Is there a technical reason that it doesn’t work on the THETA V?
It may increase your user base if it supports the V as well.
If you adjust the Android manifest to specify that it works with the V, people will be able to download and test it on the V. Alternately, you can make an apk of the plug-in available for private testing to see if it works at all, then add that “V” support string to the plug-in so that it appears in the store that way.
Some modification would be required for sure. Do you know the number of cameras sold? 
I would probably have to get a V camera but it has different gpu, encoder, etc. it was a bit hard to get here. I looked around and did not see any V available in Hungary, all sold out. 
I will think about it and look if could find. Initially I developed code accordingly, at least I thought so, but there were so many changes on the road. Also when I sent initial plugin to Ricoh they rejected as it wasn’t working on V. In H.264 it should work in most basic setup, but not sure about heating, etc. you could borrow me a device to spend a day or so to see, but I don’t know anyone using it nearby. May be after I finish this update in next round I will include in a few weeks…
Thanks for the update. the V is end of life and no longer sold. So, it’s likely not worth a serious development effort. I was just hoping it would work without modification.
no, sorry. It should work only with modifications and I’m not sure about effort yet.
About WebRTC, I’m actually looking into SRT, which should provide higher quality. Support is lower, but I would like to see how it works and latency. Now with RTMP and RTSP I was able to get about 7-8 second latency, compared to 30 seconds at youtube, etc.
What I see would be the top requirements further:
I could draw a nice chart, but clearly, these requirements are in contradiction with each other.
As the “best” solution to try, I’m trying to stream SRT from Theta Z1 (it seems more reliable and higher quality compared to WebRTC and slightly probably higher latency, but it should be manageable).
Now clearly need to transcode SRT to web, to make it watchable on any device through browser, so I’m planning to do SRT from camera to my server and from my server to distribute I’m going to deliver WebRTC from server to viewers. This could provide around 1 second latency, but same quality and FPS and security as with RTMP/RTSP. This is still just an estimate it’s not final yet at all, I mean this 1s latency…
But is there a real difference: 1s latency compared to 8-10s ? I’m thinking about use cases and I’m not sure. 1s is not 0.1-0.2s, it’s much more… So, for controlling a robot it’s just not good enough, will have to see, but it will require some WORK to be able to see and measure at all.
I think the use case is to sync up interactivity for things like chat or even sports gambling. I’m not sure what the market is for this type of group interactivity. Theoretically, it’s when you want people to ask questions that can be answered as the event is happening. There might be a simultaneous broadcast TV.
yes, online auctions, interactivity, gambling for sure. I don’t think there is any other 360 camera that could work by itself, handheld and small, etc. without additional equipment. I think SRT will also work, I’m just not sure if in this update or next one.
I do not know what the market is for interactive 360 video. However, I think you should consider contacting existing companies like Millicast, Boxcast, maestro, broadpeak, Anevia, and many others.
Once you have a way to sell your plug-in, you could ask them if they will help you promote it if you implement support for their platform as an option. If they respond with interest, then consider producing a bare bones feature versions. For more comprehensive features, they would need to pay for your development.
This is just a brainstorm for ways to help you get more exposure for your plug-in and for your FlowTours business.
thanks, Craig! will reach out to them once plugin is there and doing tests with wider audience.
hi! to give an update here… I was able to make SRT work, I was able to stream from Hungary location to my server in Germany and view the stream on multiple devices in Hungary. Latency was around 1-1.2 seconds, however for some reason I see some buffering issue, video wasn’t smooth enough, but it was probably related to spatial audio. It’s worth working on this approach. It’s powerful, because I transcoded SRT incoming source into WebRTC like I suggested a few days back and it worked.
About H265, now I’m streaming via RTMP, H265 encoding, only 12mbps upload bandwidth and running in warm room for 2.5 hours already.
I’m not sure when to stop these improvements, when I open a “door” I have to go in and look around a bit… But I would really like to pack these great improvements into my plugin and send to Ricoh.
Another note: I think I will be able to correct a bit stitching issue in very good looking HDR mode in camera directly, hope will not loose much FPS (or not at all).
Ricoh sohuld build the same cam again, lens is great, but with newer snapdragon GPU, capable of 8k encoding (or at least 6k) with 30-50% more cpu power. And supporting camera2 API…could gain some more FPS if no preview is needed and I couldn’t find a way to encode without preview with camera1 api.
I didn’t get to adaptive bitrate yet…
bye