Ina Cisco presentation a couple of weeks ago, which was held by an ex-Acano guy I think, he hammered a lot on prefering 720p60 over 1080p30. According to him, this is because the quality difference between 720p and 1080p is minimal on screens below 55", whereas occasional framedrops can really be disturbing on 30fps but not on 60fps, where it will still look smooth as butter.
I wanted to see how this would work in a real-life situation, by trying to set a preference of 720p60 on all video devices (SX10/20, MX300/700 mostly) where possible. All video devices are connected to CUCM, and we are using CMS as the meeting platform. When trying to tweak the fps, I couldn't find this setting anywhere in CUCM. I also cannot change this setting using TMS, because TMS can't really change many settings on CUCM-registered devices.
Where I would agree that you can't see much of a difference between 720 and 1080 on smaller screens, I think the difference between 30 and 60 fps is also not that noticeable in typical business use cases where there isn't a lot of movement. My suggestion to you would be to try it where you can and judge for yourself.
On your SX20 and MX's, you can change the setting on the codec, go to configuration, video, input, and you should see a section "Optimal Definition", and you will need to set the 60fps threshold to 720p. I don't believe this setting exists on SX10's, if memory serves they don't do 60 fps. Note you may (not sure) have to change something in CMS, I'm not sure if 60 fps is enabled by default.
Do you know if there is a way to configure the optimal definition tresholds from CUCM instead of on the endpoint directly? This used to be centrally managed by TMS, but now we have migrated from VCS to CUCM, provisioning of devices is fully done via CUCM. It would really be annoying if we need to manually change this on all devices.
The argument for one over the other will largely depend on the actual content of the video - ie, what you are trying to transmit. If it's fast moving video, then the 720p60 will be better, but if you need a clearer image of something that is relatively still, then the 1080p30 would be the way to go.
If 1 - 2 people take up most of the screen, smaller movements are more noticeable, whereas in a larger room fitting more people in the picture, the movements are less noticeable and a higher resolution offers more pixels per-person, so to speak. This however gets much more complex when speaker-tracking cameras zoom in on participants in larger rooms, where I would tend to suggest 720p/60fps where possible.
I have to use that resolution because the FLIR camera is also connected to a Raspberry Pi through an adapter (HDMI / CSI) that only supports a 720p60 or 1080p25 hdmi input. So the only input resolution of the adapter compatible with the FLIR is 720p60.
3a8082e126