People love to hear how cool a protocol is, and EEs and CompEs are interested in the OSI model, and robustness, and collision detection, but in the end, laypeople want the answer to the question: How fast is it? I'm happy to present a fun little program that answers those questions. The CALCnet2.2 whitepaper is two sections from expanding from its current ten pages to completion, and I'm happy to have this program to help me generate graphs. It performs two-node unidirectional and bidirectional network saturation tests. In the following screenshots, I'm using 255-byte frames, the maximum per-frame size for CALCnet 2.2. The figures shown are bytes, ##/## means (data bytes)/(total bytes including headers). I will be testing with a variety of frame sizes and network members for the whitepaper, and I'll present my findings here as they progress.

Unidirectional test:


Bidirectional test:
Quite nice, Kerm. I look forward to seeing how it turns out Smile In my playing around, it's not quite as speedy as I thought it was going to be, but then again, it's a calc network Wink
merthsoft wrote:
Quite nice, Kerm. I look forward to seeing how it turns out Smile In my playing around, it's not quite as speedy as I thought it was going to be, but then again, it's a calc network Wink
The TI-OS is reported to achieve about 1KBps, and BELL self-reports 1.33KBps, so I consider the speeds decently respectable, considering that this is a network rather than a point-to-point protocol and has to deal with all such inherent dangers. Smile There's lots of pieces of timing that I can adjust that would drastically increase the throughput; the only danger is that as I adjusted timing downwards and removed some of the safety margins that I have built in, large, complex networks down to a simple-two calculator network would gradually get less reliable.
*bump* So it's maxing out around 3500 baud, which is not blazingly fast, but which I consider extremely impressive for a network protocol and the amount of resilience and checksumming that CALCnet2.2 provides. Check out the whitepaper draft for more:

http://www.cemetech.net/projects/techreports/tech004_calcnet22.pdf
That looks cool. The speed should probably be fast enough to play some moderately complex games, although maybe a large RTS that needs to send about 400 bytes of information every few frame would push it a bit too much Razz

Keep up the good work on this Kerm. Smile
DJ Omnimaga wrote:
That looks cool. The speed should probably be fast enough to play some moderately complex games, although maybe a large RTS that needs to send about 400 bytes of information every few frame would push it a bit too much Razz

Keep up the good work on this Kerm. Smile
Cheers DJ! If there was sufficient coordination between the various nodes in a 2-, 4-, 8-, or even 16-player game, and packets were well-crafted, I could certainly see this working even for an RTS game. Smile My thought is that anything that would work well over the TI-OS linking protocol would work equally well with CALCnet2.2, since it has the same order-of-magnitude throughput, but of course that's just conjecture since I haven't tried porting any existing games to CALCnet2.2 yet. OOOOHHH TETRIS!
More importantly, how is the latency? That matters far more to games than bandwidth.

Also, you need to make pretty charts.
Kllrnohj wrote:
More importantly, how is the latency? That matters far more to games than bandwidth.

Also, you need to make pretty charts.
I made pretty charts, didn't you look at the CALCnet 2.2 whitepaper? Razz It has lots of charts and colorful graphs:

http://www.cemetech.net/forum/viewtopic.php?t=5185
  
Register to Join the Conversation
Have your own thoughts to add to this or any other topic? Want to ask a question, offer a suggestion, share your own programs and projects, upload a file to the file archives, get help with calculator and computer programming, or simply chat with like-minded coders and tech and calculator enthusiasts via the site-wide AJAX SAX widget? Registration for a free Cemetech account only takes a minute.

» Go to Registration page
Page 1 of 1
» All times are UTC - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement