Hmmm... I'm at a loss to understand how burning speed would impact on quality of the resulting media created. I recently switched to using a DVD burner that runs at 2.4x DVD scale or roughly 25 Mbit/sec which is way faster then the 1x CD speed mentioned. I assure you the quality of both video and audio are excellent at that burning speed. Before that I used various CD burners running anywhere from 2x to 12x. (CD scale) No problems. I shutter to think how long it would take to burn a full 4.7GB DVD disc burned at 1x on the CD scale, if my DVD burner would let me, I don't think it even would drop down that slow.

Naughtybird, not trying to be argumentative, on the surface it just don't seem to add up that slower burn speeds result in higher quality output. Can you expand on what you said and enlight us on what you base your comments on?

I burn audio CDs at 4x even though i have 16x and 24x burners. The reason has nothing to do with quality. I've compared 4x and 24x burns in several CD players, and ripped the tracks back to .wav files for comparison, and there is no difference at all. They appear to be bit for bit identical.

The reason for using 4x is that many cheaper players have trouble reading audio CDs burned at faster than 4x. The CD player in my car skips badly on an 8x burn, and won't even recognize a 16x. When i was burning 16x recordings of our church services, many people borrowing the CDs complained of skipping or "NO DISC" errors. Others played them fine with no troubles at all. When i switched back to 4x everyone could play them.

Chienworks, you are right about "play-compatibilty" of audio-CDs burnt at higher speeds. I never had trouble with audio-CDs burnt at 1*, but I had sometimes with CDs burnt at 4* and higher (which I just did for preview).

BillyBoy, I can´t enlighten that. It is pure fact that audio-CDs sound slightly different if you look at burning speed and manufacturer brand. Every audio-pro knows that, it has often been discussed on nuendo.com, but I found that out by myself much earlier. Some brands sound dull, some really crisp. And burning at higher speed rates ends up in a sound lacking "body", although everything seems to be there. I´m with you, there are just pits in the CD-ROM, but the sound difference is there. But please remember that my comment was only relating to audio-CDs, I have no experience with DVDs burnt at higher speed rates, neither with audio nor with video. Up to now I have no DVD burner.

If I were you I would check this one time with critical material. For what I and others found out up to now I would use 2* and 4* even for DVD only for preview, not for a "real" master. But perhaps DVD players behave different than just Audio-CD-players. I will have an eye on this.

Naughtybird

>>>Some brands sound dull, some really crisp. And burning at higher speed rates ends up in a sound lacking "body", although everything seems to be there. I´m with you, there are just pits in the CD-ROM, but the sound difference is there.

at first glance this seemed like insanity because the 1's and 0's should be identical no matter what CD-R brand or speed is used. I did some searching and found a page that has a viable explanation for the difference in sound quality. it appears to be simply a difference in error rates. since an Audio CD (unlike a Data CD) can be read with errors in the data stream, it left is up to the D/A converters to deal with the error correction, which would account for a difference in sound. here is a quote from the page:

"There are many factors involved here, such as what type of burner the CD-R was created on? What speed was it burned at? What type of software used to burn the CD-R? Last, but not least, what type/brand of CD-R is being used? All of these things combine to create a certain amount of errors, or patterns of errors, in the data stream, and in my humble opinion, that is what people hear as the differences between CD-R’s. Some people say gold CD-R’s are “sweeter” or “warmer”, some say the green ones are more “brittle”."

http://www.village-buzz.com/ArticleBody.asp?Article=80

I don't know if it's correct but I can't think of anything else that would make sense. in the digital world, the data is either there, or it isn't. if there are gaps in the data even if it's just one bit, the data is invalid. so the differences in sound quality between CD-R brands are differences in how many of these gaps occur (or how *invalid* the data is) which would depend on the quality of the player as well. Burning at lower speeds ensures the data will end up as error free as possible (as read by the CD player).

interesting food for thought.

Thanks a lot cheesehole, also for the link.

I already heared explantions like that, and others too.

Everybody working "seriously" in the digital media world learns sooner or later that it is a whole new universe. Analog had it´s rules, and so has digital, but we still have to learn the phenomenons of this new medium. And it is still new, although the CD is over 10 years old. We are just reaching the finer art of treating audio and video.
Someone else found out that he got a much punchier sound by just resampling a 16 Bit / 44.1 KHz audio signal to 32 Bit(Float) / 96 KHZ. He did this conversions in different combinations and at last he got a new signal with much more body at 16 Bit / 44.1 KHz.
Did anybody tried this with video? Exporting a DV signal as a series of uncompressed bitmaps or avi, importing this again, rendering this again to DV or mpeg2 to a new master file, and importing this again?
I would like to hear something about this.

