next »

[guest]

That would be the best solution, but the SLDB doesn't provide loop point time. (And even if it did, both song length and loop point times would have to be way more accurate than it is now.)

OK, but now when eg. XMPlay's always loop is active, looped song is just restarted from beginning instead to be smoothly continued.

[Fraggie]

That's because the plugin doesn't know where is the loop, so it can't jump to the loop point when the song ends.

You can enable "add fade-out time for looped songs" and set fade-out. This way when the song ends, it will be looped automatically (by the sid player program), and the looped part will play for the fade-out time. You could also fiddle around with "repeat short songs x times" to achieve correct looping.

(Maybe guys from HVSC could add loop point feature in their song length detector...)

[Fraggie]

http://www.c64.org/HVSC/MUSICIANS/M/Maktone/Golden_Years.sid
has distorsions e.g. from 0:13 when SID filters are enabled.

Frankly, I don't hear anything wrong with it... Does someone else get any distortions?

[guest]

http://www.c64.org/HVSC/MUSICIANS/M/Maktone/Golden_Years.sid
has distorsions e.g. from 0:13 when SID filters are enabled.

Frankly, I don't hear anything wrong with it... Does someone else get any distortions?

Compare eg. with sidplay2w.

[Fraggie]

I'm sorry I still don't get any significant difference. The song is rather loud, so you may try switching off EQ Reverb and Auto-Amp. You can also poke around with surround and panning. Please also make sure you're using the latest plugin version.

If nothing helps, send me your xmplay.ini (via e-mail or PM).

[guest]

Set 100% pan separation, disable groovy surround, enable SID filters and compare right channel.

[Fraggie]

Ok, I hear the problem now. It appears when SID filters are enabled and external filter is disabled.

Until I find a solution you can enable external filter. With auto setting it doesn't take much more resources, so I don't see a reason to disable it in the first place.

[guest]

Plugin doesn't load songlengths for:
http://www.c64.org/HVSC/MUSICIANS/B/Blues_Muz/Reel_Fishing_preview.sid
Also aren't subsongs 33-44 VBI?

[Fraggie]

It seems SLDB entry for this file is missing! (At least I couldn't find it in Songlengths.txt.)

Why do you think these subsongs should be VBI? If I remember right, the PSID file format supports speed information only for the first 32 subsongs. All further songs get the same speed as subsong 32 (which is CIA timer interrupt in this case).

[guest]

It seems SLDB entry for this file is missing! (At least I couldn't find it in Songlengths.txt.)

I've found in HVSC 46:

; /MUSICIANS/B/Blues_Muz/Reel_Fishing_preview.sid
a201d88d9c2ca0067f9e93202cfc4a30=1:34 1:57 1:02 0:44 0:40 1:07 3:05 1:08 1:09 0:22 1:10 1:21 0:04 0:02(G) 0:01(G) 0:01(G) 0:01(G) 0:18 0:18 0:18 0:18 0:02(G) 0:02(G) 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:08 0:02(G) 0:01(G) 0:02(G) 0:01(G) 0:01(G) 0:02(G) 0:02 0:01 0:01 0:16 0:16 0:13 0:15 0:15 0:15 0:01(G) 0:04 0:04 0:01 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:02(G) 0:02(G) 0:02(G) 0:01(G)

Why do you think these subsongs should be VBI? If I remember right, the PSID file format supports speed information only for the first 32 subsongs. All further songs get the same speed as subsong 32 (which is CIA timer interrupt in this case).

sidplay2w displays VBI for 33-44 and CIA for 45-57

[Mr.Eeh]

I know it's a bit off-topic, but I gotta tell you all sidaholics, that one guy recorded the whole HVSC into mp3 on both sidchips..

check www.6581-8580.com

[Fraggie]

It seems SLDB entry for this file is missing! (At least I couldn't find it in Songlengths.txt.)

