Film-Tech Forum ARCHIVE: DTS6 output level

This is topic DTS6 output level in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=006052

Posted by Ivo Noorlander (Member # 1315) on 04-10-2004, 04:51 PM:

Just a couple of months after DTS stops service on the DTS6 my unit starts acting funny! Great timing. I noticed that the levels were much higher than Dolby Digital on Big Fish. When I measured with the test disc and an AC meter, though, I couldn't get the levels up to 300 mV. They don't go any higher than two forty-something. Can anyone make sense out of this? Are the screws on the front the only way to adjust output signals?

Also is there a way to get the signal out digitally like on the 6D?

Posted by Gordon McLeod (Member # 33) on 04-10-2004, 06:37 PM:

Check and verify your meter can accuratley read 300mv @1Khz

Posted by Mark Gulbrandsen (Member # 72) on 04-10-2004, 06:48 PM:

Besides verifying your meter you also need to be aware that the DTS 6 has balanced outputs...something the later DTS's skimp on.
So be sure you're reading across the secondary or + and - of the output...not say the plus to ground.

Personally I still have never setup a DTS this way anyway.... I've always used the pink noise on the test disk and then match it to the B chain level at 1db resolution on the analyzer SPL meter. This method compensates for any possibilities of differences between gains in the channels of the processor which sometimes does happen. The other advantage of setting up the DTS levels this way is that levels are so well matched that it makes the reversions all that much more difficult to hear...as least if your "A" chain is of very high quality.

Edit addition:
BTW...DTS had not stopped supporting the 6 and they are doing their best to continue to support the 6 for quite some time to come. Some sort of internal upgrade kit for the 6 is in the works but is not yet nailed in stone. All other parts are still available for the forseable future.

Mark @ CLACO

Posted by Gordon McLeod (Member # 33) on 04-10-2004, 09:34 PM:

Tone is the accurate way to align the unit accoustic pinknoise take other variable into account

Posted by Antonio Marcheselli (Member # 373) on 04-10-2004, 10:27 PM:

In my experience I saw a DTS6D set to 300mV that had a different SPL read on an R2 RTA from dolby ones.

Many dB, not just few ones. But definitvely the 300mV way take to better results.

What do you think?

Bye
A

Posted by Mark Gulbrandsen (Member # 72) on 04-10-2004, 10:47 PM:

Antonio,
Might have been a problem in the Cinema processor or with the R-2.

I've long learned that using pink noise and either my Ivie or R-2 analyzer and reading the SPL meter of either during DTS output level adjustment produces just about perfect results in matching a DTS's output level to the "B" chain level and that the DTS will better match the analog level during a reversion.

This would also probably work fine with tone as Gord indicates as long as the "B" chain is properly set up before hand. But by golly gee we certainly don't set the "B" chain balance with tone do we....

I will say that the DTS levels are usually VERY close, sometimes requiring no adjustment at all, but sometines some fine tweeking on them can produce a more "invisible" reversion than had you not tweeked it.

Mark

Posted by Steve Guttag (Member # 268) on 04-11-2004, 09:29 AM:

Mark,
I don't agree on the pink noise of DTS to set its level. Buss level is buss-level...if the 1KHz tone is acurately recorded...the level is right...no room for error. Without exception, when done this way there has been zero difference in level between DTS digital, Dolby Digital and analog. The only real purpose to those trims in a DTS unit is to allow the DTS player to match the buss level of a particular processor.
DTS Rev C disc has good tones to use (including subwoofer) as does the latest DS disc (I forget its revision but it also has a tone for the subwoofer channel unlike previous DS discs).
Pink noise as a level setting tool has always been a bit flawed. It is a random noise (hopefully) and provides a wavering level. With pink noise, you are dependent on every step in the process of its making...the generation of the original white noise source, the conversion to pink noise via LPF, the proper band limiting...then there is the digitizing of the signal (for a DTS disc), then comes the meter stage...you again must have a meter that has accurate filters and a uniform sensitivity across the spectrum being measured (dB level measurements are meaningless without a well defined spectrum)...lastly the method of obtaining the measurements comes into play...do you hold the meter at chest height, ear height, point at the speaker playing, point straight ahead, point at the ceiling, is your body affecting the results...I've yet to find two different people to come up with the same numbers.
As to the other statement of not using tone for setting room levels...since room levels are set after B-chain tuning (unlike digital processor levels which come before B-chain tuning), a single frequency band would be affected if that particular band was adjusted. That is, if 1KHz tone was used, if you adjusted the 1KHz EQ control, it would, in effect, be setting the balance of the channel. Pink Noise wins in this area since the entire spectrum is excitd and the log average of all represented frequencies are used.
Using a pink noise card or the processors internal pink noise generator is more accurate than using a pre-recorded disc since for a certainty, in either processor generated noise you have the exact same signal sent to all channels so even if the Pink Noise is inaccurate, it will be uniform across all channels and balance is maintained. Using a DTS disc, one is dependant on how it was recorded and digitized. It is certainly possible that it too is accurate and I don't dispute you have had excellent results but it opens the possibility of inaccuracy.
Certainly the same argument can be made about using a tone method though recording a single tone is a much less complex signal to digitize and easier to maintain since when recording pink noise for level, one must again band limit it and average out the level where as a tone is a steady state signal and inherently band limited. One of the many things I like about the XD-10 is that right out of the box, the levels are set to the 300mV buss level and may, if the processor so dictates, be adjusted in .5dB increments up or down.
Steve

Posted by Mark Gulbrandsen (Member # 72) on 04-11-2004, 09:45 AM:

Steve said.....

