Compression Uncompressed

[SouthSIDE Glen]

I don't know if it's a user error, but your file won't work in Firefox (1.5.0.3), for me.

Eraos, we in fact did test it in FF 1.5.0.3 on two different OSs, and it worked fine both times. You might want to try re-installing it.

G.

 

[fraserhutch]

Thanks for your work on this, Glen.

 

[SouthSIDE Glen]

I don't know if it will confuse the intended audience, but you may also want to touch on upward compression.

It would also be cool to have some audio examples.

I did have to arbitrarily pick a cut-off point for topic materal; I could have easily written another 42 pages. But I tried to keep in mind that this was an introductory article and hitting someone who is still learining compression ratios with more advanced stuff was somthing I was trying to avoid.

I'm not sure just which thing you mean when you say "upward compression", though. Give me a quick kick in the noggen as to where your at with that. Maybe there's something basic I've neglected to include that could go into an update.

And yeah, I struggled back and forth with the audio examples idea for a long time, even as I was writing it. It honestly came down to a server bandwidth issue more than anything for the time being. If this thing is popular enough, the next edition - which will probably be on a new server account by that time - might be expanded to include some audio A/B stuff.

G.

 

[masteringhouse]

I'm not sure just which thing you mean when you say "upward compression", though. Give me a quick kick in the noggen as to where your at with that. Maybe there's something basic I've neglected to include that could go into an update.

Upward compression is essentially the opposite of downward compression (what you are describing in the article). It takes audio below the threshold value and raises it instead of lowering it by the given ratio. Low passages are raised while leaving the transients unaltered. It's good for "beefing up" mixes especially drums.

 

[kennycarvajal]

Thank you!

Very good stuff Glen.
Thank you!

Regards,
Kenny

 

[SouthSIDE Glen]

Upward compression is essentially the opposite of downward compression (what you are describing in the article). It takes audio below the threshold value and raises it instead of lowering it by the given ratio. Low passages are raised while leaving the transients unaltered. It's good for "beefing up" mixes especially drums.

I have to admit that that is a new one on me. I'm not sure how that differs from applying normal/downward compression and then applying just enough makup gain to bring the peak levels back to their input values? (Which is somethng I do cover in the text.)

Also, which compressors have upward compression capabilities; i.e. where the threshold basically means the reverse of what it does in standard compressors?

EDIT: Oh, wait, if you're talking something like the kind of thing Neodynium does (or whatver Roger Nichols is calling it now) that is a graphic multi-level compressor where you can "squeeze" the different layers of dynamics from the bottom or from the top, then I know what you're saying. I just figured that was probably more for the "Compression 221" class than it was for the "Compression 114" class .

G.

 

[michaelst]

Glen...

Excellent reading. Well written and informative.

I'm having trouble with the 'picture' windows opening to a set size without regard to resolution. In other words, some of the windows cut off infomation at the bottom.

I'm running 1280x1024, 32bit.

 

[SouthSIDE Glen]

Glen...

Excellent reading. Well written and informative.

I'm having trouble with the 'picture' windows opening to a set size without regard to resolution. In other words, some of the windows cut off infomation at the bottom.

I'm running 1280x1024, 32bit.

Thanks for you kind words.

Screen resolution in and of itself should not be an issue. What might be happening is that your browser may be set to adjust font sizes from the default on the browser end.

If you're running a recent version of IE as your browser, pull down the "View" menu and click on "Text Size". Make sure that is set to "Medium". If it is already, then try setting it temporarly to "Smaller". You may have to re-boot after that for the setting to actually take effect.

If you're running Firefox or Netscape as your browser, the menu selections and procedure are the same, but the actual setting you would need in the "Text Size" menu is called "Normal" instead of "Medium", and "Decrease" instead of "Smaller".

If you don't want to chage those settings, or they don't seem to work for you for some reason, you can always work around it a bit and point your browser window (or a new browser window) to the "\content" folder in the notebook's home folder and pull up each figure individually as a normal browser web page instead of a pre-sized pop-up window.

HTH,

G.

 

[Tonio]

Very nice work Glen. Great beginner, and intermediate write up. with some dabble of good pro level goodies!!
I haven';t read the whole thing but so far so good!!

Are you concerned about typos? I found one@ the opto section description about opto's being show(instead of slow) than VCA's.
Not a nit pick, Just wanted to note that in case this coiuld be of some importance to you.

Otherwise great stuff.

T

 

[SonicAlbert]

I'm not saying you're wrong or anything like that, Al; if you're having a hard time reading it, you're having a hard time reading it. But I have to be honest and say that this is the first time I have received such a mention on that color scheme. It does make me wonder if perhaps the gamma calibration on your monitor might be mis-aligned?

I'm having a hard time reading it. Actually, I'm not reading it at all because it kills my eyes after a very short time. Which I find unfortunate because you put a lot of time into it and I'd like to read what you have to say. My monitor is calibrated properly (Apple Cinema display).

There's a book you should read called "Type & Layout", by Colin Wheildon. It's an absolutely brilliant book in my opinion, and quite unique. After studying the concepts put forth in "Type & Layout", I highly doubt you will ever use colored text on black background again.

