codec/playback

[Rafael Amador]

I think NONE is one of the worst choices you can make for editing.
What kind of footage are you using? Which will be the final distribution format?
Rafael

PPC G5 2x2Gh 4GbRAM/BlackMagic SD/PMBP 17″Core2Duo 4GbRAM
JVC DTV-17″/FCS2/AE CS3/COMBUSTION/SHAKE

[Simon Webb]

Are you looking at the rendered image on a client monitor or on the computer screen?

[Josh Snider]

we are using footage off of 3 hour DVCAM tapes, captured in at 720×480. But that is shrunk down and cropped to fit into a small kinda TV box within a template made in photoshop. we’ve viewed it both on an external monitor and on the computer screen. but the final output is a downloadable .mp4, about 200MB in size, so it makes more sense for us to view it on the computer, in Quicktime.

The whole template is 744×398. We have tiffs exported from powerpoint, being displayed to the side of the video, which might be at about 512×384, but I haven’t checked the sizes of the video or slides to be honest.
The main problem is the smaller text, anywhere it comes up on the screen, whether as part of the tiff slides or as text components of the photoshop template.
but, again, it just doesn’t make sense to me that it would look normal pre-render, and then crappy after rendering….let me know any ideas

thanks

[Rafael Amador]

Change your sequence codec to 8b Unc, or better to 10b Unc. If you set it to 10b Unc, change the rendering to High Precision. Make a test.
Rafael

PPC G5 2x2Gh 4GbRAM/BlackMagic SD/PMBP 17″Core2Duo 4GbRAM
JVC DTV-17″/FCS2/AE CS3/COMBUSTION/SHAKE

[David Bogie]

I can’t tell from you post what you’re complaining about, the work you’re doing during the scaling operation or after the compression to .mp4. Of course everything is going to look terrible as a highly compressed .mp4.

If Raf’s assistance hasn’t helped get a better image, try going through this one step at a time for the rest of us slowpokes.

You’re capturing at DV.
You’re pulling your DV into a custom sequence that is 744×398 because that’s your client’s weird spec.
You’re scaling your DV to fit a little monitor graphic that’s 512×384 and part of your custom sequence. That will introduce significant quality issues since you are throwing away about half of your pixels.
You’re bringing in full rez tiffs from PP that have small text and you are placing them into your custom sequence. That will introduce significant resolution issues since text cannot be rendered to any video or motion codec directly from tiffs and maintain anything that resembles the original resolution. It is a function of the conversion from still image to video.
You are then rendering to an unspecified codec for your custom sequence. If you have not specified a codec, it is probably being rendered to DV.
Then you’re compressing the whole thing to .mp4?

You’re imposing scale reduction, pixel conversions, and pixel averaging at almost every step of the process.

bogiesan

[Josh Snider]

hey bogiesan, what i’m having trouble with is NOT the export, it has to do with how the image looks AFTER rendering, but PRIOR to export.
This step below is what I’m having the most trouble with:
‘You’re bringing in full rez tiffs from PP that have small text and you are placing them into your custom sequence. That will introduce significant resolution issues since text cannot be rendered to any video or motion codec directly from tiffs and maintain anything that resembles the original resolution. It is a function of the conversion from still image to video.’

Maybe the question is, what format should we save these in, if tiffs are no good? (the tiffs are exported from both PP and Photoshop)

Raf, we tried a bunch of different compressor options in the sequence settings. Uncompressed 8 and 10 bit and None all worked the best. In fact, None looked just as good as 8 bit Uncompressed. So we went with that over 8 bit, so as not to eat up too much rendering time. But we’ll go with 8 bit Uncompressed if that’s what you feel works the best.

Hope this all makes sense. thanx, josh