First off, I am a new user on this board, and a relatively new EMPEG Mark II owner (I got mine used). For now, I just installed the EMPEG as my primary source in my Honda CRX... but that is just temporary because I didn't want to wait any longer to start using it in the car (and I must say I really love this thing!).

I plan to add my OEM CD player back into the system (primarily for IASCA competition)..... the CD player has already been modified from the original installation - I removed the IC ampflifier and associated parts and added a discrete preamp output section that bypasses as much of the OE circuitry as possible....

My plan was to feed the OEM CD deck preamp outputs into the EMPEG aux input.... that way I don't have to mess with the OEM deck CD changer interface. For my intended usage, this would be ideal anyway since the the EMPEG will be the primary source deck 99% of the time, and the CD only used for competition or to play the occasional CD I haven't ripped and encoded yet. Therefore, I want the CD player out of the signal path when it's not use, so I can jjust leave it off and keep the audio path as direct as possible. That way, I can also use EMPEG's nice EQ on both sources as well.

But the problem that I came across is that I noticed that EMPEG specs indicate the frequency response is only to 18 kHz on the aux input for some reason.... does anyone know if that is real, and if so, what the roll-off rate is..... and why there is this limit? I would guess it is due to the the ADC used in the Phillips DSP chip or something in the analog path between the aux input connector and the DSP input (anti-aliasing filter, etc)? Has anyone experimented with ways around this (I2S directly into the DSP, or improved filters, etc)?

I also noticed that the output frequency response of the EMPEG is not specified anywhere I could find.... hopefully there is not something inherent in the Phillips DSP design that limits everything it processes to an 18 kHz bandwidth limit. I doubt that would be the case since I have heard this is supposed to be a pretty good DSP, which is why I am surprised by the aux input freq response in the first place.

I guess I will find out about the output BW soon enough when I go RTA my system at a local shop with ripped pink noise on my player in a couple of weeks.

Thanks for any info!

Sorry about changing the subject, but what encoder are you using to make your mp3s? As far as the 18khz cutoff, most people cant hear much above that level anyway. The sound judges at competition might, but those frequencies would be drowned out by bass I think.

Sean

I have been using LAME at 256-320 vbr for MP3 encoding. I have experimented with several encoders, and this seems to fair pretty well at higher rates - most of the time I can't tell the original from the encoding (Not that I am a 'golden ears' or anything, but as long as I can't hear the diff, I am happy :).

As far as the 18 kHz limit, yes you are correct that most people can't hear above that anyway... and I don't know whether or not I could hear the difference of a music track before and after a 18 kHz filtering.... but I have heard the effects of the 16 kHz filters often used at 128 or 192 and below in MP3 encoders, and that is usually quite noticable in A-B type testing (depending on the content). It is not unllsitenable or anything by far - but I want to avoid doing things along those lines if I have the choice.

For competition, it is also questionable as to whether or not a judge would hear the difference during subjective judging. But if the roll-off is too steep (more than 6 dB between 1/3 octaves) you will fail the objective part of the testing (RTA measurements)..... which was my immediate concern.

"Tanstaafl" on this BBS is doing precisely what you're talking about: Feeding a CD into the aux-in of the empeg and competing in IASCA with it. He is happy with the results, although you have to be a little careful with EQ curves. By default, the empeg doesn't artificially boost the bass frequencies the way most consumer CD players do, so the CD player will tend to sound warmer than the empeg until you've managed to create separate EQ presets for the CD player and the empeg.

As far as the 18k aux-in limit in the specs, I don't know why it says this. If it's true, then it must be a limit of the DSP they're using. Easy enough to test this assumption with the proper CD...

___________
Tony Fabris

Excellent. Make sure you are using 3.88, because 3.87 doesnt handle freqencies above 16khz properly. What command line settings are you using? I lowpass at 19.5 because I can't hear anything above that anyway and it saves a lot of bits.

I don't know much about competition, but I think someone on the board competes. Maybe he could comment on your problem.

Sean

Again OT, but I use MusicMatch at 160kps and it defaults to 18khz cut off... I had to adjust the slider to 20khz...

32GB (JUST UPGRADED!) Mk. II BLUE
Detroit, MI USA
www.PfeifferBeer.com

I guess I will find out about the output BW soon enough when I go RTA my system at a local shop with ripped pink noise on my player in a couple of weeks.

Unless you are just insatiably curious or can get the RTA analysis done for free, save your money. The MP3-encoded pink noise track will bear little if any resemblance to the frequency curve of the original .WAV file, and in any case (as you know full well) IASCA will not allow you to use the empeg as your source unit.

