Topic: ADI-2/4 Pro Intersample Overloads

Hi!

I know that based on the measurements of Archimago, Intersample Overloads are not present in the ADI-2 Pro FS R BE which is excellent!

How about the new ADI-2/4 Pro SE? Can somebody from REM confirm, that the same countermeasure is taken to eliminate Intersample Overloads?

Re: ADI-2/4 Pro Intersample Overloads

I was asking Michael Stipe about this just the other day. wink He said the 2/4 has 2.5dB of headroom above zero for just that reason! Manual page 34:

"OVR appears for all analog outputs at +2.5 dBFS, since the ADI-2/4 Pro SE has this digital headroom."

https://musicwall.app/hermetech

3 (edited by Bernstein 2023-04-25 22:49:49)

Re: ADI-2/4 Pro Intersample Overloads

Babaluma wrote:

I was asking Michael Stipe about this just the other day. wink He said the 2/4 has 2.5dB of headroom above zero for just that reason! Manual page 34:

"OVR appears for all analog outputs at +2.5 dBFS, since the ADI-2/4 Pro SE has this digital headroom."

Not exactly: i thought you need at least 3dB to have enough headroom for Intersample Overload, but you gave me the idea to look into the manual more thoroughly.

In 8.6 SRC there is 3dB headroom mentioned in the SRC to avoid Intersample Peaks (which I think are meant to be the Intersample Overloads?!).

4 (edited by KaiS 2023-04-28 08:02:00)

Re: ADI-2/4 Pro Intersample Overloads

Just measured:
ADI-2/4 Pro SE has 3.0 dB headroom above 0 dB volume setting to handle ISPs of a full-scale signal.

In most practical situations a volume lower than 0 dB is used, which further increases the DAC’s headroom.

ISPs at the output, BTW, don’t show up on the meters, because they only exist in the analog domain, and the metering is in the digital domain.


As already mentioned, the SRC has a -3dB option to handle ISPs.
These would show up on the (pre FX-) meters, as SRC is comparable to a DA-conversion in this regard.


REMARK:
ISP = Inter Sample Peak = Inter Sample Overload:
Signal level higher than the specific digital domain signal suggest, caused by the reconstruction filter in the DA conversion or SRC process.