Har-Bal Users Forum BBS

OK, now you are insulting me and my business ethics too, I'm done with you, if you want to make a business of impersonal, automated and imperfect mastering, go ahead. But I am not going to sit here and read you insults and jabs at myself or my friends any more.

A little constructive criticism for you, maybe go to your local technical college and look into an "interpersonal communications" course...take it twice. You tend to get much better response from people if you know how to talk to them.

Hey Dan,

Thanks for the support but I think a little more calm on all sides is in order.

Though I beg to differ on "new users" interpretation for what is necessary and the reasoning for it, I haven't fealt insulted at all.

On reflection, I can say that in my first experiences with this technology (before I had the current solution that is) I was seduced by the promise of the automated approach. Only after sufficient time and sufficient number of applications to various sources, did the reality sink in. It's just like having a new toy. Initially everything is great until you get to know the weaknesses.

Hopefully we've averted a war of words!


Paavo.

Well for the mastering purists;
You know i never took any personal attack on you, you know that also.
But you just cannot take some of my suggestions and my explenation why, so you disagree... well if all you can tell me in your (until now) replies that you dont agree, give not a lot of help here.
I can reply also to everyone, and state that, but you never hurdled up why and how, you dont know how to do that eigther, you only get offended and then just write about that > and not the subject ?
You dont give us anything but your no's , no mind that.
I dont get offended at all, but i get this reaction with purists all time.

For ALL;
Wel you can only bee insulted if you are reading my post wrongly and build up angryness with it aside, i dont connect anything with it myself.

The Programmers;
But every move you make Can Be Automated and be done by a computer, specially if it comes 2 a computer.
I am a programmer also. and i know, companies spend billions around the world just to automate things for their 24H biss.
Once somebody told me that, i should look for automation inse humans, videotape them while doing things, and find out what they are repeating.
That is automation, if you can solve the equation.
I just hope we arise from the Analog thinking world toward Digital smart programming and clear automated sound, with out the discussion what is best for me or you, just program it all what you can, every dumb or smart thing has a function, as long as it needs repeating, there is a necessety for it.... if you ever think that you would like to prefend users from harming by keeping thing out of your program.
Still program it, but make a preference option of it, so that the user will have to enable it first.


And yes i posted a new suggestion;
Under Mr.Victory next post...
So be aware or dare to read it, hahaha

Have a nice day

Sorry Paavo, but his tone and accusing me of wanting to jab my clients for unnecessary hours of work I find to be a personal insult to my business ethics. Add to that he is accusing you and Earle of not knowing your business, when I know that you do, well, I find it offensive and insulting...and just rude and a poor way to communicate.

That said, I will just stay out of his threads, as I have no desire to argue in your forum, but I do have a desire to stand up for myself and Har-Bal when being insulted.

I am no "purist", but I do believe in giving my clients the best product I can, every time. As far as I am concerned, Har-Bal automtes as much of the process as I personally feel comfortable automating.

I read in your post you would not comment any more...

What greater thing is there for two human souls that to feel that they are joined... to strengthen each other... to be at one with each other in silent unspeakable memories.
~ George Eliot

To Dan:

Appologies accepted and not required. I'd just rather things don't degenerate into a slanging much and I perfectly well understand where you're coming from.

To "new user":

If you re-read my posts you will see that I did give reasons as to why I think the simplistic implementation of "matching" is a bad idea. It has nothing to do with a "purist" debate and everything to do with theory. Simply put, there is no such thing as a "one size fits all" spectral balance for all tracks that have ever been recorder.

All excellent recordings have "similar" characteristics but they are rarely if ever "exactly" the same. I give you a simple experiment to prove my point. Find two recordings that you consider excellent (preferably from different CD's) and analyze each in turn with HarBal. Then use one as a reference and then design a HarBal filter for the other to give an exact match. Now listen to the HarBalize version compared to the original. Chances are it will sound inferior to the original, because the wrong processing was applied to it.

To do the processing correctly requires "intelligence" and programming "intelligence" is generally not easy to do. When we have time we will work on an automated approach but we have more important things to do first.

In as much as I do not want to see the "simplistic" approach to matching included in HarBal, you can brand as a "purist" approach. But in reality a purist approach would be no EQ at all. My position in this regard is that I do not want the name HarBal associated with inferior equalisation. If you have a feature like that users automatically expect that the result of using it should be superior and if it isn't then they conclude that HarBal is inferior.
This is not speculation. We know this from experience! We used to have a "quick start manual" in which users mis-read it as saying they should do exact matching to the reference. Many of those that did were dissappointed and requested refunds because they didn't use the product correctly.

Since dropping that and forcing users to actually read the manual the refund request rate has reduced coonsiderably.

