the Samsung. (Unfortunately the Samsung manual does not say much.)
The drawback of the Samsung approach is that the monitor has no way to tell the application (e.g. ChimeraX) that your head has moved to the side (that is what OpenXR does). So when you move your head to the side you just see the same view direction of the 3D object only skewed. It's not a disaster, you can rotate the molecule if you want to see it from the side. The Samsung manual says only Windows is supported since they include some crappy software that runs on your Windows machine. I'm not sure if it is needed to handle side-by-side input.
I'd love to see the Samsung display. It uses a 4K panel like the Sony and Acer. Like those displays it must be using a lenticular grating to send some pixels to your left eye and some to your right eye so the actual resolution in 3D is more like HD (2K). Still that is pretty good. (With old technology LCD shutter glasses you would get full 4K resolution since the two eye images use all screen pixels and just alternate in time.)
If I had one of these displays I think I could make ChimeraX render full-screen side-by-side. Since you can't drag the ChimeraX graphics into a separate window I would try to add code to create a separate window and render to it as an option. My lab does not have money to spare to buy such a display (NIH funding problems), so don't expect to work on this.
This Samsung is less than half the price (27” for 2K USD) compared to the SONY in the review.
Let me know if you can score one of these to test/develop for :)
There is no ChimeraX capability to make the graphics pane full screen, so it will not work with a 3D display that wants full-screen side-by-side stereo format. The graphics pane is not a docked panel like the other ChimeraX tools. The ChimeraX main window is a Qt QMainWindow which treats that center area specially and it cannot be detached into a floating window. I'm not sure how hard it would be to allow that.
We are also working on glasses-free eye-tracked 3D displays such as the Sony Spatial Reality and Acer SpatialLabs displays. They use OpenXR and only work on Windows and only with 1 person viewing, and costly $3000-5000, so they have some limitations. Here's a description of trying the Sony 3D display with ChimeraX
Hello ChimeraX team and users,
Does ChimeraX support this monitor?
I read a very positive review about this monitor and may consider buying one if it works with ChimeraX.
The review says that this monitor needs a full-screen side-by-side output. Just as many of the sub-windows of ChimeraX can be docked and undocked (e.g., log, Models, etc.), is there way to ‘undock’ the main graphics window to make it full screen? Or, is there a way to display the graphics window on a secondary monitor as a full-screen output?
_______________________________________________
ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu
To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu
Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/