 

[SouthSIDE Glen]

Are you concerned about typos? I found one@ the opto section description about opto's being show(instead of slow) than VCA's.
Not a nit pick, Just wanted to note that in case this coiuld be of some importance to you.

I think that one has already been reported earlier in thiis thread by Iron Flippy, but yeah, thanks for pointing it out. Sure I'm interested in heaing of typos so I can correct them. If that is the only one typo out of 35,000+ words in there, I'll be a happy man. But if theres a few more, please fell free to let me know.

G.

 

[SouthSIDE Glen]

I'm having a hard time reading it. Actually, I'm not reading it at all because it kills my eyes after a very short time. Which I find unfortunate because you put a lot of time into it and I'd like to read what you have to say. My monitor is calibrated properly (Apple Cinema display

There's a book you should read called "Type & Layout", by Colin Wheildon. It's an absolutely brilliant book in my opinion, and quite unique. After studying the concepts put forth in "Type & Layout", I highly doubt you will ever use colored text on black background again.

I know that light on black went out of style with the Apple/Windows GUI. For me, light-colored backgrounds are far more of an eye strain than a well-selected light on black. I know I am kind of the oddball with the old-school black background (don't forget, we went 40 years with black backgrounds on our computer terminals with no issues), but there is more than that going on here.

I have read more books on the subject over the past 12 years than I could possibly count, and I don't choose my color schemes arbitrarily or just because they look pretty. There are technical and ergonomic reasons behind each specific color chosen (I have been at this racket for quite a while.) I gotta be honest, Al. I'm picking a robust, bold font there (Arial 10pt bold, with Verdana 10pt Bold as the backup) at 75% saturation in 2 of the 3 component colors. If you are having trouble reading that, there is a problem somewhere beyond the page design itself. If it's too low contrast for you to read mostly comfortably, I'd next ask if maybe your copy of the OS is missing the proper font to represent/replace Arial Bold. I could see if you were not getting a bold font that even on the best of CRTs or LCDs that it'd be tough to read. That would make sense. But if you're getting a bold font and you can't read it because the contrast is simply too low, then I'd highly suspect a display problem.

If I boost that color any brighter on any of my three monitors here (2 tube, one LCD, one tube with dual brightness settings) it's almost like shooting a green laser into the eyeball, it actually hurts after a a sentence or two. Remember that although we may have anywhere between 32,000 and scores of millions of colors available to us (depending upon our video cards), the web browser standars themselves limit us to only ~216+ colors on our font selection pallate. The next step up in the browser pallate is just waaaay too bright.

Here's a test. I have attached a screenshot of the page as it appears on my system. This will do two things: first it will allow you to compare fonts. If your Mac is missing or mis-choosing Arial Bold or it's Mac bold sans serif equivelant, this comparison may show it. Second, your browser should represent the jpeg screenshot fairly faithfully as far as color depth and contrast, not limited to the couple of hundred of colors in font display. This might give you a better look as to how the font brightness and contrast should come out. If the screenshot still loks dark and un-contrasty on your screen, then I'd have to honestly say the problem is somewhere on your side of the firewall, unfortunately.

The image is an unretouched screen capture using the Windows inherent print screen clipboard and Adobe Photoshop. Other than cropping the image borders to a smaller area and using some light JPEG compression to fit HRs 64K bandwidth restriction, I have made absolutely no adjustments to the captured image. What you see is what I got.

G.

[EDIT:] You can print out the text if you wish by just selecting "print" from the main content frame. It will print out as black on white.

Then again, Al, I seriously doubt that there's anything I've written that you don't already know. You know your audio stuff, and what I have here is fairly basic. But I can understand your desire to read it anyway.

 

[SonicAlbert]

Southside, that's exactly how it looks to me. I'm 100% sure I'm seeing the pages exactly as you intended them to be seen. Really, really take a look at "Type & Layout", it's a very interesting book purely aside from how it can help you get your message across in print.

Incidentally, I did my own web site in light text on black background as well. If you look at my resume page, you'll see that the colors are in a similar vein to yours except they are brighter.

http://misterpotts.com/resume.html

While I do like the way it looks from a visual point of view, I'll never do it again. I've also had a few complaints about the brown text on black background used on other pages, and I'll certainly be fixing that as well.

 

[masteringhouse]

IEDIT: Oh, wait, if you're talking something like the kind of thing Neodynium does (or whatver Roger Nichols is calling it now) that is a graphic multi-level compressor where you can "squeeze" the different layers of dynamics from the bottom or from the top, then I know what you're saying. I just figured that was probably more for the "Compression 221" class than it was for the "Compression 114" class .

G.

Yeah I kinda figured it wasn't intro stuff, but as a note I would possibly mention that compresssion doesn't always mean a reduction in dynamics though that's what it usually used for. There's also compression techniques like de-essing and ducking that you might want to tease people with for the Compression II (the sequel) article.

Keep up the great work G.

BTW, many people equate upward compression with parallel compression (as in BKs book), or upward expansion. IMHO they are different, but I don't define the terms.

