Is Direct Drive Really Better?

Phd, the SP10MKII is an upgrade from the Technics SP10MKII that has a 28lb bronze and aluminum platter and an outboard power supply with speed control.

"Speed-stability is the crucial issue, not speed-accuracy within reasonable limits."

Zaikes ... I agree with you. If speed accuracy were more important the CD vs LP debate would have died long ago.
Speed stability is what really counts.

Regarding the limit of discrimination this reference has it at 0.5%
http://www.indiana.edu/~emusic/acoustics/pitch.htm

So you could argue that the difference between 0.001% and 0.01% is academic. I would even argue that the orchestral players are not able to tune their instruments to better than 0.1% so you are fooling yourself if you think a turntable needs to be substantially more accurate.

Since speed stability problems cause direct distortion to the waveform they are much more noticeable.

Doug ... how does an absolute speed error break harmonic relationships ? If we are 1% fast then a lower A is 444 Hz (instead of 440Hz) and an upper A is 888Hz instead of 880Hz. The doubling of frequency per octet relationship is still maintained, just shifted slightly 1% up in frequency.

I would agree with Sean on this. The standard of 440Hz for A is a rather recent development as any performer of Early Music will tell you. But a minor third has always been a minor third, so it would seem (and here I risk the wrath of Raul) that pitch stability is more important than accuracy.

Hi: SM?, good point. Btw, from a pure objective approach the " numbers " are what define which audio device is better not if we can hear those " numbers ".

Here are some of those " numbers " and you will be the best judge:

Manufacturer wow&flutter speed accu.% signal/noise ratio

Belt Drive.

Basis Gold MK3...0.02...... 0.02...... 90 db
Kuzma Ref............0.05...... 0.08...... 83
Michel Gyro..........0.05...... 0.01...... 80
Sota Cosmos.........0.025..... 0.02...... 87
SME 20..................0.05...... 0.01...... 85
Voyd.....................0.004..... 0.001..... 80
VPI TNT...............0.01...... 0.01...... 90
Well Tempered ......0.01...... 0.001..... 84
Roksan Xerxes.......0.02...... -----..... 83
Linn 12/lingo...........0.03...... 0.01...... 70
Goldmund Ref........0.01...... 0.01...... 85
Micro RX5000.......0.02...... 0.015..... 84
Maplenol................0.03...... 0.003..... 80
Walker....................----...... 0.002..... --
Continuum...............----...... 0.006..... --
Townsed Ref...........0.02...... 0.01...... 85
Acoustic Signat........0.02...... 0.006..... 85

Direct Drive.

Rockport.................0.01...... 0.01...... 98
Exclusive.................0.015..... 0.001..... 95
Denon DP100.........0.02...... 0.002..... 90
Yamaha GT2000X..0.0025.... 0.002..... 85
Technics SP10MK3.0.015..... 0.001..... 92
Technics SP10MK2.0.02...... 0.001..... 86
Denon DP80............0.02...... 0.002..... 80
GPA Monaco............----...... 0.002..... --

These are the best TT " numbers " . The Idler Wheel have really poor " numbers " and the Rega and the like very poor too, on many others there is no single " number ", sorry.

Btw, it will be nice if Teres and Galibier could give us their " numbers ".

Regards and enjoy the music.
Raul.

Raul: Sorry, but the only thing one can "judge" from that list is that you spent much time and effort collating those numbers. You can list claimed specs til the cows come home however, but for many reasons it still doesn't amount to an argument (scientific or otherwise), or illuminate anything about what we hear or which drive method is "best". It just doesn't. Appreciate your witholding the "wrath of Raul" however!

Doug: Hey baby, don't go all brain-masturbatory on us now! ;^) I have to side with SDT99 and Jyprez on this one. I too have never understood the 'damage to harmonic relationships' argument, since everything changes proportionally and in unison. A 1% deviation from the way something was performed on a recording is still well within the range of variation in tempo and tuning encountered in live performance, as well as the unique harmonic structures possessed by individual examples of instruments. Personally, I think you usually have to go to a 2% - 3% deviation for the music to start sounding a bit 'funny' in absolute terms, but I grant you that system fidelity and program material selection will have an impact on that threshold. I can't quibble with your opinion of what you hear, but my own suspicion is that if you didn't play a recording with which you were already familiar, or didn't play one in close proximity to hearing it at the correct speed, a 1% steady-state error, while audible in relative terms, would not call attention to itself in isolation. This is something that of course could be blind-tested for pretty easily, and I'd lay money on getting negative results.