basic workflow clarification

[blue_max]

Having helped me to get my BFD earning it's keep - and it already makes a difference - could I just ask for a little workflow clarification.

Are these steps correct, or is there a better way? Or any detail that would improve it.

1. set up REW and get the auto filters setup. (I am guessing you would start to manually adjust them later on).
2. load the filters.
3. re-balance the sub (usually increase the sound level on the amp).
4. Re-test and fine tune until happy. Presumably carrying out 1-3 each time.

5. Test with main speakers and be prepared to start 1-4 again.

6. THEN consider adding a house curve.

Supplementary question. Can I use, say, channel 4 for movies and channel 5 for music (maybe without a house curve).

Gotta love this forum!

Graham

[brucek]

Can I use, say, channel 4 for movies and channel 5 for music

Sure. I use program 5 for 5.1 HT (movies) and program 2 for 2 channel stereo (music). Makes it easier to remember...

brucek

[blue_max]

Oh, ok, that makes life easier. Thanks for the suggestion. I don't want to lose sight of my workflow question as that is what I really need help with, but I have noticed that the rs newstyle analogue meter has rather different readings between the 70 setting and adding five, and the 80 setting subtracting five. Is it clear which is the (more) accurate?

Graham

[brucek]

70 setting and adding five

That's what I use to set up the 75dB for my ECM8000, but if you're using it to measure too, you should switch to the 80 scale, so the needle has lots of headroom when measuring. At that point you're using the RCA output and not the needle reading.

brucek

[Otto]

Hi Graham,

I'll try to offer my $0.02 here...

blue_max wrote:

1. set up REW and get the auto filters setup. (I am guessing you would start to manually adjust them later on).

Yep, take a measurement, hit "find peaks" and then "optimize pk gain and q" (something like that, I don't have it open right now...).

2. load the filters.
3. re-balance the sub (usually increase the sound level on the amp).
4. Re-test and fine tune until happy. Presumably carrying out 1-3 each time.

Pretty much. I don't think you will really need to re-set the level of your sub, though, until you are all done and convinced that you like it. I generally turn my mains back on after I get pretty close to the curve I want and make sure they integrate nicely. At this point, you may want to play with sub delay and levels.

5. Test with main speakers and be prepared to start 1-4 again.

You really shouldn't have to do too much iterating through 1-4 after turning on your mains, especially if you are crossing them over at the same frequency as the sub (a different method would be to run you mains full range and then cut your sub in on top of your mains, but most people will, for example, cross sub and mains at 80 Hz (or 60, 40, or whatever)). I understand that there can be some trouble at the crossover frequency as the main and sub signals recombine at your listening position, but I've never had an issues with it; mine is usually very nicely integrated from the get-go.

If you do find that you have a dip or hump at the crossover frequency, I'd first start playing a little with delay for your sub, especially if you have not before. It that doesn't work, you will have to adjust it with some more filtering through the BFD.

6. THEN consider adding a house curve.

This is step #1 if you want to use a house curve. Actually, this is step number #1, period. In general, REW calculates filter values based on a target curve that's defined by you. That target can be flat, or it can be a house curve, or it can be some wacky thing that no one would ever do. Whatever it is, you need to tell REW what it is before REW is able to calculate filters.

So, yeah, if you want a flat curve for stereo and a house curve for movies, you'll have to do two separate measurements, filter calculations, tweaks, etc.

I have also used preset "2" for stereo and preset "5" for movies. However, most recently, I've just been using a slight house curve for either, and I don't go back and forth anymore. Actually, last time I did this, I did the whole thing with the mains on. I wasn't able to use REW's automated filter calculating abilities, of course; I did everything by hand, tweaking, trial-and-error style. Sounds great, and the mains were integrated from the beginning. Surely not the most common method, but it worked for me.

You can go at this for hours and hours, and it's a lot of fun. Good luck!

[blue_max]

brucek wrote:

That's what I use to set up the 75dB for my ECM8000, but if you're using it to measure too, you should switch to the 80 scale, so the needle has lots of headroom when measuring. At that point you're using the RCA output and not the needle reading.

brucek

Thanks brucek, I understand your rationale. When not using the rca output and playing the pink noise from the amp, I would have expected the two methods to measure the same - does this just show how inaccurate the rs mic is or is there some other explanation?

Graham

[blue_max]

Thanks Otto, great to get your view and surely worth more than the $0.02!

Interesting perspective on the house curve. I always assumed you would go for flat and then apply the curve and set the filters. Doing it this way would easily allow the curve to be changed for experimentation. Aiming for your house curve from the off would mean quite a lot of work if you changed your mind about the curve. One to experiment with I think.

I was thinking that to help fill the dips, you would increaase the volume of the sub and then filter down more. Then when you measure next time, it's based on the new sub level. As this has the previous filters applied (or should you always turn the filters off to re-measure?), it would generate a different graph.

Just one other thought. Phase (0 or 180 on mine). Would you just alter this if you had issues, or is it best to work round it using 0 - maybe adjusting the delay to stop any interference with the mains. I don't really notice any difference with the phase either way.

It's all becoming much clearer!

Graham

[Otto]

blue_max wrote:

Thanks Otto, great to get your view and surely worth more than the $0.02!

No problem, and glad to help.

Interesting perspective on the house curve. I always assumed you would go for flat and then apply the curve and set the filters. Doing it this way would easily allow the curve to be changed for experimentation.