Here's a good little discussion on the topic:
http://www.chrismilne.com/uadforums/viewtopic.php?t=4631&

 

[Dogman]

I have just started the article, but wanted to thank you Glen. So far, some good, and interesting info.
Ed

 

[SouthSIDE Glen]

There's also compression techniques like de-essing and ducking

It's funny that you mention that. In one draft I did throw a little tease in mentioning side-chaining and ducking in passing as stuff that wouldn't be covered in this introductory treatment. I don't remember for sure off-hand, but I don't think that passage made it to the final draft.

I do however touch on one particular de-essing technique using an MBC in the MBC section. But I don't really cover de-essing in its full glory here.

The main idea here was to try to cover with realtively full explaination the basic questions we always get in this forum like "What's the best compression ratio?", "What about RMS levels?", "Which compressor type is better?", "How do I tweak compressor setting other than presets?", "Should I EQ before or after I compress?" and "Should I compress during tracking, mixing or mastering?" You know, the usual suspects . There is no real black belt stuff in here.

But I do remain curious, Tom, how you feel about my original question: what is the difference between "upward compression" and "downward compression with makup gain"? I might be missing something, or maybe it's some subtle coloration cause-and-effect difference, but on paper (for wtaht that's worth) it would seem to me that they are at the end of it all really performing the same function.

G.

 

[SouthSIDE Glen]

Southside, that's exactly how it looks to me. I'm 100% sure I'm seeing the pages exactly as you intended them to be seen. Really, really take a look at "Type & Layout", it's a very interesting book purely aside from how it can help you get your message across in print.

Incidentally, I did my own web site in light text on black background as well. If you look at my resume page, you'll see that the colors are in a similar vein to yours except they are brighter.

http://misterpotts.com/resume.html

While I do like the way it looks from a visual point of view, I'll never do it again. I've also had a few complaints about the brown text on black background used on other pages, and I'll certainly be fixing that as well.

I've visited your very nice website before (that's where I picked up the nickname "Albie" ). I wasn't going to mention anything to you here about the fact that you had used a black background, as I though that would have too "campaign politician-ish" .

I guess it's one of those things that depends a lot on personal tastes, and maybe even personal eyesight, much like the choice of nearfield monitors can be for the ears. For me, the teal-on-black is very comfortable and non-fatiguing. Combined with the standard formulas; that 0-192-192 is the brightness equivalent of 50% monochrome saturation (0% + 75% +75% = 150/300, or 50%) means that there should - for most eyes - be plenty of brightness and contrast there.

Plus there's the fact that an old fogey like myself grew up with blackboards instead of today's whiteboards and with alphanumeric computer monitors (monochorme and color) ind not graphic GUIs; black backgrounds just kind of come natural to me. On the other hand, I cannot read long texts on the average Internet page with a white or "tool tip yellow" background for very long because the glare of the background gives me eye strain headaches before very long. This is even true in Word. As I was writing this article in Word, I'd often have to break to give my eyes a rest from the glare, or if I was on a roll, actually put sunglasses on if I wanted to keep typing .

I will admit that there is a LOT of misue and abuse of black backgrounds, they can be easy to mis-design. Ive seen people use 25% brightness saturation (0-128-128 in any order) or even 50% (128-128-128 gray, for example) and non-bold fonts on black backgrounds, and they are almost impossible to read. On the other end of the spectrum (pun intended ) there are those whose pages look like Haloween candy with orange or full-on yellow (255-255-0 or greater) text on black. Those I usually refuse to read for more than a sentence.

I'll certainly consider changing the pallate somehow in the next edition, Al. I don;t take your opinion lightly. But frankly, I'm to ready to get my time back for other projects to want to spend any more time revamping this particular edition.

And next time I head up to Borders for a coffee and a Mix magazine, I will definitely look up the book you recommend too. Sounds interesting.

G.

 

[JeffLancaster]

For what it's worth, I found the color scheme to be very appealing, and had no trouble reading the text. But everyones different...

Glen, I read the entire article in chunks over the course of the last couple of days. Really enjoyed it and I think it helped to clarify a few points for me. I'd love to see an "advanced edition" follow-up to this some day. Thanks again.

 

[lurgan liar]

I love this article Glen....ALl we need now is someone to do an EQ one !!!

 

[SouthSIDE Glen]

My thanks to everybody who has responded so positively to the compression notebook over the past week, both in public and in private. I am gald it has been so useful - or at least entertaining - for all you folks. If it can preemptively answwer a few questions from someone, it's worth it.

So far only the one typo reported earlier in this thread has been reported and confirmed. If there are others, please don't hesitate to let me know. Rep points for every successful find .

Also, regarding the post earlier about someone not being able to load images when using Firefox 1.5.x, my beta tester has reported to me that he got wind of some situation with an on-line game where the same thing was happening with just that version of browser. There's some obscure data security setting when dealing with local files somewhere deep in Firefox's preferences settings that was causing the problem, and my tester is pretty confident that the same thing may be happening here. When he gets back from his Memorial Day holiday, he'll track down that info and pass it along to me, and I'll post the solution here.

G.