Finally: The Truth About Fosgate Surround Designs and DPL-II

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

deepsky4565

400 Club - QQ All-Star
Since 2002/2003
Joined
Mar 17, 2002
Messages
451
Location
Spokane, WA
I have just finished researching Fosgate units for synthesizing surround from stereo, of course with lots of seperation, and a real quad like experience.

The Tate 101a used the first DES, directional enhancement circuitry. This was a huge advance for SQ decoding, and also for processing stereo. The first circuits were slow, however, and can lead to pumping type artifacts. Most of the time they would only be obvious to critical listening, especially of one or two channels at a time. Sometimes it is downright unlistenable.

The 3601 was not a step forwards, as nearly I can tell, and I haven't auditioned one. They have nearly universal negative reviews, at least for our purposes.

The Fosgate models 1 and 2 were an advance, and have powered back channels.

Model 3 and 3a, and then 4 and 5 have very excellent performance, as I think they aproach the performance of a Tate, but with very minimal pumping. The 3a was the most expensive of this sequence, and commands a higher price. Harman Kardon bought Fosgate Audionics, and rebranded these units, as did their subsidiary label JBL. I have a model 4, and love it. It has replaced my Tate for synthesizing surround, as the pumping is nearly negligble for most music, and it gives 5.1, not just 4.0.

The next step in Fosgate's design was 6-Axis. I just received a Citation pre-amp which has the 6-axis circuit. I believe only Citation 5.0 and 7.0 preamps had 6-axis. There was also the car decoder, the Harman Kardon CSP-1, and possibly the same unit branded as an Infinity. I have not hooked my 7.0 up, but in research, including lengthy conversations with Bob Popham, who worked for Fosgate, this is the ultimate Fosgate synthesizer. It is even faster, which should reduce pumping even more than the models 3-5, and boasts a much better seperation between channels. It also has lots of parameter controls. I will be posting reviews of the Citation 7.0 in the near future. The 5.0 came later, and used cheaper components. I do not know if this affected the 6-Axis circuitry, though. The 5.0 did offer Dolby digital and optional DTS however. It is interesting that the reviews at the time said the lack of digital decoding on the 7.0 was not a handicap, because the reviewers felt that 6-Axis sounded better. One other note on the 5.0, is that it offered the digitally decoded dolby digital and DTS through a 6-Axis Encoder to be recorded on the stereo analog outputs! The primary drawback to both units is the lack of external inputs. MSB technology offers a $400 5.1 input upgrade to the 7.0.

Now to the interesting part, that I have just uncovered. Just before the launch of Dolby ProLogic II, there was a media blitz, at least in audio circles, of this new wonder to be released. The extremely glowing commentary of DPLII was mouthwatering at the time. The thing I most focussed in on was its purpose to convert stereo music into 5.1 surround. It was said to be Fosgate's ultimate acheivement, as it incorporated a new concept of feedback logic to acheive incredibly accurate decoding, even better than 6-Axis. Well, after more research, at first it seemed that DPLII was an improvement of 6-Axis. I bought one of the very first receivers that offered DPLII, and I liked it, but it didn't ring my bell like I thought it would. The Tate was much more satisfying. Unfortunately, I did little surround listening for the last year and a half until recently, as I was building a dedicated home theater, and music listening room. After setting up the Tate again, finally, it just was so much more pleasing to me as a quadgeek than DPLII. What's the deal here? DPLII was supposed to be so awesome in its abilities as Fosgates ultimate, yet his ancient Tate ran dircles around it (pun intended). Well to make a long story a little shorter, Roger Dressler of Dolby Labs finally admitted that they wanted some accuracy and dynamics issues resolved for the new DPLII, and to accomodate them, Fosgate reduced the front to rear seperation. I do like DPLII, but if you are like me, and like an aggressive surround experience, DPLII should not be confused with Fosgates' earlier designs. This is why I just bought the 7.0, and again, I'll let you quaddudes know the score. I haven't heard much about the 6-Axis design in quad circles, but I think it's probably because the Citation units were so expensive, about $4k. Now they run about 10-15% of their original price.

Thanks for listening!

Dan in Spokane
 
Thanks for the good research. Your explanation of the lack of real quad surround from DPL- II is why I don't care for its sound because it does not have the front to back seperation of the old Quad decoders.

I had a Lafayette full logic decoder which was much better than it's Fosgate 3600 replacement, which was not impressive at all in Quad. Unfortunately I never heard the Tate.

The Ambisonic SSP-1 is a great decoder but does not have logic circuits and is not really impessive in action movies, but can produce real Quad sound in SQ & QS..

My new ISP HTD 6.1 has fast logic with no pumping and the best directionality I have ever heard. Front to back sepreation is great too. Produces great Quad from stereo. Sounds better on Dolby sources than Dolby . Music mode sounds very natural. This decoder gives a great Quad sound from all sources.

I am interested in hearing how well the Fosgate 6 Axis decoder gets Quad sound from Stereo.


 
Interesting.

