Theoretical SQ question. Optical to Bit One vs. Pioneer P99

[pentoncm]

What would sound better between the Bit One and the Pioneer p99? The pioneer has just about as much tuning ability. I would think that if you were not to use the optical input of the bit one and just went with RCAs, the pioneer would have a leg up because there would be one less conversion in the chain between digital/analog. There aren't that many HU's on the market that even have optical outs, maybe they will be coming? I think the ultimate setup would be to use a deck with an optical out to a bit one to an audison Thesis amplifier. The Thesis line has optical connections so everything would stay digital until the power amp stage of the amplifier. I have no idea how much the top end Audison line costs, probably outrageous $.

Also, the new Alpine H800 has a optical in, hopefully more of there decks will have optical outs in the future. I think right now its limited to their nav units.

[MacLeod]

Not sure what youre asking.

If youre gonna use the Pioneer for processing then you dont need the Bit One. If youre going to use the Bit One for processing then running optical is the only way to go. Itll have zero noise and keep the signal digital til it gets to the Bit One. If you run RCA's from the head unit to the processor, youre gonna get a higher noise floor and some slight hiss at higher volumes.

[pentoncm]

I was asking, HU w/optical and Bit One vs. P99 by itself. I imagine that both would be fairly equivalent SQ wise. They both have 32 bit processors, and 24bit DAC's. The major advantage I see with the Bit One is the Butterworth/L-R filter types. The p99 manual does not specify but I'm assuming they are Butterworth filters as they go 6/12/18/24/30/36 and L-R filters are typically only even order.

[arun1963]

The bit one or any processor over the p-99 any day. While the p-99 gives you a lot of dsp, it does not let you set eq for each driver independently. That is a big + when tuning around the xover points. The eq on the p-99 is independent for L/R but summed for all drivers on either side. Let's assume you're crossing your mids and tweets at 2.5khz. How you set the 1.25-5khz on one side would effect both your mids and tweets. With a processor you can set this range separately for the mid and tweet. The other advantage of a processor is that it would give you a lot more xo options.

Again it depends on how far you want to go. For 99% of folks the p-99 would be more than enough though.

[pentoncm]

The bit one or any processor over the p-99 any day. While the p-99 gives you a lot of dsp, it does not let you set eq for each driver independently. That is a big + when tuning around the xover points. The eq on the p-99 is independent for L/R but summed for all drivers on either side. Let's assume you're crossing your mids and tweets at 2.5khz. How you set the 1.25-5khz on one side would effect both your mids and tweets. With a processor you can set this range separately for the mid and tweet. The other advantage of a processor is that it would give you a lot more xo options.

Thanks for pointing that out to me about the L/R being summed on either side. I did not pick up on that. I know from using the Bit Ten that there is a huge difference when Eqing around a crossover point. If you have a woofer that is in the door panel and a tweet that is aimed opposite headrest, the woofer response is decreasing most likely around 3 -4 kHz so eqing that section seems to dig into the blending and won't make that much of an impact for a response peak. On the other hand 3-4 kHz the tweeter can probably deliver perfectly 60 or more degrees off axis so its really strong and more of a contributor to the peak. Cutting on the tweeter makes a much bigger difference. The tough part is finding the balance, lets say you want to cut -4db out of 4kHz right at the crossover point. Where do you do it? I've had better luck cutting -.5 to -1db out of the woofer and -3.0 to -3.5db out of the tweeter. Sometimes you have to just listen to the woofer or just the tweeters to find out which way to go.

How do you feel about the Alpine H800? I like the idea of a 10 band parametric (in addition to 5 band that is for the sub) because there are less points of modification so you can generate a smoother curve.

Are most crossover systems built on butterworth? For instance HU's and processors when they don't specifiy? Butterworth is a steeper filter which is a benefit in a way, but it gives you a peak at the crossover point which i don't like. You could go with a higher slope L-R if you want a steeper slope but flat crossover response. And Butterworth just sounds muddier in the midbass section.

[arun1963]

With door mounted mids, a typical 6.5-6.75" driver will beam at ~2.5khz. The whole range from ~1.25-3khz will be hotter from the further but more on axis mid. So you would want to cut this range maybe 2-5db for the far side, depending on the frequency. Now assume you're running your tweet low, say 2khz. As you rightly mention, the tweets don't have this problem in this range. Its an overlap range with one set of drivers behaving very differently. A typical 1" tweet will beam around 12.5-13khz. With a summed eq you would have to sacrifice some amt of tonality to keep the image from skewing to one side. A processor would give you better tonality in this range for sure, better imaging too.

