It's arguable that we'll ever need 24bit recording. At a certain point there are frequencies in either direction that the human ear can't hear. 24bit just gives you more headroom when mixing and therefor makes it easier to have like 24 or more multitracked instruments sound good. With 16bit you have less room so you have to carve out your spots more precisely.

16bit can sound very good - because of the magic called dithering - dither is only a random 1bit noise signal added on when a 16bit file is processed, but it stops quiet sounds going grainy when they are down to only 1bit - the noise makes them 2bit and the magic of human hearing is that we can still hear the sound inside the noise but tend to ignore the noise (like you don't often notice the hiss from an amp).
24bit doesn't need dither, but it's added when you knock it down to 16bit.
A good dither noise makes 16bit as good as 20 or better!

An advantage of 24bit for recordists is that you can record quieter - I always tended to get the level as close to clipping (0dB) as possible in 16bit, especially vocals where you might not want a grainy sound. With 24bits, I'll record as low as -12dB knowing I have no fear of clipping (don't have to stop and do another take) and still have clean recording - and if I have to boost it, it'll still be clean.

1bit is all very well, but it's impossible to process, unless you convert it to 16, 24bit or whatever first, and then you have to convert back to 1bit again.

Back to the digital portastudio - I don't know that one, but they usually require you to spare 2 tracks to bounce your mix to, bounce ahead to a new song or record out to a "master" 2track. Reason is that 8tracks on a CF is pushing the recording speed for flash memory - although it's getting faster all the time, fast enough cards were probably not around when the recorder was designed, which might have been started 3 years ago.

I mean, the 404 is fairly recent, but only allowing 6 stereo samples playing at once and a max 1GB card size is probably the most a CF card could do when Roland designed it.

Jim

I'm pretty sure why some Compact Flash cards work and others do not has solely to do with the reading / recording speeds the SPs can or can't handle.

That is to say, a lot has changed in the last 5 years or so, going from say 32Mb cards as standard up to an insane 8Gb and even 16Gb range of cards now, all with incredible reading speeds and this has been a very fast development.

Probably simply too fast for the SP series to fully take advantage of yet, on the other hand they've been able to make sure sample loading and so on is lightning fast.

As always, every extra or more complicated feature than the previous comes at a performance price. I'm thinking Compact Flash cards haven't been the real bottleneck, but rather internal memory and processing power.

i have the dp02cf and messing with it the other day for the first time and im hella confused on what to do i tryed puting some loops and what not over from my pc but i got them in and cant seem to put them on the tracks haha anyway im a noob i guess because i cant seem to find a way to get the thing to work lolz

I don't know this DP02CF, but it may be that you can record your mix on it by bouncing "forward" to a new "song" on the card. Check in the manual, should be something about exporting a finished song.

The big advantage of 24bit recording is you can have modest record levels and still get a sound that doesn't get grotty when it's processed with compression and eq later on. I record a lot of acoustic performers and they always will sing louder on the take than when we did a dry run to set the record levels - result = nasty digital clipping. So, with 24bit I can set at around -12dB on the meters (that's about halfway) and not worry about clipping. With 16bit record I'd have to set about -6dB and cross my fingers (and anything else crossable) during the take.

But... none of that matters if your sounds are already processed, 16bit will be fine. Vocals will be fine too if the singer is recording wise and knows to sing over the mic when they go loud instead of right into it.

Converting from 24bit to 16bit shouldn't make the sound worse. Most music these days is worked on at 24, 32 or even 64bit before final mastering to 16bit. There's a clever thing called "dither" and as simply as I can explain...
Dither adds noise to the audio. This makes small detail sounds louder because the mixed in noise boosts the overall level. The human brain is very clever at picking out signal from noise while ignoring the noise itself. Result - details are heard just the same on the 16bit version as they were before. Very clever stuff.

Jim

Still some peep using these?

Anything I should know? I'm about to grab one and and it looked pretty decent and fun.

Looking to get into live situation and... you know... BURN A COUPLE OF CD lolz