Topic: choosing an ADC

[Brian Skalinder]

Oops, I had my numbers screwed up on the AD-20 price, both in my post and in the n00b post.  Looks like retail's around $300.  I was looking at a (high, at that) price for the PS-2 and mistook it for an AD-20 price.

N00b and previous post corrected.

At any rate, IMO price isn't the issue in this comp given my previous post.   

[bagtagsell]

i am thinking about getting my own ADC.  i will still hold off on mics for a bit.  i am recording onto a jb3.

I would invest in some mics.  You are patching right?  Can you regularly get a digital patch?  Getting an adc won't help you there.  Save some money and buy some mics.  IMO. 

If you are having a problem with an optical patch, buy a hosa box to convert a coax to an optical signal

[jaguaracer]

Not true!!

Thanks for clearing that up! I believe I put them in one of my other posts (or was about to) the question that I wasn't sure if the truncation process eliminate important bits that significantly altered the recording.

[charles]

i am thinking about getting my own ADC. i will still hold off on mics for a bit. i am recording onto a jb3.

I would invest in some mics. You are patching right? Can you regularly get a digital patch? Getting an adc won't help you there. Save some money and buy some mics. IMO.

If you are having a problem with an optical patch, buy a hosa box to convert a coax to an optical signal

Or...here's an idea. Buy some Nak cm100's or cm300's ( $150-300 ), and an AD-20 ( $150-320 ). Start with that combination. That way you can get accustomed to using mics > a/d > jb3. The Naks  run on an internal battery,  so you won't need the PS-2. Then, when you're ready to spend a little more, you can get the Nak's modded ( $150? )...this will make them run on 48VPP and make them small enough to stealth with....and purchase a PS-2. This might be a sensible route to take in your situation.

[thierryhenry]

Coresound Mic2496 has gotten many mixed reviews, I know Steven (sml42 here) despises it, literally hates it. I don't know if it's worth the $500 Core Sound charges. PS-2 > AD-20 is a very combo. Although like someone said, once you put those together it's pratically the size of a UA-5, depending on how STEALTH you go, many people have STEALTHED a UA-5 in a bag before, I think that's my next purchase.

[jeromejello]

ok y'all thanks for the info... but i have now dug up even more questions about adc's:

first of all, i respect the amount of hassle that coresounds has seemed to dish out.  but both the mic2496 and the ad-20 are coresource (or is the ad-20 just sold by them and its really zeph) so i cant really say either sucks in quality without saying both have the potential (esp on the cust service end) to suck.

second, and here is where i am most confused.  i get that the jb3 is 16 bit only.  but it seems every adc is 24 bit (with the exception of ad-20, which is 20bit, and i will address that later) + the 8bit info.  so if both the ua-5 and mic2496 are 24 bit adc, how is it that one is better than the other for the jb3 if there is still going to be a certain amount of truncing?  does the ua-5 dither the 24 bit to 16bit?  i thought the mic 2496 dithered as well if you chose the right switch config - 44.1 and 1x (vs 48 and 2x = 96)  even the ad-20 is going to trunc 4 bits of data (on top of the 4 bits that it already zero'd) to get it into the jb3.  i sorta get where the 16 bit and 20 bit arguement is going on the ad-20 faq, but it seems that its not the complete or whole story.

finally, the modded version of the ad-20 which basically adds the ps-2's function to the ad-20 with the addtion of an extra 9v.  is this doable on my own? with the assistance of someone here?  or do i have to shell out the redonkulous cash to get the modded box.

i am sure i will have more questions, thanks for your help and patience.

[charles]

first of all, i respect the amount of hassle that coresounds has seemed to dish out.  but both the mic2496 and the ad-20 are coresource (or is the ad-20 just sold by them and its really zeph) so i cant really say either sucks in quality without saying both have the potential (esp on the cust service end) to suck.

They are not made by the same people.

[charles]

but it seems every adc is 24 bit (with the exception of ad-20, which is 20bit,

Not neccesarily true. The issue here is that the word width (bit) for the core 2496 is not adjustable. With a unit like the ua-5 or the v3, your word width can be slelected up to 24 bit.  The AD-20 is only 20 bit. The core2496 is only 24 bit. In other words, if you run a ua-5 into a jb3, and you run the ua-5 at 16 bit, nothing will be truncated. Starting to make sense?

[jeromejello]

but it seems every adc is 24 bit (with the exception of ad-20, which is 20bit,

Not neccesarily true. The issue here is that the word width (bit) for the core 2496 is not adjustable. With a unit like the ua-5 or the v3, your word width can be slelected up to 24 bit.  The AD-20 is only 20 bit. The core2496 is only 24 bit. In other words, if you run a ua-5 into a jb3, and you run the ua-5 at 16 bit, nothing will be truncated. Starting to make sense?

starting    i was reading more about the ad-20... and due to the s/n ratio of the ad chip it gives 16.5 bits of data which is why apparently it is a good match for 16 recorders like the jb3. 

from an info file:

The digital output on the AD-20 is a 20 bit 44.1 kHz S/PDIF signal.  Because the dynamic range of the AD-20's A/D is about 103dB A-weighted or 98dB unweighted this equates to about 16.5 bits of resolution before the natural noise of the A/D can be heard.  Because the natural noise floor of the system hangs right on the 17th bit, the AD-20 also works great as a 16 bit recording device without the need for additional dither (noise) to be added to the signal.  You might say the AD-20 has its own natural dither.  Had the S/N of the AD-20 been beyond the 17 bits then additional dither noise would need to be added at the 17th bit to avoid excessive quantization noise. 

i am intrigued to see how much of the mic2496 is actually available... i assume around 21.5 which would be less of a clip then first appears.