Topic: Tascam DR-70D 4-channel audio recorder (Part 3)

[Ultfris101]

FWIW - To correct an earlier allusion I mentioned that the D70 is based off the hs-p82 - after reviewing the p82 schematics, the p82 has discrete transistor based Pre-Amp coupled to AD8599 opamp

The hs-p82 does not use the ne552 op amps - instead it uses a discrete transistor network in front of AD8599 op-amps

Whoops - I missed that comment earlier.  Thanks for clarifying that.  Clearly, Tascam must be using "HDDA" as marketing, and it doesn't describe any particular circuit or chip.

I swear that a couple months ago I found a place on the Tascam website that said something (one of the USB audio interfaces I think) used the ne552 op amps and stated that it had this in common with the HS-P82. By deduction I concluded that the DR-70d which has the ne552 op amps has this in common with the HS-P82. I can no longer find this text so I'm wondering if they made some changes or I just misunderstood.

The potential for four HS-P82 quality channels was quite enticing but also a bit surprising and now it makes more sense I guess.

[Joshua_G]

It's possible that both versions are correct. That is, in few of their products, for each mike channel there is a discrete transistors front-end, followed by an OpAmp chip. So, "HDDA" is what they call the discrete transistors front-end - which doesn't exclude a following OpAmp chip - which replacing by another one improves the subjective SQ (Sound Quality).

If this is the case, "HDDA" is both a marketing phrase and valid technical consideration, which (potentially) improves both technical specs and SQ.

[MakersMarc]

So will this thing phantom power four mics at the same time?

[willndmb]

So will this thing phantom power four mics at the same time?

yes

[MakersMarc]

Thanks!

[noahbickart]

http://bt.etree.org/details.php?id=579901

Tedeschi Trucks Band
May 18, 2015
SummerStage, Central Park; New York, NY

Schoeps mk41v (DIN)> Vark KCY> Naiant PFA> Bussman Modified Tascam DR-70D @ 24bit 96 kHz> Macbook> Sound Studio (Tracks & Fades)> xACT (Flac)

Recorded and Transferred by Noah Bickart <noahbickart@gmail.com>

[2manyrocks]

What is your impression of the results of the mod compared to the stock unit? 

[noahbickart]

What is your impression of the results of the mod compared to the stock unit?

I never compared them side by side....

[2manyrocks]

Did you at least get a pretty Busman sticker? 

[vwmule]

So, generally, this deck has exceeded expectations. I've only used twice but it's easy and issue free.

[voltronic]

So, generally, this deck has exceeded expectations. I've only used twice but it's easy and issue free.

It has some significant issues with it that Tascam should address (see link to FAQ in my sig) but I agree that it's generally pretty easy to use given the features it has.  I'm also really impressed with the quality of the built-in pres.  I posted a sample of a large choir recording I made this week, and at least for a live concert with audience the pres are plenty quiet enough.
http://taperssection.com/index.php?topic=172679.msg2144032#msg2144032

[voltronic]

Something very interesting I noticed regarding the gain knobs that I haven't seen mentioned before: The knobs don't appear to function as continuous digital encoders as I thought they would.  Instead, you turn them until they hit a threshold which activates a programmed level change in the software to the next "step" of gain.  It's not a continuous control, even though it feels like one.  So, the result of you moving the knobs is similar to analog stepped pots but without the detented tactile feedback.

Evidence to support this:

1. While monitoring the output with headphones, set a gain pot to zero.  Very slowly bring it up.  You won't hear anything for the beginning of its travel, but then around 7:00 / 8:00 the channel will suddenly come to life, as though you had a console with the fader up but you had just disengaged the mute switch.

2. With an empty room, I've been able to set my levels of all 4 channels pretty equally by eye using the room tone / steady HVAC noise and watching the meters.  If you turn the pot past a certain point, the dB meter pops up a couple segments and stays there.  Even if the pot position of the other channels isn't exactly the same, I've fount that as long as you're in that range of "wiggle room" then you're getting a matched level.

Can anyone verify this?  If I'm right, then this would somewhat lessen the need for ganged gain controls as some of us here have called for.  We still definitely need a per-channel numeric level though, and I would also like to know exactly what the gain steps are in the software.

Hoping Tom Duffy from Tascam will be able to weigh in on this - I've had no luck getting a response from him after numerous attempts.

[Jonmac]

It's fairly easy to measure the gain steps, inject a steady tone ( 1kHz ) into the input, set the machine to record and slowly increase the gain until you reach 0dB.

Examine the resultant waveform in your DAW and you should see the steps and be able to measure the size.

Here's a test I did on the Zoom H6 and found that the gain is adjusted in 0.6 db steps, and not continuously as I first thought.

https://dl.dropboxusercontent.com/u/70230944/Attenuation%20Steps.jpg

[Ozpeter]

the gain is adjusted in 0.6 db steps, and not continuously as I first thought.

Wasn't the original idea of a dB that one dB represented the smallest level difference the human ear was thought to be able to detect?  In which case, 0.6dB steps would sound smooth.  But I may be wrong.  It has been know to happen.

[Jonmac]

the gain is adjusted in 0.6 db steps, and not continuously as I first thought.

Wasn't the original idea of a dB that one dB represented the smallest level difference the human ear was thought to be able to detect?  In which case, 0.6dB steps would sound smooth.  But I may be wrong.  It has been know to happen.

Oh yes, it sounds very smooth, you can't tell that the gain is adjusting in steps, it was only after doing the test that I found this out.

The gain controls have been very well implemented by Zoom.

The stepped stud faders used by the BBC up until the late 60's, were in 2db steps as it was then thought that the human ear couldn't detect a change of 2db or less.