Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion Meridian - Systems +3dB "LevelPerfect" Tweak Performance Improvement Poll

Mine have a ' Width' setting allowing them to go anywhere they like

Would be interesting to hear M viewpoint on all this too

~M~

I agree on that. Im almost afraid to make this tweak given that I am on the verge of V6. George will this editor work with V6 files?

If you send me a V6 I'll let you know.

One person in Poll says V6 sounds better too.

It's the same MSL file.

Like I say every where verify in MConfig the levels are what you ask for and measure them. The G91 accepted the change but ignored it.

Ok, I think I may have found the secret for the -3dB headroom.

It was already in the Wiki from an older test another awesome test if I do say myself because it was really valuable for this new bit of information.

http://www.meridianunplugged.com/wiki/index.php?pagename=TrifieldCenterEQCurves

Trifield Center EQ needs approx ~3dB of HeadRoom, especially when set to EQ3.

If you run Center EQ set to "Flat" it will not clip with the +3dB Tweak.

To test this I used a Stereo Full Scale Test signal at 3KHz (right where EQ3 boosts the most). The system was set having ALL channels at +3dB. Then I recorded center channel with Center EQ = Flat (no clip) and Center EQ = EQ3 (clip). Then I set preset -3dB on Center with Ceneter EQ = EQ3 (no clip). This is good !!

This is really not a good reason for losing 3dB on everything else (even when running Center EQ Flat in Trifield).

The only DSP Preset that does this is Trifield (Music DSP is a derivative of Trifield and has it too). Cinema etc. does not (but I didn't measure it). Only so many hours a day my wife can stand this.

Discrete movies are not "Processed". That applies to Lossy and Lossless.

So even if your a 90% Movie 10% Music person your movies pay a price for this capability in Trifield. Or if you listen in 2 Channel, you still pay for it. If Meridian must shift levels to make head room like this they should do it just in that Preset. In fact they could process the Center -3dB then tell the DSP Speaker to boost it +3dB later.

You even pay the price if you only have TWO full range speakers playing a CD !! (like the G91 test I did).

Keep this in context that we are talking awesome vs really awesome.

So if you love Trifield I suggest making sure Center EQ is set flat and run +3dB in speaker calibration on as many channels as you can. Or you can also try leaving Center EQ Default (EQ1 or EQ3 if that's your thing) and set center in Preset to -3dB in Trifield (and Music) only and run the +3dB in speaker calibration on all channels you can.

I'll try to update Wiki with this info soon.

Keep in mind also this is still no excuse (in my opinion) for doing speaker calibration in digital domain in first place. The -3dB is a seperate concern.

I will update my program to flag this so you can see the total resolution loss or clip risk for each preset. Basically if I see anything but "Flat" I will treat as if it was 3dB louder.

By the way this *might* (I said MIGHT) give a hint as to why analog sounds better at -4.0dB.

I've uploaed all the Recorded WAV files and Input Test files for the above testing.

Including a very well recorded Full Scale song I recorded on the Center Channel with the +3dB Tweak in both Center EQ = Flat and EQ3.

http://softronix.com/pictures/levelperfect/

14 Folks with DSP Speakers Say Large or Small improvement. 0 Say No improvement.

So what are you folks gonna do if Meridian says we are all nuts

Blame you of course

I'm not setup to do an A/B and don't trust my memory, so I am relying on the general consensus.

When I compare Direct DSP mode (no upsampling) between G68 and Direct to DSP-6000 instantly and hear no change I know something is a lot better because it shouldn't change. Which anyone with DSP speakers can compare instantly.

Hi George

Are you suggesting that direct mode should be seen as a digital pass-through and not coloured by the processor?

Thanks
Tom

Hi Tom,

No that's not the case if you have MRC or any other calibration or tone adjustment.

However, it should be the least processed.

I always found (568.1, 568.2mm and G68) that any processor vs SPDIF direct to DSP was different. What George has uncovered is a big reason why that is the case.

Thanks George for finding out.

The sound with +3dB (play with 3 volume increments lower) seems to smoother with voices maybe even smoother when the voices are at low levels.The piano and accustic instruments sound more "natural" with a little more body.

Robert