www.bigbrainz.com and s3.amazonaws.com
To evaluate WebGL performance we strongly recommend a simple "Quick Check." This involves walking into a lab and simply clicking on the link on roughly as many computers as you expect to be loading concurrently and observe whether the game loads well enough. This will give you an excellent, real-world assessment. Alternately, you can consider the following and make an educated guess. Note up front that Firefox will currently reload significantly faster than other browsers.
Startup time is dictated by Download Time + Load Time. That is, how long it takes to download the content from the internet PLUS the time it takes the computer to then read all that code, load everything into memory, and start drawing it on the screen. We’re fairly accurate at showing “Download” time on screen. Load time is more complex and can include some things that are hard for us to track, measure, or display such as parsing code, or instantiating objects and such. Download time will depend on per-computer bandwidth. Load time will depend computer speed.
As mentioned, download time is determined by per-computer bandwidth:
At 1 Mbps our game will download in 2 minutes.
At 1 Gbps it will download in 1/10th second.
Individual computer bandwidth is determined by the school’s total bandwidth divided by the number of users concurrently downloading content. Note that once we download a game we have almost zero bandwidth impact—until we load the next game.
Example #1. If a school has a Gigabit connection and one student, they will download the game in 1/10th of a second. If the same school decides to start 1,000 students simultaneously at 10:00, it will take roughly 1,000 times as long for everyone to download, or two minutes.
Example #2. If a school only has a 5Mbps connection and one student, they will download the game in 24 seconds. If the same school decides to start 30 students simultaneously, it will take roughly 30 times longer than that to download.
Ideally the browser will cache the assets when it downloads them so that future downloads will be instant. If downloading is not fast enough the first time, try doing it a second time to see if caching is working. If it is, the download portion of startup should be instantaneous.
There are two types of caching. The core game is cached by the browser. We have little control over it. Generally the browser will try to hold on to often-used content so we SHOULD be cached effectively. But there are some practices that can set a browser to automatically clear its cache constantly.
Once you load the main game, each individual game is cached in a different location (Unless you’re on an ARM processor). It’s possible that some schools may have configurations that would wipe this out as well.
Once a computer has the content cached or downloaded, it needs to parse the code and instantiate all the objects. A blazing fast workstation will load a game instantly. The slowest chromebooks and cloudbooks we’ve tested load in about 30 seconds.
WebGL browser features are evolving very rapidly. As an example, version 29 of Firefox ran our game at 1 frame per second, while version 42 ran it at 40 frames per second. Expect performance to increase significantly throughout 2016 and 2017.
As of today Firefox caches parsed code and therefore can generally load the game up to six times faster than other browsers. This is obviously not an option on Chromebooks, but we strongly recommend using Firefox anywhere else.
We are developing a tool that should let allow some rudimentary remote troubleshooting. So far it will let us view download and load times.
Wireless routers and access points are notorious for not delivering the full performance that a school expects them to. If the access point is too far away its connection can degrade or be lost entirely. Also keep in mind that different wireless technologies have different maximum throughput.
802.11b = 11 Mbps Max (2.4GHz)
802.11g = 54 Mbps Max (2.4GHz)
802.11n = 600 Mbps Max (2.4GHz & 5GHz)
801.11ac = 1300 Mbps Max. (5GHz)
do I play?
Please watch the Walkthrough video at www.bigbrainz.com/Schools.php.
Progress reports are now available inside the game. While playing, simply press Esc--> Progress
Research study from Brigham Young University: link
Massive results research: link
What’s is “unlocking” all about?
Our research has shown that unless schools have four simple things in place, they will not get their money’s worth—i.e. 250 pre-tests, but only a handful of post-tests. “Unlocking” is just a simple 4-step checklist that ensures you have the simple things in place that will ensure your students are successful.
How do we unlock?
In the new version, just press the big, red "Unlock Deluxe" button. If you have to stay with the old version for some reason, just give us a call and we can walk through the four steps with you--or just email us that you've completed the 4 steps.
What are the 4 Steps?
- Have you designated someone to have ownership of fluency success?
- Has the Principal seen the Principal’s Video?
- Is Timez Attack installed and running smoothly?
- Have you scheduled Timez Attack?
(roughly 30 minutes/week . . . until Finished)
Can we switch to the new version if we already started our students on the old version?
No problem. The new version will figure out exactly where students should be within the first 5 to 10 minutes.
What if it’s not feasible for us to install the new version?
We understand that your technology requirements or limitations may keep you from installing the new version right now. No worries—just give us a call and we can walk through the unlocking process over the phone.