Measurements of Nakamichi Dragon Cassette Deck

[egellings]

This is a detail measurements of the iconic, Nakamichi Dragon Vintage cassette deck. It was kindly brought to me at a meet at our local audio store (Gig Harbor audio).

For those of you too young to know , the Nakamichi Dragon has the reputation of being the best cassette deck available at the end of the cycle for the format. It came out in 1982 and retailed for USD $2,499. That would be $6,500 today's dollar so quite a lot of money. I was too poor to afford one at the time so it was a pleasure to get my hands on one finally for this test.

We were testing it in a dark room and this is just with the light of a phone so please forgive the poor lighting:

View attachment 18584​

Unlike digital products and amplifiers, we are at the mercy of calibration tapes for testing such products. The owner had a few of them and that is what I used for testing.

I unfortunately did not capture the results of 1 kHz tone but here are some other frequencies.

315 Hz:
View attachment 18585

Oh boy. We are so used to SINAD (signal over distortion and noise) of 90+ that numbers like 41 dB seem so, so low!

Likewise distortion components are just 40 to 50 dB down from our main tone as opposed to 90+ in digital.

Let's jump up to 3 kHz:
View attachment 18586

Channels are mismatched in both phase and amplitude. The Dragon is supposed to auto-calibrate the phase but clearly it is not able to do so.

Increase in frequency has also increased our distortion.

Widening of the 3 kHz tone at the bottom shows random jitter/tape speed variations.

Lastly here is 12.5 kHz response:
View attachment 18587

Granted, the levels are low but 4.5% distortion??? Phase and amplitude errors followed us here too.

Conclusions
Even though this is not a thorough test and the pedigree of test tapes is unknown, these results are more than depressing for those of us who cherished this marquee audio product. Worst of the worst digital products have performance that is hundreds of times better. Oh, well.

-------------
As always, questions, comments, recommendations, etc. are welcome.

If you like this review, please consider donating funds to support these reviews using:
Patreon: https://www.patreon.com/audiosciencereview)
or upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).

It still might have Big Tone, though.

 

[ronniebear]

Youtuber VWestlife has video reviews of the TEAC double cassette deck, the TASCAM double cassette deck, and the TASCAM combo CD player/cassette deck. For each of these decks, his reviews include his own technical measurements of wow-and-flutter, speed accuracy and frequency response. Based on Vwestlife's reviews, the performance level of each deck is somewhat better than the official TEAC specifications. Watch the three videos below: these decks aren't on the same quality level of higher-caliber decks from the '80s and '90s, but similar to lower-cost entry-level tape decks from the '80s (i.e. models originally sold for around US$150). Draw your own conclusions, these seem to be the best currently-available brand-new component cassette decks.
New TEAC W-1200 cassette deck - Detailed review
New TASCAM 202ᴍᴋVII cassette deck - Detailed review
New TASCAM CD-A580 CD/MP3/cassette deck review

​

 

[Count Dacula]

Good to know - there are better decks in terms of performance and calibration options, but nearly all vintage cassette decks are three to four decades old and aren't all that easy to get correctly repaired.

I have a NiB SONY ES deck that I have no idea what to do with. PM me if anyone has any thoughts.

 

[witwald]

...the difference in straight tape and using Dolby C is obvious and important. I don't know how the THD changes with and without Dolby C (I would think not much). But noise floors are much better with Dolby C and it usually improved the high end response on machines vs no Dolby C.

Below is a set of frequency response and noise measurements for a Nakamichi ZX-7 cassette deck when using Nakamichi ZX Type IV metal tape. This was published in the brochure for the ZX-7. As mentioned above, Dolby C improved the high-frequency response as compared to Dolby B, as well as significantly improving the signal-to-noise ratio over a wider frequency range.

 

[anmpr1]

Below is a set of frequency response and noise measurements for a Nakamichi ZX-7 cassette deck when using Nakamici ZX Type IV metal tape. This was published in the brochure for the ZX-7. As mentioned above, Dolby C improved the high-frequency response as compared to Dolby C, as well as significantly improving the signal-to-noise ratio over a wider frequency range.

