Reel To Reel vs Hrx files vs Vinyl

[Steve williams]

I have never been able to sit down and compare the same music in all 3 formats but know that there are many here who have. Particularly interested in comments from Bruce_B who has his own mastering studio.

I have been able to compare tape with vinyl but never the three together as I have only heard HRX alone with no comparison

What are member's thoughts about the 3 formats in terms of the leader of the pack. From recent posts it seems Frantz is in the HRX camp.

 

[MylesBAstor]

Don't look at me I've only done vinyl vs. tape! What I think is impressive and faithful the newest generations of tts are to the tapes.

 

[Bruce B]

This is one area where the source has alot to do with it. I've heard things sound the best on each format. Now we're talking the best of each format.

TT = Rockport or EMT 948 w/45rpm media

Digital = Pyramix or Sonoma workstation through either DAD AX24 or Meitner DAC8IV using DSD 128fs or PCM at 32/352.8kHz

Tape = Studer A80 with Flux ER heads and modified circuits

The majority of sources, tape wins every time, if the source is exactly the same. I was in Japan about a year and a half ago and did a recording of a Jazz trio. I took the feeds from an SSL 9k J mixing board. We recorded in 4 formats; 1/4" tape on an ATR machine, DSD5.6 on a Korg, DXD using Pyramix and DAD AX24 and lastly Pro Tools at 24/192

The Pro Tools take sucked. It was lifeless and 2 dimensional. The DSD5.6 was really nice. The bottom end was a little light though. The DXD was better still with a good coherent balance from top to bottom, though the top seemed a little extended. The tape was very "musical". Was very balanced from top to bottom but felt that the transient snap advantage was given to the DXD and DSD. The hardest thing for digital to reproduce is the low stuff like reverb tails and room ambience. I want to close my eyes and envision how big the room is. Redbook CD can't do this.

Now here's the dilema. Mastering engineers want the sound very clear and precise with no coloration. If there is going to be any processing, we want to make sure that a 1dB adjustment is because the music needs it. Which leads me to coloration and harmonics. Analog and transformers are best at these. Tape wins out with vinyl a close second. This is what makes things "musical". The other end of the spectrum.

Each format has it's own sound, as most of you know. On a really good Studer with SM900, tape hiss is a non-issue. Same with vinyl. A properly set up system with an extremely clean record, surface noise is a non-issue. With DSD128fs and DXD, I feel bandwidth is a non-issue as well. I've made digital transfers using the 2 formats from tape and alot of the times I can't hear the difference. Now granted, most people don't have the room or equipment to detect the difference between a $10k converter and a $15k converter. But if you want that last 1%, tape wins every time!

 

[Steve williams]

Thanks Bruce

I knew you would enlighten us with your post

 

[MylesBAstor]

This is one area where the source has alot to do with it. I've heard things sound the best on each format. Now we're talking the best of each format.

TT = Rockport or EMT 948 w/45rpm media

Digital = Pyramix or Sonoma workstation through either DAD AX24 or Meitner DAC8IV using DSD 128fs or PCM at 32/352.8kHz

Tape = Studer A80 with Flux ER heads and modified circuits

The majority of sources, tape wins every time, if the source is exactly the same. I was in Japan about a year and a half ago and did a recording of a Jazz trio. I took the feeds from an SSL 9k J mixing board. We recorded in 4 formats; 1/4" tape on an ATR machine, DSD5.6 on a Korg, DXD using Pyramix and DAD AX24 and lastly Pro Tools at 24/192

The Pro Tools take sucked. It was lifeless and 2 dimensional. The DSD5.6 was really nice. The bottom end was a little light though. The DXD was better still with a good coherent balance from top to bottom, though the top seemed a little extended. The tape was very "musical". Was very balanced from top to bottom but felt that the transient snap advantage was given to the DXD and DSD. The hardest thing for digital to reproduce is the low stuff like reverb tails and room ambience. I want to close my eyes and envision how big the room is. Redbook CD can't do this.

Now here's the dilema. Mastering engineers want the sound very clear and precise with no coloration. If there is going to be any processing, we want to make sure that a 1dB adjustment is because the music needs it. Which leads me to coloration and harmonics. Analog and transformers are best at these. Tape wins out with vinyl a close second. This is what makes things "musical". The other end of the spectrum.

Each format has it's own sound, as most of you know. On a really good Studer with SM900, tape hiss is a non-issue. Same with vinyl. A properly set up system with an extremely clean record, surface noise is a non-issue. With DSD128fs and DXD, I feel bandwidth is a non-issue as well. I've made digital transfers using the 2 formats from tape and alot of the times I can't hear the difference. Now granted, most people don't have the room or equipment to detect the difference between a $10k converter and a $15k converter. But if you want that last 1%, tape wins every time!