The problem here is that if you start with a flat curve, it's implied that you already have filters running. Then, if you come back and re-measure to go for a house curve, REW won't know that there are already filters engaged -- therefore, the new filters that it calculates for the house curve won't work properly because once they're loaded, the original ones are gone, and its starting point thereby changed.

Aiming for your house curve from the off would mean quite a lot of work if you changed your mind about the curve. One to experiment with I think.

Nah, it's not too bad. REW is very good about hitting the target, so you don't really need to iterate through the steps much (unless you want to ). Just load a new target, mesure, let REW calculate filters, load the filters, remeasure to ensure you're happy with the result. Sometimes I find myself doing a million passes and tests and remeasures, etc. I think I do it because it's fun for me, but I know in the back of my head that it's probably really not all that fruitful to spend that much time on it. In the end, I usually just have to set it, leave it and live with it for a while. Then, it might be wise to come back in a week or a month if there's something I don't like...

I was thinking that to help fill the dips, you would increaase the volume of the sub and then filter down more. Then when you measure next time, it's based on the new sub level. As this has the previous filters applied (or should you always turn the filters off to re-measure?), it would generate a different graph.

Yep, you don't want to apply much boost (many would say no boost, but I do a little sometimes). If your sub has the headroom, just crank it way up and quash down those peaks!

Your right -- once you start to remeasure with filters engaged, you get different results. REW currently has no way of knowing what filters you have so it just assumes that you are doing a filter-free (i.e., no processing in the BFD) measurement. If you allow it to calculate filters at that point, your response will be all screwed up if you use them. Usually, after the first pass where I've let REW do the calculation, the following tweaking is minor enough to just do it by hand. Once you do it a few times, you'll have it down. And although the graphs on the screen won't be 100% correct, I still find them to be a useful guide to getting me where I need to be.

Just one other thought. Phase (0 or 180 on mine). Would you just alter this if you had issues, or is it best to work round it using 0 - maybe adjusting the delay to stop any interference with the mains. I don't really notice any difference with the phase either way.

I don't have a phase control on my sub, so I haven't been able to play around with that. If I did, I'd probably start by doing two sweeps: one with the switch at 0 and one with it at 180. Those initial two sweeps would be with the mains on. Then I'd start doing my measurements and filter settings. Get that all loaded up and remeasure with the mains on with the phase switch in both positions again. Pick the one that looks the best around the crossover point. If you have any funky interaction, especially around the crossover point, you may be able to tweak it out a little with sub delay from your pre/pro/receiver. I haven't ever had any trouble around my crossover point...

It's all becoming much clearer!

Excellent. Have fun!

[blue_max]

Thanks Otto.

I think it may be wise to delete the filters and re-measure from what you are saying.

You are probably right about the phase - I will do that.

All my kit is in the lounge, which is also our bedroom (don't ask!). We have two kids too, so finding time to indulge in my hobby is none too easy. My computer in the lounge can't quite see my wireless network (it not a laptop), so that makes things even more difficult. I don't need fun - I need results fast!

It is interesting to actually SEE the results though. I hope this thread is helping others - I am learning so much from it.

Graham

[brucek]

When not using the rca output and playing the pink noise from the amp, I would have expected the two methods to measure the same - does this just show how inaccurate the rs mic is or is there some other explanation?

I tried it with my old analog meter and comparing between the 70 and 80 scale, I get a 1dB higher reading for 75dB pink noise and sine waves on the 80 scale. Not too bad...

brucek

[Malice]

Graham,

Did you get your Midi issue resolved? Just curious!

Bob

[blue_max]

Hello Bob,

Guess you missed the reference in my 'calibration confusion' thread. Yes thank you, you were dead right about reversing the cables.

Maybe I think about these things too much, but I seem to have bass in spades now. I have a house curve, but at serious levels, I am clenching my buttocks for the neighbours!

This whole endeavour has made a huge difference. It has really brought the bass to the fore. I may have been over zealous with the house curve (30 6.0), so maybe toning it down a touch may be preferable.

At least I know how to make the changes now, so thanks to all the help on the forum. You have all been very patient.

Hope I can be of help to the newer members, now I am a veteran!

Graham

[PeteD]

blue_max wrote:

Just one other thought. Phase (0 or 180 on mine). Would you just alter this if you had issues, or is it best to work round it using 0 - maybe adjusting the delay to stop any interference with the mains. I don't really notice any difference with the phase either way.

Remember that the setting for the distance of the sub from the listening position or the delay (if your receiver has it) is a phase adjustment. This does not have to be done at the sub for many people.

Pete

[blue_max]

PeteD wrote:

Remember that the setting for the distance of the sub from the listening position or the delay (if your receiver has it) is a phase adjustment. This does not have to be done at the sub for many people.

Pete

Thanks for that Pete. In fact, I just did a sweep with it both ways and it made quite a difference - much more than I could tell by listening. The sub is quite close to one seating position. Switching to 180 seemed to remove a deep trough I was getting.

When all the measuring is done, it still has to be down to your ears to be the final judge. It would be nice to just know it was as good as it could be and sit back and start to appreciate why it was. I am sure I will be tinkering again, but I need to live with it for a spell.

Graham

[Malice]

blue_max wrote:

I am clenching my buttocks for the neighbours!

Graham

Very friendly are they?!

[blue_max]

Malice wrote:

Very friendly are they?!

But not for long I suspect. Probably could have phrased it better.

Graham