I've found in HVSC 46:

; /MUSICIANS/B/Blues_Muz/Reel_Fishing_preview.sid
a201d88d9c2ca0067f9e93202cfc4a30=1:34 1:57 1:02 0:44 0:40 1:07 3:05 1:08 1:09 0:22 1:10 1:21 0:04 0:02(G) 0:01(G) 0:01(G) 0:01(G) 0:18 0:18 0:18 0:18 0:02(G) 0:02(G) 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:08 0:02(G) 0:01(G) 0:02(G) 0:01(G) 0:01(G) 0:02(G) 0:02 0:01 0:01 0:16 0:16 0:13 0:15 0:15 0:15 0:01(G) 0:04 0:04 0:01 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:01(G) 0:02(G) 0:02(G) 0:02(G) 0:01(G)

Right. I took the wrong database. Anyways, you probably noticed that MD5 fingerprint from the SLDB is different to the one calculated by the plugin. That is probably because of the speed issue you also mention.

sidplay2w displays VBI for 33-44 and CIA for 45-57

I'd say sidplay2w is wrong with this.

Here's why. There are two speed setting interpretations for tunes above subsong 32. First method uses speeds from songs 1-32 for songs 33-64, 65-76, etc. Which means song 33 gets speed of song 1, song 34 gets speed of song 2 and so on. This method was used by the orignal PlaySID. When using the second interpretation, all songs above number 32 get speed of song 32. This method is used with PSID2NG sids unless a flag states the old method should be used. (Actually the first method is a result of a bug in the PlaySID. It was intended to behave like the second method!)

To make this long story short -- all PSID v2 files should use the new method unless it is explicitly specified in the header. The old method should be used also for PSID v1 files. Your file is in PSID2NG format and flags in it's header don't say that the old method is better.

I'd guess that sidplay2w and HVSC's songlength tool aren't completely PSIDv2NG compliant. (This new (correct) speed interpretation was introduced to libsidplay2 in May 2006, while the latest sidplay2w version is from June 2005.)

I know it's a bit off-topic, but I gotta tell you all sidaholics, that one guy recorded the whole HVSC into mp3 on both sidchips..

A great source of reference!

(Timbaland could've waited a bit, with this site composing would've been even easier for him.)

[Knurek]

(Timbaland could've waited a bit, with this site composing would've been even easier for him.)

Never thought I'd see someone mention Timbaland and composing in one sentence...

[deus-ex]

Never thought I'd see someone mention Timbaland and composing in one sentence...

[Fraggie]

Sorry.

[guest]

Some MUS files aren't played - check eg. several tunes in BJ_Pools directory from CGSC:
http://www.btinternet.com/~pweighill/music/CGSC_v113.7z.001
http://www.btinternet.com/~pweighill/music/CGSC_v113.7z.002
http://www.btinternet.com/~pweighill/music/CGSC_v113.7z.003

[saivert]

On a Pentium III 700MHz system the XMLPlay program consumes 50% CPU time when playing SIDs.
sidplay2/w (stand-alone SID player for windows) only consumes around 12%.

How come? Both are using full C64 environment and the same engine. Is there any overhead when playing through XMPlay's output engine?

[Fraggie]

Switch the plugin to mono mode.

[saivert]

I might have seen this the wrong way. I thought the XMPlay SID plugin was written badly such that it was inefficient. I have used Process Explorer (which is way better than task manager) and discovered that in fact sidplay2/w has the same CPU utilization during playback, but still sidplay2/w does not make the mouse cursor lag.
I think this is down to XMPlay SID plugin not yielding enough to other threads/processes during playback. It kinda holds on to the CPU for far long while sidplay2/w more easily lets other processes use the CPU too.
This should be easy to fix.

[saga]

If i check "Repeat short songs [   ] time(s)", the actual value will be saved when XMplay quits, but the checkbox before that option will be unticked again when I restart XMplay, so songs' won't be repeated anymore