From my admittedly long time ago experience, B Dolby was subjectively better than C, in spite of the latter's spec improvement. Why that was for me?, I can't say. I never used top tier decks, usually mid-range, so perhaps on a better machine it would have been better.

FWIW, the best sonic improvement to my ears were Type IV 'metal' tapes, that were vastly superior to what went before, even on moderately priced decks.

Toward the end of the cassette popularity run, Dolby S was reported to offer a significant sonic improvement, but I never heard an S deck.

 

[sergeauckland]

From my admittedly long time ago experience, B Dolby was subjectively better than C, in spite of the latter's spec improvement. Why that was for me?, I can't say. I never used top tier decks, usually mid-range, so perhaps on a better machine it would have been better.

FWIW, the best sonic improvement to my ears were Type IV 'metal' tapes, that were vastly superior to what went before, even on moderately priced decks.

Toward the end of the cassette popularity run, Dolby S was reported to offer a significant sonic improvement, but I never heard an S deck.

The main reason why Dolby B was often better than C was that line-up for C was far more critical than for B. Dolby mistracking, i.e. the difference between the encoding and decoding was more audible with C as it covered a much wider bandwidth.

When I sold Nakamichis in the mid 1980s, I would line them up on sale to the customer's choice of tape. Factory line-up was done on Nakamichi's own brand tape, that was if I recall, manufactured by Maxell, but I may be wrong on that. If asked, I would recommend That's Tape, as at the time I thought they had the 'best' performance, and I encouraged customers to buy at least a box of tapes, ensuring they were from the same batch..

The difference between a properly aligned machine and one used straight from the factory on even good tapes like TDK or Maxell was considerable. Dolby B tracking was just about acceptable, Dolby C (mis) tracking was quite audible. I had a set of Nakamichi test tapes and BASF frequency response line-up tapes as well as an example of Nakamichi's own tapes. Factory line up using Nakamichi tape was pretty good, all spoilt when a user chose another brand of tape.

Tape in general was a fussy medium, we would line up our studio machines every morning, and there were marked differences between batches of nominally the same (usually Scotch 206) tape. That was running half-track stereo 15ips on Ampex professional machines. Cassettes had no chance with their much lower speed narrower track and tiny dimensioned tape transport if using tape different to the one they were lined up on. This didn't matter a lot on 'normal' domestic tape machines, especially the auto-reverse ones where there was no chance of azimuth being consistent, but on a Nakamichi, 'perfect' line-up made a very worthwhile difference.

S.

 

[restorer-john]

From my admittedly long time ago experience, B Dolby was subjectively better than C, in spite of the latter's spec improvement. Why that was for me?, I can't say.

Probably poorly aligned or badly set up decks. Not your fault, but Dolby C was light years better than the best B type encoders. A good deck with Dolby C, switchable MPX and good Rec/PB alignment was incredibly good, even at 1 7/8" IPS.

 

[MRC01]

Below is a set of frequency response and noise measurements for a Nakamichi ZX-7 cassette deck when using Nakamici ZX Type IV metal tape. This was published in the brochure for the ZX-7. As mentioned above, Dolby C improved the high-frequency response as compared to Dolby C [correction: B?], as well as significantly improving the signal-to-noise ratio over a wider frequency range.
...

What I find interesting about that graph is that it suggests the advantages of C were only at high recording levels, with B & C having the same frequency response at -10 dB and lower. Back in the era of cassette tape before the loudness wars, most of the music would be at these lower levels.

My experience too was that Dolby B sounded better than C. This could be due to its greater sensitivity to tape head alignment and perhaps also bias adjustment fine tuned to the particular tape.

 

[anmpr1]

What I find interesting about that graph is that it suggests the advantages of C were only at high recording levels, with B & C having the same frequency response at -10 dB and lower. Back in the era of cassette tape before the loudness wars, most of the music would be at these lower levels.

My experience too was that Dolby B sounded better than C. This could be due to its greater sensitivity to tape head alignment and perhaps also bias adjustment fine tuned to the particular tape.