I look at two things when managing a crossover zone. How many octaves and driver dominance across this range. I want to keep it simple so I need this blend over the least number of octaves. I typically run all my drivers on steep slopes. So crossed at 2khz, with both mid and tweet at 24db/oct, I'm looking at a range of about 1.5 octaves with the xo point in the centre. Steep slopes also seem to give the best imaging. I also don't go for staggered or overlapping xo.

If my tweet is cut at 2khz, the low end of the body amd mass from the tweets is at ~ 1,25khz. This is what the tweet is pulling up as the low end. 1.25khz is set for the mid, 2khz is set partly for the mid and partly for tweet, 3.15khz is set for the tweet.
Listen to the mids and tweets separately. Clean up the snow from the tweets and hear the mids a bit dull in the upper mids. Then blend the two set of drivers. I hope this made sense.

[arun1963]

I have no hands on experience with the h-800, better ask Mac. Linkwitz vs Butterworth? With one the cut off starts at the xo point with the other it starts before the xo point, wherein the the xo point is about 2db down. That's akin to getting a steeper slope. Set it to what you prefer and get on with the tuning.

[pentoncm]

Both filters start cutting before the crossover point. Butterworth is defined as -3db at the crossover point and is steeper because of this. L-R is -6db at the crossover point and gives you a flat response.

Here are what my curves look like for the woofers

[arun1963]

Was looking at your setting. Curious how you arrived at these. Are these for mids and tweets or only the mids? Looks like you're crossing around 80hz.

[pentoncm]

Those curves are very preliminary. I am having more trouble out figuring out the near woofer. Those are the curves for the woofers only. The tweeters i am still working on. They are crossed at 60Hz on a 24db/ocatave L-R filter.

Is there anything that looks out of whack or anything? Any suggestions?

[pentoncm]

Just ran RTA with pinknoise using a droid and the AudioTool App. Put right next to woofer gives very flat response. Changes quite a bit when put at ear level.

Curve looks decently balanced, some things need shifted. Near woofer, 200 and 250 down more, 315 up, 500 to 800 down, 1.25 to 1.6 down, 4 khz down.

Far woofer looks alot better. 160 down,200 and 250 up, 315 down, 600 to 800 down, 1 and 1.25 down.

[pentoncm]

New updated EQ settings after I ran the EQ analysis. Definitely sounds much clear and near woofer is much closer to the far woofer in sound.


I checked out the Alpine H800, only Butterworth Filters.

[arun1963]

You can't reject the h800 just cause it has butterworth i/o linkwitz .

There's a big difference between speaker vs ear level response in a car, thanks to your environment. I posted graphs here in one of the threads. Measure at ear level one side at a time. I'm assuming that's what you did. I'm not sitting and listening in your car, so I can't say for sure but see if any of this helps.

1. Cut a bit more in the 80-100hz range. 80hz will make your mid bass boomy. You want it tight and snappy. 50-60hz is where you have the rumble and impact in mb. Cutting a bit at 80-100 would help bring this out and keep the mb tight. I would measure 160hz again. It's normally much hotter from the far side, but maybe it's different in your car. Measure 315hz again too.

2. Most cars will have a null around the 400-600 region. So you can raise this range relative to whats below. You have this range 1-4db's lower on the far side, just check that again. I remember that I had to measure many times and it was frustrating to come up with different numbers most times .

3. The range you want to cut the most is 1-4khz. Cut this range a bit more than you have it currently. Make sure ~ 1khz and 2.5khz is balanced for L/R. 1-2.5khz deep cuts and then start bringing back at 3-4khz. A flat response in this range generally helps.

4. If you're using the same eq for the tweets then cut at 6-8khz to reduce sibilance. Cutting 5khz reduces listening fatigue and pushes the stage a bit further away from you. 10& 12 khz is the shimmer on cymbals, 12khz is also 'air' If you roll off 16-20khz it will make the lower end more prominent. If you can run these flat the you know you have dialed in whats below properly.

More than anything else, 'hear' the sound and how it changes with the tweaking. Get a good reference level where you want your sound to be and then try to get the sound there.

[pentoncm]

Thanks for the help. The reason that I have not cut the 800, 100Hz regions is that they don't sound boomy to me. This is probably due to the Linkwitz filter, as you can see those areas have already been cut partially by the crossover. I am going to change my settings to Butterworth on the Left, and Linkwitz on the Right. The far side sounds like there is more bottom end anyway and it could be a way to balance it. Also, there is a dip in response around 3.2 kHz to 3.5kHz on my near side and its pretty significant. I'm going to move my tweeter crossover from 4000 to 3250 kHz and this will create a boost at the crossover point with the Butterworth.

[arun1963]

Well, GL with your tweaking. Everyone in this hobby goes from tuning with the brain do the tuning with the ears.