"The only real purpose to those trims in a DTS unit is to allow the DTS player to match the buss level of a particular processor."

And That was pretty much my exact point..... as I said a couple of posts above yours.....
This method compensates for any possibilities of differences between gains in the channels of the processor which sometimes does happen."
Hence... Unless you are trying to exactly match the output of another input source digital processor for what ever reason tone should be used for that purpose. Other than that reason my method of tweeking the DTS output works just fine. In my case its normally with the R-2 in which I use Mic 1 to ultimately set all room levels and then run the P.N. on the setup disk and make any minor corrections to the DTS output levels if they are needed. As for a DA20 its fixed and you are obviously stuck with channel imbalance if the processor is a bit off in levels. My mics needless to say are in the same "THX" confoguration every time I do this so perhaps thats why I get very consistant results although I have gotten pretty much the same results with my Ivie and a single mic. The whole idea is not to tell when a reversion takes place and with the method I've described thats what you get.

I totally agree on the tone issue and not using it for measuring anything AFTER B chain alignment nor do I use any film or disk as a P.N. source for setting up any "B chains". Not sure why you thought I was doing that one since I stated above that I use the P.N. card or source(an Ivie P.N. generator in my case) to setup the B chain. Then use the disk P.N. to set or tweek the DTS levels if needed.

Mark

Posted by Steve Guttag (Member # 268) on 04-11-2004, 10:06 AM:

Ah, hear ya...
The buss level I was referring to comes from some processors that operate at say .775V as opposed to Dolby's quasi-standard 300mV or in the case of some CP-50 installations where, depending on channel you may be at 500mV, 300mV or 170mV.
The way your post reads (to me) is that you are concerned that on a channel by channel basis you are trying to compensate for small inaccuracies on a uniform buss level...like, for instance on a Panastereo CSP-1200's digital input that is referenced to 300mV...to compensate for their 1% tolerence resistors with a DTS recorded disc that has no published tolerence.
Again, I don't dispute that you achieve good results if you use pink noise to balance the DTS processor to another just that it is inherently less accurate than matching buss levels prior to equalization.
Steve

Posted by Mark Gulbrandsen (Member # 72) on 04-11-2004, 10:16 AM:

I guess what I like to do is not necessarily worry about the "exact level" hitting the buss, I'm much more interested in obtaining perfect results(channel to channel as you said) down in the room and during a reversion. Thats all. Agreed though on using tone to set up a buss level if other than 300mv and I usually use the RMS AC voltmeter on the IVIE to do that.

Mark

Posted by Alan Haigh (Member # 1781) on 04-11-2004, 04:33 PM:

Mark's theory ought to be fine, and I'm sure he acheives good results in practice. However it relies on the assumption that the pink noise on the DTS disk is the same as the pink noise on the cat 65. First of all which cat 65? they are all different. Secondly, even with a "perfect" cat 65, there is a small difference, due to historical reasons. The DTS unit is an A chain device, so should be lined up with tone to "bus" level, just like analogue optical.

Posted by Mark Gulbrandsen (Member # 72) on 04-11-2004, 04:52 PM:

Of course we could also say that no tone and no AC RMS voltmeter are going to be exactly the same day in and day out either... Better chance though if left on continously, not moved around, and both are calibrated on a regular basis as specified by the manufacturer.
One will also find that every pink noise card made by the same manufacturer and others as well as all test film, CD roms, and your favorite fuel that you put in your gas tank every time you fill up. Thats part of life but sometimes minor compensation for some things can easily be delt with....

Mark

P.S. Many of us over here do not use the CAT 85 series(I think you meant 85, not 65), we use the C.E. SG-1 card instead. Its noise is quite a bit better utilizing a 7 pole filter and longer running digital generator although the latest Dolby card may possibly be as good. However, there is also a convenient tone generator built in for other test purposes and the proper noise gating facilities for acoustical measurements(THX).
http://www.componentengineering.com/signal_gen.html

Posted by Steve Guttag (Member # 268) on 04-11-2004, 08:50 PM:

The other nifty thing on the SG-1 is the ability to inject a signal in if you put the noise/tone switch in the middle position...you can then put out say a phase-checking thump on a channel by channel basis.

Steve

Posted by Ivo Noorlander (Member # 1315) on 04-13-2004, 01:07 PM:

Thanks for the insight everyone. I guess my problem was the balanced outputs. My setup works fine on 6d's unbalanced outputs. I made a little switch device so I wouldn't have to use those little pins on the DTS output for measuring. I did however use one single pin (13) for the return signal, so there you go.

I do still wonder if the 6 can put the signal out digitally like the 6D.

Posted by Mark Gulbrandsen (Member # 72) on 04-13-2004, 06:41 PM:

Donno about the digital output capabilities but the 6 has a much better analog section than the 6D does. The XD-10 beats em all though.

Mark @ CLACO

Posted by Alan Haigh (Member # 1781) on 04-19-2004, 03:07 AM:

Thanks for the correction - yes cat 85 of course. I use a SG-1 too, which is probably why I got the number wrong. The pink noise difference is about 1.7 dB (a theoretical difference - could be more or less in practice). This is why DTS moved to using tone for line ups rather than pink noise. Although having said all that Dolby have changed their level very slightly with the release of the CP650 and Digital mastering unit, so the difference is only about 0.5dB now. Not enough to worry about. As you were!