Crossfade feature in Music Disc Creator?
#1
Posted 29 October 2006 - 10:09 AM
I currently have EMC 8 which seems to have a bug (never resolved by Roxio) with the "crossfade" feature. Unlike previous versions (eg; EMC 7.5) it is NOT possible to execute a STRAIGHT OVERLAP crossfade (without the previous track fading out and next track fading in). Does anyone know if this feature has been fixed or improved with EMC 9?
Any comments appreciated!
-Mike
#2
Posted 29 October 2006 - 10:27 AM
Attached image(s)
---------
System 2: HP DV7 laptop, Turion II Dual Core 2.4Ghz, 4GB RAM, 640GB hard drive, ATI Mobility HD4650, ATI HiDef Audio, Windows 7 Home Premium 64bit.
Gary Russell
TNUSA
#3
Posted 30 October 2006 - 08:15 AM
ggrussell, on Oct 29 2006, 01:27 PM, said:
Gary, in Version 8, even set up as you show it, Music Disc Creator actually does a fade-in. It was one of those acknowledged bugs that had a thread at one point. Someone realized that the non-fading option could allow a clipping situation, and decided that they wouldn't allow it. I haven't had/made a chance to test it on Version 9 to see if it was fixed.
Beware the lollipop of mediocrity. Lick it once and you'll suck forever. - Brian Wilson
[GIGABYTE GA-MA785GM-US2H MB | Athlon II X3 440 (3.0 GHz) | 2GB DDR2 RAM | 1-500GB HD (C: XP, G: Win7, D: - Apps, E: data & apps), 1-500 GB HD Data) | 2 - LiteOn DH20A4P DVD burners | External Dell QFlix DX-20A6Q DVD +/- writer | Windows 7 | Creator 2010 | Tektronix Phaser 850 solid ink printers | Epson R220 Photo/Disc printer | Ricoh GX 5050n dye sublimation ink | Epson Workforce 1100 printer
#4
Posted 30 October 2006 - 09:49 AM
---------
System 2: HP DV7 laptop, Turion II Dual Core 2.4Ghz, 4GB RAM, 640GB hard drive, ATI Mobility HD4650, ATI HiDef Audio, Windows 7 Home Premium 64bit.
Gary Russell
TNUSA
#5
Posted 30 October 2006 - 10:08 AM
ggrussell, on Oct 30 2006, 12:49 PM, said:
Then maybe it's fixed! I'll have to try it when I get home.
Beware the lollipop of mediocrity. Lick it once and you'll suck forever. - Brian Wilson
[GIGABYTE GA-MA785GM-US2H MB | Athlon II X3 440 (3.0 GHz) | 2GB DDR2 RAM | 1-500GB HD (C: XP, G: Win7, D: - Apps, E: data & apps), 1-500 GB HD Data) | 2 - LiteOn DH20A4P DVD burners | External Dell QFlix DX-20A6Q DVD +/- writer | Windows 7 | Creator 2010 | Tektronix Phaser 850 solid ink printers | Epson R220 Photo/Disc printer | Ricoh GX 5050n dye sublimation ink | Epson Workforce 1100 printer
#6
Posted 04 November 2006 - 11:06 PM
#7
Posted 05 November 2006 - 03:10 AM
d_deweywright, on Oct 30 2006, 01:08 PM, said:
It's definitely not fixed. I crossfaded two tracks that have hard endings and beginnings. I had it set up as the top image below shows, with no fading, but when you listen to it, it sounds like the bottom settings. In fact changing it to the bottoms settings, with no fade-out, but a fade-in sounds exactly the same.
Beware the lollipop of mediocrity. Lick it once and you'll suck forever. - Brian Wilson
[GIGABYTE GA-MA785GM-US2H MB | Athlon II X3 440 (3.0 GHz) | 2GB DDR2 RAM | 1-500GB HD (C: XP, G: Win7, D: - Apps, E: data & apps), 1-500 GB HD Data) | 2 - LiteOn DH20A4P DVD burners | External Dell QFlix DX-20A6Q DVD +/- writer | Windows 7 | Creator 2010 | Tektronix Phaser 850 solid ink printers | Epson R220 Photo/Disc printer | Ricoh GX 5050n dye sublimation ink | Epson Workforce 1100 printer
#8
Posted 14 November 2006 - 09:03 AM
d_deweywright, on Nov 5 2006, 03:10 AM, said:
You are correct. This is definitely NOT fixed. In fact, this has not worked correctly since version 6 which I am still forced to use at home to have this feature. Subsequent versions have deleted the sub-second timer; now the fade-in issue is still present. I have asked Roxio many times... each time it has been acknowledged to be a bug, but it has not been fixed. The whole "clipping" discussion makes no sense either since this works beautifully in version 6.