Since PL II has several "modes" in its current incarnation, perhaps they need to offer a new mode - full separation - for those who wish to take advantage of the new technology and don't want the Dolby tradeoffs.


 
Sorry, the new DPL-II even in the most extreme will not come close to offering the seperation found in the earlier Fosgate units. Believe me, I've tried, and Roger Dressler of Dolby labs himself confirmed this. I wish it did.

Dan in Spokane
 
OOps, didn't see that word "new" in bmoura's reply. Yes, I personally asked Roger Dressler for this option in the future. He declined to respond. I am sure that at least some of Fosgate Audionics top people also saw my plea for this in the future. Again, no reply, but maybe that is better than hearing no. We can hope.

Dan in Spokane
 
Interesting read, and I would tend to agree on the separation issue regarding most material. However, have you listened to properly encoded QS material via DPLII? The separation for this application is pretty near discrete, and better than what the Tate can achieve with even the best SQ source material.

 
Well, since Dolby owns the technology and patents behind PL II, Dressler is the guy to talk to/appeal to on this.

Although chatting with folks like Jim Fosgate and Charlie Wood at Fosgate/Audionics couldn't hurt.

In fact, Fosgate/Audionics now licenses and makes PL II products. So who knows, perhaps they could sell a PL II product with the original full separation option in it.


 
Cai,

DPL-II does seem to do a very good job on encoded material, Dolby or QS. I agree the technology is a step forward, but the front-back seperation on normal stereo is what I'm dissapointed in.

Dan in Spokane
 
Dan, I'm in complete agreement with you. I just wanted to make sure the that the positive aspects of DPLII were not overlooked.

 
I've had the Citation 7.0 preamp up and running. Basically it is more similar in seperation to DPL-II than to my Fosgate Model 4. There are many parameter controls, but none seem to give me the front-back seperation like the Tate and Model 4. The Citation is truly audiophile quality in build and sound, but I got it because it was supposed to be even better than my Model 4 for synthesizing. The Citation sounds great, and the surround mode really is good, but not in the ballpark for us quadguys. I'll let you guys know if I find any secret settings that may unlock something, but I've been through all the obvious ones. I have a multichannel switch so I can switch units on the fly, and direct comparison to the Model 4 shows that the matrix is just more front-centric. Very smooth, just not a lot of discrete stuff going to the back channels. With QS test signals, back channel information is sent there, but like with DPL-II it takes encoded signals to be strong enough to be discrete to the rears.

Cai, I'm hoping to bring my Model 4 over to have a showdown with your QSD-1. These are the two that I consider to be the best, but want a side by side comparison. I tried Circle Surround and DTS: Neo, and they aren't close. I'm not knocking those who like them, they probably haven't had a chance to listen to these units, or don't like or need a quad-like discreteness in their surround.

Dan in Spokane
 
Dan, I agree with your comments regarding CS, but wonder if you or anyone else has had any experience yet with CS-II to be able to say if it has any improvement with respect to rear-channel discreetness?? I have alot of experience listening to CS-I but have never heard CS-II. CS-I is no match for QS VM's ability to produce a focused and discreet holographic soundfield from stereo.

Anyone got a link to a CS-II review??

John


 
My comments were based on CS-II with a software plugin. It sounds good, if you haven't heard what I consider a real discrete synthesizer. I like a soundscape with discreteness, not just sound all around.

Dan in Spokane
 
Hey Dan, I'd love to do a side-by-side comparison. Next time yer in the city, just give me a holler. Is the Model 4 a stand-alone decoder or is it a preamp/receiver?

 
Until recently, my only experience with Dolby Pro Logic II (DPLII) has been in my car. However, I recently purchased a new processor/pre-amp for my home that offers DPLII decoding (among other things).

Last night was my first chance to really run DPLII through its paces. The first thing I tested was the QS decoding capabilities of DPLII, since so much has been said about this. With my trusty "Quadrafile" CDR, I ran the various channel tests... hmmm... not very discrete. I then changed the DPLII parameters from their factory defaults to the following:

Center Width (range 0 to 7) : 7 (this effectively turns the center channel off)
Dimension (range 1 to 7) : 4 (this offers equal front/back balance)
Panorama (on/off) : on (more on this later)

With these settings, the QS channel tests performed brilliantly. I then performed the same tests with the RM encoded tracks. I always thought that QS and RM were identical. Well, this little test proved me wrong. The channel tests for the rear channels were playing in the front channels! I then switched Panorama on and the RM tests then performed accurately. The QS tests performed equally well with Panorama turned on or off, but the RM tests performed correctly only with Panorama turned on. This indicates to me that QS and RM are not identical, but I digress...

I then played some QS musical material that I was familiar with, using the settings outlined above. As far as I can tell, DPLII decoded the material perfectly. It sounded really good! I then played some stereo material to see how it did with that and I must say that I was pretty impressed. It sounded a heckuva lot better than it did in my car! Stereo material performs in a similar manner to RM material in regards to the Panorama setting. You need to have it on for the best “quad” performance.