Top tier decks usually featured auto bias. I'm sure that that by itself would help any tape. For my use, mostly to make dubs for my car, it didn't matter, one way or the other. I never considered the medium really hi-fi, but it was sure convenient as hell. And given its physical limitations, was amazing that it sounded as good as it ever did.

Higher end Naks, ReVox, Tandberg et al. were always impressive machines. I remember when the Nakamichi 1000 came out. Few could really wrap their heads around the price, and sophistication. My limit was decks like the Pioneer CT 9191, Teac 450, and such. My last deck was a mid range Denon, that was nice, but after a few years the logic controls went haywire, making it unusable. By then CD had come out, and that was pretty much that. Even for cars.

 

[MRC01]

Top tier decks usually featured auto bias. I'm sure that that by itself would help any tape. For my use, mostly to make dubs for my car, it didn't matter, one way or the other. I never considered the medium really hi-fi, but it was sure convenient as hell. And given its physical limitations, was amazing that it sounded as good as it ever did.

Higher end Naks, ReVox, Tandberg et al. were always impressive machines. I remember when the Nakamichi 1000 came out. Few could really wrap their heads around the price, and sophistication. My limit was decks like the Pioneer CT 9191, Teac 450, and such. My last deck was a mid range Denon, that was nice, but after a few years the logic controls went haywire, making it unusable. By then CD had come out, and that was pretty much that. Even for cars.

Denon made great tape decks. I bought one cheap on eBay and restored it myself, and it measured better than the Nakamichi Dragon that Amir meaured a while back.

Denon DRM-540 Cassette Deck Measurements

Seeing @amirm 's review of the Otari reel to reel tape deck motivated me to do some calibration and measurements of a Denon DRM-540 that I bought on eBay for about $50. I don't have a scope or fancy testing gear, so I used my ESI Juli@ sound card. That should be more than sufficient for an old...

www.audiosciencereview.com

 

[dlaloum]

What I find interesting about that graph is that it suggests the advantages of C were only at high recording levels, with B & C having the same frequency response at -10 dB and lower. Back in the era of cassette tape before the loudness wars, most of the music would be at these lower levels.

My experience too was that Dolby B sounded better than C. This could be due to its greater sensitivity to tape head alignment and perhaps also bias adjustment fine tuned to the particular tape.

The noise was highly audible and much better ( reduced ) with Dolby C.

On an auto bias Revox, results with many tapes were excellent, I tended to prefer basf chrome.

 

[musicforcities]

My father had a Nak 700ii it was beautiful to look at and use with a thick slab of brushed stainless and touch sensitive buttons. With all the fidget alignment buttons behind a door.

But I could never get decent sound out of it. No high frequency at all. Everything sounded like a cheap type I tape. I used all the right settings. And every top tape brand and type: TDK SA, SA-x, Maxwell, etc.

Unless you aligned the heads for each and every tape. Commercial tapes were a nightmare. Even those recorded on the machine often needed alignment the, for example from side a to side b. When perfectly aligned—and I mean perfect—it sounded very good.

Perhaps it was the sideways loading tape orientation and the door: gravity was working against consistent alignment in a way it doesn’t in top loading or the more common front loading.

Or maybe something was broken. But as far as anyone know it was in proper running order.

One of the oddest experiences I have ever had with a piece of hi if that supposedly was great, measured well if one was to believe the published specs and stereophile review. But jeez what a fiddly beautiful beast it was.

Also took up tones of space due to it vertical form factor. But it was sort of sculpture so who cares about that.

 

[musicforcities]

the pedigree of test tapes is unknown,

Even if the tapes were mint condition and stored decently many of the once “great” cassette tape formulations age relatively poorly due to degradation, oxidation, issues with the binder etc. storage in heat and humid conditions will cause faster aging of course.

Inside the machine belts freewheel bearing lube drys out, the lube and gearing for auto align mechanism also dry out or get dirty , tape heads get dirty and oxidize, etc etc.

But even a perfect Nak Dragon using type IV was limited by the tech spec of the cassette format…the SNAID was just never going to be great. I heard them back in the day and they sounded great until one played a well mastered CD or reel to reel or a vinyl record on good turntable.

