192 & 2" Transfer Test Ideas aka: Return Of The PT Jedi

[DaveCarlock]

In restaging the infamous Mixerman test, I'm looking for test ideas. The first question of the day is:

What is a comparison format we can all agree on?

Mixerman used dB converters into a Panasonic SV-3800. Can this be improved on in anyone's minds?

Other 16-bit alternatives would be an ext clock capable stand-alone CDR.

A 24-bit alternative would be the Tascam 24-bit DAT, but there's no real way for anyone to hear it in 24-bit outside the studio.

DC

[DaveCarlock]

Next question:

Once we agree on a comparison format, how can we most accurately present it to those interested for study?

Options I can think of:
Audio CDR copies
PT sessions
RADAR sessions
DATs

Mixerman and a few others didn't approve of comparing files inside PT, though I'm not sure if I can agree with them. If a listener's system was first tested with tones to prove that a properly calibrated 192 spits out what is already inside the session (via the internal tone generator), then there shouldn't be any more of an issue comparing all the files in one format than in another--beit DAW or DAT or a consumer CD player.

But aside from my agreeing or disagreeing, can anyone think of other format alternatives that wouldn't "color" the results?

Because the failure in Mixerman's test may be linked to the interaction of a 2" machine, transferring in AES files would bypass the possibility of an impedance issue.

DC

[DaveCarlock]

Can we agree that using any "favorite" high quality clock while transferring all sources to the final comparison format is acceptable?

It shouldn't really matter what the clock is as long as it's generally respected by most and is consistently used on all sources. The alternative is to clock the comparison format, presumably DAT or stand alone CDR, with the respective clock of the source. For instance, clock DAT/CDR converters on the HD mix with HD clock and clock DAT/CDR converters on the RADAR mix with RADAR clock. The question arises here, though perhaps truer to the sonics of the individual formats, how can you then fairly compare the 2"? By relying on the clock of stock converters of the DAT/CDR? Somehow I thought not...

What are your thoughts everyone?

DC

[DaveCarlock]

Will jitter be a variable when changing the comparison format so others can hear the test?

Some concerns had been expressed on jitter from the AES transfer, but they seemed to be discounted by Lynn Fuston's extensive personal testing as well as supported by testing reported on the Digital Domain website.

http://www.digido.com/jitteressay.html

The general idea was that the final clock used when pressing a glass master will override the jitter induced anywhere along previous transfer stages. In other words, if a "favorite" clock were to be introduced at the final stage of a session, it would not be necessary to use the "favorite" clock in previous AES transfers.

Can I assume that since multiple tracks within MDRs use one clock, and using a "favorite" clock on the bounced two track replaces any previous "jitter spec" with its own "favorite jitter spec" on playback, obsessing over the use of a "favorite" clock during individual tracking is redundant. To the experts: is this true?

DC

[DaveCarlock]

Here's a multi-question question:

Does anyone have an issue with sending the comparison audio files out via the internet?

Will this induce any more or less of a variation in the sound than listening to a CDR?

Does anyone have any empirical evidence that sending audio files through the net will affect their sonics at all?

If so, does anyone have an optimum method to deliver audio files via the net to preserve the integrity of the audio?

[shaggy]

My comment about 'sharing' over the internet:

Fine. As long as the data is transfered FTP. Not TFTP or UDP- they're both 'what you get is what you get' type of service.

Wav files? (zipped or stuffed?)

Also, Sony's entry into SACD DVD line- is the DVP-NS500v

It'll play 24/96 DVDs as well as 16 bit SACD. I've listened to Waters 'In the Flesh' recording (SACD) on this machine. And it is impressive. In reality, it's the least expensive Pro player at the moment.

[DaveCarlock]

Thanks for your ideas Shaggy. In order to make this accessible to as many people as possible, I think we're going to stick to CDR, although I'd love to try out the Sony box sometime.

The plan at this moment is to eliminate DATs this time and mixdown through the same dB Technologies converters straight to a CD recorder. The dB Technologies clock will be used on all the mixes to CDR with audio transferred through an AES connection. Here we at least have a consistent clock to compare with. I may also run a mix with the dB converters synced to the HD clock just for "fun". As well, with the RADAR clocking the dB converters. I have to make sure this is possible first.

Once we've got the master in CDR format, I'll copy it into Toast format and allow everyone to download the files and burn your own CDRs. That will save us time over here and maintain a consistent format we can all listen to.