I then started doing comparison tests between DPLII and my trusty Sansui QSD-1. The most noticeable difference here is that the QSD-1 is much, much smoother and more pleasant on the ear. I'm sure this is due to the QSD-1's analog operation versus DPLII's digital operation. I thought DPLII sounded good until I listened to the QSD-1. There is no comparison. The QSD-1 sounds many times better.

As far as decoding QS material goes, they are very close, although I will have to give the QSD-1 the edge for a slightly better, yet noticeable, discrete presentation. When it comes to decoding stereo material, again, DPLII and the QSD-1 are very close in decoding performance when the QSD-1 is in QS mode. However, when the QSD-1 is switched to surround mode, it completely blows DPLII out of the water.

DPLII is really, really good, and is certainly a viable option for anyone wanting to enjoy decent stereo-to-quad synthesis or decoding of QS/RM encoded material. Just remember to adjust the DPLII settings for optimal discrete performance! However, for those who demand the best, you can simply not go wrong with the Sansui QSD-1 or other Vario-Matrix analog decoders.

 
Just an update in my quest for the ultimate in stereo to quad (or 5.1) performance. I recently acquired a Harman Kardon AVP1A which is the Fosgate Model 3A rebranded. This was supposedly the last and greatest Fosgate unit before the Six-Axis circuit found in the Citation processors. As you can read above in the thread, the Citation was a big letdown in terms of surround depth, much more like Dolby Pro-Logic II. Not bad though in any other way.

I had the Model Four, and thought the 3A (AVP1A) would be even better. It is definitely a more expensive model, as it has side and back channels, and 4 sub outs! And it has more control over inputs. Well to make a long story short, it definitely has the same matrix, as in it does great for a real discrete surround presentation, but it very noticeably has more artifacts than the Four. The Four just does a great surround presentation, without calling attention to itself.

So far, the Fosgate Model Four is in my opinion the best hands down at doing what I want, making stereo sound like a DVD-A discrete mix. I only rarely detect artifacts, but it is always with very difficult mixes, and it only makes itself known when you know what you are listening for. I think if the matrix used in the old "rock mode" on these Fosgate units were combined with the technology in DPL-II, it would be virtually perfect. The 3A is unfortunately too obvious at what it is doing, a little like the Tate, but not nearly as bad.

I have yet to hear a Model Five, but I have read where it is reviewed as not as good as the Model Four, and was a less expensive unit. I don't expect it to be as good as the Four.
That leaves only the QSD-1 as a potential contender against the Model Four. I'll be visiting Cai, and his QSD-1 this Summer, and have a face off. Thanks Cai!

So, my advice for anyone like me is to get a Fosgate Model Four. Don't even bother with anything else for synthesis, especially when they only run about $50-100 on ebay! I think it should command prices like a Tate or a QSD-1, given its excellent performance. If anyone out there has one or gets one, let me know what you think!!!!

Dan in Spokane
 
Cai,
What are you using to decode in DPLII? Receiver or outboard?
And what do consider good test QS/RM records? What's the "Quadrafile" CDR? Finally you say to turn off the center channel for QS/RM, but I thought I remembered hearing that with SQ the center channel had one of the four original signals. So I guess this is not the case here.
Thanks,
Tom
 
Thanks for the report, Dan! I'll look forward to doing the quad shoot-out next time you're in town.

Tom, I use my pre-amp (Anthem AVM20) for DPLII decoding. Pretty much any QS/RM recording with a "more discrete" mix is good for testing. For info on the Quadrafile CDR I am referencing, check out this link on my website:

<a href="http://www.greatgig.com/quad/quadrafile.htm" target="top">Quadrafile - The Ultimate Test Disc</a>

You are confusing SQ and the original DPL (not DPLII). DPL was based on the SQ matrix. To get five channels with a center channel from a quad matrix scheme, they rotated the soundstage a bit to create a center channel and then split one of the remaining channels into two mono channels. The two mono channels became the rear channels and the remaining three became the fronts (left, center and right). This is why SQ material sounds so funky with the original DPL... it shifts the soundstage out of whack.

 
Ok, so you guys are saying that DLPII decodes QS/RM accurately. How does it do with SQ? What about playing stereo through DPLII compared to a later model Sansui "Varo-Matrix"?
 
DPLII does not accurately decode SQ material. It does an okay job with it, but the channel allocation is not correct.

If you read back through this thread you will see that my impression with DPLII (with parameters adjusted as I described) is that it synthesizes surround from stereo in a manner very similar to a Sansui VM decoder in <span style="text-decoration:underline">QS</span> mode. I find the Sansui VM decoder in <span style="text-decoration:underline">surround</span> mode to offer a much better stereo-to-surround presentation than DPLII.
 
Please refresh my memory: What's the difference in channel placement of a SQ recording when played back in DPLII? And is the channel placement correct in QS/RM unlike with SQ? Playing some of my QS recordings it almost sounded like the rears were the mains and fronts were the rears. When playing a QS/RM recording in DPLII is the center channel shared with another channel? Because if I decreased the center channel it sounded like I was missing part of the mix.
I need to get a copy of the Quadrafile disc!
 
Back
Top