TMPGEnc's DC component precision setting

[DRP]

Can someone who knows please tell me what the DC component precision setting in TMPGEnc means and what it does? Options are 8, 9 or 10 bit but I can find nowhere any recommendations as to which setting will yield the best quality results. Nor can I find anywhere an explantion of what it does so I can make up my own mind.

90% of my encodings are of downloaded AVI files and all I want to know is which of 8, 9 or 10 bit is the best setting to get the highest possible quality.

[redwudz]

Here's maybe the clearest explanation:

DC component precision

The DC component represents the average brightness of a block (8x8 pixel region). In MPEG-1 video, this is permanently set to 8 bits, but MPEG-2 allows higher precision. It is important that the DC component is accurately represented, so a setting of 10 is recommended for MPEG-2 sequences, unless you are encoding using a low bitrate.

(From http://www.dvdhelp.us/index.html?html/tuttmpgencadvvideo.html~mainFrame)

Or this explanation, a little more complex:

DC component precision
10 bits

This controls the precision of the discrete cosine values used in the image compression (this translates the image into a series of interfering frequencies and lets the program eliminate frequencies that aren't visually relevant - this is similar to the compression used in JPEG images). Higher values results in better quality, at the expense of encoding time.

(Sorry, lost the source for this.)

Seems 10bit is almost always best.

[mistycat]

If you type dvd - hq on google, there is a guide for configuring TMPGEnc where it states that 10 is the best setting for quality. I have also read that using higher values on lower bitrates can decrease quality.

[jimmalenko]

Or this explanation, a little more complex:

DC component precision
10 bits

This controls the precision of the discrete cosine values used in the image compression (this translates the image into a series of interfering frequencies and lets the program eliminate frequencies that aren't visually relevant - this is similar to the compression used in JPEG images). Higher values results in better quality, at the expense of encoding time.

(Sorry, lost the source for this.)

Seems 10bit is almost always best.

That's from http://dvd-hq.info/Compression.html

[DRP]

Thanks guys, appreciate it

[FulciLives]

That's from http://dvd-hq.info/Compression.html

Although a bit "old" that guide is still an essential guide for those that wish to learn the in's and out's of TMPGEnc Plus.

- John "FulciLives" Coleman

[adam]

Higher DC precision is only better assuming an unlimited bitrate. The setting essentially sets how much of the high and low frequencies are discarded. Higher DC precision discards less. It is keeping more data, but that data must be represented and that takes up bits. Unless you are using a high bitrate, using a DC of 10 will absolutely kill your quality.

My general rule for DVD encoding is 4mbits or less use DC of 8. Anything up to 8mbits use a DC of 9. I rarely use 10.

Here's what the CCE manual has to say about it:

In general the optimum settings cannot be determined since changes depend on the details of the picture and movement intensity. For pictures with intensive movement and fine bumps, a low number of bits is appropriate. For smooth pictures with little movment, a higher number of bits is appropriate. A low number of bits may be specified for use at a low bitrate up to 4 Mbps, and a high number of bits may be specified for use at a high bitrate.

[DRP]

Thanks for that. I was actually wondering what is considered a "high" bitrate. FWIW I am usually producing SVCD (even on DVD), so I am mostly in the ~2500kbps range. In this case I guess my use of 10 bit DC precision is overkill. I shall revert to 8 bit and see if I notice any difference.