I've often heard that it never sounds as good when the multitrack is transferred from tape to digital but was wondering if anyone has found an A/D converter that retains the sound exactly after the A/D transfer. Is it possible to have the digital files sound exactly the same (to the best of human hearing ability) as when played right off the tape deck?

For me: Forssell AD

Digitizing audio without changing its sound quality is the ideal goal of almost every AD/DA converter, just like the ideal goal of almost every camera is to capture light exactly as it is in the real world. There is no single product on earth that does it with 100% accuracy, but the better designs do it with more accuracy than the worse ones, and these days the better designs can do it very very well.

I would trust your ears more than what you hear people write online.

I would say the critical thing to AVOID is the breed of converters that have added transformers involved - eg Burl Mothership, JCF Latte - that additional analogue "transformer sound" (whatever it is) is more likely to have an audible difference than any high end converter that's designed without this intended flattery.

Any differences between the high end converters are likely to be very small, and possibly indistinguishable - I certainly couldn't tell the difference between the playback from a desk and a Lavry converter when level matched.

I can't think you'd go wrong with any of the higher end converters - Avid IO, Apogee, Prism, plus I guess Mytek, Forsell etc (I've not used those 2).

You would have also to have a capable D/A... Forssell D/A is yet to be beaten for me, SPL madison is also really great, awesome bang for the buck (for D/A side only).
Because you'd be comparing analog tape and A/D/A.
Dealing with A/D, TI pcm4222(Antelope Pure 2 has one, but filters are apprently not set for maximum fidelity) or old Foscusrite AD245 are outstanding and not too expensive.
Otherwise Pacific microsonics HDCD or MSB converters, but much more pricey.

Well, if you're going AD then you're going to have to come out DA. So, you've already got two stages and more variables there. And then you have the digital-recording medium/platform itself. And more cables. Already, the amount of variables is reaching astronomic proportions.

If you're really serious about this, I'd contact some serious mastering houses and inquire about their services, and/or also pick their brains a bit. And I'm sure none of them is going to just give you an AD solution, because they all know there's much more in the path than that.

I'd start with STERLING SOUND - Audio Mastering Studio - New York City

My 2¢.

My short answer would be, if I was going to transfer multitrack to digital, I'd use Lavry.

Im wondering also do the antelope clocks (ocx, isochrone 10m (now mx)) make any noticeable improvement on the high end converters like the lavrys and forssells, or would using them in conjuction with say an apogee symphony bring them up to par with the lavrys? Wondering if anyone has a lot of experience testing these different converters along with the antelope clocks which on the one hand Ive heard great things about, on the other hand I see a lot of them for sale.

But neither the DA, nor the digital recording medium actually affects the sound of the transfer...yes it's important the DA is neutral so you can be certain what you're transferring, but it's not "fixed" into the transfer.

And the capture medium, providing it can spit out timestamped WAV files, is truly irrelevant.

FWIW, the JCF Latte's ADC is completely transparent, transformer-less, it does not change the sound passing into it. The DA on the other hand, is exactly what you said -- A heavy transformer vibe. But the ADC is not colored. It is wide, clear, no "lens" effect - Meant to Capture all the vibe without mutating your analog signal.

My Crane Song HEDD fits your needs sound wise.
They have upgraded their converters this last year so they sound even more transparent , but still have the analog vibe .

Weiss Engineering Converters. No specific "Sound" at all, thats a good thing for conversion.

Worked with These at the facility i work. would be nice to afford These in private life...

First, how many channels are the tape recordings? 2/4/8/16/24 channel? From a time is money standpoint and possibly from a "how good is the tape's condition?" standpoint (needing to bake tapes and/or tape shedding), you will want an A to D converter with enough channels to do a transfer in just one pass (some of the above suggestions are only 2 channels which sucks if you have a marginal condition 24 track tape). With large channel count converters you have limited options. The next rub is that we have never had a large converter shootout to determine which converters are the most transparent so whatever you read is likely just a person's very limited experience with just a few converters, hopefully via blind testing against a monitered stereo live source. There is also the rub that many manufacturers choose the "less than best" filtering options because as quality there increases, so does latency, which for most is a more important factor to keep as low as possible.

As for clocking, the current rule of thumb is that the main A to D converter should be master clock as that is the most important stage. Any converter that you use in Slave mode via exterior clocking is only as good as it's PLL circuit which in most cases is never as good as using the converter's own internal clocking. Some however prefer the "sound" of less precise clocking so because it's fairly easy to change clocking configurations, it's worth some experimentation time. Clocking threads here will have more info including links to an SOS magazine article/testing which explains the finer details for most to comprehend. Though I have not seen any testing done regarding this, just thinking out loud seems that you will probably have more acurate clocking if BOTH the A to D converter and D to A converter reside in the same unit sharing the same Master clock (neither converter is using the PLL circuit).

My 2 cents,
not a scientific test, but my pal has an ongoing sample base utilizing the same source material file(beautiful big DR, and wide freq range), DA/AD looped back into his system, so far within the tested units, it shows no existence of a "transparent" AD or DA, that comparison base is around 15 diff converters so far, ranging from Weiss to Ross Martin, I've listened to all of them on my system that utilizes a custom built DAC, they all show audible diffs, and the re converted files do NOT null out against the source file, or each other in tests.

there is a correlation to the better sounding ADDA rigs to my ears, the units with less channel count per converter chip, that have beefy PSU's and implementation, stand above the more and lesser units respectively.

I'd recommend getting something with a great, overbuilt power supply system, lower channel per chip count, that you personally like the sound of, and use it.

Have you been sure to time/level align the files? Or used audio diff maker to do that automatically?

