Anthem D2/D2v/AVM50/AVM50v/ARC1 tweaking guide - Page 1215

Exactly! I don't mean anything like stop listening to great music just because it was recorded in 78 rpm, but only when something is unacceptable because it's not according to the standards of its day by a long shot. Look at the fuss that just one dissolve is causing:

http://www.hometheaterforum.com/t/31...ory-in-blu-ray

We're always on the lookout for demo material and once in a blue moon we also find something that's broken. I can think of one fairly recent example of something with a crazy LFE channel. Wouldn't it be worthwhile to fix the movie and re-release it if all its other aspects are spectacular? I can't imagine going back to 80 Hz global crossovers for the sake of crappy content, wherever it may be. If 80 Hz is right for LFE to prevent directionality and focus issues, it would also be right for redirected bass because the sub doesn't know or care which channel the content is coming from.

Very few are broken if you ask me. But it’s not so black and white as being broken or not. Not all is perfection even for movies.


Yes. So if a content maker wanted to create an LFE signal that met this requirement, using his mixing console, what filter order and corner frequency would he use? A typical 120 Hz filter will not meet this requirement because filters have slopes. As shown in the graph, a 120 Hz 4th order LPF (green) is only down 6 dB at 120 Hz. The 80 Hz filter is down 14 dB (red) -- better, but still not optimal.




It depends on how the guidelines are interpreted. Here’s what the DD Encoding manual says:
The 8th-order 120 Hz elliptic (blue) certainly meets the requirement, but it is not intended to shape the audible spectrum, just to limit the extent to avoid aliasing and minimize bitrate. The "sound" of a brickwall filter is rather odd when it is in the audible passband. Ideally the LFE spectrum would be defined by the 80 Hz plus the elliptic.


Just to be clear, I was describing an LFE sub limited to 80 Hz. No crossovers. And yes. Ideally LFE should be filtered at 80 Hz (and limited to 120 Hz), and the filter should be in the signal path being recorded, not in the monitor path.

Let me mention something else. The heyday of theatrical 5.1 and LFE really came with the advent of digital film sound, even though they originated with 70mm mag. Not many people realize it but DTS theatrical was a 5.0 signal path, with the LFE spliced into the surrounds. The cinema processor split the signals with 80 Hz filters: HP to the surrounds; LP to the subs. This forced movies to be monitored with LFE limited to 80 Hz so that the results would be consistent on the stages and in theaters with any of the digital formats. But even then the recorded LFE tracks were not uniformly confined to 80 Hz. This became rather apparent at home when comparing a DTS track to the same soundtrack on DD (rare as that was). The "deeper, tighter" DTS bass was easily replicated by filtering the DD LFE to 80 Hz. Should we hear what the mixers heard and what was played in the cinemas, or what was recorded in the track?

So as you can see there are differing practices involved, and the results vary as a result. Depending on who you ask, the question of whether it is correct to reproduce everything delivered in the LFE channel has different answers. But unless the answer is "yes, always" then some sort of LFE filter option is needed, even when when ARC is running.

************************************************************ ******

1) In Setup > Level Calibration, zero out *ALL* The lines

2) In the first line, set test mode to Manual

3) Scroll down one line to Test/Noise Level. The test tone should be coming from your Left Front speaker. Adjust that Test/Noise Level line to yield 75dB SPL as measured using your SPL meter at ARC mic position #1.

4) Scroll down to either subwoofer line. The test tone should be coming from your combined system of subwoofers. Leave that line at 0dB. Check to see if the combined SPL from your system of subwoofers is roughly 75dB SPL measured from the same spot as above. If not you should adjust the output of your subwoofers using their own controls -- leave the subwoofer line in Level Calibration at 0dB. I'm sure you have your preferred method of doing this which leaves all your subs balanced with respect to each other.
************************************************************ ****

With respect to the quote above, when I measure 75db at the "Test/Noise Level" line for the Left Front speaker, I set this to 75db. Then, do the rest of the speakers matter as far as getting them to 75db besides the sub. The sub is at 75db when I set the Front Left but the remaining speakers are not all at 75db. They are close at between 73 to 76. Do I leave those values, for the remaining speakers at 0 and let ARC do it's thing or do I also line adjust each speaker to 75db? Note, I have set my speaker distances prior to doing this part.

