Emax

It actually makes no difference to my sound design as I only have Emax II and Esi 2000 samplers. 

For live use I sometimes used to save banks on the Emax II in the compressed format because it was faster to load one floppy and convert to 16 bit than it was to load two floppies. In the live performance context there was little audible difference between the 16 bit banks and the 8 bit compressed banks.

I also loaded some sound library disks from a friend's Emax into the Emax II and found the quality was excellent. Comparing the same bank played back by the Emax I and Emax II side by side I actually preferred the Emax II sound, although they were both good...

--- In emax@yahoogroups.com, ss <ssws1@...> wrote:
>
> Hi Tristan!
> 
> May I ask how that effects your work, your end sound designs?
> 
> This is getting very interesting because I can now see the Emax I, II,  
> EII, EIII, & EIV differently in terms of individual sound qualities in  
> ways I never
> did before.
> 
> I used to do a lot of these conversions with SoundHack...
> 
> 
> On 22 Jul 2009, at 08:03, tristanupton wrote:
> 
> > Well, actually you do get the equivalent of 1MB of 16 bit samples in  
> > the Emax I. A 1MB (16 bit) Emax II bank can be saved on a single  
> > floppy as a 512kB (8 bit) Emax I bank and loaded into an Emax I.  
> > Likewise, a 512kB Emax I bank can be loaded into a 1MB Emax II.
> >
> > Your description sounds like how the Sequential Prophet 2000 stores  
> > 12 bit data in memory. The Emax only has 512kB of sample memory but  
> > can store 18.8 seconds at the 27.7k sampling rate (18.8 x 27.7k =  
> > 512k). The Emax floppy disk format also only allows for storage of  
> > 512kB of sample data. For the method you describe to be correct the  
> > Emax would need to have 768kB of sample memory to provide this  
> > sampling time and it would be unable to store this on a single  
> > floppy disk!
> >
> > My understanding is that the Emax digitally compresses each 12 bit  
> > linear sample word from the ADC to an 8 bit compressed sample word  
> > using a non-linear algorithm (similar to A-law/u-Law companding).  
> > The sample data is stored in memory and saved to disk as 8 bit data  
> > but the Emax converts each sample word back to linear 12 bit on the  
> > fly for output to the DACs and for sample dumping etc. I believe the  
> > Emulator II works in much the same way except its data companding  
> > occurs in the ADC/DAC chips.
> 
> 
> 
> 
> ***************************************
> - This communication is confidential to the parties it is intended to  
> serve -
> 
> 
> 
> 
> 
> 
> [Non-text portions of this message have been removed]
>