 

[witwald]

Top tier decks usually featured auto bias. I'm sure that that by itself would help any tape.

Quite a large number of higher-tier cassette decks provided adjustment controls for tape sensitivity, bias and tape-head azimuth. On the Nakamichi ZX-7 three-head cassette deck, these were manual controls, but the calibration process was quite straightforward to accomplish, and bias and sensitivity settings were individually retained for Type I, Type II, and Type IV cassette tapes.

Other three-head cassette decks, such as the AIWA AD-F770/AD-F990, enjoyed a fully automatic microcomputer system that adjusted bias, equalization and sensitivity.

Even some lower-tier cassette decks, such as the Harman Kardon CD391 two-head cassette deck, had manual record calibration and fine bias trim controls on the front panel that could be easily adjusted for the tape type being used.

 

[pseudoid]

Frank, if you do ever decide to sell I would be interested, to add to the things I never use but like to look at from time to time.
BW Keith

Did @Frank Dernie ever purge it from his inventory?

 

[USER]

Realistic SCP-18 Stereo-Mate Portable Cassette Player Measurements

Seeing as this is the de facto member-submitted cassette player measurement thread and seeing that portable cassette players are back in fashion, I thought it would be good to post these results here.

This player was the lowest-end model Radio Shack sold in the early 1990s so it has the potential to act as a baseline with which to judge any portable player made today (I hope to measure one soon, though not that FiiO as I nor anyone I know would purchase one for the going price). These results should be the poorest that are reasonable to expect and demand and I think it is more than fair to call out anything that performs worse. Certainly there was way worse back then, but Radio Shack was a popular and smart choice for the budget-minded consumer. They made good and reliable products. And so I think it makes a good ASR baseline. Note that I got this because it was in NOS condition, super cheap ($15 shipped), and because my personal experience showed me that the belts Realistic used were phenomenal. I have measured 3 SCP-32s that were in excellent condition and they all had wow and flutter that beat spec! This isn't the junk Sony used that turned to goo decades ago. So if there is anything that will provide the best opportunity to own something that works reasonably close to like-new without a major restoration project, this is it.

Again, this was the entry model. Controls are extremely limited. Play, stop, fast-forward. That's it. Specs are as follows. Attached is a scan of the manual for posterity.

Wow and flutter measurements (all my measurements) were made using professional ABEX test tapes. I also used an E1DA Scaler and Cosmos ADC. The player was powered by battery. I got 0.3V out of the headphone out at max volume using a 1kHz test tape. Pretty good for these!

Given the limitations of the player the number of measurements I could take were similarly limited. Any tips, corrections, or advice are appreciated.

Beats spec! (As expected from Realistic.) WRMS results were generally stable. Speed could be further adjusted, though this is likely close to the best I would be able to get. Update: I took out the belt to double check things and I noticed that it has a kink from resting taut against the motor pulley for all these years. So in good condition the portable performs better than these results show. I am going to try boiling it and giving it some rest to see if it returns to normal shape. Still, specs leave a lot of room for wear and tear.

1kHz and 12.5kHz results after azimuth adjustment. 12.5kHz Lissajous. (You want a nice 45 degree angle line with these.) A bit of instability all around. I wasn't expecting to get better given the quality and as correction is limited.

Finally, 1kHz RTA. Not able to hit that 47 dB SNR, at least with REW and this test tape.

This last one was with nothing playing. A lot of noise.

SINAD pretty much compares to all the other decks I measured being only a few dB worse but please don't mistake this for good performance. The tape itself likely dominates SINAD, though the differences between them are telling. In actual use the hiss is very audible and gets in the way. More, hum is noticeable, especially when there is silence and particularly with sensitive IEMs, which is what you would want to use with classic portables today. Unlike the hiss, it can become much less noticeable when music is playing. More, the soldering on these things is not of the highest quality. That said, it seems to be coming from the motors. These older plastic models did not have adequate shielding. I will try out some nu-metal tape in the future to see if these models can be improved. At some point I am going to try to check the capacitors to see if any are faulty and do further work to see if the SNR can be improved, but I do think that most of this was always there and that it was a good purchase for this purpose. I do read of similar complaints of the newer portables so this may be very close to what we get with them--namely, low quality during any era and comparable to a $20 player (~$40 today with a lot of inflation).

