Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[HarryBrillJr.]

Chris Hindle wrote on Fri, 08 August 2008 07:52

Geri O'Neil wrote on Fri, 08 August 2008 07:26 There's a preset in the Greybox settings for a KF760 hang with KF-730s underneath as nearfields. Have you tried the settings in this preset for just the 730s? I wonder if they took into account that there's only 3 or 4 or so 730s hanging on the bottom as opposed to a setting for 16 730s and adjusted the preset accordingly? Just a thought... Geri O I would expect that preset to take into account the low/low-mid "contribution" of the 760's...... note - I have no horse in this race, nor am I ever likely too...

Actually frequency shading is not usually the best way to go.  You should read Bob's book.

[Peter Morris]

The 730’s will load automatically (from memory) with the crossover set at 59Hz. Another EAW thing I don’t get, when almost every time I use it at 80Hz to 100Hz.

The subs will also automatically load without the correct delay setting to align them with the top boxes (don’t get that either)… the subs need about 4ms delay.  FWIW the sub processing time is 3.13 ms the KF730 more like 7 to 8ms.

These can all be accessed from the front panel of the UX8800

You may find a shelving filter set at about 200Hz +2 dB to boost the 10s and either shelving filter around 7K – 2 to -3 dB to reduce the HF  or a 6 dB low pass filter at 20K about – 8dB, will improve the sound quality.

[Langston Holland]

Peter is right, if you load the KF730 preset you will get the full range high pass for the lows by default. Below is this default in purple and the typical 81Hz high pass for sub use in cyan. Why anyone would deploy a system of this caliber without measuring stuff first is beyond me...

For fun, I've included a somewhat busy screenshot to point out an amazing detail of the UX processor. Obviously, changing the high pass frequency and slope of the low bandpass will affect its phase somewhat, including the higher frequencies where the crossover to the mid/high bandpass occurs. That guy with the pony tail actually had the UX change the phase of the lower portion of the mid/high bandpass to compensate. Amazing. Not really that necessary IMO unless you're after perfection. The green trace results from the full range default that you see in the purple trace. Move the high pass up to 81Hz and the slope changes from 2nd order BW to 4th order LR and the blue phase trace results.

[HarryBrillJr.]

If I'm flying the subs next to or behind the mains I'd prefer to overlap through the crossover.  That's free headroom.  You will need to apply INPUT EQ to both the subs and the mains (assuming you are driving the subs from a different output) to correct for the overlap gain.

Flying the subs next to, behind, or ground stacking, will all require different delay times.  How can EAW tell you in advance where you are putting your subs?  How are you coming up with these times?  An "impulse alignment" will provide a different result than a "crossover alignment".  This all assumes you are measuring in the first place.

[Peter Morris]

The KF 730’s and SB 730 are designed to fly as part of the same array. The SB730 can be used as support for a ground stacked array – so the time alignment is usually known.

You would expect if you used the array shown below,  when you down load KF730 + SB730 they would be time aligned.

no excuse!

[Mac Kerr]

Peter Morris wrote on Sun, 10 August 2008 06:17

The KF 730’s and SB 730 are designed to fly as part of the same array. The SB730 can be used as support for a ground stacked array – so the time alignment is usually known. You would expect if you used the array shown below,  when you down load KF730 + SB730 they would be time aligned. no excuse!

Given the array in the picture, when you measure the phase response through the crossover, and the impulse response, are they both off? There are 2 well known ways to align subs to mains, which return different timings, which one are you referring too? Have you measured both?

Mac

[HarryBrillJr.]

Peter Morris wrote on Sun, 10 August 2008 05:17

The KF 730’s and SB 730 are designed to fly as part of the same array. The SB730 can be used as support for a ground stacked array – so the time alignment is usually known. You would expect if you used the array shown below,  when you down load KF730 + SB730 they would be time aligned. no excuse!

Yes, I agree with that, but how do you know it's not aligned?  How are you making your measurement?  I'm not so much questioning your data, just curious how you came up with it.  There are several valid approaches.  They may have used a different one than you.

[Peter Morris]

The time issue arises because the processing time is not fixed through the processor.  The subs use IIR filters and the delay time is 3.13 ms.  For the “top” boxes, which are “focused”, FIR and IIR filters are used (and some other tricks).  These take much longer to implement.

If you put a mic on the front grill of the speakers what you will find is:-

SB850, 750 etc about 4ms delay
KF 750 - 11.6 ms
KF 730 - 8.2 ms
KF 760 … similar to the 750 (I have measured it but can’t remember)

You can also measure it directly from the UX8800 outputs and find the same.

One of the ideas behind the UX8800 was to eliminate the user errors, to use a configurable black box, hence the “grey” box name.  It’s a great idea.  Put in the specifications of your amplifier and select the speaker boxes you need and every thing else is worked out for you.

I’m happy to accept that in many situations you will not know the sub alignment, and will need to Smaart things, BUT this makes it mandatory, and as such, defeats the black box idea to some extent.

For example a 730 rig of 2 x SB730’s and 4 x KF730 a side ground staked as designed – put the amp settings in the pilot program, down load the settings from the Pilot into the UX8800 and your sub will be out of time/(phase). – My argument is that this configuration and process should work straight out of the box. Smaart should not be needed in this case.

With respect to your point about there being 2 ways to aligning the subs - given that GF produces a very flat phase response, and the sub crossover function is LR which is 6 dB down at the crossover point, I would argue that there is only one way to align the subs – for maximum summation at the crossover point, so that the subs and lows have exactly the same phase response and are time aligned (more or less).  The net result in smart (phase response) looks like a text book plot of a 24 dB LR filter + the port function of the sub.

[HarryBrillJr.]

BUT, I happen to know some of the guys that were/ are involved with setting this stuff up prefer an impulse alignment in many cases.  Is the bulk of the sub energy flat phase and the bulk of the main energy flat phase?

On the subject of latency, I think they should have all settings default to the same latency, whatever has the most.  This is the only way to ensure that speakers play nice together when used together.

I'm not arguing for or against EAW.  I just know the guys that are creating these settings, and they are not sans clue.

[Peter Morris]

Its not an alignment preference,   4 ms is almost 180 degree out at the crossover frequency.  Initially I though the polarity was wrong.

….the EAW engineering guys are great.  My issue is how the UX8800 has been delivered to the market …  Accuma Labs … and….

Bottom line is: - 18 or months after its release, you can not just turn it on and have it work as it should….

[ProSoundWeb Community]