Thanks for an interesting post!

First, was that a stock ATR? I take it you only had access to 1/4" machine; would you prefer or do you think the results would be slightly different with a 1/2" machine.

Second, I think one could have predicted that tape would gracefully round transient attack. JGH interviewed KOJ back in 1984 in Stereophile (vol. 7, no.4) where Keith talked about time smearing (related to frequency dependent length of the magnetic gap -- and that Keith claimed can be compensated for electronically since it's fixed in time) and presence edge smearing. Keith defined presence edge smearing as, "a particle to particle creeping or print through of steep transient information. Another problem is the way a playback head distorts a steep wavefront by anticipating it's arrival at the gap......."Because the head's pole faces are extremely long in comparison to the gap width, that steep wavefront will start inducing magnetism into the head a fraction of a second before the wavefront actually passed across the gap. A complex phase shift occurs because the short and long wavelengths are reproduced at slightly different times."

BTW, I think this is one of JGH's best interviews and it's filled with fascinating information about recording technology.

 

[rhopkins]

My listening group has done this a few times now and the answer always comes back the same. Tape is the best source and it's easy to hear every time from the first note. Vinyl comes in second and depends largely on the LP how far back it is in this horse race. Digital (and it doesn't matter what-HRX, SACD, up sampled or plain Jane CD) from a spinner or a hard drive is so far back it's in the corner.

Digital varies wildly so sometimes it can impress. Certainly the HRX titles are some of the best but they will never touch good vinyl and they don't have a prayer against tape.

This has been my experience. I love tape and vinyl. Digital, not so much.

 

[mep]

Well, there are more than a few in this forum who will pounce on your post and tell you that the reason you prefer analog over digital is because of all of the analog distortions that somehow sound more pleasing then digital and fool you into thinking that analog actually sounds better then digital. There are 4,000 measurements and 10,000 graphs that they can whip out to show you how misguided you are. That said, I am in your camp Rhopkins.

 

[audioguy]

+1

I've not heard tape but have heard great vinyl and have heard digital as good as it gets (at the time) and analog still wins. And you are right: the 10,000 measurements say I'm wrong but I will go with my ears/brain/heart on music.

 

[microstrip]

TT = Rockport or EMT 948 w/45rpm media

Bruce,
My eyes focused on this line. How does the Rockport compare with the EMT948? BTW, were you using the standard grooved rubber mat in the EMT?

 

[Bruce B]

Bruce,
My eyes focused on this line. How does the Rockport compare with the EMT948? BTW, were you using the standard grooved rubber mat in the EMT?

This is a modified unit with custom vdh cart. We've used rubber, felt, cork and acrylic mats.

 

[rhopkins]

I stumbled upon this today and thought it was interesting. Scroll down to the little graph near the bottom. The text reads as follows:

"We found a graph at Merging Technologies that tells more than the proverbial thousand words. In this graph MT compares the impulse response of analogue playback to various forms of digital playback and shows that only full DSD can compare to analogue. It is this lightning-fast ‘unclipped’ impulse response that makes analogue sound much more ‘live’ than digital."

http://www.6moons.com/audioreviews/feickert3/blackbird_3.html

 

[FrantzM]

I stumbled upon this today and thought it was interesting. Scroll down to the little graph near the bottom. The text reads as follows:

"We found a graph at Merging Technologies that tells more than the proverbial thousand words. In this graph MT compares the impulse response of analogue playback to various forms of digital playback and shows that only full DSD can compare to analogue. It is this lightning-fast ‘unclipped’ impulse response that makes analogue sound much more ‘live’ than digital."

http://www.6moons.com/audioreviews/feickert3/blackbird_3.html

Hi

I resisted .. I really resisted but I am fallible so I caved in, I took the bait .. This graph is NONSENSE ... Such impulse response is not possible with ANY of the current analog reproduction system .. simply can't; that would suggest a response flat in the GHz from this "analog" system... Phono ... tapers off around 30 Khz and would not reach 100 Khz .. neither would tape when the bias is right there anyway ..So scratch these two what is left? Cassette, eight track? TWhat is that "analog" system? Please do tell us ..
Just before bowing out this is the impulse response of a 4 GHz Optical receiver ... 4 Giga Hz is 4 Millions Kilohertz

Even this Optical receiver has some kind of overshoot and far from the Merging technology perfect analogue device ... bowing out ...

 

[microstrip]

We've used rubber, felt, cork and acrylic mats.

And the winner was?
BTW, did you modify the motor control electronics or only the phono circuits?

 

[amirm]

Hi

I resisted .. I really resisted but I am fallible so I caved in...

I win Frantz. I post my reply and then deleted it without hitting post . I actually spent half hour trying to find the source of that article and found nothing even on MT's site. I think by analog they mean just a pulse generator.

