AES Journal Forum

It is true, David, CD´s and even DVD´s are overcompressed nowadays, and we mastering engineers (some of us at least) are constantly trying to show the client why this should be avoided. Sometimes we succeed, though, and we are able to obtain a good quality sound that is never overcompressed. Check organizations such as Turnmeup.org for more info on this subject. Regards!
Andres Mayo
Mastering Engineer

Dear Shane,

It is not possible to list the best or "most accurate" equipment for mastering because it really depends on your preferences. But I can tell you that plug-ins based mastering is certainly more limited because you depend on the CPU variable resources as much as you depend on the bad or decent quality of the plug-ins algorithms. I personally prefer hardware based mastering, either digital or analog, but the kind of processing where you can rely on the processor itself to accomplish the task it was primarily designed for.

Hope it helps.

Andres Mayo

Hello, Shane.

I'd like to add some comments about your question concerning hardware and "plugin-based" mastering.

The main thing to remember here is that mastering, just like mixing, recording and arranging, is still a "human-based" job. This is easier to understand when you think about mixing, for example. There is a lot of people that work whithin the realm of their workstation, using its routing and edition capabilities, using plugins, of course, but above all, using their own skill, knowledge and good taste in order to accomplish the task.

There are very few that still use analog consoles with digital tapes these days and the ones that still use analog tape are even fewer. Some use their digital consoles for mixing and some only use them as control surfaces.

Some others combine the best of analog and digital worlds by using a "summing device", as they call it these days. They make stereo submixes within the workstation and then use the many analog outputs that they have in their audio interface to feed such device with 4, 8 or even 12 stereo submixes. Some of those devices, like the ones designed by SPL, the pioneer on this field, allow the user to add some analog processing to each stereo channel and then "sums" or "mixes" the result.

They record the result in another track of their workstation using a pair of inputs on the same interface or on another.

But I have seen grammys going to people that made their mix within the workstation, and to people that use summing devices as well. I even had an album from Dave Grusin that was mixed while it was been recorded straight to DAT tape, so no multitrack was involved in the process, digital or analog.

I think that any of those strategies may give you good results, as long as you know your equipment well enough, you have enough knowledge about audio, good taste, and above all, you know exactly what are you trying to accomplish with your mix.

Something similar happens with mastering. You can achieve great results with software or hardware, digital or analog. As Andres Mayo said, it's more about personal preferences. You are more likely to achieve good things when you feel comfortable in your environment, which often happens when you're working with the tools of your preference, which in turn are often the ones that you know the most.

About actual differences between digital hardware and software, there are some facts. First of all, as Andres Mayo states, there´s the fact that the hardware device has its own processing engine which frees some resources in your computer. Second, the design strategies. Within the computer, you may have managed code or native code, the second being harder to do but faster than the other. Whatever the case is, the software application is run instruction by instruction, in a sequential manner.

In the hardware world, there are two main approaches: sequential and concurrent. This means that, when you use microcontrolers or even microprocesors, you're still running sequential instructions that are often programmed using software anyway. If you use, PLD's CPLD's or FPGA's, that's a whole different story, because you can be running any number of processes at the same time, and they don't even need to be related to each other. In the end, though, this has nothing to do with the QUALITY of the processing being made by the device.

As in the software case, it all depends on the algorythm.

Sometimes the designer prefers to use the hardware approach just because he/she doesn't want to be pirated. The hardware design is a black box that cannot be easily copied or instantiated.

Other than that ... 1 and 0 are always 1 and 0, if you know what I mean.

Hope this helps.   :)

 Dear Guillaume,

I don´t know what kind of music we are talking about, but certainly a crest factor of 25-30 dB is quite a lot to ask for. I would say on first sight that you should be more than OK with your mixes, but on the other hand I have no idea what the mastering engineer had in mind when he requested that. If you want me to double check the mix for you, I can do it for free if it helps you in any way. Please write me offline at andres @ andresmayo.com

Good luck,

Andres

Hi Guillaume,

I am glad I can be of help. Best regards,

Andres