Once version 1.1 of the software comes out, you can put the pink noise in the empeg as a .wav file, and it will then be very interesting to see what the different frequency curves are from the empeg and from the CD player.

Just as a side-issue, I have the 1999 IASCA tracks in my empeg in both .wav and .mp3 and have been doing A/B comparisons. It is too early to say anything for certain, but my first impression is that the .mp3 tracks seem just a bit "thin" and bright sounding compared to the .wav tracks, but the differences are so slight that I wouldn't want to swear to it. At present, I don't have the ability to play CDs in my car, so I can't make it an A/B/C comparison. I should have that ability shortly -- new competition car is nearing completion.

I, too, think it is a good idea to route the CD player into the Auxilliary Inputs of the empeg -- that is how I am doing my new car -- not only because this gives you the ability to use the empeg's 20-band parametric equalizer, but also the rather clever loudness setting that can be adjusted in 1.5 dB increments.

Therefore, I want the CD player out of the signal path when it's not use,

To do this right, you need to look into the Sony XA-39 switch so that you can play the CD player with the empeg out of the car. Do a bbs search on XA39 and XA-39, and you'll get a few interesting ideas about this.

Will you be competing in the "Street" class or the "Ultimate" class in IASCA? I have the same concerns about the RTA that you do... only my concerns are more from the low end of the scale since I am presently running only a single 10" subwoofer (the new car will be two 10" subs) and I have had problems getting much response down in the 20-40 Hz range. My 4" MB Quart coaxials are so good on the high end that I've not had any problems there. :-)

Check your email -- I've sent you a little something you might find entertaining.

tanstaafl.

"There Ain't No Such Thing As A Free Lunch"

Unfortunately, I encoded most of my MP3s some time ago with earlier versions of lame.... I have been upgrading as new releases become available, but I haven't gone back an re-encoded the older stuff yet (but intend to eventually). I just added another hard drive to my home system which should give me enough spce to archive everything losslessly (using MAC) so I don't have keep re-ripping my CDs everytime there is a major improvement in encoding technology..... eventually, I'd like to just play back WAVs or losslessly compressed files on the EMPEG as well (maybe in another year or so.....when laptop drives are avaiable in sufficeint size at reasonable cost).

This is the command line I have been using typically
-b 256 -m j -h -V 0 -B 320

I haven't experimented too much yet with turning joint stereo off..... but I intend to do some comparisons with it onn and off before going back and re-encoding all my older MP3s. See if there is a trade off to be made with stereo separation (imagin, soundstage) and MPEG artifacts.

Thanks for the reply.....

Well, the RTA analysis is not free, but only $20 for a session at a local shop. I want to dial in my whole system, I just replaced the front sound stage (new speakers in new locations due to added kickpanels) and I want to get everything dialed in. I plan to have the CD player installed by then as well, so I can also do pink noise testing with it to get evrything pretty close, and then look at the pink noise geenrated from the EMPEG to see if is similar.

Decoding the pink noise MP3 back to WAV and looking at an FFT of the pink noise (using MATLAB), it seems to actually be pretty close to the original, more so than I execeted (the pink noise still extends out to 20 kHz too).

And yes, unfortunately, you are right that IASCA won't let me use the EMPEG as a source. If so, I wouldn't even be putting a CD player back in the car. I asked them about the possibility of competing with WAVs ripped from their CD stored on the EMPEG - the answer was maybe in the future, but definately not this season (and it sounded like probably not next season either). I even offered to bring a laptop with a CD-ROM drive and rip the actual judge's copy of the CD while he/she was watching, letting him/her name the files, and then download them to the EMPEG - still no. Too bad.

Thanks for sending me your system diagram, very cool. Sounds like a lot of work to accomodate your MB quart front sound stage in the dash - but I bet it was worth it :).

BTW, I also have an XTANT amp (a previous generation model - 3300ix) and love it.

In the manner you have wired your sorce units, the source selector gives you the options to a) listen to the EMPEG directly, b) listen to the CD player directly, or c) listen to the CD player through the EMPEG... is that right? With the EQ curves flat, do you hear a difference between (b) and (c)? (i.e. to rule out MP3 part of the EMPEG player, so the primary difference is with and without the EMPEG DSP in the audio path).

I was considering a source selector as an alternative to routing the signals through the EMPEG. I was just going to make one with a DPDT relay controlled by simple diode/transistor logic based on the remote outputs of the OEM CD and EMPEG. I didn't really want to add additional compoenents though, unless I had to. I will check into Sony's source selector as well, I was not aware anyone made something like that for car use.

