The first new quality cassette tape in ages. Much has been written about it
here and
here and
here, but so far no objective assessment has been done. Thanks to the cooperation of
the Bandjesfabriek I found myself in a position to change this, leading to the publication of this article, on Cassette Store Day of all times!
Methods
The test method is described in detail
here.
The test field
Anticipating the arrival of the tape sample I had to decide which cassettes to compare to. A given is Maxell UR (1994), simply because that is what I am using for calibrating all my machines to. It is or should be close enough to the contemporary IEC I reference, and thus serves as a standard for compatibility.
Historically I was mainly a type II user, and as such I do not have many premium or super ferrics. In 1983 I purchased quite a few Maxell XLI and XLI-S, but these really are not the best examples of this breed. I have a lot of UDI (1985), but these have worn badly and exhibit gross distortion when recording their first few minutes. I have NOS UDI-CD (1996): excellent, but they demand a high bias and thus deviate too much from the norm. In the end I settled for UDI-S, an advanced premium (i.e. non-doped) ferric from 1986. I only have two or three of these, but they are excellent performers and still in good shape.
Finally I added BASF FE-I (1988). It should be close to the norm. Also, given RTM's heritage there might be a family link between FE-I and Fox. We shall see.
What then with present-day Maxell UR? That is a very
pertinent question. Today's UR is about the only tape that is reliably
available, and any newcomer must at the least be a significant
improvement over UR. But I don't have any of them, and I cannot go out
in the street and buy one. Rumours are that today's UR, which is
technologically not related to the classic Maxells, is inferior to the
old tapes. So after this page was published I purchased new URs, and succeeded in measuring them in November. I also added tape comparisons from another Tapeheads member. As such this article was significantly updated on 10/11/2018.
The references
Maxell UR C90 (1994)
Relative bias: (reference)
Relative sensitivity: (reference)
THD @ Dolby level: 0.42%
MOL400(THD=1%): +3.4dB (compression = 0.2dB)
MOL400(THD=3%): +5.3dB (compression = 0.6dB)
MOL1k(THD=3%): +2.9dB
SOL10k: -1.3dB
SOL15k: -7.3dB
Bias noise: -49.5dB, -52.5dB(A)
Dynamic range: 57.8dB
Maxell UR C90 (1994) with Nakamichi Cassette Deck 1
Relative bias: (reference)
Relative sensitivity: (reference)
MOL400: +2.6dB
I include this measurement to show the impact of the differing bias setups on MOL: +5.3dB on the BX-300, but only +2.6dB on the CD1.
Maxell UDI-S C90 (1986) with Nakamichi BX-300
Relative bias:0
Relative sensitivity:+0.1dB
THD @ Dolby level: 0.55%
MOL400(1%): +2.2dB (compression = 0dB)
MOL400(3%): +5.2dB (compression = 0.3dB)
MOL1k(3%): +3.6dB
SOL10k: -1.2dB
SOL15k: -7.4dB
Bias noise: -51.0dB, -54.0dB(A)
Dynamic range: 59.2dB
This is a very compatible tape that improves over UR with a lower noise. A new premium tape should aim at this sort of performance.
BASF FE-I C60 (1988) with Nakamichi BX-300
Relative bias: -1
Relative sensitivity: +0.2dB
THD @ Dolby level: 0.67%
MOL400(1%): +1.2dB (compression = 0.0dB)
MOL400(3%): +4.3dB (compression = 0.3dB)
MOL1k(3%): +3.5dB
SOL10k: -0.9dB
SOL15k: -6.9dB
Bias noise: -48.2dB, -51.3dB(A)
Dynamic range: 55.6dB
Maxell UR C90 (2012-now) with Nakamichi BX-300
This is the version of Maxell UR that is presently available in Europe. It has the 'globe' wrapper and states to be made in Malaysia, and distributed from Hungary. The shell has four screws, no hum shield, and appears rattlier than older 'real' UR.
Reducing the bias relative to UR 1994 resulted in very smooth and flat response curves. However, undulations in the upper octave suggest a less than ideal tape-head contact.
