Customers and potential customers are starting to ask questions like 'is MegaPEG.X' better than product 'Z'? Before I get into what I think, and what I'm finding, I want to start by talking about competition, what it is, and what it ain't.

If a person were to take some time and research my personal side, they might find that I am descendant from the last living tradition of warrior-knights that walked the planet, and lived by a code of conduct. There is no question that there can be fierce competition among warriors; the question is rather about the code of conduct.

You won't find me slagging anyone else's product with unsubstantiated claims and spewing filth in online forums, and definitely not posting anonymously. Eventually I expect all such barbarians to venture onto the jousting field where they will be dealt with accordingly.

Today I took a look at the output of a product I'm going to call 'Nigel'.

I grabbed a difficult piece of camcorder footage, and straight outta the box, at 29.97i 7.5 Mbs, Nigel had a definite quality edge. In both cases, I simply grabbed the nearest template that was available in MegaPEG.X and Nigel and went to town.

I'm not really that concerned about speed, because MegaPEG.X is definitely competitive. In fact, encoding 720x480x24p we clock in around 3:1 on a 1.2Ghz machine, which is a fine number for the quality you can get at that speed.

So, back to Nigel. Like I said, at 29.97i 7.5Mbs, nearest preset, Nigel had a bit of a quality edge. At about 8am this morning, I started doing short test encodes with MegaPEG.X, using it the way that it is designed to be used, applying iteration after iteration and studying the output of each run with MPressionist.X Pro.

After about an hour of experimentation, MegaPEG.X had the quality edge, and I was running the MegaPEG.X bitrate at about 2 Mbs lower than the original template. Unsatisfied with the output of either Nigel or MegaPEG.X, I had switched to our TypeB de-interlacing, which performs very well, and unlike a standard 'blend fields' algorithm, does an excellent job with Titles and CG (Computer Graphics).

Right now it is about 1pm, and I have the MegaPEG.X bitrate down to 3.8 Mbs, and I'm still running at 29.97i, with de-interlacing and DCT type manually set to progressive. And there are no artifacts in the file. I mean if you get up real close to the monitor you can see the edges of some of the more tightly compressed frames, but these edges will disappear as soon as an interpolated frame resample (as would be the case in display for TV or playback in DVD Player.app) is applied.

My test encodes are running at the 2nd to highest quality setting for progressive frames in MegaPEG.X - the great thing about tuning with our new product is that once I get the settings dialed, I can simply bump the quality up to the highest precision, and let the damn thing run overnight. The file will have the same GOP pattern, the same general bitrate and compression profile, but the file will be smaller and the picture quality substantially higher.

Experiment in the day, compress final footage at night. Sound familiar?

There is no question that Nigel is a worthy jousting partner. But when it comes to the needs of a professional compressionist, Nigel runs short not on quality, but on the ability to be adapted to the task at hand.

I expect that after I apply our 29.97 to 23.976 time resample and apply 2:3 pulldown flags, I will squeeze out another 1-2Mbs from the bitrate. Of course, at this point, I'm doing that just for kicks.