Best settings for rendering photo slideshow in VS10?

Machineman · Post by **Machineman** » Wed Oct 25, 2006 3:16 am

I'm making some slideshows using the timeline and creating a video file. The video file will later be added into a DVD menu.

They are 3megapixel jpgs and Im adding some music. When I play the created video file on my PC the slideshow pictures don't look as crisp as they do in viewed as a pciture.

Settings
MPEG files
24 bits, 720 x 480, 29.97 fps
Qualoity 100
Upper Field First
(DVD-NTSC), 4:3
Video data rate: Constant 6000 kbps)
LPCM Audio, 48000 Hz, Stereo
Smart render
Perform non-square pixel rendering

Do I have the settings wrong for pictures?

sjj1805 · Post by **sjj1805** » Wed Oct 25, 2006 3:26 am

Smart Render will have no effect because they were not videos in the first place. LPCM is also a large uncompressed audio format taking up lots of disk space and so you may wish to use AC3 (Dolby) instead. the smaller file size will release more room on your completed DVD for the video file.

If your pictures are not as crisp as you would like try resizing them with
PhotoImpact or your favourite picture editor so that they are the same size as your final output. whilst VideoStudio can accept larger file sizes it has to resize them - a job better carried out by a picture editor rather than a video editor.

Post by **Ken Berry** » Wed Oct 25, 2006 3:53 am

If you are only making a video based on still images, then you might also want to experiment with using Frame Based as the field order. That being said, this is the first time I have heard of anyone using Upper Field First for a slideshow. I know people use Lower Field First successfully. But anyway, try out the Frame Based idea... There will be no harm is you later include it in a DVD which has, as separate titles, videos using another Field Order...

daniel · Post by **daniel** » Wed Oct 25, 2006 11:41 am

Although I readily admit I never even tried using frame-based, I've always felt insecure about this.
Since I consider every second of animated title, or transition (in a slide show they make the bulk of the size) or dynamic filter (brightness, pan/zoom etc) to be producing a moving video, I always used field-based (lower first by pure random choice).
Even so, there a places where a very high output rate is needed (7MB not always enough). I must add this is PAL so a little more points than NTSC.

Anyone has a fact-based suggestion in this frame-vs-field for "improved" slide shows?

BTW the pictures themselves have always been perfect, even resized by VS8/9/10 from 4MPixel non-4:3 input. The only time when I have problems is with 720x576x24 maps that I include in the show.

Post by **Ron P.** » Wed Oct 25, 2006 1:23 pm

The reason for Frame Based vs UFF or LFF, is that still images do not have seperate fields like video clips. An image has one whole frame. Video clips have the 2 fields and when it is painted on the screen, the fields are interlaced to make 1 frame.

Just think of the UFF and LFF as 2 rectangles, each containing information that when they are lined up under each other make up a picture. With Frame Based it uses the whole image, not parts of it.

I have used Frame and LFF for slide-shows, and have had no problems. It seems that Frame based gives me a little better quality though. Generally when I'm going to use LFF, I've got photos mixed with video, and the video is digital (DV) so LFF rules..

sjj1805 · Post by **sjj1805** » Wed Oct 25, 2006 3:17 pm

Thanks Ken for spotting that point I overlooked in my original response.
Yes you should use Frame Based for still images.
Another tip is to apply the anti-flicker filter.

VS10+ - this is found under
File | Preferences | General tab.

VS9 - This is found
Share tab | Create Video file | Options

Field settings are determined according to the source of your video not its final output.

Normally you find that
Digital sources (Camcorders with IEEE1394 cable) are Lower Field First.

Analogue sources (Like TV cards) are Upper Field First.

Still Images (Scanner - Digital Camera) are Frame base.

daniel · Post by **daniel** » Wed Oct 25, 2006 4:05 pm

OK, but think of a transition or a title, if you use field-based you can SEE that the two fields are generated (and yes I assumed anti-flicker ON) andr the moving part is split in time in 50th instead of 25th.
So the movement appears smoother because the turning page, for instance, has the two fields instead of just one frame.

Of course for the parts where your image is static you are losing CPU time but I think of it as CBR vs VBR. It's most of the time a useless waste, but when it counts, it has the capacity to show it's best.

This is clearly ignoring the particular case of a CUT transition slide show with fixed titles only fading. There your data is constant and no second field would help anything.

Again this is my perception but it's only a feeling and I would like to get hard facts or sound theory to settle the point. Up to now nothing has convinced me. I guess when I'm idle I could take the time to really test it.
Could happen one of the next years...

EDIT: this is all assuming TV viewing of course.

Trevor Andrew · Post by **Trevor Andrew** » Wed Oct 25, 2006 4:46 pm

Wow, wow, wow, slow down

This guy asked a simple question regarding a slide show.

I am sure he was expecting a simple explanation.

What he has got is a discussion within the forum about the prose and cons of ¡K¡K¡K.. come on.

A bit complex for me, I am now getting confused between uff-lff-fb sm and all the other stuff-----

Lets keep the discussion to the Op¡¦s question.

I feel we are going off on a tangent

Answer his question

But don¡¦t make it into a discussion between yourselves,

Sorry another senior moment

Trevor

sjj1805 · Post by **sjj1805** » Wed Oct 25, 2006 7:15 pm

Daniel,
You are going far too deep and as Trevor points out taking this off topic.
We have been simply letting MachineMan know which settings to use,
nothing more, nothing less.

Post by **Ken Berry** » Wed Oct 25, 2006 11:04 pm

I am not sure I totally agree, with great respect to my friends Trevor and Steve. Sure, we are trying to answer the OPs question. But it is common practice in this forum for other posters to ask legitimate follow-on questions in an additional post in the same thread. So I have no problem at all with Daniel's questions...

That being said, I have no 'fact based' answer to those questions!!

I can only say that with just about every home DVD I make, I include at least one slideshow. When I do this, I invariably use Frame Based.

The slideshows also invariably include a variety of transitions - predominantly cross-fade, but a selection of others just to add a bit of interest. Throw in a few pan-and-zooms for the same reason. End destination is also playback on TV. And I have never had one failure. Crisp photos in the final show, and no problem at all with any of the transitions used. But that's just one person's experience. But hey, I can -- and do!! -- live with that!!!

Machineman · Post by **Machineman** » Thu Oct 26, 2006 12:19 am

Thanks for all the great responses. I originally had it upper field first becuase I was using the same settings as my camcorder source mpg video. But no video will be mixed with these photos under the same renderd video file.

I'm not using any transitions so I just now made a file using frame-based. Anti-flicker is also enabled. The video file still doesn't look that great when viewed on my PC (WMP) but looks better on my TV so I guess its OK now.

Will frame based be OK for standard analog TV's and DVD players without progressive scan?

sjj1805 · Post by **sjj1805** » Thu Oct 26, 2006 12:27 am

MachineMan

The Field Rendering is decided upon the SOURCE not the OUTPUT.
See my post 6th up from this one.