It was brought to my attention via a post at HTF that some receivers/processors run the LFE channel through the low pass filter of the crossover. Thus, if I were to set my mains at 40hz (which is how I have them currently), then I may be wasting the bass above 40hz in the LFE channel. Is this something the 950 does?? I guess I had figured that the LFE channel was a discrete signal that was passed to the sub, and then the frequencies below the crossover point from each speaker was passed along to the sub. I have Paradigm Studio/100's as mains, and if I set them to a higher crossover point (e.g. 80hz), I feel like there would be no benefit to having purchased the 100's over the 60's or 40's. Can anyone help??

I know that, for the six channel analog input, the subwoofer channel has a fixed 120 Hz LPF. I'm not sure if it's global but I think it is.

My understanding is that the 120 Hz low pass is also applied to the LFE channel from DD/DTS/etc sources as well. Entirely different from the low pass to the sub for the center, mains, and rears/surrounds.

Correct, this takes place in the decoding stages along with the 10db LFE boost.Some soundtrack however already "filtered" before encoding,and DTS at half bitrate start to roll off at 90hz.

Sorry, but...incorrect.

The 120 Hz. LFE LP filter is only used when in 'Dolby Configuration 0'. This is all sats set to 'large' and sub to 'yes'.

All other configurations (which include any speaker(s) set to 'small') cause the direct LFE path to be disabled. The bass signals are then sent to a summing module and then to the LP filter of the selected X-over.

So, to answer the original question, yes, if you set the X-over to 40 Hz., you will lose 1.5 octaves of the LFE...assuming you feed 1 sub through the SW output of your 950.

As far as having no benefit in purchasing the 100's, this is not-to-worry. If a X-over point of 80 Hz. is selected, then the mains should have flat response to 1/2 to 1 octave below that point (60-40 Hz.). On the other side, if the sub is LP filtered at 80 Hz., it should have response capability to 120 Hz., thus fairly preserving the LFE signal at the same time.

BoB- Unless you know something I don't, I don't believe that that is correct.

The 950 uses individual low passes on the signals from the center, L & R, and rears/surrounds, that sum with the LFE 0.1 channel which is low passed at 120 Hz. The reason why I suspect that this is so, is that the Sony TA-E9000ES had an individual setting for the LFE low pass freq, plus the fact that Scott confirmed the individual low passes for the main speaker channels. ??

KCB:

Simple enough to verify:

Play a digital multi-channel disc.

Set all 3 X-overs to 120 Hz.

Shut the 5 sat's amps off and just listen to the sub.

Switch the FL/FR X-over to 40 HZ.

If you still hear 120 Hz., you and Scott are correct.

I've studied Cirrus' block diagrams and read their papers. Unless they built something different from what I've read, the LP is determined by the mains filter when sats are set to 'small'.

Tell me what you think?

" [Ex.: Dolby Configuration 1 (all sats set to 'small', SW to 'yes')], Bass Manager 505 is enabled with the following paths enabled or disabled. The direct LFE path to the output through summing module switch 805 is disabled and the path through summing module 802 and Low Pass Filter 803 enabled. The direct LFE inputs to the Processing Modules are turned off."

IOW, the LFE 120 Hz. filter is used when all sats are set to 'large' wherein the LFE has a direct path to the SW output, through the filter. As soon as you set sats to 'small', this path is disabled and the LFE is routed to a summing module, summed with redirected bass, and then sent to the selected LP filter.

The listening test is the easiest way to tell if you're getting 120 Hz LFE to your sub. The difference between 40 Hz and 120 Hz is something anyone can hear rather easily.

I truly am baffled by the resistance I get, everywhere I say it, but I'll say it again. How hard would it be to have a seperate LFE output with NO filter that feeds a seperate LFE sub? Every sub has it's own filter. Include a switch to sum it for those who just don't care or only have 1 sub. LFE discrete: 'yes' or 'no'. (DUH)

BoB- Looks like you are correct. I keep making the mistake of *assuming* a piece of gear operates in the way that *I* would design it...

I'm assuming here (!) that the LP 803 is the main L & R's?