You can leave the values, for the remaining speakers, at 0. Just setting the LF to 75db and the sub to 75db is good. Once you have set them to 75db, you can run ARC, and ARC will take care of things from there.

Did that ninja12.

ARC is very picky software. On my Dell laptop, which I ran ARC with no problem before and same 3.0.2 version, it would not even kick off. It would get to the point where it would recognize the unit and turn it on but not come up with the prompt to have the mic at position #1 to and hit OK to start the measuring sequence.

Went to my Macbook Pro, running Windows 7 bootcamp, and it went okay there as it has before as well.

Funny thing, good that I have two laptops though. I wish ARC would have some error logging so that we can send stuff to Anthem when it seizes up or goes into a loop (I've had ARC just stop on a random speaker, regardless of the listening position and never move onto the next speaker because it would not stop emitting the test tone). I wrote to Anthem on that but they never did get back to me with a solution or any thoughts as to what might be causing this.

On the run of ARC I just did, an error came up with to the effect of too much noise detected at surround right and would I like to repeat the test tones to which I said YES and it continued okay from there. I'm always leary, however, when ARC does not complete without error. Lately, I find I just hate running ARC because it's not very reliable and when errors happen it's a very tedious and time consuming process.

Attached are my latest ARC results, what do you guys think?

Your RF has a dip that ARC couldn't fully correct; but, it's within 3db of full correction so I wouldn't worry to much about it. Other than that, your charts look very good. How does it sound to you?

The good news is you won't have to run ARC
again Charts look very good.
John

He could lower Room Gain a bit to help with that. Which would just be a re-Calculate and re-Upload. No need to re-Measure.
--Bob

Thanks folks for all the help. Yah, it sounds pretty amazing.

On the video side, I have my Video 1 and Video 2 settings as follows:

S-Video OSD: NTSC
Preferred: HDMI
Resln: 1920 x 1080 p24 (Video 2 is set to 1920 x 1080 p60)
Color Space: HDTV
Data: YCbCr 4:4:4
Output: 12 bit
Letterbox: Black
Sync: Normal
Component 2 Out: Passthru

Exact same settings for Video 2 save p60 as shown above.

I use Video 1 (p24) for my Sony BDP-780. I also used it for my Panny BDT-310 but it's probably a mute point/setting for it as I have the main out of the Panny going to the projector. I also use Video 1 for my PS3.

I use Video 2 (p60) for my Sat TV and Apple TV.

What's the general rule of thumb for p60 vs p24 usage based on my devices for example? Can I use p24 for iTV as I have the new iTV that does 1080p now.

On my Sony 780, just in the menu screens of the player, I notice lag/ghosting when moving from left to right of the menu screens but does not seem to happen when I go up and down.

Playback looks amazing but every now and then I get HDMI handshake issues. Do I have these settings, for the most part, correct? If it helps, I am using an Epson 6010 onto a Firehawk G3 (90 diagnonal).

Guess I'm bored today thus really nerding it out with my D2v

Since going to 2.13g, I've had about 3 or 4 full/complete audio dropouts where I've had to turn off/on my D2v in order to get audio back. I've experienced these dropout only when changing sources or leaving the unit idle with no audio playing and then coming back to use it. I have not experienced an audio dropout or hiccup, however, while the unit is playing. Still, I never experienced so many audio dropouts (in such a short time) with version 2.10. I don't understand why Anthem has not released a non beta of > 2.10; particularly, when there are fixes to be addressed.

It has been a while that I did not run arc and today I ran ARC and I also got the error 0x4.

I think that 0x4 is when it could not hear anything from the 4th speaker in the sequence. I have a 4.1 config (no center) and it was doing Fron-Left, Front-Right, Surround-Left, and nothing.

