I wouldn't have 20 read relays. They are only used for picking up notes that tag you. You only need maybe 2. Gossip reads from other people's write relays, you don't get to choose where it reads from (the people you follow do).
Discussion
reduced to 4 relays. faster.
but still a lot of "RECOMPUTING", around tens of seconds to a few minutes to load a thread/previous post. I mostly give up.
after 30 min online, it becomes faster to load threads/previous posts, but not always..
it is not just a internet issue because image fetches are fast (a few seconds at most), I never leave them on auto, always click when I want to see them
I know asking is easy and you are a lone developer.. but..
separate tabs for each request with specific status/icon messages for each one would be the best.
Or at least some queue drawing or general statistics to give me an ideia if I should give up, get a cup of tea, etc.
just a single status message on the bottom left is confusing, I forget to clik on it and it is not clear to which request it is refering to.
I am in South America using Free VPN hosted in Europe or USA.
I've put some work into performance improvements on the lmdb branch. I will stabilize the format soon. If you try it now, you will have to wipe the lmdb driectory and reimport again later.
I also added a status bar to the bottom that shows the total number of events, and the HTTP activity (in flight / total being sought). That helps a bit to see what is going on... if your event count is climbing, then it's bogged down processing incoming events. That is common when you first startup after a long period and there are lots of events to catch up on. If the HTTP count is very high, then it's pulling down lots of images. I should also add relay count I guess.
You might also want to set your FPS to around 5. Maybe 10 if it can handle it. Don't try to run a high FPS. I may change settings to cap FPS at 30 instead of 60. I've been using 10 for a long time and it's smooth enough for me.