Ggs samusko. At first i was trying some stuff i saw Scatt do recently from when he came first but thought i wasn't doing it right cos i was struggling but then my usual tricks weren't working either. Your Cloud was untouchable and Sonic was also really hard for me to hit. You should go to a tourney i reckon you'd do quite well!
Now im nervous about getting 13th at UQ again lololol a few people are picking up Cloud.
I did some Apache Bench benchmarking on the old server against the new server. I don't really expect everyone to understand this but in general you want smaller numbers here to prove that the site is now loading faster. It definitely feels faster to me.
Tests used 100 requests with 10 concurrent users at a time.
Connection Times (ms) min mean[+/-sd] median max Connect: 209 237 32.2 223 364 Processing: 2937 10216 2481.0 9840 16519 Waiting: 967 8430 2483.6 7979 14808 Total: 3145 10453 2483.5 10064 16756
Percentage of the requests served within a certain time (ms) 50% 10064 66% 11253 75% 11799 80% 11973 90% 14032 95% 16358 98% 16500 99% 16756 100% 16756 (longest request)
min mean[+/-sd] median max Connect: 208 221 10.4 217 246 Processing: 215 1890 354.2 1985 2032 Waiting: 215 1156 551.4 1126 2024 Total: 429 2111 355.5 2208 2248
Percentage of the requests served within a certain time (ms) 50% 2208 66% 2225 75% 2228 80% 2236 90% 2240 95% 2242 98% 2246 99% 2248 100% 2248 (longest request)
Also did a bunch of more manuals test, loading of the Online page ten times in the browser and recording the average load time and latency. Here's another good result!
OLD SERVER Average page load: 2.1 sec Average latency: 1.75 sec
NEW SERVER Average page load: 0.57 sec Average latency: 0.53 sec
PAGE LOADS SEEM TO NOW BE UNDER ONE SECOND YES THIS IS EXCELLENT I'M SO HAPPY
I must give a big thank you to Kira who spear headed this entire operation and who without I wouldn't have been able to complete this <3
I don't think the lowest tier is an option based on site traffic, so it'd probably be the second tier option at around $250 a year. It's also an ongoing cost too instead of something that's a one off. It's definitely on my list of things to do but I got a few other things to pay for first before I start investing more in this site on stuff like that.
cAKE said: Suggestion: Like Leaderboard? Kinda like the badge leaderboard but it would just be something fun and small
When you implement a feature on a website you need to think beyond the design and functionality of the feature and into what kind of human behaviour and dynamics it will evoke from your users.
For example, the primary reason Elo was added wasn't actually to rank players because if it was it would be considered a failure to me. It was actually implemented to give instant objective feedback to players after tournaments about their skill progression, to encourage TOs to submit and thoroughly tag data and to hold higher level players accountable for when they do things like sandbag in pools with low tier characters. Even if the national rankings aren't completely accurate, because these are the goals of the Elo system in my eyes I consider it successful.
I can't really think of any kind of success which would come from a likes leaderboard other than people making spammier post and trying to whore themselves out for likes which already happens to a degree already. Having a leaderboard for this kind of thing would only encourage that behaviour further and benefit a select few people who would be at the top of the leaderboard and who probably already know they're quite popular on the site already.
I do have some loose plans for a popular content section which shows posts and shoutouts what have gotten a high number of likes but I think the leaderboard idea may be a bit too far for the time being. Thanks for the suggestion though, I just wanted to explain my reasoning so you can understand where I'm coming from.
Linoone said: I think that if you take a game you should lose less Elo
I completely agree with this and honestly this is how the Elo system was first designed. If I went 2-3 with Jaice then I feel like I should lose less Elo than if I went 0-3 with him.
The problem with this is that not every TO submit tourney results with the score data included; some do, but others just record win/loss and there's no way to get this data after the tourney finishes. What I wanted to avoid was a situation where some tourney results are more valuable than others (the ones with scores recorded) because it would be arbitrary from a player's perspective whether their 1-2 loss to a top player would be counted as it would be all down to the luck of if the TO remembered to record this data.
I decided that I had to take the lowest common denominator for both of these situations which was Elo ranking a match as a single unit instead of by a game breakdown. While I agree with you in theory, in practice it's just not practical or fair.
LachlanF said: I know it's an idea that has been posted before, but an ability for users to upload directly into the comments for when photos are being used, without going through the QuickDraw images, and potentially some way of making uploading photos from a mobile device easier.
This is a nice to have but is kinda tricky because it involves AJAX file uploads. I'll put it onto the wish list.