Naughtybird

Interesting topic for sure, but I think we're more in the realm of subjective rather then objective and reproducable results given same source material, burning device, media, and ultimately the same playing device. While some may think they can detect differences between a recording on gold verses green CD's, (who knows, maybe they can) has anyone bothered to do lab tests and analysis the wave patterns to see if they are perceptible differences? To take it to the nth degree surely a recording when listened to will sound "different" depending on the acoustics of the room, then on to how many people were in the room which may effect the acoustics, even to climb out on a limb maybe some listners had a bad head cold and their normal hearing wasn't what it normally is. Where's the end?

I don't know how loose standards are for a A/D converter to pass a data stream and accept X numbers of errors which may result in a noticable pitch shift or other loss of overall quality or what process is being altered that dropping down a burn speed results in some device like a car CD player suddenly being able to read a disc it could not before burned at a higher burn rate. I guess a slower burn speed by result in less errors being passed but if so why haven't I seen any of the major media remanufacturer place warning labels on their blank media saying for best results do not exceed such and such burning speed?

Red Book CD digital audio has a couple layers of error detection and correction, and all errors are rated by what level they occured at, and whether they were correctable or not. The uncorrectable errors get replaced with interpolated data, which hopefully you don't hear as a glitch.

Interestingly, the lowest level errors, due to misreading of the pits, occur ALL THE TIME, usually at a low rate, and are nearly always corrected, except in cases where they also coincide with other problems, like dust or scratches.

The tolerance of the pit lengths gets worse at higher burn rates, as does the centering within the 'groove', introducing more of these low-level errors. Therefore, this increases the chances of read errors.

The high-end CD error scanning devices will print a picture of the error rate across the disc, as a disc-shaped image. Although you can see every scratch and piece of dust, the whole disc is just covered with dots indicating little errors that get corrected.

Final masters, stuff you really care about, etc., should always be done at low speeds to keep the pit tolerances accurate. That's just my opinion. I'm not a CD expert.

///d@

>>>Someone else found out that he got a much punchier sound by just resampling a 16 Bit / 44.1 KHz audio signal to 32 Bit(Float) / 96 KHZ. He did this conversions in different combinations and at last he got a new signal with much more body at 16 Bit / 44.1 KHz.

this whole theory about the error rates is based on the fact that Audio on a CD is treated differently than Data on a CD or on a PC. errors are not allowed to occur on a PC. just one inverted bit will screw up an entire flow of execution and crash programs or corrupt files.

running resampling algorithms *will* affect sound quality because data is being added to the sound file. different algorithms will result in different sounding files. some may be preferable to others. I'm not surprised that it sounds different, but I'm with BillyBoy on this one... we are entering the realm of *subjective* whether it sounds better or worse or punchier or whatever.

>>>Did anybody tried this with video? Exporting a DV signal as a series of uncompressed bitmaps or avi, importing this again, rendering this again to DV or mpeg2 to a new master file, and importing this again?

different compression/resampling schemes will result in different looks. interpolating data then throwing away data then interpolating then throwing away then interpolating then throwing away... eventually you'll end up with something that someone will look at and say, "that looks better".

I think it's important to differentiate these ideas from the CD-R quality phenomenon. these ideas are simply methods of introducing the effects of different compression / interpolation schemes into your source file. they have nothing to do with error rates since errors are not allowed to occur in files on a PC.

thanks for the ideas though!

Cheesehole,

of course you are right, resampling is a different thing than CD-burning errors.

I just wanted to show that staying in the digital domain doesn´t automatically mean you are not loosing or changing data.
And this changing of data is done in different ways than those we all know from analog.
New treatments, workflows and attitudes have to be found to work in pure digital.
I used Tascam DA-88s for eight years, and after a short time I could hear the error detection code related to the fact that the digital recording was done on tape with dropouts.
But when I changed to DAWs I didn´t expect something similar there.
Did you guys ever tried a dither plug after equing an audio track?
Try it, I will do so this afternoon. In the ASIO world it does a great effect, let´s see what it does in the .wav driver world.
Ready to learn something new every day.

Naughtybird