Is There Any Shuffling Option ... Like in winamp

Support requests related to player usage

Postby mcgyver83 » Tue Jul 10, 2007 3:08 pm

I have the same problem...some song are preferred on other...mah...
I don't know why... :-(
mcgyver83
 
Posts: 8
Joined: Tue Jul 10, 2007 1:43 pm

Postby Banana » Sat Aug 04, 2007 4:46 pm

@guest

You describe what I refer too as perfectly random. Now, as I'm continuing to argue, S1mp3's are NOT perfectly random, so there is no 1/20 probability for all 20 songs, but different ones for each:

The point is it IS NOT random in that sense that the player does favour certain songs over others. Try it a week long. You will see that statistically some songs are played much more often than others. Repeat it and see you get the same result. You must not add/remove songs during this time.
Banana
S1 ++ poster
S1 ++ poster
 
Posts: 383
Joined: Wed Mar 28, 2007 6:10 pm

Postby xyzzy » Wed Aug 08, 2007 4:19 am

Banana wrote:The point is it IS NOT random in that sense that the player does favour certain songs over others. Try it a week long. You will see that statistically some songs are played much more often than others. Repeat it and see you get the same result. You must not add/remove songs during this time.

This behavior is inherit in playing totally random songs. We, as humans, don't have the capacity to track random behavior over a vast number of iterations and will start to interpret short term variations in the playing frequency of each song as patterns that are not really there.

Here is a small random song simulator written in Perl that simulates 100 random plays from a selection of 20 songs as mentioned above:
Code: Select all
#!/usr/bin/perl
use strict;

my %results = ();;
my $range = 20;
my $plays = 0;
my $max_plays = 100;

do {
   my $random_number = int(rand($range)+1);
   $results{$random_number}++;
   $plays++;
}  until ($max_plays == $plays);

foreach my $song (sort {$a <=> $b} keys %results) {
   print "Song $song has been played $results{$song} time(s)\n";
}

And here are the results from when I ran it the first time:
Code: Select all
Song  1 has been played  4 time(s)
Song  2 has been played  4 time(s)
Song  3 has been played  4 time(s)
Song  4 has been played  8 time(s)
Song  5 has been played  6 time(s)
Song  6 has been played  3 time(s)
Song  7 has been played  5 time(s)
Song  8 has been played  3 time(s)
Song  9 has been played  5 time(s)
Song 10 has been played  5 time(s)
Song 11 has been played  1 time(s)
Song 12 has been played  7 time(s)
Song 13 has been played  2 time(s)
Song 14 has been played  2 time(s)
Song 15 has been played 13 time(s)
Song 16 has been played  9 time(s)
Song 17 has been played  5 time(s)
Song 18 has been played  6 time(s)
Song 19 has been played  7 time(s)
Song 20 has been played  1 time(s)

Is this totally random? Yes. Would it seem random to a human listener? Not so much. This simulated player appears to love song #15, while song #11 and #20 hardly got played at all.

Now, with as few iterations as 100 we see that the most played song got 13 times as much play as the least played song. If we increase the number of plays from 100 to 100.000, then each and every song will converge towards getting 5% playtime. However, a human can't keep track of that much, so we will still be imagining patterns that aren't really there based on a much more recent subsample in which the variation in the number of plays will be comparatively larger.
xyzzy
 
Posts: 2
Joined: Wed Aug 08, 2007 3:32 am

Postby Banana » Thu Aug 09, 2007 1:27 pm

Well, seems somebody should make tests with 100000 occurences. I know random doesn't mean that strange things do not happen (it is possible for example that the same song is played 100000 times), but I also know that S1mp3 have especially bad firmware and that creating random numbers on a computer isn't that easy.
Banana
S1 ++ poster
S1 ++ poster
 
Posts: 383
Joined: Wed Mar 28, 2007 6:10 pm

Postby xyzzy » Thu Aug 09, 2007 4:14 pm

Heh. Without programming information on the Actions chips I can't say for sure. It's of course possible that they screwed it up.

If someone really wanted to test, I guess they could make 20 distinct really short music files using a test tone generator, hook the player up to some music analyzer software via line-in on a computer. That should get to 100.000 samples within a couple of days and give an automated way of counting how many times each were played... :)
xyzzy
 
Posts: 2
Joined: Wed Aug 08, 2007 3:32 am

hey

Postby hiddenlink » Fri Aug 10, 2007 8:07 am

again, as i've said before, random is not random if it picks perfectly.
mp3 Player fanatic
lovers of killer smiles (",)
iPod nano Clone 1gb v. 9.0.48
i cannot modify its firmware, can anyone give me the tool.
User avatar
hiddenlink
s2 poster
 
Posts: 115
Joined: Tue Jul 04, 2006 2:04 am
Location: Philippines. The side of the Underground

Postby Banana » Fri Aug 10, 2007 2:09 pm

:shock:

What are you trying to tell us again, hiddenlink? By the way, if it "picks perfectly" (what do you mean by it at the end?) it still CAN be random, because random INCLUDES the possibility to "pick perfectly" (wathever this means to you).
Banana
S1 ++ poster
S1 ++ poster
 
Posts: 383
Joined: Wed Mar 28, 2007 6:10 pm

hey

Postby hiddenlink » Sat Aug 11, 2007 5:58 am

i dont think youre getting it the right way buddy. the technical examples of xyzzy explains so perfectly what RANDOM means.

when it picks perfectly, then it is not random, it is selection.

now try this, maybe this explanation will works for you,

get a big bowl, small stones and a handkerchief.

now number all your stones, put it in the bowl. now tie your handkerchief into your eyes. pick one stone from the bowl, have somone to list the stone number, then place it back again in the bowl, repeat the process. (note: dont forget to stop coz random picking is infinite.)

well i dont know if everybody get the point of xyzzy coz it is so technical, scientific and mathematical. maybe that example is for the highly intelligent persons only, say the genius ones. bwahahahaha
mp3 Player fanatic
lovers of killer smiles (",)
iPod nano Clone 1gb v. 9.0.48
i cannot modify its firmware, can anyone give me the tool.
User avatar
hiddenlink
s2 poster
 
Posts: 115
Joined: Tue Jul 04, 2006 2:04 am
Location: Philippines. The side of the Underground

Postby Banana » Sat Aug 11, 2007 10:12 am

Of course I'm getting it right :D You're supposing that the S1mp3's follow exactly the theory, THIS I'm challenging. I'm saying, there are imperfections IN the player that make it not really random.

I newer told xyzzy was in any way wrong by the way. And all I said is that S1mp3's are very probably not random, just as I can prove my Sigmatel based player is not random at all (even not considering the putting back issue). You can easily bring the later one into a position where you can show that the next song played depends on when you hit the stop button. THAT is all my point.
Banana
S1 ++ poster
S1 ++ poster
 
Posts: 383
Joined: Wed Mar 28, 2007 6:10 pm

hehehe

Postby hiddenlink » Mon Aug 13, 2007 1:42 am

of course, s1mp3 follow the theory. what the random function in s1mp3 is the same thing, exactly, as the random defined in the wiki's, in the dictionaries, in the encyclopedias.

speaking of sigmatel based players, they are not in the same way as random, coz they are not actually random, they are the same as what iPod's do. FYI: iPod's chip is Sigmatel. So whatever the behavior of your Sigmatel-Based player is the same as the behavior of the iPod's.
(note: don't misinterpret behavior as to function coz functions differ with other players with the same chip but, BUT, behavior is the same to all of them.)(note again: what i mean with behavior is the way these chips play music from a set of functions fed in them. Don't you notice that your Sigmatel-Based player has the same noise quality with that of the iPod's? And also shuffles songs the way iPods do?)
mp3 Player fanatic
lovers of killer smiles (",)
iPod nano Clone 1gb v. 9.0.48
i cannot modify its firmware, can anyone give me the tool.
User avatar
hiddenlink
s2 poster
 
Posts: 115
Joined: Tue Jul 04, 2006 2:04 am
Location: Philippines. The side of the Underground

Postby Banana » Mon Aug 13, 2007 8:19 am

@hiddenlink
Following theoretical assumptions does not equal correct implementation of the random number generator!
Banana
S1 ++ poster
S1 ++ poster
 
Posts: 383
Joined: Wed Mar 28, 2007 6:10 pm

Postby IRONHEART » Tue Aug 21, 2007 8:24 pm

I would like to know if mp3 recognizes shortcut for another archives of mp3 like a mp3 archive. This can be a dummy question but this may solve the point to make a random way of all mp3s. If can recognize sorry for the question but i've never tested it but can anyone awser my question to solve my doubt.

if my english is bad. sorry again but don't guilt me. guilt the google translator.
IRONHEART
 
Posts: 3
Joined: Wed Aug 01, 2007 1:59 pm

Postby milad » Wed Aug 22, 2007 11:15 am

hi

NO S1 firmware can't recognize shortcut so you can't use this trick

bye
Model:Apacer AU822
Mainboard:K2TJ1061006305
Chip:2085H
Battery:230 mAh 3.7v
Flash Chip:K9K8G08U0M
Firmware:ACTS16D_37 3.5.38
Zhaleh Theme: http://forum.s1mp3.org/viewtopic.php?t=6725
User avatar
milad
Moderator
Moderator
 
Posts: 858
Joined: Mon Oct 30, 2006 9:42 am
Location: Iran-Sanandaj

Re: Is There Any Shuffling Option ... Like in winamp

Postby Sanley » Wed Jun 09, 2010 8:55 pm

hi, can anyone tell me how to put the shuffle on??
Sanley
 
Posts: 1
Joined: Wed Jun 09, 2010 8:48 pm

Previous

Return to Player usage

Who is online

Users browsing this forum: No registered users and 1 guest



cron