Relative bias: -1
Relative sensitivity: +0.7dB
THD @ Dolby level: 0.48%
MOL400(1%): +3.1dB
MOL400(3%): +5.6dB
MOL1k(3%): +2.9dB
SOL10k: -1.3dB
SOL15k: -8.3dB
Bias noise: -47.3dB, -50.3dB(A)
Dynamic range: 55.9dB
Maxell UR C90 (2012-now) with Nakamichi Cassette Deck 1
In this case keeping the bias at the same level as UR 1994 gave the flattest response. Curiously the upper octave was stable this time.
Relative bias: 0
Relative sensitivity: +1dB
THD @ Dolby level: 0.68%
MOL400(1%): +1.2dB (compression = 0.0dB)
MOL400(3%): +4.0dB (compression = 0.3dB)
MOL1k(3%): +0.8dB
SOL10k: -0.4dB
Bias noise: -47.1dB, -50.2dB(A)
Dynamic range: 54.2dB
Shortly after doing these measurements I decided on aligning a CR-1E I am refurbishing to these URs. I thought it fitting that a budget Nak would be tuned to give the best with today's all-present budget tape. But I had to give up: no two places on the tape measured the same in sensitivity, and regardless of bias the highest treble was lost. The frequency curves invariably looked a bit ragged. Thinking that the deck had issues I changed over to UR 1994 and all was well again. Today's Maxell UR may give a reasonable magnetic performance, the real problem with this tape is that it simply is too inconsistent to give a quality result! (I later tried again by mounting the UR tape in the shell of an older model, this helped a bit.)
RTM Fox K7
This is what you have been waiting for. Do note that what I received is a sample, probably of pre-production status, and as such not guaranteed to contain the actual final tape formula that you can purchase from November 2018 on. But it will be of course be very close.
Alignment
Putting Fox in the BX-300 and running a frequency sweep with the bias control in the central position (i.e. Maxell UR 1994) the new tape exhibited a strongly rising treble, starting below 10kHz. Bias was then increased to +2 on the control knob (which runs from -5 to +5), yielding at -20dB a supremely smooth response out to 10kHz, followed with a (typical for my BX-300) 2dB peak at 20kHz. The -3dB point was at 24kHz, but this is also typical for this recorder.
The sweep at -10dB was also very flat, with -3dB at 20kHz. At 0dB 13kHz or so was reached, again without any undue undulations. The +6dB curve showed a remarkable lack of compression below 500Hz.
Increasing bias to +4 pushed down the 20kHz peak, while the range below 10kHz remained very smooth, i.e. no suckout developed in the lower treble. In the end I settled for bias +2 for the remainder of my evaluation.
In Cassette Deck 1 bias was set at +4.The resulting response was less ruler-straight than with the BX-300, but this is a property of this deck.
RTM Fox C60 (2018) with Nakamichi BX-300
Relative bias: +2
Relative sensitivity: -0.2dB
THD @ Dolby level: 0.38%
MOL400(1%): +3.4dB (compression = 0.2dB)
MOL400(3%): +5.9dB (compression = 0.8dB)
MOL1k(3%): +3.0dB
SOL10k: -1.0dB
SOL15k: -7.5dB
Bias noise: -47.2dB, -50.6dB(A)
Dynamic range: 56.5dB
RTM Fox C60 (2018) with Nakamichi Cassette Deck 1
Relative bias: +4
Relative sensitivity: 0.0dB
THD @ Dolby level: 0.65%
MOL400(1%): +1.3dB (compression = 0.1dB)
MOL400(3%): +4.3dB (compression = 0.4dB)
MOL1k(3%): +1.1dB
SOL10k: -0.6dB
Bias noise: -47.5dB, -50.7dB(A)
Dynamic range: 55.0dB
Comparative measurements by Tapeheads member SR2245
I sent my second Fox sample to SR2245 in Germany, who performed a series of measurements (on a big Pioneer deck, a CT-939 or so) and comparisons to competing tapes as well as to the actual IEC standard.