Finally, I attempted to show something related to the frequency response as best as I could. Given limited options I recorded a track and compared it to the CD version. Note that this is bonus info and not to be taken as final because of the many variables that are unaccounted for. That said, for the most part, the sound should be similar in the most sensitive areas. A little boost between 6-10Hz, though not much, and a drop out in the highest and lowest frequencies. I would bet that the boost is related to the mastering as this was a Dolby tape. The drop outs are likely the limits of the portable and medium. But not bad! This does leave me with one question: is that small bump above 6kHz responsible for me thinking that hiss was a little louder than usual on this portable?

 

[knobtwiddler]

Youtuber VWestlife has video reviews of the TEAC double cassette deck, the TASCAM double cassette deck, and the TASCAM combo CD player/cassette deck. For each of these decks, his reviews include his own technical measurements of wow-and-flutter, speed accuracy and frequency response. Based on Vwestlife's reviews, the performance level of each deck is somewhat better than the official TEAC specifications. Watch the three videos below: these decks aren't on the same quality level of higher-caliber decks from the '80s and '90s, but similar to lower-cost entry-level tape decks from the '80s (i.e. models originally sold for around US$150). Draw your own conclusions, these seem to be the best currently-available brand-new component cassette decks.
New TEAC W-1200 cassette deck - Detailed review
New TASCAM 202ᴍᴋVII cassette deck - Detailed review
New TASCAM CD-A580 CD/MP3/cassette deck review

​

Aiwa and Sharp (as well as others, but I can think of these 2 brands OTOH) both made boomboxes that had 0.05% RMS Wow + Flutter. The Aiwa ones retailed for around $150USD (and you got a tuner, amp + speakers into the bargain). No modern deck gets anywhere near 0.05% today. My understanding is that ~ 0.15% is as good as it gets (below 0.1% is considered inaudible, so don't play sustained piano notes with a modern deck) - and this would be on a good example (Techmoan measured W+F on a modern walkman @ over 0.2%). If boom-boxes from the early '80s had around 1/3rd of what a modern standalone deck can offer, it should give an idea of just how bad the modern offerings are.

Back in the day, only a handful of firms made transports. Alps (Yamaha and others), Sankyo (Naka, Teac, NAD etc.) were the 2 largest OEM mechanism suppliers. Sony and Technics made their own transports. Generic Sankyo mechanisms could get under 0.03% W+F. The Dragon started out using the 'classic' Nak mechanism, but moved over to Sankyo in later years. The Dragon is unique, in that is has DD capstans on both sides - only Revox (and one or two more obscure, smaller OEMs) fitted DD capstans on both sides. By doing this, Nak dialled the capstan belt out of the equation, hence the Dragon's unimpeachable W+F spec (0.019%). The Revox B215 has 4 DD motors. Nak never made a machine with DD reel motors, which creates a service headache in terms of motor flat-spots and idlers (later ones used gears, which were noisy, but lower in maintenance requirements).

To my knowledge, the only mech in production today is the Tanashin, which on a good day might achieve [highly audible] 0.15% RMS... In contrast to the Quartz-Locked DD motors from days of yore, the Tanashin is entirely belt-driven and has more in common with one you might see in a 1960s shoebox recorder.

Due to hole eccentricity, none of my turntables can approach my Naka CR-7 for W+F performance - and I have several test records. I don't doubt that the TT platters on their own have stellar W+F if you measured it optically, but hole eccentricity means I cannot find a record that measures below 0.05% RMS, whereas my CR-7 happily outputs 0.025%. It doesn't suffer from rumble or surface noise. With NR switched on, you could argue that it gives most turntables a good run for their money.

I get the impression that Philips never envisaged Compact Cassette evolving that much beyond where it was in the mid '70s. I think they felt that serious buffs would stick to R2R. An arms race between the big corporate players such as Sony and Matsushita ensued, eventually giving the format quite listenable (albeit awful, from an objective standpoint) performance.

