| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
HVSC SID Playtime Database ?
I am currently coding something Sid related Tool for PC and was wondering if there exists a Database (csv,xml,or whatever) that contains the default Sid Subtune Playtimes.
At the Moment i use Siddiag to recieve the Sidtimes, but this very annoying and time intense. If there would be a DB-File where i could search with the MD5 of the Sid that would speed up Time detecting a lot.
Anyone knows about the existance of this or something simmilar ? |
|
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
what about Songlengths.txt? |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
Thanks, didnt even know that existed.
Shame on me |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
Don't worry, 99.9973% of users don't even know DOCUMENTS/ exists :P |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
I just checked the md5 hashes,
e.g.
; /MUSICIANS/H/Hubbard_Rob/Commando.sid
a08c08aa1c74990c7d753e8c1d2f023e=3:57 1:02 0:06(G) 0:02(G) 0:01(G) 0:01(G)
i tried different MD5 Tools and always get
c4c5ff8cfefdf683c50e66775cfac1ee
Is the md5 calculated in the songlengths.txt without the Sidheader probably ? |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
It is calculated by actually playing it, not by summing the file bytes, else it would change even by fixing the title field, or replacing it with a relocated version.
I'll let Wilfred explain _how_ if he wishes. Maybe it's time to update Songlengths.faq with some more infos. |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
Hmmm sad, so there is no real easy or fast way to detect the Sidlength from songlenths.txt with just creating a md5 hash. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
uh - the md5 is of course calculated from the binary and not "by playing it". see here: https://tnsp.org/hg/sidinfo/file/263cd25749a1/sidlib.c |
| |
iAN CooG
Registered: May 2002
Posts: 3132 |
hmm ok, I was probably thinking on the hashes in sidknown's database.txt |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
It might be a good idea to have playing md5 as default , but beeing able to switch to file (bytes) md5 in siddiag. That would help me a lot right now. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
wat |
| |
Mr. SID
Registered: Jan 2003
Posts: 421 |
There's a lot of discussion about calculating the MD5 in this thread (at the bottom): Track Times in SID files? |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
I already saw that thread, and also understand the complains about md5 beeing more acurate when Playing a Sid.
But for my current Project i need the md5 created on the Filecontent (summing Bytes), which is completely different. I just wanted that because i didnt want to start siddiag all the time for each single Sid i want to have in my Database.
Nevermind, i have created the md5 Database with the Layout i wanted myself, instead of complaining it isnt there, so i dont have to lick anybodys Balls or wait Days/Weeks until it`s done (if ever)
And for the ones that are not found due better or older Rips, Header Change or whatever i index them with siddiag on runtime.
But thanks for guiding into the right Direction :-) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
why dont you just use the functions from the source i posted above? lol |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
lol ? probably you dont expect there are people not coding c , i am not into c at all.
but instead of porting that code to my project i decided to use 17 lines of code to merge sidlenghts.txt and my previous created hvsc67.md5 into a new songlength.md5
as previously said, thanks for guiding into the right Direction :-) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
/o\ |
| |
TheRyk
Registered: Mar 2009
Posts: 2062 |
Fred lives here
he knows best.
Contact him with whatever issues about songlenghts.txt |
| |
Mr. SID
Registered: Jan 2003
Posts: 421 |
Again, in case the message didn't get through yet:
The MD5 in Songlengths.txt is based on the .sid file contents, not on the playback result. |
| |
lA-sTYLe
Registered: Sep 2003
Posts: 63 |
Sorry to disappoint you.But You can Check yourself. They are Not
atleast not with Files i tested. But will to be 1000% sure i will test again this evening |
| |
Fred
Registered: Feb 2003
Posts: 284 |
The MD5 hashes in the Songlengths.txt file are based on the content but not on the full content. You need to know which fields in the SID header needs to be included in the MD5 calculation and how to handle them and in which order, in order to get the right MD5 hash.
The MD5 calculation was done in the time that we had to compare a lot of SIDs against the HVSC collection to filter out double entries. There was a need that SIDs with a different title, author or released field would produce the same hash. The same MD5 calculation was used for the Songlengths database since the tooling was already there.
For a lot of people the calculation is confusing and unclear. Having the MD5 hash based on the full content makes more sense and we're considering it for changing it since there is no need for having this kind of MD5 hash for the songlengths database. Changing it means also that current applications and tools need to be updated. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
i miss engraved64 in this thread |
| |
Compyx
Registered: Jan 2005
Posts: 631 |
Quoting Groepazi miss engraved64 in this thread
lol, that would seriously derail this thread. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
this is not usual method |
| |
Fred
Registered: Feb 2003
Posts: 284 |
As of HVSC#68, we have 3 different formats for the song lengths:
- Song length database (old format)
- Song length database (new format)
- SSL files for players that run on C64 hardware
Have a read here for details: http://hvsc.de/download/C64Music/DOCUMENTS/Songlengths.faq |
| |
Compyx
Registered: Jan 2005
Posts: 631 |
There's something a bit unclear in that document:
Quote:Each song length is of format:
mm:ss
where mm and ss are numbers and mm never starts with a 0.
I assume this is supposed to mean that mm should not have leading zeros, but the way it is worded now means you cannot have song lengths of under a minute. |
| |
Fred
Registered: Feb 2003
Posts: 284 |
You're right. It should never have a leading zero. Thanks for noticing. I will update the document in the next update. |
| |
Compyx
Registered: Jan 2005
Posts: 631 |
Cool. Pedantry has it uses :) |