After having check all cabling, and everything was connected correctly. Then I figured that I forgot to uncheck "Rear" in ARC, so it was just trying measure Rear{L or R} which I don't have.

Thanks Mstanic ... i guess i'm suprised, despite the many that have installed beta FW in the past, that no one else here contributed to answer your own question for i suspect many would like to know the answer as well. Sad indeed ... and thanks again for taking the time to answer(your very own question!).

I have 7.1 and it gets the error on the 8th speaker - The Sub.

I agree in principle but as mentioned in practice things may not work ideally when redirected channels must be summed with LFE and room correction is to be run on top of it, all with a crossover algorithm created to prevent audible rounding errors plus a room correction system that uses twice the processing power as the industry average for home systems and about as same as pro systems. In an ideal world one fast DSP chip would handle everything with no compromise but that has to wait for another time. The D1 was the first processor to use dual Motorola chips, having been planned with room correction in mind, and for that we had to create a way of making one chip talk to the other. It wasn't as easy as it seemed in the beginning and that alone delayed the project more than could have been anticipated. Later, that end evolved into the current dual core Freescale chips with the HD decoders and 8-channel 192 kHz capability, but the perfect system is still elusive. How bad is this? I say hardly bad at all. Hypothetically, others can claim that they have the ideal system but frankly I'll believe that when I see it and wouldn't give up ARC for anything else on the market today. (Yes that's obviously biased even though I just work here, but just try to pry it away...).

So when these cards are dealt, the lesser of two evils has to be decided on. Again, engineering is about finding the best tradeoffs, and there will always be tradeoffs. To filter LFE for the sake of the odd soundtrack that could use it, or do as was done to avoid taxing other things? Some time ago I sampled the LFE output of several big ticket movies, running them through a spectrum analyzer to see what the upper bandwidth limit was (the opposite of what owners of big subs look for), and while there was plenty up to 120 Hz, I didn't see anything beyond that. Not to say it doesn't exist, just that it's usually in hiding.

So in the end we picked our poison and what's done is done because it's impossible to please everyone and everything, and more importantly because as you and I agree cases that could have used the other option are far from being great in number. Now this makes me wonder, what's a good and practical reason to disable the encoder's 120 Hz filter anyhow? That's even if the LFE track really has nothing above 120 Hz. Just asking in case anyone reading this knows.

I'm not worried. The highest my sub's LPF goes is 100 Hz.

I've not seen any beta tester report that can be connected to this problem description. If you haven't, please send a report to tech support including model numbers of everything involved, connection types, audio formats, video formats, direction of switching when problem appears, how often it happens, and D2v serial number just in case build date has anything to do with it. This is the only way anything can be done about it.

That's a communication error where ARC is trying to tell the tone generator to stop sweeping but the command isn't getting through. I've heard of that with USB adapters in the mix, never with straight serial. As for the communication error with tech support (happens sometimes), please just re-send using the online template instead of straight e-mail if more help is needed, or give Piero a call.

I did just that... Put some shelving material under the stand to raise it a good 2 inches and also allow me to slide it around while making measurements. The extra height didn't affect the response at all. I could go 2 more inches but the next size up in actual speaker stands that I can buy locally is 22" which would put the top of the speaker too close to the bottom of the screen.

The only thing that helped slightly was moving it about 1" closer to the wall. New graph attached. I think I am stuck with the response of this CENTER speaker in this room (middle of almost square room).

First attachement is my "convenient" front-left corner position. I found this to give the most low bass using quick measure.

Now, in the second attachment, I managed to move it to the left side about 1/4 of the way into the room. In the third chart it is 1/3 of the way in. I think it lost too much at 50 Hz so I put it back in the corner for now.

Given that the sub's LPF kicks in at 100 Hz, should I push ARC a bit to get as much out of the mains as possible, leaving the sub for the bottom octaves only?

Thanks,
Stefan

The room correction will have no knowledge of the LFE filter, be it in the recording studio or in the AV processor, so that's not an issue.