Thanks for all the info.... I don't have my system info updated yet, but you can see what I had last season and early this season at my website if you want....

http://www.mrmunsell.com

under audio, then CRX audio. Of course, quite a few changes have been made since the page was last updaed. I have the same amp rack, with 1 XTANT now. All the Morel speakers are gone..... replaced with 3 way DEI components (woofer in door, mid/hi in kickpanel) - no more rear fill. No more DEX-P1R. The sub and sub-encloure is the same still. Just haven't had time tor revise my site lately. W

hen I came across a used EMPEG for a good price, I ditched all the PC-based MP3 stuff I was working on that is mentioned there now :).

Thanks for all your insight into this. Glad to meet someone else already competing with an EMPEG in their system.....

do you hear a difference between (b) and (c)?

Don't know yet... the car is still a work in progress. It is pretty much complete except for the custom subwoofer box to hold the two 10" Xtant subs. That box is being built using the large trim panel (that extends from the rear door and over the wheel arch back to the tailgate) as a base. I give up a 4" wide strip of my rear cargo floor between the back of the wheel arch and the tailgate to the subwoofers -- the only compromise to vehicle utility that I made. (Well, actually... if I get in an accident that triggers the fuel pump shutoff relay, I will have to remove the sub box (4 screws) to reset the relay, and I'll have to remove the sub box if I want to replace the rear shock absorbers.)

I figure it will take me literally months to tune and tweak the system to my satisfaction. As you saw in the diagram I emailed you, there is an enormous amount of tunability in it.

Here's a hint for anybody interested in seriously tuning a car stereo: you can't rush it. All you can do is listen for a few days, and finally decide that, "Yes, there is too much sub-bass and the treble really is too bright..." Then you make a small change, and then listen for a few days more. If you try to tune it for an hour at a time, at the end of the hour you will be able to convince yourself that anything you do sounds great. Then the next morning you turn it on, and it sounds terrible. YMMV, but I have found that after 20 minutes, I can no longer make even a semblance of objectivity.

tanstaafl.

"There Ain't No Such Thing As A Free Lunch"

Greetings!

you can't rush it

There is a really great ear training course that I have started (in my infinite spare time...). It is called Golden Ears, by Dave Moulton. It is a completely CD-Audio based ear training class designed for audio engineers and musicians. It teaches you how to identify types of effects, delays, etc., including EQ and boosts and their exact frequencies. This class recommends something similar - only listening for short periods of time, and coming back later.

Paul G.
SN# 090000587 (40GB Green)

I was considering a source selector as an alternative to routing the signals through the EMPEG. I was just going to make one with a DPDT relay controlled by simple diode/transistor logic based on the remote outputs of the OEM CD and EMPEG.

This is what I have done in my install, a large 10A DC DPDT relay with a few diodes and it works perfectly OK, with no noise that I can detect. Cheap, as well.

One of the few remaining Mk1 owners... #00015

Wow, that Golden Ears course looks awesome. I want it.

Wait, it's 120 dollars! I still want it, but can't have it right now. Grr.

___________
Tony Fabris

If you want, I can send you some additional information. The CDs are a lot of A/B comparisons of different forms, comparisons and examples of various effects, delays, EQ, etc. done with pink noise or music. They are really quite nice. I have not really gotten that far with them yet. I just do not have the time at the moment. If you are interested in more information, let me know.

Quote from manual:

How good can your ears get?

A couple of years ago, a colleague of mine, who likes to complain that he is going deaf in one ear and can't hear anything high in the other, knocked the proposed CBS Copycode scheme (which was allegedly inaudible) right out of the ballpark by identifying it 100% of the time in controlled double-blind tests conducted by the National Bureau of Standards. Once he astounded me by identifying, by brand and model, the side microphone I had used in a middle-side stereo recording I was playing for him, and noting (correctly, as I found out to my chagrin) that it was not functioning to spec! And yes, he has done these drills!


Paul G.
SN# 090000587 (40GB Green)

tfabris: Wait, it's 120 dollars! I still want it, but can't have it right now. Grr.

Uhhh, look again, Tony. $120 is only the first installment. $220 total price. Sigh....

pgrzelak: If you want, I can send you some additional information.

I would be very interested in additional information, particularly some kind of first-person review of the product -- how easy was it to use, how effective was it, was it worth the money, etc.

I have bookmarked that site, and I imagine that I will talk myself into the purchase in the next few days...

tanstaafl.





"There Ain't No Such Thing As A Free Lunch"