|
IECI Y348M |
RTM FOX 2018 |
Ferro Extra
1992 |
TDK D 1994 |
UR 2016 |
TDK AR 50 1992 |
| Empf. 400 (dB) |
0 |
-0.5 |
+0.1 |
-0.6 |
-0.9 |
+1.2 |
| Bias % |
0 |
-3 |
-6 |
+5 |
-2 |
-3 |
| MOL400 (dB) |
+4.8 |
+3.3 |
+4.6 |
+3.7 |
+1.5 |
+7 |
| SOL10 (dB) |
-3 |
-3.5 |
-2.9 |
-3.5 |
-3.3 |
-2.9 |
| Rauschen (dB) |
-52 |
-52.5 |
-53.5 |
-55 |
-52 |
-56 |
| Tiefendynamik (dB) |
56.8 |
55.8 |
58.1 |
58.7 |
53.5 |
63 |
| Höhendynamik (dB) |
49 |
49 |
50.4 |
51.5 |
48.7 |
53.1 |
The levels in this table are referenced to 0dB DIN, or 250nWb/m (DIN). This is 1.2dB higher than the 200nWb/m (ANSI) zero-level I use on this blog. Bias in the table is as read on the deck's scale which runs -20 ... +20%.
These measurements confirm our earlier findings re bias noise, but on MOL Fox now shows a larger difference with UR '2016'. This was initially puzzling, but in August 2020 we found that SR2245's UR contained indeed a different tape than my own sample, with a lower MOL.
Discussion
My decks are set up for UR 1994, which appears to have a 400Hz sensitivity very close to the IEC norm that was in use until 1994, reference tape R723DG. While formally obsolete, many decks still around today were built to this norm. Compared to this standard Fox has a normal or very slightly below-normal sensitivity, allowing successful Dolby operation on non-calibrated decks.
In October 1979 the IEC adopted BASF lot R723DG of Ferro Super LH-I as the first IEC Type I primary reference tape. Ferro Super LH-I was meant to operate at a higher bias than customary in Europe so far, echoing the trend towards enhanced ferrics that originated a few years previously in Japan. As such Ferro Super LH-I has not much in common with 1978's Ferro Super LH. I have one LH-I: it looks fine, but the magnetic layer barely holds any signal these days! So ... no measurements.
At the end of 1994 IEC reference tape Y348M was introduced, behaving the same at 400Hz, but with a 1.8dB higher 10kHz sensitivity than R723DG, and thus demanding a higher bias (all else held equal). Fox is closer to this newer reference than UR is (UR seems to sit somewhere between R723DG and Y348M at 10kHz). On decks using the post-1994 IEC reference, Fox will be a good match, while on older decks Fox will be slightly under-biased, which is not really a bad thing.
Bias noise, regrettably, is a bit on the high side. While it is indeed 0.4dB better than today's UR, it is at the same level as old budget tapes (Maxell UL 1980, TEAC CDX 1990, Goldstar HP, ...): serviceable, but still 1-3 dB shy of established 90s cooking ferrics like UR, PDM FE, or BASF FE-I, and a full 4dB above the golden age premium ferrics.
When it comes to MOL and SOL, Fox can take quite serious levels, as shown by the BX-300 measurements. Even on the CD1 a MOL400 of +4.3dB over Dolby level was attained, and while this is
better than old UR or FE-I on that same deck, it is only 0.3dB better than one version of the current UR.
Concluding: Fox is a basic ferric tape that seems close to the most-recent IEC norm, and delivers about 0.7dB more dynamic range than today's Maxell UR as shown with my Naks, and some 2dB more dynamic range as shown with the Pioneer deck. Its main advantage, however, may well be that it is much more consistent than UR.
Listening
I never intended this article to become subjective, but one can't break-in the world's first new tape in ages without listening, not? So I gave it a quick spin, using a TEAC UD-501 DAC, Nak BX-300, TEAC HA-501 headphone amp, Sony MDR-MA900 headphones. Without Dolby, peaking at +5 or +6dB above Dolby level, the sound is solid, stable, and surefooted. Noise is entirely acceptable for well-produced pop (I used Sade), but gets intrusive during quieter moments of jazz (the intro of So What?). Switching to 70 microseconds equalisation helps significantly, as does Dolby B. I did not try any classical music.
Edit: much later I wanted to do a project with the Fox. Something electronic, with heavy beats. It came out a bit disappointing, dynamically softened, especially in the bass. I re-did the recording with That's CD/MH and that was appreciably better. But then it should, not?
December 2019 update
It appears that RTM released a new version of Fox at the end of 2019. Allegedly MOL and sensitivity have gone up slightly. There is a little bit more information
here.
INDEX OF ALL CASSETTES