NB - I checked the path of a Sony TC-K777ES2 from the mid 80s on my AP. It gave 0.0033% THD+N combined on loopback (this was just the path looped back - you don't want to think about THD+N coming back off tape...).

 

[USER]

Janryzone JL-101 Stereo Cassette Player Measurements

Here we go. So the following are some measurements of the Janryzone JL-101, which is one of the cheap portable cassette players currently being sold. This was purchased as a generic player so it likely is sold under numerous names (including Burtowic Cassette Player if I am correct). The reason I decided to try this one first is because it is marketed as using metal flywheels and thus as higher quality compared to its competitors. Honestly, it's a convincing argument. The price ranges from $15-$30 so it is comparable to the Realistic player I recently measured, though slightly cheaper when taking inflation into account. In the end I think it is fair competition.

Obviously no specs are given for these. The headphone out put out 0.085V at max output (with 1kHz signal).

These are the wow and flutter and speed measurements out of the box. Again, I used professional ABEX test tapes and an E1DA Scaler and Cosmos ADC. The player is powered by battery here.

Good news and mildly annoying news. W&F is a bit better than the Realistic but it runs fast. It should be noticeable. Luckily that is adjustable.

I can get speed deviation close to 0 but it seems to always be off again when a cassette is reloaded. This I think is a fair, representative result. Periodicity at 0.13Hz (wow), which I didn't see in the Realistic.

1kHz oscope and lissajous results. Some channel imbalance.

12.5kHz oscope and lissajous. Much worse performance. I may be able to adjust the azimuth settings some so but I am not expecting much better. Likely not worth the risk. FR should be more limited.

Noise and distortion measurements at 1kHz, max volume. Comparable to what we had before. I am guessing the test tape dominates these. Results when nothing is playing seems better in the higher frequencies than the Realistic.

SINAD is a little better at the volume I used with my IEMs.

Again, as with the Realistic, I could hear the motor hum and, more intrusively, hiss. The sound was a little smoother tonally on this one though and I might prefer it. This seems to be what you get with these cheap models. I'd need to listen to and measure a refurbished, mid-tier, vintage model to know whether these are at all worth it.

Finally, I tried to see what I could do to show the frequency response. So here is the comparison of recordings of the same track. The Realistic shows a wider response. The 12.5kHz azimuth measurements above make sense in this context. Noise and headphone compatibility/drivability issues not withstanding, the Realistic should sound noticeably better.

---------------------------------------------------------------------------------------------------------------------

HOWEVER, there is USB out as the player is marketed as being useful for digitization. For these measurements it is no longer powered by battery and is bypassing the headphone out.

The FR is marginally improved this way, though still worse than the Realistic. Clearly the headphone amp chip is of extremely poor quality.

It is hard to compare the noise and distortion results as the output level isn't the same as I am not using the E1DA Scaler. Noise and that 60Hz hum seem buried in the idle measurements.

I then decided to see what happens if I used a Topping HS02 isolator in order to remove the impact of my computer as much as possible.

It is hard to see the difference, but measurements of the player on idle do show minimal impact. Above you see SINAD improvement of nearly a dB.

So if you want to eke out better performance, here you go.

Finally I redid the other measurements as the FR results hinted at a difference in the higher frequencies. The only thing of note in the 1kHz results is that the lissajous was shockingly stable. Not sure what that means.

12.5Hz results. And, again, odd lissajous results.

Finally, for the geeks: check out these wow and flutter results from 2 different test tapes.

The first is my ABEX TCC-114. The second is my TEAC MTT-211NA. Obviously they fluctuate a little bit but they are effectively giving the same results here. Pretty darn cool.


