Sound Devices 788T now supports 192 KHz

[Frank Kruse]

Originally Posted by myke2242

Sorry guys I didn't realize that what we were talking about till now. My concern is the processing power these limiters take as it would be more cycles then the analogue input limiters.

I still dislike the 788 limiters. I perviously own a 744 which had great sounding limiters. Many of those who owned a 744 and moved up to the 788 were disappointed with the limiters for many reasons.

Do you have the recent version installed. The first version of the limiter clicked and crackled but that was fixed many years ago.

[Frank Kruse]

Originally Posted by myke2242

Sorry guys I didn't realize that what we were talking about till now. My concern is the processing power these limiters take as it would be more cycles then the analogue input limiters.

I still dislike the 788 limiters. I perviously own a 744 which had great sounding limiters. Many of those who owned a 744 and moved up to the 788 were disappointed with the limiters for many reasons.

Do you have the recent version installed? The first version of the limiter clicked and crackled and was unusable but that was fixed many years ago in a firmware update (I think in v1.76)

[mgoorevich]

I owned 744t for years and now have (3 years) a 788T.
I love the limiters on 788T and don't think they are worse than 744 ones.
The sound of 744T is a bit different. May be this explains why people feel the difference in limiters behaviour? Also the A/D dynamic range of 788T is 123 dBA while 744 has 114 dBA.
With this dynamic range I recorded several feature films without engaging limiters for a single time on my dialogs. Never got any distortion and I record at normal levels.

744 limiters:
Analog (pre-A/D converter), dual-stage optocoupler and FET, -4 dBFS threshold, 20:1 limiting ratio, 5 mS attack time, 200 mS release time

from Sound Devices site:


The 788T’s limiters have been enhanced with firmware revision 1.76. The ability to adjust limiter characteristics provides the sound mixer with advanced control more typical in studio gear than that found in location sound recorders.

[myke2242]

Originally Posted by mgoorevich

I owned 744t for years and now have (3 years) a 788T.
I love the limiters on 788T and don't think they are worse than 744 ones.
The sound of 744T is a bit different. May be this explains why people feel the difference in limiters behaviour? Also the A/D dynamic range of 788T is 123 dBA while 744 has 114 dBA.
With this dynamic range I recorded several feature films without engaging limiters for a single time on my dialogs. Never got any distortion and I record at normal levels.

I do have 3.0 installed but haven't tried the limiters yet, probably won't. There are many reason why people don't like them and It's probably subjective as well. To me personally the limiters don't sound as transparent and smooth as the ones in the 744. The bigger issue is that we won't be able to over come the lack of Functionality above 48k due to technical limitations of the 788. In other words even if you think they sound awesome you won't be able to use them above 48k. Most of my recordings take place 96k and above.

[Fokke van Saane]

I still think 192KHz is overkill.

[cmaynes]

Originally Posted by mgoorevich

And it's time consuming to tweak and set the machine up to 192kHz.
I hope the new 3.x.x firmware will address these tasks to the machine to adjust it automatically.



Well.. you won't need limiters at this SR anyways as the only use of 192kHz IMHO is for ultra quite atmos with using mics going up to at least 50kHz and future pitching down for different sound design purposes.
Limiters are needed for loud signal and 48Khz 24 Bit is more than enough for that job.
SD788T has wonderful pres and fantastic AD converters. My bet is that NO ONE could hear the difference between 96kHz and 192kHz at this level. It just increases the marketing value of the unit.

I use SD limiters quite often, in fact they are a staple of my workflow- I trust them implicitly actually.

As to the value of high sample rates, I can hear the difference between 48k and 96k. I cannot hear much difference between 96k and 192 though. Why would you think limiters would not be useful for sound design? that is a curious statement to make I think... except for the fact that they are still unavailable on the 788t.... which even though SD claims the limiter circuit is pre a/d on the 788, it makes ZERO sense that they would still be disabled above 48k....

[Frank Kruse]

Originally Posted by cmaynes

which even though SD claims the limiter circuit is pre a/d on the 788, it makes ZERO sense that they would still be disabled above 48k....

The 8 limiters are pre A/D but digitally controlled by the processor. I guess at 96k and above the 788 needs every milligram of processing speed so the limiter is disabled. Yes, it's unfortunate but the machine doesn't seem to be able to do it at 96k and above.

A mono Nagra is not stereo and it never will be. Wether it makes sense or not it's simply a restriction of the hardware.

[mgoorevich]

Originally Posted by cmaynes

I use SD limiters quite often, in fact they are a staple of my workflow- I trust them implicitly actually.

As to the value of high sample rates, I can hear the difference between 48k and 96k. I cannot hear much difference between 96k and 192 though.

Same here. I can live happily with the most critical stuff at 96KHz 24 Bit.

Why would you think limiters would not be useful for sound design? that is a curious statement to make I think... except for the fact that they are still unavailable on the 788t.... which even though SD claims the limiter circuit is pre a/d on the 788, it makes ZERO sense that they would still be disabled above 48k....

Ok I need to clarify it . The higher sample rate was designed to increase aliasing frequency and following filtering of it (according to Nyquist theorem).
E.g. in 96kHz, everything above 48Khz must be filtered. There are very few mics that go that high btw. So logically you would use SR of 96 KHz to record signals with ultra high frequencies contence . At this high frequency the energy of the sound waves is usually very low. A way lower than the threshold of the limiter if you'd insist to use it.
Thus is my statement: limiter is needed for the higher energy lower frequencies sounds, where SR 48KHz is more than enough for superior quality capturing.

Now, practically it would be great to have limiters at 96 KHz so one just can record everything at higher SR without thinking twice what content is going to be recorded.
So I can admit that my arguments are based more on the theory than on a real life

[myke2242]

Originally Posted by Frank Kruse

The 8 limiters are pre A/D but digitally controlled by the processor. I guess at 96k and above the 788 needs every milligram of processing speed so the limiter is disabled. Yes, it's unfortunate but the machine doesn't seem to be able to do it at 96k and above.

A mono Nagra is not stereo and it never will be. Wether it makes sense or not it's simply a restriction of the hardware.

Interesting thought about your statement. I am well aware of the issue. i find it interesting that SD figured out how to do 192k with the limited available processing power that is left in the 788. I'm not a hardware designer or dsp eng but why couldnt SD simplify the limiter parameters to have a on/off and threshold settings and make attack, release etc static like the 744. that mite free up enough cycles to make limiters available at 96. or even if you only access to 4 limiters that you could assign to any 4 channels would be useful.

[Frank Kruse]

Originally Posted by myke2242

or even if you only access to 4 limiters that you could assign to any 4 channels would be useful.

hm. 8 tracks but only 4 limiters...not sure.

probably the same outcry would happen. why not 5 limiters?! why 4 limiters? can't we have 4 tracks at 48k with limiters and the other 4 at 192k without limiters? why not enable the EQs on 4 tracks at 96k first instead of having limiters. etc. etc.