So what about your earlier statement that: "The Metaxas machines measure as well as the vintage machines."
The quoted figures from the Nagra and Revox are at 15 ips. At 30 ips, the figures are lower. Typical figure with the Nagra T is 0.022%, with 0.032% as the upper limit of acceptability.
.Are you now saying that intermodulation distortion also does not matter to sound quality as well ?The amps with the lower THD and SNR are usually not the best sounding.
If you chase the specs, you are often doing so at the expense of sound quality. Usually, it's because you have feedback loops (like global negative feedback) that end up doing more harm than good. Look at DarTZeel's specs, they are terrible - comparatively. You can argue specs all you want over at Audio Science Review
"This fact forms the basis for the weighting curve shown here. The weighting curve is misleading, inasmuch as it presumes inaudibility of flutters above 200 Hz, when actually faster flutters are quite damaging to the sound. A flutter of 200 Hz at a level of -50db will create 0.3% intermodulation distortion, which would be considered unacceptable in a preamp or amplifier."Wow and flutter measurement - Wikipedia
en.wikipedia.org
"Ideally, flutter should be measured using a pre-recorded tone free from flutter. Record-replay flutter will then be around twice as high as pre-recorded, because worst case variations will add during recording and playback. When a recording is played back on the same machine it was made on, a very slow change from low to high flutter will often be observed, because any cyclic flutter caused by capstan rotation may go from adding to cancelling as the tape slips slightly out of synchronism. A good technique is to stop the tape from time to time and start it again. This will often result in different readings as the correlation between record and playback flutter shifts. On well maintained, precise machines, it may be difficult to procure a reference tape with higher tolerances."
So the measurement is essentially meaningless if you try to compare from one machine to another.
The reason why THD is a poor measurement is because it fails to distinguish between consonant (harmless) and dissonant (harmful) harmonics. This is equivalent to measuring total cholesterol for cardiovascular disease risk. That's why we now measure LDL (bad) and HDL (good) cholesterol separately. This problem is specific to THD as a measurement and not applicable to anything else. If you strip out the consonant harmonic component and only measure the dissonant harmonics, the figure will correlate very well with perceived sound quality problems. This has been shown over and over again with panels of blinded listeners. And to say that W&F measurements are useless because TDH is useless, is like saying measuring blood pressure is useless because measuring total cholesterol is useless. This is poor logic if there is even any logic in it.
Interesting, thanks.
I meant 0.07% - add an additional zero to my numbers! I measure weighted RMS and DIN. Ampex used RMS back in the day, but like I mentioned, the actual numbers for a well maintained machine could be better than spec. The 3M mastering machines (M56/M79) had an even larger flywheel and a capstan well over an inch in diameter with their closed loop drive system. The resulting wow and flutter numbers were extremely good. The Technics consumer machines also used a closed loop system.
I think we need to separate the 'spec' from actual measurements. Unless we have actual measurements, we just don't know. I agree though that a measured 0.09% is not good enough, especially at 30ips, and with modern motor control no less. My 30ips machine, which is far from a Studer, actually measures at 0.06% DIN. We also don't know what proportion is wow verses flutter. Flutter is far more offensive to *me* but measurements lump that together with wow, and wow doesn't bother me as much.
The older Ampex specs tended to be conservative, and the machines actually measured better. I have no idea how conservative or not the specs of the machine which is the subject of this thread are. My 354 is specified for 0.15% wow and flutter, but I routinely measure 0.07%. The only area where the old Ampex specs were 'optimistic' was frequency response at low frequencies.
W&F figures published by European manufacturers are generally DIN unweighted. Other figures are specifically stated when published, and Ampex usually publish several standards, but they differ only slightly.
There are very few people left in the tape department at Nagra, shocking as this may seem. Now that Hubert Bartels has fully retired, we Nagra T owners are on thin ice. These machines are definitely not for DIY types. I have heard that Nagra will start manufacturing the QGB adapter again, given the large number of Nagra IV-S out there. These machines are indestructible and easy to service. They are excellent recorders, but were not designed for high quality playback. I think eventually someone will come up with an upgraded repro card. I told them that they can simplify the Nagra T; get rid of the time code, sync and complex editing functions, and use the rest of the design for a new "audiophile" tape machine. However, they told me that some of the parts are no longer available, and will be too expensive to re-manufacture.
I had a conversation with Jeff Polan @NextGen about a plug-in replacement upgraded Nagra IV-S repro circuit and its something that's on his backlog of projects. It's something we plan to revisit later this year. I'll let Jeff comment on more details about it, including potential limitations he's identified that may necessitate a custom power supply if done exactly how he would want to do it. With that said, I've already pledged a loaner / test unit for development of the repro card when he's ready.