In the end I'd say it is comparable to the low end Realistic from 1993. It does some things marginally better, like noise and distortion for the most part, though the hiss in both makes this largely irrelevant when using headphones. If you are recording via USB this may go to the new portable as hum seems to be lessened. Wow and flutter is improved, not insignificantly, but unfortunately speed deviation is worse. You would really need a W&F test tape to improve stock performance as this was too fast (which is a complaint I have read about with respect to the more expensive new portables.) Finally the frequency response, likely the most important thing here, is worse. If we are talking archiving cassettes, perhaps EQ based on these results may help make this worth it over the Realistic. The big question is if the middle-class portables from the golden era actually performed any better.


Stay tuned for another cheapy. And again, any questions, concerns, corrections, or tips regarding my measurements are encouraged. I'd love to do better here.

 

[Count Dacula]

I have a NiB Sony ES TD that I have no idea what to do with.

 

[USER]

"Cassette to MP3 Converter" Portable Cassette Player Measurements

Here are measurements of another no-name portable cassette player currently in the market. What's interesting about this one is that it does the MP3 conversion itself. You simply insert a USB stick and set it to record. However, based on a peek at the internals this portable should provide insight into the general performance of the low-priced portables in the market today. This one costs around $20. So about $10 in the early 1990s. (Crazy.)

Once again no specs. Here are the internals. As you can see the flywheels are plastic. The button layout seems similar to those of many other cheap portables. I'm going to assume that they are mostly the same thing. More, this comes with the same accessories as the other portable: same pouch, same USB power cable, and same earbuds (!). Even the boxes are pretty much the same. I'm guessing the vast majority of these are made by the same manufacturer.

Unlike the previous player, the USB port is strictly for power and MP3 output. Output voltage is 0.105V at max volume (using 1kHz signal). Again, I used professional ABEX test tapes and an E1DA Scaler and Cosmos ADC. The unit was powered via USB.

For this measurement I used my cheap dupe wow and flutter test tape from eBay as I always do during my first go around. Have to use protection, lol. The results are close to the real thing, though usually it gives a slightly "better" results. It is less accurate the better the results are. Not surprising. But here it is usable.

YIKES!!! The scraping noise coming from the motor spindle wasn't a good sign. Numbers were fluctuating wildly. Weighted quasi-peak was higher than 5%. I quickly decided to stop before I ruined my test tapes.

I tinkered with it but did not see any results that told me it was worth risking my good tapes. For some reason as I was getting ready to send it for electronics recycling out I decided to play the second side of the cassette in reverse. Hmmm. It works. I guess something is off in the build. These results were taken after some adjustment.

These results are worse than the other cheap portable. Speed deviation changes a bit after each insertion and I would not say it is better than the other. (The cassettes are not held firmly on these modern players.) It seems to me that you want those copper flywheels. At the very least they signal higher quality components all around.

Noise and distortion at 1kHz seems comparable to what we have seen. Again, the test tape is likely dominating noise. Higher frequency issues seem attenuated.

Idle measurements. USB and battery power. 60Hz hum is the biggest difference.

1kHz oscope and Lissajous. The results are similar to what we have seen.

12.5kHz looks rougher. We should expect a similar drop-off in the frequency response. Azimuth is not adjustable here so this is what we have.

Finally, here are comparisons of the frequency responses of my test track. First is a comparison to the Realistic, the second a comparison to the copper flywheel cheapy via USB out. Is that noise shaping? Anyways, no bueno.

I listened to it briefly. Hiss is the best of the three and the motor hum is much, much better. Doesn't matter though. It doesn't sound good on my IEMs. Dull and lifeless. The FR points to what is wrong.

I think it is obvious that the portable with the copper flywheels is better than this and ones like it. Perhaps there are build issues here making the results worse than those of a good copy as I am playing it in "reverse." This would point to build quality problems that very clearly tell us to avoid these altogether. Still, I wouldn't expect this to compare to the other one even if this was a perfect copy. In the end this doesn't meet reasonable low-end standards from the 90s. I do understand why this seems to be marketed as useful for digitizing voice recordings. And perhaps this is (barely) good enough for that. But again, go for the other one with USB out if you want something cheap. But also do get yourself a reasonably priced dupe wow and flutter test tape as you are going to need it to adjust speed.

--------------------

Bonus update: the earbuds that come with these Chinese portables are garbage as expected. Super, super sensitive though.