That is my opinion and as the developer of this product that is the way we will go forward. My appologies if you feel dissappointed. By the way, just cos every other EQ plug product has what you are suggesting does not make it right. If correctness is implied by pre-existence then HarBal shouldn't exist at all.

Regards,


Paavo.

People who think they might get offended - Dont read this further.
----------------------------------------------------------------------------

Ok i agree it is your program, hoping that other plugin manufacturers are catching on to this.

But this means your target is purely directed towards a speacial users base, the ones who actually will know about music and mastering. Thats why i use purist words, cause your userbase would be expanded towards video-users and unexpirienced users also (the non-purists) , if you did.
I disagree with it cause it would be very easy to program and could not take you more time than a couple of hours
Until nobody does, the advantage goes to the purists...
And the non-purists loose out on it until time comes to save them.
I think the userbase would be 10x or 100x bigger with harbal matching, cause every beatmaking human would use it for sure!, so i dont think its conna be a long wait for them, the recording industry is opening up in a fast matter... so its harbal or someone else who will get away with it.
But surely you will know what i mean by now that harbal directs a purist userbase only.
Unpure = Humans / Musicians dont case about EQ, they want to play!

Thay way i see it, its a matter of time that other manufacturers are catching up..... it will be either habal or some else who does this feature.
I guess ill wait and see.

Thats my op.

Thank your for your reply and as i programmer and sound lover, i agree with you, its your program and i buy it...

New user wrote:People who think they might get offended - Dont read this further.
----------------------------------------------------------------------------

Ok i agree it is your program, hoping that other plugin manufacturers are catching on to this.

But this means your target is purely directed towards a speacial users base, the ones who actually will know about music and mastering. Thats why i use purist words, cause your userbase would be expanded towards video-users and unexpirienced users also (the non-purists) , if you did.
I disagree with it cause it would be very easy to program and could not take you more time than a couple of hours
Until nobody does, the advantage goes to the purists...
And the non-purists loose out on it until time comes to save them.
I think the userbase would be 10x or 100x bigger with harbal matching, cause every beatmaking human would use it for sure!, so i dont think its conna be a long wait for them, the recording industry is opening up in a fast matter... so its harbal or someone else who will get away with it.
But surely you will know what i mean by now that harbal directs a purist userbase only.
Unpure = Humans / Musicians dont case about EQ, they want to play!

Thay way i see it, its a matter of time that other manufacturers are catching up..... it will be either habal or some else who does this feature.
I guess ill wait and see.

Thats my op.

Thank your for your reply and as i programmer and sound lover, i agree with you, its your program and i buy it...

Our friend

Let me step in here for a minute.

What you are advocating is something that will compromise the integrity of our company. We have created and presented a tool that not only works but teaches the end user as well.

Most of our customers (and there are many) can now intelligently articulate what is going on with their tracks because they have a greater understanding of the sprectral content of a song.

We never want to remove humans from the equation, especially when it comes to music. This forum serves as a thinktank for those individuals truly concerned about mastering the craft of eq'ing.

We agree that you can't just simply load the application and start working effectively without reading the manual. But if that were the case, then everything we have worked for up to this point would have been in vain.

We are not interested in selling a product that everyone will come running to purchase and them not gaining a better understanding of EQ'ing.

We can go to bed peacefully at night knowing that we are doing our part to REALLY help those individuals who have a real thirst to learn.

There can be no substitute for real knowledge. When I read the posts in this forum I can feel the excitement from those individuals who have gained real incite.

With your knowledge you should be contributing and answering questions for those engineers who are up and coming.

How can you advance without true understanding?

That is why my friend we are here 24hrs a day 7 days a week.

Peace

Earle

We can go to bed peacefully at night knowing that we are doing our part to REALLY help those individuals who have a real thirst to learn.

There can be no substitute for real knowledge. When I read the posts in this forum I can feel the excitement from those individuals who have gained real incite.

Can I get an AMEN!

Well stated, Earle.

more like 3 days .... and 4 days off....
Well thats about it!, i just read this now. last week i got 2 days off extra, so i dont know what you are saying? Oh that thingy about sound ?
I dont work like that, i just see now in har-bal if it fits the ref....
Just like a car stearing and voila, it sounds fine with me.
Thats why i started to see if i could make 2 filterfiles come closer, and made a program to adjust the filter towards matching.

