How reliable is the Analogue Productions Test Record sweep?

Ultimate Test Record. I have it. A very, very smart acquaintance told me:

I think regarding the HFN, I remeber him saying it was as if someone manually turned a freq. knob for one of the sweeps, it was that obvious.

 

I'm thinking that if "purity" of test signal is your goal, would would almost have to do that. You could either record all the signals to a tape (thereby introducing other issues), or you could twiddle the oscillator controls as the record is being cut. I suppose you could do it with computer software and feed the cutting lathe with a (gasp!) DAC!

It seems there is a slight market for a new, state-of-the-art test record! Perhaps a double 12 inch set with sweeps, pink noise, VTA, the whole enchilada! An audiophile entrepreneur could sell hundreds!

 

I think for a test LP, using an DAC would be fine, unless someone can explain to me why it wouldn't be.

Full sweep frequency from 2 Hz to 20 kHz would definitely be one thing I'd want on it, I'd also want something that can be used to adjust peak gain for needledrops and ofcourse an analog jitter wow and flutter test and channel balance. I think in many respects digital master -> DAC -> cutting would be advantageous because the digital master could be made available to compare wav forms with what is produced when played.

Perhaps those who know what is useful in a Test LP could come up with what will be on it as digital masters, then someone with the right contacts could start a kickstarter project that those of us interested in could fund so that it can be cut and pressed without financial risk? With digital version available for scrutiny first, problems in the mastering could be found and fixed before it was cut.

 

Pink noise is very, very useful, too. To me, that helped me more than anything else, since that way I could see specific stylus alignment, L/R balance, and overall frequency response at a glance.

 

Do you have some screenshots of specific measurements with different carts/setups?

 

No, that I don't have, since I haven't done this for several years. But I did do it with the main setup I use, simply to verify performance and tell me that the front end isn't doing anything wacky to the signal. I did discover that the left channel was down about .5dB compared to the right, but I compensate for that during the transfers.

 

The best way I have found to use the pink noise tracks in a test LP is to generate a reference pink noise file on the PC and level match it to what is recorded from the LP. Then a simple A/B comparison will quickly reveal the frequency response of the phono chain. As I stated further up the thread, I understand "3dB per octave, 10dB per decade" but I can't always "see" it. That's why having that reference is critical for me, anyway.

 

That's a good idea - I'm going to try that.

 

Gee, I just look at it on a frequency response spectrum graph. You can see very quickly if the pink noise is rolled off or otherwise non-linear:

 

When I see the computer generated curve, I'm like, "yeah, that looks right". But my "pink noise curve memory" just isn't as good as yours. And I doubt it ever will be.