Mixerman really doesn't want to have people trying to critically monitor this test through PT. Fair enough. Again, I'm looking for the right combination of test parameters that will sit well with a majority of people, Mixerman and myself. With that goal in mind, I know we're going to have a great test by the time we have it all together.

Timeframe update:
We cannot get our schedules together by Monday (we both need some time for our families on the weekends) and after that I am booked out of town till the following Weds. So at the moment, we are looking at Thurs May 2nd. In the meantime, Mixerman will cut some drums to 2" in advance to give us more time on test day.

Keep offering ideas, I want this to be a solid test--beyond criticism.

DC

[Bushpig]

OK GUYS!!! LISTEN UP!!!!

I originally joined in with the first "192 Bass-Lite" thread way back on about page 2 and I stated CATEGORICALLY that I agree with what Mixerman reports to hear with a loss of low-end when passing audio through Pro-Tools. But as all my experience with PT up 'till that post had been only on PT Mix systems, I thought I'd shut up and wait 'till I had a chance to hear the new PTHD gear which I have now had a chance to do. (More on this later)

I have for years been bugged by exactly this same phenomenon as Mixerman describes when using ProTools in any and every different interface situation I have encountered. I have spent years doing the traditional transfers from 2" to PT just as Mixerman describes and I have exactly the same feeling as him (bottom octave missing or screwed with). I also bought a Radar II 24 bit system with full AES/EBU to Protools capability and experienced the same sound change!! I have tried Apogee convertors, 888/24's/ digital tranfers of 2 track material, multitrack material, analogue desks and digital desks all hooked to ProTools. You name it, I've probably done it!! Every time the same result. I described the sound change that I hear, on another similar thread about 6 months ago (which gave highly detailed test procedures etc.. Unfortunately this thread went missing which means I can't refer you to it). I describe the sound I hear as like an upside down triangle. Less bass which is also less powerful like it's been "twisted". Top end more harsh and brittle. Boxier sounding overall and changed depth/stereo image.

I have a different theory to propose here and it can be verified by everyone in a far simpler way than spunking tons of cash on a studio and 2" tape etc.

We have a very simple and fundamental question to ask ourselves and the folks at Digidesign and it has a simple yes/no answer!!

After we all conduct the following simple test and answer this question HONESTLY (including you Dave Clementson and all at Digidesign), then we can branch out into the why's.

Question: DOES PROTOOLS CHANGE THE SOUND OF A BASIC STEREO AES/EBU TRANSFER?????

What I am getting at is as follows. I believe, and according to many other people including some on this thread, Radar II, Sony 3348 and other digital formats sound better than ProTools!!! That is my statement and here's why I think that is. A lot of people here are saying AES/EBU transfers are error free, provided you use decent cabling and observe good clocking rules. I agree. I believe the only essential difference between a ProTools rig and any other "simple" digital recording device is the addition of the TDM architecture and this, I believe, is the only thing that could let Digi down in the sound department. A bold statement I know, given I am not a digital audio boffin, but follow my easy test now and be VERY HONEST WITH YOURSELVES!!

I have an Apogee PSX-100 which is a combined A/D and D/A unit which is very flexible, but hopefully many of you will be able to recreate this with your own gear as it is a basic primitive "real world" application. I listen through my D/A all the time as it means I will be hearing the resultant "CD output sound" at all times.

So, take any PT Rig, Mix Plus or HD and setup a simple AES/EBU loop through your PT interface from your A/D convertor and returning to your D/A. You know the deal, decent cabling and observance of clock master etc... Just create 1 stereo track in a new Tools session (any sample rate or bit rate, it doesn't matter just as long as it matches what you are sending from the A/D obviously), and "record ready" that simple track. Don't record or playback anything, just have it sitting there passing audio through via AES/EBU from the A/D. Now, with the PSX-100 I'm lucky. If you hold down the D/A input selector switch, you can readily flip between the DIRECT output of the A/D side of the unit (ie: the "original" sound) and the returning feed from ProTools. Feed any signal of your choice into the A/D (a CD you know or your current mix) and just sit there and switch back and forth from direct signal to PT signal (Forget that there's an A/D conversion going on and just listen for comparitive difference. With this test there's no DAT error correction to worry about clouding the issue or tapes to swap)

NOW REALLY, REALLY LISTEN TO WHAT YOU HEAR!!!! Focus on a bass drum or something low for a few flips back and forth. Listen very hard to vocal sounds and the "SS's". Are they smoother or nastier or more smeared or is the stereo image different?? Don't concern yourself with descriptive terms like "better" or "rounder" or "warmer"!! This is a test of PT's ability to truly pass numbers in to numbers out. If you hear ANY change in sound WHATSOEVER, no matter how slight, then you have to agree with what I am saying.

