My analysis, for what it's worth...

Violet lines: uptrend since September
Green lines: short term downtrend
blue: very short term uptrend
yellow: resistance

the yellow line at ~2372 satoshis defines an important zone of resistance. Right now the price is stalled at around 2300. Buyers and sellers are at a crossroads. It's a moment of tension but look for a resolution by Monday. This week, smart people will be hearing about bitshares for the first time. They will take their time to study it before investing any money. That's why I have doubts about whether "this is it". I don't anticipate a frenzy into bts on Wednesday. I think the price might sink because a lot of people intend to pull their coins out of poloniex on Monday so they can have have them in their wallets for the snapshot on Tues.

If people pulled their coins out of the exchange wouldn't this mean lack of supply shooting price up?

 

Central bank interest? I couldnt find anything about that

We have updated the port used by graphene.bitshares.org to use 443.

We have also added improved user error messages (better than grey screen)

Thanks it works now.

Consider gzipping the css/js files and using a CDN (cloudflare is free, but perhaps the added optimizations are worth it to pay for especially RailGun, they reduce the time to load the page considerably and colocate the content for you so less load on your nginx servers).

Anyways consider the static files to be served by CDN and your index can be sent from server (bypass the CDN) so you can preserve cookies.

Currently its 2.5M to load the site which takes 4.42s gzipped it would be a fraction of that. Cloudflare can gzip on the fly for you, although best to pre-gzip and store as your release build.

Check the score: https://gtmetrix.com/reports/graphene.bitshares.org/6HVYreVy

Based on that you should:
1) Enable Gzip
2) Leverage browser caching (set a long time to expire these files to leverage local cache)
3) use a CDN for colocated loading.

https://bitsharestalk.org/index.php/topic,18751.msg242103.html#msg242103

can we run 36 core witnesses and see if we can hit 100k tps?

At that point, id like to try to set the block time to 10min and see what the new TPS is. Just for fun while we are there.

guess you got your answer CfB!

are we able to set up a bunch of witnesses with 36 cores (with > 10Mbit/s bandwidth) and test out higher TPS? We can pay $100 for a month for say 17 witnesses for a cost of $1700 to prove to the world we can do 100k TPS on a stable testnet.

Hi Guys,

I'm trying to understand the bottleneck for TPS on the BTS chain.

If I used bts purely for settlement reasons and didnt care for 1 second confirms, lets just say I used 10 min blocktimes would I be able to achieve a higher TPS than faster blocktimes today? If so, what sort of increase are we talkin about?

I'm assuming with 1 second blocktimes you have to start worrying about the latency to propogate blocks to majority of delegates so it will start to become a bottleneck to process transactions.

launches with gray screen, I'm guessing you tried to change the websocket port to 80 but it didn't work for some reason?

Since it looks like the html files are served locally, you can also try port 443 which is usually open behind firewalls.