On 26 Sep 2005 at 22:29, Kurt Gnos wrote: > And, David, I must confirm Lee, you don't have much of a whim of > knowledge considering manners and modern sound technology and > certainly don't know what you don't know
You have completely missed the point. You are more interested in telling me how poor my $150 soundcard is in comparison to "modern" samples, but I've listened to demos of the "modern" stuff and for the price and the performance costs, the emperor has no clothes -- when the move was made away from dedicated DSPs for sound towards soft synths and processing by the PC's CPU, a lot was lost, and the reason for it was simply because PC makers were too damned cheap to spend the money on $25 chips. The results leave me underwhelmed in terms of sound quality. And from where I sit, the whole industry made a huge wrong turn sometime around 2000 or so. I don't by any stretch of the imagination consider my soundcard's sounds to be musically satisfactory -- they are obviously not -- but if I compare its solo strings to the basic sound of a lot of the soft synth solo strings available in the sample sets, I don't think it comes out too poorly. And it does not require a multi-GHz CPU and gigabytes of RAM. Indeed, it worked just fine in my P120 with 128MBs of RAM until a few years ago. I was even able to record to WAV files on that machine. None of the "modern" sound solutions have anything like that kind of efficiency. Yes, the soundfont approach can give you much more flexibility of different samples for different musical purposes, but at the cost of making your MIDI files 100% tied to that sound sample. If you're producing only WAV-based output, that's not really an issue, but I still believe in distributing a general MIDI file. Given that no synthesized sound is ever going to be as good as real musicians, I'm unsure that the time it takes to get the better results with the more advanced soundfonts is worth the time and effort. I'm producing the MIDI files and the MP3s not as any kind of substitute for actual performances, but as a way to get some kind of idea of what the music might sound like. There's a lot that can be learned from listening to those files that can't easily be learned by studying a score. There's also a lot that I learn in producing the MIDI performances, however rudimentary they may be musically speaking (and however inferior they obviously are to performances by real musicians), that ends up informing the decisions I make in editing the music. Most of the sources I work from are problematic (or even defective) in one way or the other, having both errors or inconsistencies in the text, and those "defects" and inconsistencies need to be resolved in any edition that's going to be used successfully for a live performance. Figuring out how to get a rudimentary MIDI performance helps me figure out where I need to add editorial marks, like dynamics, articulations, bowings, etc., and it's the process of producing a MIDI performance that allows me to do that. You or someone else might spend a lot more time or use completely different tools than I do, and the results would be different. But, likewise, if you were performing it with live musicians, the results would also be different from a performance that I would be involved in. My MIDI files are like rough sight-reading sessions, whereas yours may be more like performances that result from hours of rehearsal. That's fine -- you have different aims. I'm just trying to get a whole lot of music into performable condition, and attempting along the way to add enough life to the MIDI output that you can get a bit of a clue about the musical content, at least enough to tell if the piece of music is interesting or not. >From where I sit, all MIDI performances, no matter how elaborate your sound setup, are completely unacceptable as a replacement for even a half-assed live performance (Hi, Dennis!). There is so much that even amateur musicians do (consciously or not) that is almost impossible to get into a MIDI file generated from Finale (if you're inputting the MIDI directly into a sequencer, you can do things that output from Finale's notation cannot do as well), even using Human Playback (which I don't have, as I'm using Finale 2003). So, you're basically comparing one very imperfect facsimile of the music to another facsimile, made with less versatile tools. That's why in my original post I allowed that perhaps given that it was a rudimentary synthesized performance using not-very-realistic string sounds that the difference between 192K and 128K was not really relevant, as either was wholly inadequate compared to even poor quality live recordings. I don't know if we agree on that, because you seem to be arguing that my soundcard and my sequencing are both so awful in comparison to the results one can get from "modern" sound samples and sequencers that it doesn't make a difference, and that if I *were* using the newer stuff, it *would* make a difference. Well, I beg to differ. And I don't think it's helpful to criticize when you aren't accounting for the reasons why I prepare my files the way I do. The only reason I am putting up the MP3s is that I've discovered that the substantial majority of people out there have PCs whose MIDI sounds are vastly inferior to the sounds I'm accustomed to. If I thought that playing the MIDI files directly would get a better result than I'm hearing, I'd not bother with the MP3s, but I know that's simply not the case. Most PC soundcards are wavetable-based, or the systems depend on soft synths (QuickTime and the Microsoft soft synth), and all of those are COMPLETELY INADEQUATE, even moreso than the sounds I get from my own soundcard. I've heard them -- they are all dreadful, in many cases, hardly better than the old days of FM synthesis. If I had better sound output, I'd be thrilled. And it's also true that I could actually do a lot more work to tweak the output from my own soundcard (there are a lot of different options for the samples I do have, and I also actually have two hardware synths available, the one in the MP3s I posted, and another one that is more appropriate to popular music, though it does have better-sounding brass instruments), but the MIDI performances are means to an end, not an end in themselves, so I don't see it being worth the extra effort. Also, as I said, I want to keep my MIDI files vanilla GM so that those with low bandwidth (or better basic samples than my soundcard) can still hear something reasonable. I certainly lack the time to create one GM MIDI file and one that is tied to my particular sound card -- there's just way too much music for me to get through to be able to manage that. Anyway, I've gone on too long answering a post I intended to ignore in the first place. But I feel there's an awful lot of assumptions behind the criticism levelled at me that are completely off base. And I wanted to clarify that. -- David W. Fenton http://www.bway.net/~dfenton David Fenton Associates http://www.bway.net/~dfassoc _______________________________________________ Finale mailing list Finale@shsu.edu http://lists.shsu.edu/mailman/listinfo/finale
