(vitalforces = editor3333)

You're right. I'll have my money ready.

I can confirm that native YUV editing is much faster than YUV-RGB conversion editing. I used to use a Canopus (non-RT) Raptor DV card with both the RGB-based Premiere (5.1 & 6.0), and the YUV-kernel Canopus Edit. Canopus Edit was about twice as fast as for the same render as Premiere (this was on a Celeron 400Mhz), because, (so I'm told), the YUV signal from the video didn't have to be converted to RGB.

FWIW.

It would take total program rewrite. That's what Sonic EPM said once. If it happens, great. In the end though, Vegas does so many things so good I'll wait for a RGB>YUV render.

Craig H.

A program rewrite? Nah--if it's not broken, don't fix it.

Choosing between having an internal "canvas" in YUV planar or RGB scalar isn't just an efficiency consideration. The two scales and palettes don't quite fit eachother without providing additional bits per vector for rounding errors.

Your eyes are more YUV than they are RGB! Camera technology varies slightly with the detector and filter system so isn't in particularly one or the other by definition. The back end of the CCD is the video encoder which typically follows the broadcast definition of video and heads down the complimentary color - YUV route.

There are compromises going the YUV route, and some of these can be unravelled if your AVI stored form is RGB during/after acquisition. However 8-bit YUV deserves 9-bit, maybe 10-bit in RGB to reduce noticable contouring.

So you test Vegas against this gamma/linearity situation. Have yourself generate a set of color ramps and run this through Vegas. Then render the output of the timeline and then import this back into Vegas. - With a good pipeline, you won't see RGB or YUV conversion troubles. Some FX and filters are mathematically more simple to render in one canvas type or "space" more than another. Some, it makes no difference.

Premiere Pro has a YUV pipeline mode. Chances are that some 3rd party plugs will break if Adobe didn't also expose their RGB mode.

The same considerations come along when you choose 3D effects to actually manipulate in 2D, by constraining them in your controls. Chances are that this is less efficient, but it all depends on the implementation and how the host software looks ahead in predicting the best way to handle this. It is too abrupt to just say RGB editing is worse.

The monitor at the end will have RGB phosphors, substrate or gases. Whether you connect with component Y Cr Cb phonos, Firewire , Y/C or SCART-RGB. It also will not have the exact same gamma as the model and settings of the TV that was behind it on the production line it came from.

Non-square pixels, interlacing and standard definition isn't the final word. So if going YUV will truly buy Vegas something - will it be wasted effort in the longer term. So Vegas is slow then? - not on my PC it isn't.

Sub-pixel sampling? Higher numbers of bits per pixel? Well - you might want to test it against Premiere and other YUV competition.

Where an NLE doesn't handle these issues well - it should be pointed out in the test reports/reviews. So why isn't it? Why wasn't Premiere before Pro regarded as a toy before it got a YUV pipeline? I'll tell you - because it doesn't always matter that much.

If your target is DVD or DVB broadcast - then you've got more things to worry about than what mode/plane the NLE is operating in!

....which brings me further off-topic to ask why B&W DVDs don't have 24bits of precision when the movie is running. You don't actually need the color resolution in B&W movies - or do you? Also, if you only load up luminance with B&W 8bit data on an MPEG-2 DVD, why wouldn't you be able to record nearly 20 hours of best quality onto it..... Standards aren't always very smart - are they?