Topic: SD 788t at 96K or Zaxcom Fusion 12 at 48k - which will have better audio?

[AB52]

I am debating this. I know this is a difficult question to answer and ideally I would have both and figure it out myself - but the likelihood of that happening is slim. 

I can get an SD 788t and later get another one and they will sync sample accurate - and have 16 tracks at 96. Cost about $12,000 plus.

Or I can have 12 tracks at 48K with a Zaxcom Fusion 12 for about $8500. (I only need 12 tracks)

Will there be any significant sound difference between the SD at 96k and the Zaxcom at 48K?

Zaxcom is suppose to have better pres (for the 8 channels - the other 4 are line in)

Thanks for help with this.
AB

[page]

Does the material and end medium justify 96k?

If I'm dropping that sort of coin, it better be a requirement for me to care about it as a factor, and short of taping bats or sound effects for movies, I can't think of anything I need above 24khz.

[flipp]

If you only need 12 tracks another option is a 788t and a 744t for about the same coin as the Zaxcom; but that still leaves which sounds better unanswered.

[AB52]

I understand. Plus the 788t will sync with another 788t sample accurate.  It will not be sample accurate with a 744t - though they will sync.

And in response to the other post - the 48k recording on the fusion (when all 12 tracks are engaged) - will likely be a world beat type situation or rock.  Very rarely classical.

If you only need 12 tracks another option is a 788t and a 744t for about the same coin as the Zaxcom; but that still leaves which sounds better unanswered.

[page]

I understand. Plus the 788t will sync with another 788t sample accurate.  It will not be sample accurate with a 744t - though they will sync.

And in response to the other post - the 48k recording on the fusion (when all 12 tracks are engaged) - will likely be a world beat type situation or rock.  Very rarely classical.

two things:

1) True, they won't be sample accurate as far as start/stop, but they will be clocked together. In my experience, I've seen multiple 722s that were c-linked and were not truely sample accurate either (the slave started late and ended early, but it was only off by an 1/8th of a second in both sections, and it was clocked).

2) Given that information, I don't think the lack of 96khz is a stopping point. Pre-amp quality, ADC quality, along with user quirks would easily take precedent over that to me.

[AB52]

Thanks Page.  I suppose even the are not sample synced - as long as they go into a DAW and can be aligned - that would be fine.

The tough question is whether the SD at 96K actually sounds better than the Zaxcom at 48K.

I agree with you about the mic pres, converters, the internal clock, etc all probably matter more.

THANKS again for your help.
AB

[notlance]

A couple of things:

1. SD claims that multiple non-788t boxes linked together will be sample accurate.  Quoting from their "788T Linking with Multiple 7-Series Recorders" Sound Note:

"Linking multiple 702/702T/722/744T Recorders
When two or more of the 702, 702T, 722, or 744T recorders are linked, all linked inputs are sample and phase-accurate. No compensation is required between units."

http://www.sounddevices.com/notes/recorders/linking-multiple-7-series/

2. I think the Zaxcom will run 12 channels at 96k.  Looking at page 156 of their latest manual, they claim recording at 192k limits you to 6 channels.  I guess the assumption is you can record 12 channels at 96k.  However, the manual is a bit vague about this point.  Like the 788t, the Zaxcom limits the availability of digital effects at higher sample rates.  That little DSP must be giving its all at 12 x 96k.

And, for whatever this is worth, it has been my experience that a well designed digital system sounds as good at 48k as it does at 96k.  There are many, many other factors much more important than sampling rates.  I guess it is nice to be able to point out that one's recorder will go to "11".

[AB52]

All true - except that when I called Zaxcom (TWICE) - they said 12 channels are restricted to 48k.  But that may not matter like you say.
  And while the non-788t series recorders with by sample accurate with each other, not so with the 788t and one of them. Though a 788t will be sample accurate with another 788t.

Still do not know whether I will go with the 788t or the Zaxcom.  It changes every day!

A couple of things:

1. SD claims that multiple non-788t boxes linked together will be sample accurate.  Quoting from their "788T Linking with Multiple 7-Series Recorders" Sound Note:

"Linking multiple 702/702T/722/744T Recorders
When two or more of the 702, 702T, 722, or 744T recorders are linked, all linked inputs are sample and phase-accurate. No compensation is required between units."

http://www.sounddevices.com/notes/recorders/linking-multiple-7-series/

2. I think the Zaxcom will run 12 channels at 96k.  Looking at page 156 of their latest manual, they claim recording at 192k limits you to 6 channels.  I guess the assumption is you can record 12 channels at 96k.  However, the manual is a bit vague about this point.  Like the 788t, the Zaxcom limits the availability of digital effects at higher sample rates.  That little DSP must be giving its all at 12 x 96k.

And, for whatever this is worth, it has been my experience that a well designed digital system sounds as good at 48k as it does at 96k.  There are many, many other factors much more important than sampling rates.  I guess it is nice to be able to point out that one's recorder will go to "11".

[OFOTD]

I vote for the Zaxcom.   

Much better sounding preamps make up for any loss in available 48-96-192 resolution IMO.

If I were going to make the upgrade to 8+ tracks I would start and stop at the Zaxcom boxes. 

:ducks from the Nagra crowd:

[AB52]

That is where I am at.  Heck - those pres may be so good - I may use them next to my boutique pres in my studio.

I vote for the Zaxcom.   

Much better sounding preamps make up for any loss in available 48-96-192 resolution IMO.

If I were going to make the upgrade to 8+ tracks I would start and stop at the Zaxcom boxes. 

:ducks from the Nagra crowd:

[AB52]

Interesting you mention the Nagra crowd.  The Nagra VI seems like a great recorder.  But in the U.S. - in order to get 8 tracks at 96k for instance, you have to pay $8200 for the recorder and then $1000-$2000 on a great set of converters to match the level of audio quality on the other channels.  That is an expensive proposition.  I would be glad to be wrong about this.  There is a lot about the Nagra I like.

Best,
AB

I vote for the Zaxcom.   

Much better sounding preamps make up for any loss in available 48-96-192 resolution IMO.

If I were going to make the upgrade to 8+ tracks I would start and stop at the Zaxcom boxes. 

:ducks from the Nagra crowd:

[baustin]

The 788t records 12 channels, correct?

With c-link and word clock sync a 788t and 7xx are not not sample accurate? Hard to believe. Why not? What if you sync a 788t to a 7xx vs 7xx to a 788t?

[page]

The 788t records 12 channels, correct?

With c-link and word clock sync a 788t and 7xx are not not sample accurate? Hard to believe. Why not? What if you sync a 788t to a 7xx vs 7xx to a 788t?

788s can do 12 tracks, but only 8 inputs. As I read the manual, the other 4 are mixdowns and stuff like that.

As for not being sample accurate, they won't start/stop and contain the same number of samples. Known bug, and they can still do word clocking, but they will be off as far as X samples in a file from recorder A vs recorder B. To it's credit, I've c-linked a 2x 722s once before and they didn't contain the same number of samples (maybe 1-2 seconds were on one file that wasn't on the other; 1/2 sec @ the beginning, 1 second on the end) so I don't think it's a flawless process in general. SD talks about the delay to calculate and you could program your tracks to delay properly if you wanted to do that, but it's more work.