8200 vs. DSP-X products

[FFoti]

Not necessarily so! Even though the breakpoint is high, it's not an absolute frequency limit. There is "spillage" between the bands as the slope of the x-over filters is gentile. Band #5 is still doing work. We've tested this, and it still effects the high frequencies even in narrow-band situations.

-Frank Foti

> Yes, and with the "brick-wall" filter, there is a trade-off
> between brightness, distortion/ringing, and overshoot
> control which needs to be fine-tuned, especially at lower
> frequencies (like the horribly low 4.5 kHz bandwidth that
> the ITU requires in Europe).  Plus there is no point to
> having four or six bands of processing when the upper half
> of those bands will never be heard on the air!  Thus,
> adjustable crossover frequencies between each band is a
> really nice thing to have, at least for users who are stuck
> with narrowband AM audio.  For example, on the Omnia 5-EX,
> the highest band's crossover is at 4.8 kHz -- so if you're
> only transmitting 4.5 or 5 kHz's worth of audio, that
> processing band is rendered useless.

[westlife]

> There is "spillage" between
> the bands as the slope of the x-over filters is gentile.

There is a joke to be made here, but I'm not gonna touch it. :-)

______________

"This is the New York Emergency Broadcast System satellite channel.  They took the crosstown bus."

[dspxscott]

>
> Yes, and with the "brick-wall" filter, there is a trade-off
> between brightness, distortion/ringing, and overshoot
> control which needs to be fine-tuned, especially at lower
> frequencies (like the horribly low 4.5 kHz bandwidth that
> the ITU requires in Europe).  Plus there is no point to
> having four or six bands of processing when the upper half
> of those bands will never be heard on the air!  Thus,
> adjustable crossover frequencies between each band is a
> really nice thing to have, at least for users who are stuck
> with narrowband AM audio.  For example, on the Omnia 5-EX,
> the highest band's crossover is at 4.8 kHz -- so if you're
> only transmitting 4.5 or 5 kHz's worth of audio, that
> processing band is rendered useless.

Would you want the Xovers to be fixed (and if so what at), scale with output low pass filter frequency or be manually adjustable?
I know i am being lazy and could probably find it but anyone have any links to images of the NRSC low pass curve. I am looking at how many dB down the stopband of the filtering needs to be??

[FFoti]

dsanders@telos-systems.com

> Hey Frank,
> Does Denny Sanders have an email address? I wanted to ask
> him a few questions and searched high and low on the Omnia
> site but failed to find one. I spoke with him a month or so
> ago and wanted to follow up.
> Thanks,
> George
>

[westlife]

> Would you want the Xovers to be fixed (and if so what at),
> scale with output low pass filter frequency or be manually
> adjustable?

Ideally, all of the inter-band crossover frequencies should be user-adjustable.  That would allow a very useful degree of customization which most other digital processors don't provide.

> I know i am being lazy and could probably find it but anyone
> have any links to images of the NRSC low pass curve. I am
> looking at how many dB down the stopband of the filtering
> needs to be??

Here you go... the shaded area shows the official specification (-15 dB at 10 kHz, and so forth until it reaches -50 dB at 15 kHz), while the curve represents the typical response of an analog NRSC brick-wall filter.



That "bump" at 11 kHz in the stopband is not a mistake -- that was designed to allow some leeway for analog filters which have a bit of spillover in the 10-11 kHz region.

______________

"This is the New York Emergency Broadcast System satellite channel.  They took the crosstown bus."

[westlife]

> Orban 8100A goes on eBay from 1.2K to 1.8K, so I doubt
> you'll be able to get 8200 that cheap...

A completely stock 8100 goes for that much.  An 8100 with the Orban 6-band limiter add-on, or with various add-on aftermarket processing (Texar Prisms, etc.) is less common and fetches higher prices.  In fact, if you really want that kind of setup, you'll probably have to buy a stock 8100 and the add-on processing separately, because that's how they usually end up being auctioned.

And unless you are really in love with the way it sounds, I find it hard to justify spending more than $1000 on a stock Optimod 8100 when you can get a brand new DSPXmini for $1700 which will easily outperform it.

______________

"This is the New York Emergency Broadcast System satellite channel.  They took the crosstown bus."

[Russty]

The plain and simple answer is you get what you pay for.  Of course the DSP-Xtra is going to sound better than the DSP-X which is going to sound better than the DSP-Xmini.  Likewise the Omnia 6 is going to sound better than the Omnia 4.5 whic still sounds better than the Omnia 3.

Hey, If you really care about the sound of your radio station, then you will have no problem spending the extra money on a good quality top notch processor like the DSP-Xtra.  If you just want to produce square waves for loudness and don't care for sound quality, then put whatever cheap stuff you can come across on the air.  Leave the 8200 there if you want to be tight budgeted.  In itself it ain't a bad processor.  However, if you want to go head to head and play with the Big Boys...

Personally, if I were running things, I would want my stations to sound the best they possibly could, regardless of market size. I would work from the tip of the antenna all the way back to the source inputs in the studio to make sure I had the best sounding station(s) on the dial.  That's just me though.  That's why I am a technical grunt and not management.