I urge everyone to try this including and especially Dave Clementson and Co. When you guys try this and find that I'm not kidding and that I and Mixerman are hearing a manifestation of the same basic problem, ie: something inside Protools slightly changes the sound, then we can all start discussing the merits and pitfalls of various A/D setups and all the other bogey that surrounds as subjective a subject as "sound quality". I know some will want to argue that my test is not very subjective, but I agree with MM and various others who say, screw the tech garbage and the out of phase cancellation tests etc, and just use your ears on a simple, repeatable, real world test. Please. (My CDR in an identical AES/EBU loop sounds identical to the original A/D on switching)

Now the good part. Despite the fact that I still hear a slight change of sound in this test using PTHD as I do with Mix Plus, I had the good fortune to try a couple of already existing multitrack mixes through the desk like a proper session on first, the Mix Plus (printed to CDR) and then immediately swap cabling and drives and print the same mix from PTHD. OH MY GOD!! What a difference. So big was the improvement that the first time I did it I thought I had made some error in the switch over, but no, I repeated and checked everything thoroughly and printed a different mix on both rigs and the change was just awesome.

Sooooo!! Well done Dave Clementson and all at Digi. The sound of the new system is a vast improvement over the Mix Plus. Good work guys. Thanks!! As Dave already knows, I am a massive fan of ProTools and couldn't live without it, so all I want is for it to sound as good as all the other digital systems out there. But I stand by what my ears tell me when comparing like for like in the situation such as MM discusses and what my test can prove. I would urge everyone to check my post on page 2 of the older thread though, to get a sense of the "head in the sand" attitude displayed by Digidesign on this issue. I, like MixerMan have offered many Digi employees to come with me to any session I do and I'll prove my claims, but since the marketing department appears to make more of the decisions in Digidesign than the engineers (like giving us 3 segment, semi-professional, MOTU style cheap LED's for metering on a professional interface!! Send marketing to one of my sessions and I'll show them what's important on a professional recording interface. And since when did a professional recorder ever need seperate input and output meters, HUH!??!!), it appears Digi are less than likely to want to fix their "sound".

If it turns out I'm right, there should be no panic from Digidesign though. Just give us a "direct mode" I/O path that doesn't use TDM for all those sound critical applications like simple 2" tape transfers and 2 track mixes for editing. If the price to pay for having all the brilliant, life enhancing qualities that TDM provides us with, is a slight-ish degradation of the sound, I have no problem with that. Just give us the choice and be honest about it Digi!!

Enough. Please get back to me when you've all tried this test.

Cheers.

Steve Bush

[DaveCarlock]

Bushpig,

Your AES test is another good test and it should be examined seperately from the test I'm focusing on: 2" transferred in analog to a 192 in HD.

If you have any ideas about the way we plan to do this particular test, give me a shout.

DC

[DaveCarlock]

So the upcoming test as it stands now is this:
=================
2" transferred directly, without a console, to RADAR and HD w/a 192.

***The RADAR and HD sessions will be striped with 29.97FPS TC and the 2" will be slaved to each machine during transfers to allow for sample accurate waveform analysis.

***The same ELCO connector will connect the 2" to the HD rig and the RADAR rig for the transfer, and then the console for the 2" mixdown.

***PT and RADAR will both be calibrated to -18, as is unofficially the standard in most of LA.

***The console will be used to balance the mics, add eq, etc. The settings and fader levels will not change between sources: 2" playback, HD with all faders @ 0dB, RADAR with all faders @ 0dB.

***The console outputs will be sent to a dB Technologies converter set--the dB clock will be used and sent via AES to a CD recorder. The final comparison format will be a CDR.

***The CDR will be data copied into Mac Toast format and those Toast files will be available for download through the internet to those who can burn their own CDs.

***We ask that you DO NOT COMPARE the files by putting them into Pro Tools of any make. Please rely on your best CD player and monitoring system for listening tests. One might argue that Pro Tools has better converters than a consumer CD player, but it's just cleaner to keep the test subject out of the playback process. All of Waterboy's comments are noted and agreed upon!

======================

Does anyone have any suggestions about, problems with, or additions to this test itinerary? Lee Blaske? MattiMattMatt? Digi?

Speak now or forever hold your peace!

DC