 

[rhopkins]

It must be marketing then. Those guys are always trying to prove something and graphs are the easiest way I guess.

It did look interesting though.

I was having a second look at it and I'm pretty sure the original intent of the graph was to show how various digital sources compare. I think that's what it says on the graph itself. They never actually say that it's a turntable, or cassette deck or eight track, or whatever. They simply say that it's an analog signal and then show you how the various digital sources try to recreate it.

I think that's what it's saying... Maybe one of these engineering types can explain what they are doing if this assumption is incorrect.

It does look like 6 moons may have used it incorrectly though based on your comments. Never trust the media!

 

[hdplex]

The picture on the 6moon site is from Merging marketing page for DXD.
http://www.merging.com/products/show?product=1&page=11
which is originated from this document : http://www.merging.com/uploads/assets/Merging_pdfs/dxd_Resolution_v3.5.pdf

I guess the true origin of this picture is from this AES paper.
http://tech.juaneda.com/en/articles/dsd.pdf

Quote
"This effect is clearly illustrated in Fig. 9. The impulse responses of 4 different systems in a multichannel configuration are depicted: a 48 kHz system, with a bandwidth of 20 kHz (that is, 8 kHz transition bandwidth is allowed for anti-aliasing filtering), a 96 kHz system with 35 kHz bandwidth (26 kHz transition bandwidth), a 192 kHz system with 75 kHz bandwidth (42 kHz transition bandwidth) and an SACD system with 95 kHz bandwidth. Though none of the systems reproduces the input exactly, the DSD system shows the least artifacts. Clearly, the 48 kHz system has great difficulty in reproducing the click; due to the steep filtering it starts ringing at a -30 dB level approximately 1 ms before the click, which is very audible. Also at the higher sampling frequencies, the ringing phenomenon cannot be removed, though it is reduced significantly. Only the DSD system is very effective in suppressing the ringing effect, due to very slow filtering above 95 kHz. The price to pay for this is the increase in noise floor with respect to the other systems; however, as the noise floor contains only high frequency components which are uncorrelated with the audio, they are not perceptible. "



Fig. 9: Responses (from left to right) of a DSD, a 192 kHz, a 96 kHz and a 48 kHz system on a -6 dB block input (‘click’) of 3 ?s duration.

I remember this since there was a really heated 1 bit SACD vs linear PCM debate at the time.

 

[Steve williams]

Great first post.

Welcome to the forum

 

[RBFC]

Comparison of different sampling rates

I found the graph presented by hdplex very interesting. We've had discussions here about how DACs tend to have more distortion when run at 192kHz as opposed to 96kHz, both considered high-resolution sampling rates. Although the 192kHz rate may produce a bit more distortion, perhaps the superior impulse response (as shown in the graph) may account for some or much of the popular impression that 192kHz is preferable when available.

Given the exact same master, what is the opinion of the membership on whether 96kHz or 192kHz will produce superior sonic results?

This is quite intriguing, since the audio specialty labels seem a bit split on this one.

Lee

 

[amirm]

That impulse response looks that good simply because DSD/SACD has such high sampling rate (2.8224 MHz). The way to make the graph look good is by having a lot of high frequencies mixed together. And that sampling rate accomplishes that.

What the graph doesn't show is that at that high frequency, SACD has only a single bit of resolution. This means as you approach its maximum frequency, you have incredible amount of distortion as clearly the signal does not toggle as a binary message. Even in the audible range, there is a ton of distortion. Heavy noise-shaping is used to push that distortion up.

Real SACD systems have filters about 70-80 KHz (for the fear of blowing up your tweeters), so in reality, they never look any better than the 192 KHz PCM as that has a bandwidth of 96 KHz from impulse response anyway. Remember, as I mentioned, the graph is a function of how much high frequencies exist in the system. Filter them out and you get less sharp transients. See the difference between 44.1 KHz and 192.

Philips paper above tries to say that based on research on Hypersonics, it doesn't matter if you have a ton of ultrasonic noise. That the ear only cares if you have preserved correlated music at ultrasnoic frequencies. But that research is disputed and not accepted as fact.

All of this said, some of this clearly lands in the world of unknown. It is extremely difficult to design system performance at frequencies greater than 20 KHz. Who the heck really knows what stuff out there is good or bad, when in traditional measures we can't hear those sounds? It is like saying what color wall is more pleasant in a super dark room .

 

[microstrip]

... It is extremely difficult to design system performance at frequencies greater than 20 KHz. ...

Amir,
It is something that puzzles me. The performance of analog semiconductors has improved by several orders of magnitude during the last 25 years. Computational power of processors also increased by several orders of magnitude. ADC and DAC improvements were huge. And we still say "it is extremely difficult to design system performance at frequencies greater than 20 KHz. How can we explain it?