Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: randomness of Serial #? (Read 3710 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

randomness of Serial #?

not a request for anything, just something I had noticed pertaining the randomization of the generated serial numbers.

was going to post before and ask, Just how random is the serial number.
but I found that out already, its random enough BUT, after encoding some 3,000 plus tunes and entering the metadata into a SQL, and wondering why I had more records in the CSV than in the SQL, i tracked it down to being a problem with dups in the serial number, (which i was using as a UNIQUE ID, which I later fixed that problem by adding a second UNIQUE ID, the playtime length PLUS serial.)

now I dont have the problem, since I have been using my own serial series.

which happens to have started at 9,000,000
which also permits identifing newer encoded oggs, over previous ones.

but the question i was wondering about, is.

would it be possible to generate a minimum serial number such as a 10 digit min, OR the addition of letters.

I have tried to use lettes in a serial number, but no go, except for 1. 1 ogg I encoded, which happened to have been a 90 minute concert, and it contained the letter "e" and the "+" sign. not sure how that had happened.

i think the serial number maybe could use alittle change, since I had encountered about 20 with the same exact numbers, and its simple to figure out how this happened.

since each number is generated randomly, EXCEPT when doing a batch encode.
therefore if already the number -  1744 exists  for a serial #,  and I start to encode a batch, which the first generates its own "random" starting number 1740 and increments +1 thereafter, its easy to get an overlapping number, after 4 encodes the serial would be 1744, same as a previously encoded file.

I know its an easy "fix" for me, to just use my own serial sequence, which I am doing from now on, but since I have to encode on command line, in either *nix or dos, its a slow process to type out 50 or so song file names and not misspell a couple now and then, which stops the encoder, and I have to retype.

Now, if someone made a GUI which included the additional field for the seiral, that would be cool.

and I am sure if someone uses this for streaming, they might be unaware of colliding serial numbers, and if someone requests a song, which is marked with an identical serial number as another song, which song is going to be streamed to them?

thanks for the suggestions.

--also, by the way, waiting for RC4, need the change in Vendor date, for ID purposes in SQL. would help : )

......Locust

randomness of Serial #?

Reply #1 –
Quote
Originally posted by LocustFurnace
was going to post before and ask, Just how random is the serial number.

It's random enough that multiple streams in file have different serials.

Quote
would it be possible to generate a minimum serial number such as a 10 digit min, OR the addition of letters.

The serial number is stored as long,  so it can only store 10 digits. And letters are not possible.

Quote
I have tried to use lettes in a serial number, but no go, except for 1. 1 ogg I encoded, which happened to have been a 90 minute concert, and it contained the letter "e" and the "+" sign. not sure how that had happened.

The serial number was too long to be displayed properly and your program displayed it  in exponent notation. For example 1.4e+30 means 1.4 * 10^30.

Quote
i think the serial number maybe could use alittle change, since I had encountered about 20 with the same exact numbers, and its simple to figure out how this happened.

It's not intended to be unique, it's only used for separating multiple streams inside ogg.

randomness of Serial #?

Reply #2 –
Each serial number is unique to an Ogg Logical Bitstream page, within a set Ogg Physical pages. They are only randomized to an extent to serve the purpose of being able to place concurrent multiplexed streams in a contigious order or sequentially multiplexed streams in a non-contigious order, which will eventually become useful for having multiple concantenated streams running in parellel in the future.
budding I.T professional

randomness of Serial #?

Reply #3 –
thank you for the quick replies!

I'll just keep on making em with my own serial sequence for use as a sql Unique ID.