#9
Posted 14 November 2006 - 03:01 PM
johnstred, on Nov 14 2006, 12:03 PM, said:
No, the clipping issue makes perfect sense, but isn't the big issue someone seemed to consider it. Not sure if you're familiar with the math being done on a crossfade with the limits of the values in .WAV (or in the digital representation of music) but it definitely can be an issue. The proper solution is to offer the option of a straight cross-mix without fading, and warn that clipping could occur.
If you want to get into a discussion of why clipping can be a problem, we can, either here, or off-line.
Beware the lollipop of mediocrity. Lick it once and you'll suck forever. - Brian Wilson
[GIGABYTE GA-MA785GM-US2H MB | Athlon II X3 440 (3.0 GHz) | 2GB DDR2 RAM | 1-500GB HD (C: XP, G: Win7, D: - Apps, E: data & apps), 1-500 GB HD Data) | 2 - LiteOn DH20A4P DVD burners | External Dell QFlix DX-20A6Q DVD +/- writer | Windows 7 | Creator 2010 | Tektronix Phaser 850 solid ink printers | Epson R220 Photo/Disc printer | Ricoh GX 5050n dye sublimation ink | Epson Workforce 1100 printer
#10
Posted 16 November 2006 - 07:41 AM
d_deweywright, on Nov 14 2006, 03:01 PM, said:
If you want to get into a discussion of why clipping can be a problem, we can, either here, or off-line.
I am not a computer guru. While I believe and appreciate your comments, all i can tell you that the crossfade feature works like a charm in version 6 where you can fade in and out like in v9, but can also overlap songs without fades. Why hasn't it worked since?
#11
Posted 16 November 2006 - 04:35 PM
johnstred, on Nov 16 2006, 10:41 AM, said:
Okay, obviously you need to understand the potential problem.
Digital music in a .WAV file is represented by numbers that can range from -32768 to +32767. When you overlap two songs without any fading, you are simply adding the two values at each point. (There are 44100 points or samples every second.) If you've ever plotted a sine wave on a graph, that's how you can think of the music, except it's not simple sine waves, they wiggle up and down all over the place. As you add these values, there's about an equal chance that the value from one song will be negative when the other is positive. In this case, you're guaranteed that the sum will be greater than -32768 and less than +32767. There's also an equal chance that the two values will be either both positive or both negative. In this case, there's a chance that the sum of the two values could be greater than 32767, or less than -32768. For example, one value is 10,000 and the other is 28,000, the sum would be 38000. In this case, the actual stored value is limited to 32767, and clipping has occured. Notice, the possibility is against clipping occurring, but it can happen, and in many cases, it may not be noticeable. Also, if one song is already fading itself out, then there's even less likelihood that clipping will occur.
Apparently, when EMC 7.5 was being put together, it dawned on someone that the simple overlap could cause this clipping, regardless of the fact that people had been quite happy with the feature "forever". The decision was made that any type of overlap would be cross-faded, and thus we have the situation we're in now. I agree that it was a poor decision. The proper way to handle it would be to either a) put up a little warning box indicating the potential for clipping, but allowing the user to continue or b ) actually do the math for the overlap and only warn if any actual clipping will occur, but allow the user to continue after listening to it.
There's the math, and the (il)logic as I see it.
Hope that helps!
This post has been edited by d_deweywright: 16 November 2006 - 04:39 PM
Beware the lollipop of mediocrity. Lick it once and you'll suck forever. - Brian Wilson
[GIGABYTE GA-MA785GM-US2H MB | Athlon II X3 440 (3.0 GHz) | 2GB DDR2 RAM | 1-500GB HD (C: XP, G: Win7, D: - Apps, E: data & apps), 1-500 GB HD Data) | 2 - LiteOn DH20A4P DVD burners | External Dell QFlix DX-20A6Q DVD +/- writer | Windows 7 | Creator 2010 | Tektronix Phaser 850 solid ink printers | Epson R220 Photo/Disc printer | Ricoh GX 5050n dye sublimation ink | Epson Workforce 1100 printer
#12
Posted 21 February 2007 - 09:57 PM
johnstred, on Nov 16 2006, 07:41 AM, said:
FINALLY!!! Yes, this IS a huge issue... this worked fine in version 5 and 6. You could easily crossfade any two tracks anyway you wanted without having this issue... and no - there was NEVER any clipping! I have scores of CDs I burned with version 5 that sound beautiful, with all sorts of crazy crossfades and crosscuts on them. Why won't Roxio do something about this???
-Drude

Help
Roxio Community