I have the luxury of talking about these concepts absent the realities of any particular product's design issues.


I'm talking about making the overall reproduced sound quality better, not just dealing with odd soundtracks.


Yes. that's true. My question is whether that is what we should really be reproducing, wrt the filter diagram I posted. It's a rhetorical question.


Very little. If the content maker has provided sufficient filtering, or otherwise created an LFE signal that is constrained to <125 Hz, then there may be a desire to avoid the passband phase shift of the filter.

While adjusting your sub levels, you might wish to take note that some SPL meters exhibit a low frequency rolloff, resulting in an actually higher level being set. Example: the newer RS SPL meters are down about 5dB at 31.5Hz (digital version 3dB) and almost 10dB at 20Hz (digital 6dB). The older meters exhibit a much lesser rolloff.

This effect is evidenced by a higher post ARC trimming of the sub level as compared to the other channels.

Ben

When I had this same problem a while back Bob suggested that I turn off my firewall off in my laptop. Just temporarily turning it off during the ARC procedure, that is, not turning it off permanently.
I have not had that problem since.

Tom

I am surprised no one mentioned your "Hall Of Fame" worthy bass graph.

Did you run a full measurement to see what the corrected result is like? That would be worthwhile because this Quick Measure has the widest range of the three. You might end up losing a little at 50 Hz after correction but gain a lot at 100 Hz.

Besides, I wouldn't be remotely as afraid of EQ applying +6 dB at 50 Hz in the sub instead of the center channel, just for the sake of the center channel's driver, in case you're still inclined to change its correction target from 95 Hz to 60 Hz. I still don't recommend that but it all really depends on how loudly you'll ever play things.

And we have the luxury of making things the way we want them to be in our own homes! ;-)

I hope everyone understands that taking the audio hobby to the next step of getting into the biz is a bit of an insane choice -- as much as many in it are afraid to admit, I'm sure -- and when planning our flagship systems we start with a dream system concept the same as anyone else would. Then reality sets in and one of my favorite Albert Einstein quotes is the first thing that pops into the mind: In theory, theory and practice are the same. In practice, they are not.

So after the rhetorical questions have been pondered while considering everything that has been written and talked about on the sound quality of our gear, especially since ARC came into the mix, I can't see any reason that we should have done things differently, and I realize opinions will vary to no end just as they do on the production end. If there's a standard and it's not always followed, how can there be only one way of doing things on the playback end? Maybe one day we'll have the dream DSP chip that will allow the dream system. Maybe it'll even allow the ditching of the PC in running ARC parameter calculations as practically everyone would like. That's still just a dream with nothing on the DSP horizon to even hint at change.

Gotta go for now, the reality that work and hobby are two different things is setting in on yet another Monday morning.

Hello Bob and Nick.

Bob was RIGHT - When ARC says Error 0x04 Turn Up
SUB Volume. It really means DOWN.

With Andrew's Help today all is well. I must admit to get
ARC to run - the Sub Tones are wicked low - but then I
am use to strong BASS.

^ A low sounding bass sweep makes sense, since much of the sweep range is below the frequencies you can hear anyway.

So did you re-engage your floor bouncers?
--Bob

OF COURSE - but not while running ARC. I just used conventional Subs.

0x04 can also mean it's clipping, though unlikely if Quick Measure readings were below 90 dB.

Balancing main channel levels by ear isn't hard but setting the sub 10-25 dB too high based on test noises is not unusual because our ears just aren't sensitive down there.

----------------------------

SPL meter phone apps are becoming more popular. Be careful with them - phones have automatic gain control for their mics which throws off readings by a huge amount.

If you download such an app make sure you calibrate it against a normal meter while playing the main speaker test noise at 75 dB, and check readings again using the sub test noise. Chances are that the app will still be way off in the bass but if it holds the peak reading, it may be the more accurate one. It depends on your phone and the app, and different apps don't even agree between themselves on the same phone even after calibration. Try to memorize the margin of error if you intend to use the app afterwards.