"Never sounds as good" is subjective. I have been happy with transfers using my Apogee Symphony. For me, the short coming of transfers has always not been in loss of frequency, but in loss of transients, due to pre-ringing in AD converters. Newer, better quality converts, like the Symphony, have really diminished the amount of pre-ringing found in the FIR filters. My ears have been happy. If you want to get totally anal, I get you can use an Atomic clock to completely eliminate all jitter artifacts.

.

There seems to be a growing belief among "high-end" folks that DSD conversion does a better job of capturing audio (without committing "mojocide").

Many say that even when later converted to "standard" PCM, these captures still sound closer to the original than when going straight to PCM.

Any of you "high-end" folks wanna ring in on this?

.

This has me curious about filters. Val from qes has a button that let you choose between 2 types of filter on the TI 4220(non dsd 4222 chip).

I'm not saying the DA is fixed, but its use is part of many variables in the judgement call of what's really going on with the AD. My point is more about, where do you really make the call with so many variables?

OP was wondering if anyone has found an A/D converter that retains the sound exactly after the A/D transfer. Is it possible to have the digital files sound exactly the same (to the best of human hearing ability) as when played right off the tape deck?

I'm just saying that is a bigger can of worms—with many more variables—than might be apparent.

We could maybe ask that OP softens their question a bit, and just ask about really good, transparent, neutral AD converters.

And products by companies like Lavry, Forssell, and Mytek would come up.

You might not want Apogee, UA, Lynx, etc.. that have their own character and some coloration.

24-96 Mastering brought up a good point in this thread on comparing AD converters that provided sound samples... and they do go down the rabbit hole a bit. A/D Converter Shootout (with samples)

"Don't forget that there is also a D/A conversion in the process. You're not listening to an A/D conversion only, but to a D/A followed by a bit of copper, followed by the A/D conversion to be evaluated.

BTW, here's what the voting results look like:

I like converter 1 Crane Song HEDD 192
39 Votes - 10.32%
I like converter 2 Lavry Engineering Gold
48 Votes - 12.70%
I like converter 3 Mytek 8x192
86 Votes - 22.75%
I like converter 4 Forssell Technologies MADA-2 (prototype)
73 Votes - 19.31%
I like converter 5 Prism Orpheus
67 Votes - 17.72%
I like converter 6 Universal Audio 2192
42 Votes - 11.11%
I don't have a preference
74 Votes - 19.58%

Probably not, no. You don't want an external clock when you don't have to have one.

I guess I was assuming that if anyone has been unable to decipher any change doing a tape transfer then there wouldnt be an issue on the DA side. But I guess that should be included in the question. I wonder though is there a way to test for nulls in the analog domain the way we do in the digital domain? Like nulling the playback off the tape machine against the playback from the dac after the transfer? Or does wow and flutter and other analog quirks make such a repeatable scientific approach impossible?

Although I would say if the difference is not audibly perceivable in a blind test its not neccesary to null 100%. Ie good enough for even the best of human hearing.

Well theres one I know of coz I use it off my 2 track deck. JCF Latte, at 192 I cannot hear any degredation whatsoever, it sounds the same, its incredible. All the life, motion and depth intact. Its only 2 channels though.. you may want to check out their AD8 unit, is apparently awesome

No transformers on the Latte AD, its just hi res SAR with passive filtering. RediC

From what I gather these are the brands I see come up repeatably in other threads discussing top end converters.

Pacific microsonics
Merging
Weiss
Lavry
JCF
Forssell
Meitner
Mytek
Prism
Burl

DSD in fantastic. I pretty much print exclusively to DSD, these days, and don't even use my ATR102. The only problem is that I'm not aware of a multi-track DSD beyond stereo, and you can't really put it in a DW when you are done with it, unless you are going to convert every file from DFF to WAV.

But yeah, DSD is the ****, and I highly recommend it for final mixes.

Merging Technologies Pyramix with Masscore Extended will do 64 Channels of DSD or DXD at 64/128/256 DSD rates, using Hapi/Horus Interfaces and Premium cards. They recommend their Turnkey PC's for Masscore - but you can custom design your own machine to their specifications. The Pyramix Native Pro Software will also do DSD/DXD rates up to 16 channels.

DC filtering should better be turned off.

DSD versus PCM is another whole can of worms with people on both sides of the debate. I usually try to avoid controversial topics, but what the heck, I'll give you the position of the other side just so you hear both opinions and can make up your mind for yourself.

To totally oversimplify, DSD is basically 1-bit delta-sigma conversion. Almost all modern AD/DA converters are utilizing multi-bit delta-sigma conversion, which is kind of the next progression beyond what DSD could do. At the time it was created in the late 90's, DSD offered some benefits over 16-bit PCM (i.e. "better than CD quality"), and was a great marketing pitch for Sony/Phillips to get people to buy their Super Audio CD format. But the format never really caught on commercially, was technologically surpassed by multi-bit delta-sigma 24-bit PCM, and today in the grand scheme of things it is basically obsolete but still has some niche fans in the audiophile community. It's kind of like how audiophiles still want to listen to vinyl rather than digital files - sure, it's technically "worse" by every measurable specification, but it just, like, feels better man.

For the non-condensed version of this argument, this interview with John Siau from Benchmark is probably the most accessible resource:
John Siau: Benchmark Audio Guru | Real HD-Audio

Here at Apogee we choose not to include DSD on our products because it's practically non-existent in the greater commercial market, and we don't have much confidence that it's about to come roaring back any time soon. But stranger things have happened!

Doc, yeah, but that's a really expensive option!

Always happy to see another voice of reason regarding DSD.

I heard of converter upgrades for the advocet, but wasn't aware that the hedd 192 converters have been upgraded yet.