Bosso,

Is this the case for the 950's 6 channel analog input? I'm pretty sure that the subchannel has the fixed 120 LPF and it gets the redirected bass from the other five channels. It appears to be independent of the triupple crossover settings. I'll try your test tonight and report back.

DB:

There is no information, to date, that can lead to the answer to that question, which is, 'What happens to the global 120 Hz. LFE filter when the analog BM is engaged?'.

The test described above is for digital only because, of course, there is only 'on' or 'off' of an 80 Hz. X-over in the analog BM.
The analog BM scheme is:
1. A global 120 Hz. LP filter on the LFE signal.
2. An 80 Hz. LP(redirected)/HP X-over on the 5 sats, or...
3. An 80 Hz. LP filter(redirected),and full range signal on the 5 sats.

My guess is that the global LFE filter is left intact in both '2' and '3' above. I can't test for it because I don't derive the LFE from the 950 when in '5.1 bypass' mode.
Anyone got an RTA and some time?

Here's the way that it *should* be (according to me ): the LFE for the 6 ch bypass should be sent through full range. Most subs have a crossover applicable to the input anyway. But I bet that if the 80 Hz crossover is engaged, that then the LFE channel is low passed at 80 Hz too.

Bosso,
I haven't tried your "theory" but if it's right then the 950 falls into the chategory that Brian Florian explained in his now "infamous" article about variable crossover designs,and Scott has some explaining to do.

Hun- Not really.

The Secrets article has to do with the digital crossovers. In fact, in a recent thread Scott addressed this again, in that the Cirrus chip theoretically does do the digital crossing over correctly.

This is regarding what happens with the analog inputs. But most pre/pros have no BM on the analog inputs!

The author was asking about the digital BM in the 950.
Bosso claims that you would loose content from the LFE channel if you select 40hz or so for the mains,Scott said otherwise in a different thread.That is what I was respomding to.
The analog 6ch input is a global 80hz low/highpass filter,we all know that.

Can you dig up that thread? I only ever remember Scott commenting on the digital crossover.

I'm not sure I understand this, Hun. The subwoofer channel has a 120 Hz LPF, not 80. The other channels are 80 Hz high pass or full range depending on the position of the switch. Have you found this to be different?

KCB: I recall Scott's post as referring to the analog BM. It was a question of whether or not the 950's global LFE LP was the same as the ICBM (which is said to be 150 Hz.).

I recently asked someone if the 45A has a global LP on the LFE and he said it does, @ 100 Hz. So, what happens if the player's fixed LP is 100 Hz. (or, as in Sony players, 120 Hz.) and it sends that filtered signal to the 950's 120 Hz. filter?

Since both filters are fixed, could this be the answer to the phase problem, when switching to 6CH BYPASS?

Things that make you go...Hmmmmmmmmmmm.

BoB- Ahhh, maybe.

Because digitally, the LFE would have no filtering in the 45a. Any filter always adds a phase shift.

Ahh, but I can test that now. I have the MC-8 and am running full range everywhere, so I should *still* have that phase shift then. I'll check for it this weekend.

But Hun was talking about a 40 Hz crossover, which can't happen with the analog inputs. Maybe I'm confusing myself...

Ahh, actually don't even have to test for it. The 1st test I spotted this with was a discrete test tone CD. 2 ch. So any LFE filter in the Pioneer wasn't even in the loop. Has the be the difference between the digital and analog crossovers in the 950.

KCB: How did you spot it with a 2 ch discrete test tone CD?

Play the CD through a digital input. I have both normal and inverted inputs on my sub. With the digital crossover engaged, and playing that particular test tone freq, I pick the sub input with the loudest response at the listening position. (I also adjust the variable phase knob for max output too.) 5 - 10 Hz above and below the crossover freq yields the same result.

Then, I use the analog outputs from the player (player: all large; sub on, but doesn't even matter for this test), into the front 2 channels of the 5.1 inputs. I engage the 80 Hz analog crossover on the 950. I redo all the optimization for max output. In the end, simply switching sub inputs yields the max output in this case.