Opinion on Procoder

tygrus wrote on 8/6/2004, 5:18 PM
Well I finally got some decent renders out of the program and after doing a comparison against MainConcept, I have to say Procoder is slightly better in quality and detail. It retains a little more sharpness on my project after I give it some blur to over come interlace flicker than MC does.

The one place I did notice quite a difference was how much better procoder did in bad pics. It seem to retain much better end quality when the source wasn't quite as good.

Comments

farss wrote on 8/6/2004, 7:00 PM
I don't have Procoder but I have found that TMPGEnc works better than MC doing mpeg-1 encodes from poor sources.
Haven't really tried it with mpeg-2 as MC does an acceptable job particularly with 2 pass.
If I was really needing the ultimate result I'd either be going for a hardware encoder fed SDI or the the CineCraft Pro encoder but both of these are very expensive options.
Either way DV25 and interlaced video is not what mpeg-2 was really designed for, the whole DVD concept was developed around film conversion.

Bob.
John_Cline wrote on 8/6/2004, 8:46 PM
I have evaluated the following MPEG2 encoders by tweaking them for maximum image quality, regardless of encoding speed: CinemaCraft Encoder SP, MainConcept (both from within Vegas and the v1.4.2 standalone), TMPGEnc Plus v2.5 and TMPGEnc Express v3 and Procoder v2.0.

After careful consideration, it is my opinion (and that's what you asked for) that Procoder v2.0 is the best looking of the bunch and it is now my default MPEG2 encoder. (Well, until something better comes along.)

John
mbelli wrote on 8/7/2004, 11:41 AM

I've tested all the top ones carefully: Squeeze, TMPGenc, Procoder, Cinema Craft, MainConcepts.

My observation, Canopus Procoder produced the best looking stuff. Although, I did like the look of Quicktime and WIndows Media 9 that I got from Sorenson Squeeze, I also like Squeeze's interface a lot, simple and intuitive, one click, a bunch of formats are encoded, depending on your destination (CD, Web server, hard drive).

So basically, Procoder for MPEG 1 & MPEG2, Squeeze for Quicktime, Windows Media, MPG4, Real Video.


MB
johnmeyer wrote on 8/7/2004, 1:49 PM
John_Cline said: I have evaluated the following MPEG2 encoders by tweaking them for maximum image quality, regardless of encoding speed: CinemaCraft Encoder SP, MainConcept (both from within Vegas and the v1.4.2 standalone), TMPGEnc Plus v2.5 and TMPGEnc Express v3 and Procoder v2.0.

John, even though you don't use the Mainconept standalone encoder much, did you play around with its various settings much, and if so, did you write down settings that worked best? I am still playing around with various settings and have settings that seem to work pretty well for fairly clean DV NTSC video. However, I am constantly looking for ideas on how to tweak a little more quality out of the encode process.
farss wrote on 8/7/2004, 3:46 PM
Here's an idea on how to evaluate the results. Viewing a DVD on a monitor isn't a very scientific way to evaluate the encoder. Even putting different encodes of the same material onto the one DVD it's still difficult to do a scientific evaluation as there's a time lag between viewing each clip.
So why not drop the encoded mpeg file onto the Veags TL. Put the original material onto the track above, apply the Invert filter and set the track opacity to 50%. The resultant image is the difference between the source and the encoder output.
Just for a bit of confusion try encoding generated bars, look at the result on the scopes. Then try the A/B comparison above, interesting or confusing?
Bear in mind when you encode from generated media you are working from 4:2:2 material, it doesn't get much better than that and yet even with no motion the encoder is still 'doing things' to the image.
You can even do the above test using two mpeg files, make two from the same source but with different bitrates and compare them using the above method.

Bob.
John_Cline wrote on 8/7/2004, 4:06 PM
Bob,

Inverting one of the clips and comparing the difference was one of the techniques I used to evaluate the different encoders. This is also one of the techniques I used to compare different DV codecs a few years ago when I discovered the flaw in the Pinnacle DV codec.

John
farss wrote on 8/7/2004, 6:19 PM
John,
good to see I'm not alone in trying to get away from subjective evaluations. Just wish such approaches were more widely adapted!

As an engineer I'm almost daily staggered by the amount of inaccurate information flying around in this business. Here's the latest 'What the?'

Guys first foray into the land of DV25, normally shoots DigiBetacam. So he asks the post house best way to get the material into the suite. They insist it's composite even though he could ingest firewire into the system. We explain to him for ages that not only is the already encoded video now being decoded and then re-encoded but the same thing is happening to the audio. Now maybe as a result of the conversion some of the DV25 'harshness' is lost but so is resolution.
And when the final result doesn't look all that great they'll blame it on shooting with DV25.

Bob.