Its no prob, Izone works better with matching the functions are there, just the EQ is not so tight and the refs are less actual. I just use both harbal and ozone in conjuction to match, it takes me 2x to do this accurate, 1 test run and 1 adjustment, cause the eq from Izotope is not the same, i use harbal and have to switch back one time.
And voila, a complete match! If i import it back in harbal, i can see its tight to the ref... so that one is solved.
I am now testing firium-eq and that one works better with matching.
So its one step down, just 2 steps and a complete match occurs.
Never could do that so accurately like now, before i needed a lot of EQ and brainstorming. So harbal rules in EQ adjustments better.
This takes no less than 5 minutes now and boosts a song to oblivion, and then a few minor adjustments and voila...
Doesnt matter if its 3 hours of videotape or sound!
Like i said i needed this kind of matching, even if you dont.

btw
I might even take a look at the filter file and work around over there, it would mean a one step match for me, but i have to write a program for that one, hmmm maybe next week. Hope its written in plain text or ascii that might speed up the proggramming a bit.

Just have to put trough some videotape more and where done.

No more worries over here; thx 2 myself, i can rest my head.
The first 2 step real matching is here, and its here 2 stay.

Hears You Laterz..

You say that Izotope works better but sounds worse. I argue that part of that comes from doing exact matching. Exact matching can work well for frequencies below 1kHz but above that you can get into trouble quite easily.

If you are interested in writing an exact match utility that is fine by me. Send me an email and I can email you back some code to read the file and some explanation of what is in it.

Regards,


Paavo.

Yes , i did explore the filter settings file of harbal, but i cannot actually say its working , thats why i use PAS EQ STUDIO presets instead.
I would be nice if i would know the structure of the fileformat of the FILTER ANL files, or can read or write then. If i could accomplish that i could use harbal as the prefered EQ instead of PAS EQ STUDIO.
The GENERAL EQ LINE i found with my program could then be exported to harbal and be seen as a ref or filter. would be nice!
I you could share this info it would be very much appreciated over here, and when finishing up the program in the next month or so i could send you my programme to you, and you can see for yourselve if it works or not.....

Thx anyway.

SINED wrote:...
I would be nice if i would know the structure of the fileformat of the FILTER ANL files, or can read or write then. ...

As I said in a previous post that you didn't respond to, if you send us an email then I can send you the C++ source code used to read and write .fil and .anl files.

Regards,


Paavo.

Paavo wrote:... Exact matching can work well for frequencies below 1kHz but above that you can get into trouble quite easily.

Oh boy...what a thread

Well, folks can get over-enthusiastic for full matching automation, here's my 2 cents...

I think it's true that most every thing a human does, every key-stroke or mouse input can be automated. What can't be easily automated are the many variables and decisions leading a human to make the keystoke or move the mouse.In other words when the feedback loop between action & re-action involves human sensory inputs, artistic decisions, contextual considerations - and above all knowing when to break a rule - that is the tricky part. That's why we humans get the big bucks !

Concerning other tools that assist in or do full automated matching...I have Har-Bal, Izotope Ozone3, Firium, CurveEQ, Magix ACL2005, and GlissEQ2.

IMO a fair part of the spectrum that has a high success rate for doing a visual match is a potion of the mids. I can usually count on a major part of the region above 100Hz and below 4KHz to match my visual curves. For example if I use a tool like Har-Bal or GlissEQ I can view an average representation of my wave as well as superimpose a commercial recording I want to match or even superimpose pink noise which approximates a lot of the general rock music slopes (will be a mostly flat line or curve in Har-Bal). That gives me a visual curve to pull my curve into, like a grease pencil on an oscilloscope - but if my recording doesn't have enough HF content I just can't push up the 4KHz region using a shelf...it sounds terrible to the ear. Automation wouldn't know that.

Same with using an automated matching tool - they match the mids fairly well but the bass and HF, ugh - you have to match by hand every time if uou have a good enough monitoring situation to even hear what you're doing in the bass or highs. Point being it's fairly common for audio to have a fairly good balance of mids that is well suited to be balanced by matching - it is also common for most folks to have fairly good monitoring so they can hear the mid balance well.

But all that's just for matching broad strokes, one of the other features of Har-Bal is to be able to push down on resonances. How would automation be able to decide where I want the kick and bass EQ relationship to be ? Very tough I think especially since you have to decide do I want them both to balance, do I want the kick louder, do I want the bass louder, etc....

Here's another 'resonance' type example in the highs - save Har-Bal shows you a peak around 10KHz or 12KHz - do you push on it ? Maybe it's a bad vocal ess, maybe it's a good vocal ess. You have to listen to make a decision, pulling it out could either fix a vocal or destroy it.

My bottom line is to think that an automated matching tool could do pretty good in the mids and at best give you a couple of options for matching bass and highs - you still have to listen and make decisions and readjust curves.

I'm a Har-Bal fan BTW - maybe you can't tell - it is a very cool tool