|
Post by aurel on Jan 19, 2022 4:47:46 GMT -6
ahh you talking to johnno about sdlbasic update
|
|
|
Post by johnno56 on Jan 19, 2022 18:01:39 GMT -6
Sorry to take so long in replying... Hmm... Benchmarks between RC and SDL? Nope. Although both languages are similar, I mainly duplicated programs for each, to see if I can convert them. But no benchmarks. Is there a standard set of benchmarks to be used or are the same programs run on each and compare the times? Here is a link for Rugg/Feldman benchmarks specifically for Basic. Tom Rugg and Phil Feldman are co-authors of Basic programming books back in the early 80's en.wikipedia.org/wiki/Rugg/Feldman_benchmarksAlthough they use line numbers it should not be too hard to convert them and add a built-in timer. It would be interesting to see how SDL and RC compare to their posted list of times... Could be fun?
|
|
|
Post by johnno56 on Jan 20, 2022 13:20:25 GMT -6
this topic is about qb64 but who care they have a CEO johnno when we talking about sdlbasic rcbasic did you ever made any benchmark ? same task and then measure time for that task Well. I converted all of the Rugg/Feldman benchmarks for both SDL and RC. Compared to the tests that were run "back then" on "their current machines", I figured that it would not be a fair test, to compare my machine with theirs... After all, their fastest machine, had a 3.6mhz CPU... The test was simple. Set a 'start' timer. Perform the test. Set a finish timer. Subtract finish from start to get result in milliseconds. Each test was run 3 times then recorded the average. Totalled all 8 times and produced an average score. RC ran all 8 benchmarks with a total of 56.31 ms for a total average score of 7.038 ms SDL ran all 8 benchmarks with a total of 4.97 ms for a total average of 0.621 ms If you want to compare them with the original test, visit the link, then realise how fortunate we are that our "old clunkers" are 'way' faster than theirs... lol en.wikipedia.org/wiki/Rugg/Feldman_benchmarks
|
|
|
Post by aurel on Jan 20, 2022 13:38:20 GMT -6
well so according to that sdlbasic is faster than rcbasic but when i tried for example shooter galaxy ranger i see that rcbasic perform really well and fast ..i cannot remember sdlbasic apps and i ddon't like sdlbasic because have impact on registry in windows case .
|
|
|
Post by rosy on Jan 20, 2022 15:45:22 GMT -6
Jessu ... What is this test that lasts 50 ms ...
|
|
|
Post by johnno56 on Jan 20, 2022 15:58:06 GMT -6
Aurel,
SDLBasic is faster than RCBasic when running *those* tests... When it comes to 'overall' performance, RCBasic, is considerably faster. Because the languages are similar, I figured it would be a fair test, to see which one is quicker...
Just like most 'Basics' each one has it's 'better' abilities...
The Rugg/Feldman tests did not include... Graphics; Sound; File read/write etc... Those 8 tests plus these would give a better overall 'picture' of each Basic's performance... Perhaps we could create a set of benchmarks? Might be fun? Any thoughts?
|
|
|
Post by rosy on Jan 21, 2022 1:55:14 GMT -6
|
|
|
Post by aurel on Jan 21, 2022 6:49:20 GMT -6
i don't have sdlbasic but i think that my stupid benchmark will show how sdlbasic respond i mean how much time need to fill string array and print each number.
|
|
|
Post by rosy on Jan 24, 2022 14:08:07 GMT -6
Here is a fairly interesting test of calculating prime numbers: youtu.be/iUmWy27Yj1k?t=648The same thing starts with integers without INT and SQR
|
|