Pixelated images on render to DVD

[Bob Peterson]

How large are the stills?

[Terry Esslinger]

Were you playing it from the same playback device each time?

[Vance Bridges]

Bob: The stills are huge files both from slide scans, digital
pictures (hi) and so on–the ones that were crud on the
wide screen are the ones I did macros on from digi files
from LOCongress and evidently I wasn’t aware that I had passed
the safe zone (for larger screens) That can be remedied by my making a still of the image and having a more solid result or just backing off. (A few others did not have that problem but that was when the entire picture was shown)

But what I am most bothered by is that the DV video material on this
sample(which is perfect) was pixel-blotched like it was made with a cell phone.(I have seen video from this camera 6 feet across and it
is crystal clear–(an old Canon)
Evidently–the slowing of the image of the video clip in Vegas
and rendering from a best setting AVI to an mp2 has this result.
(Part of the problem I have with Vegas is that huge projects
crash when you try to render all out to mp2 and I try to pre-
render special effects and complex montages to a best avi and then
go to mp2.) All of this material by the way–looks perfect on a
standard sized 4/3 monitor.

[Vance Bridges]

Terry: the pixel-blotching was observed when the clip was played
on two different Samsung wide screens (and players) in the media
room. That is not to say the damage wasn’t on the smaller screens
you just couldn’t notice it. My reference monitor in the office
unfortunately is merely an old magnavox for reviews–but the clips
look really good on it–as they do on the smaller screen. Besides
overcranking close ups on the digi stills–I am at a loss on the
DV clips–Even slowing up in an Avi and going to mp2 should not ref
so much bad material.

[Bob Peterson]

On the stills, I was told in Vegas training that Vegas is not particularly good at reducing still resolution. Therefore, give it stills that are sized, in pixels, for the video. Larger than that if zooming, but not huge. That makes sense when you consider that things like Photoshop spend substantial resources on resizing images while maintaining image quality.

I’m sure I can’t help much with the video aspect. Have you tried rendering without the speed change to see if that fixes the problem? I do know that zooming is a problem for most video. I zoomed a video in an FCP class several months back, and was promptly told that zooming is NOT DONE in video because of the resultant pixelation. I concluded that zooming is not a strong point of FCP.

[Vance Bridges]

I learned a long time ago that big still images (I mean really big) need to be reduced (in another program) to work properly in Vegas.
The macro close ups of detials in these stills were done beyond the safe point so their pixels were exaggerated in the wide screen bigger screen, The video on the other hand is the problem–There was no zoom. It’was a straight shot and is perfect. The slow-mo and bit rate may play into this but so far no one knows

[Vance Bridges]

Other forums have been discovered discussing similar concerns. I will defer to them. The boards center on bit rates, settings in mp2 renders and different authoring programs. The problem really isn’t in the innate material but in the translation to mp2 and then DVD. Even in Cow this was found e.g. where one writer discovered DVDA creates a better mp2 than allowing Vegas to. For good video to translate so bad in “best” setting–means the issue is lost
in translation i.e. I am beginning to think that Vegas is great for assembling a project–but not great for its final render or DVDA. I’m shopping…

[Vance Bridges]

CRYPTIC VIDEOLOGY A WASTE OF TIME
FORGET VEGAS AND DVDA FOR YOUR DVD

I went back and zoomed in on one of the digital
still frames from an avi–I zoomed in on a cheap monitor
until it was the size it would be on a large wide screen.,
hd It was great–though a little blurry; it was clean with no overt pixilation unlike the monstrosity that resulted when I rendered
a DVD on the same picture in Vegas., which proves my point: nothing is wrong with the material.

Rendering to mp2 and DVD out of Vegas is a bag of worms and that’s why all of these forums and discussions on other websites take place. Once Sony bought the software they didn’t care about the customer. They and their designers have allowed
a fluke to be passed on–upgrade after upgrade with built in weirdware to make the customer assume they will be getting a sweet deal when in truth it is a endless wonder of what ifs and how to-s that result which keeps the producers getting the upgrade.

I repeat what was said earlier. Vegas is okay to assemble your project and produce avis- or at least an EDL. -It is not a good program to render anything to a final output–and I will repeat this message on other forums and to consumers. Also–their customer
service is a sad sad joke.

[Mike Kujbida]

Vance, you haven’t given a lot of details here so it’s hard (for some of us anyway) to offer suggestions.
I’ve personally done several DVDs that were combinations of stills and video, some of them quite lengthy, that have been shown on everything from the 15″ reference monitor in my edit suite up to a 16′ x 20′ screen in a theatre and have never experienced the kinds of issue you’re describing.

How long was the finished project?
What were your render settings for DVDA?
Was the source video SD or HD?
On your slo-mo footage, did you use “disable resample”, “smart resample” or “force resample”?

[Vance Bridges]

Mike: To answer your questions:

1) This was merely a 4 minute or so clip from the entire program–(the duration of which is 60 min part 1 & 30 min part 2 (90min) with
hopefully enough bit space left to have a 15 min footnote and a
reference list roll that can be of lesser or degraded quality render
to save space, The menus will be kept to the basics with no
spec effects. The end result will be glass mastered. So we are only
talking 4 minutes here. (Because of the complex nature of the intro
credits and hi-res dissolves in the program I found it easier to run off no-sound best-setting avis. The sound is treated totally separately.

So this sample was merely an MOS (silent) image only reference.
An mp2 at best setting was created from an avi best setting for the test. Rendering the entire program direct to mp2 has been attempted before and failed–although I have seen clips before on large monitors that went out direct to mp2 and looked good–but I avoided that direct route for this test assuming it would not degrade. Unfortunately Vegas does not allow me to assemble pre-rendered mp2s without re-rendering them–so I am left to either avi the pieces or direct out mp2 (if it works …which it won’t for the entire program)

I will have to do some bit-rate calculation before the entire program
is burned–

The DVD test was burned at 4x on a +R Memorex although I wanted a
slower speed

After reading other posts–I may want to experiment with doing the
mp2 as a two pass–another writer said he got better results with DVDA doing the mp2 (although I have no control there)

2. The clip included dissolving stills with pans and ended with video
of a man standing–his clip reduced to 72% for a slower run. It stayed at smart re-sample. ( His section and some of the still information was overtly pixilated and damaged in the DVD reviews on three separate monitors.

3)The DVDA settings were default (which may be an issue), and there
was nothing done to the mp2 settings which were rendered also at
best setting. The DVDA merely said it was NTSC wide, and in
Optimize it stayed at default BR with a max at 10.080/ BR 9.748. Total 1.586. Est size 184.5 mb default at 8.000

4. The original source was clean digital stills (hi res) and the video was undamaged standard 720/480 anamorphic wide. I correct myself in an earlier qualification where I said I may have zoomed in too close to the still material–Indeed, I looked again
and zoomed in on the monitor here and nothing broke up. No this is a
pixel-compression issue. Something is occurring once it goes out of the timeline. Whether that something is via the avis or the mp2 renders or the DVDA burn–we will see. Some DVD authoring people may look at a sample of my avi or mp2 to see what’s going on.

Its frustrating to have a beautiful program and not be able to get it on a DVD worth seeing.But I will continue to resolve the issue.

Thanks for your time.