avi to mp4 10 times as big ?? I need to upload to YouTube

[JytteC]

in my configure button I have 'target quantizer' (been using), but also 'target bitrate' makes no difference?
I do have CRF mode as an option, so I will try that now.

[jagabo]

in my configure button I have 'target quantizer' (been using), but also 'target bitrate' makes no difference?

With target bitrate you are telling the encoder you want a specific file size (remeber: size = bitrate * running_time) and don't care what the exact quality is. If you specify the same bitrate as your source you should get a file of the same size (aside from audio size and container overhead differences).

I do have CRF mode as an option, so I will try that now.

CRF and QP are both "quality" based encoding methods. You select the quality you want, the encoder uses whatever bitrate is required to deliver that quality. QP is a mathematically constant quality method (throw away "this much" detail in every frame). CRF takes into account what the human eye is more sensitive too and gives better image quality (when the two deliver the same bitrate) or a lower bitrate with the same visual quality.

[JytteC]

HA! I changed to CRF (mind you every time I went in to change the configure, it changed back to QP, so I changed it again)....
Results:
crf, TQ10 (82% quality) 500MB (bitrate 67Mbps)
crf, TQ15 (72% quality) 320MB (bitrate 43Mbps)
crf, TQ18 (66% quality) 242MB (bitrate 32Mbps)
crf, TQ20 (62% quality) 197MB (bitrate 26Mbps)

Then I compared the raw video to the last file of 197MB, back and forth focusing on different details... and any loss of picture quality was indiscernible

Now this is something I can work with! And now I know what to do.

Thank you so much all for your help!! I'm off to the races.

[JytteC]

Jagabo, thank you for explaining things to me in a way I could understand! This is the first time I've had to do these things and I was really confused.

[JytteC]

And I read up on the 'single pass' and 'two pass' encoding. I guess I should go to 'two pass' as it matters little if it takes longer.
Thank you all again for your great pointers.

[jagabo]

HA! I changed to CRF (mind you every time I went in to change the configure, it changed back to QP, so I changed it again)....
Results:
crf, TQ10 (82% quality) 500MB (bitrate 67Mbps)
crf, TQ15 (72% quality) 320MB (bitrate 43Mbps)
crf, TQ18 (66% quality) 242MB (bitrate 32Mbps)
crf, TQ20 (62% quality) 197MB (bitrate 26Mbps)

Don't expect all videos to show those same file sizes. qp/crf encoding will deliver different bitrates depending on the nature of the video. For example, shoot a video of a bookshelf with no motion. Then shoot a video zoomed in on a tree with all the leaves blowing in the wind. Encode both with the same qp/crf. The bookshelf video will be much smaller than the tree video.

Here's an example using your video. There's not a lot of motion in the cropped area but a 1 minute video with motion at CRF 18 turned out about 28 MB. The first frame of that video repeated 1800 times (also 1 minute) turned out less than 1 MB (all other settings were the same).

[JytteC]

I understand that they will be different. Night videos are also smaller in size (B/W with less going on).
It was so cool to see that cropped video. I tried to do that with one of my old (poor) videos, but couldn't figure it out, so ended up zooming in instead, and since the raw video was already poor, the end result was impossible to watch.
I'm not sure how much I'm missing out on by having the free version of VSDC, I have only seen a few times 'you can't do this without upgrading' (certain ways of zoom was one of them), but then again, I'm only just starting doing some real work with it, so I don't know. I probably should. I need to learn a lot LOL

[johnmeyer]

I have had several trail cameras. I did have a Bushnell, but returned it. Here is my harsh review:

Bushnell 8MP Trophy Cam Review

Here is one thing you should know about ALL trail cameras.

The resolution on most of the under $500 cameras, and certainly the under $100 cameras, is fake. You can do your own tests. Set the camera to either still photos or video. Set to the highest resolution. Walk in front of the camera a few times to get some test pictures. Then, change the resolution to the next lowest resolution setting. Repeat the tests. Then, put the photo or video into your computer and zoom into the high- and the low-resolution version of roughly the same image. See if you can resolve any more detail in the high-res image. In all likelihood, you will find the two photos or videos identical in resolution, but one will be much larger in file size. The reason for this is that the sensor in most of these cameras is really, really cheap, and to get to the higher resolution specifications, they simply do an up-res in the camera. This buys you nothing.

Unfortunately, pretty much all of these cameras operate this way, so you won't be able to find a better solution. However, if you verify that you don't get better quality with the higher resolution, you can save a tremendous amount of space by using the lower resolution.

You WILL find that some camera, regardless of the resolution, really don't resolve much detail. I returned another Bushnell because its photo quality was so bad. Here is my review of that camera:

Great detection range, but absolutely AWFUL pictures

I ultimately ended up with a Browning camera. Here is my review of that camera. I previously owned a Moultrie and was also happy with that.

Very good overall (4 out of 5 stars), but detection range is a little disappointing

As for getting the video into the computer, you really need to use a tool like Mediainfo to determine the exact specs of the video. Very often the file extension on these camera is misleading and you will, for instance, find .MP4 files that are not h.264. I was actually able to drop the .MP4 video files from several trail cameras directly into programs that do not natively deal with h.264.

[JytteC]

I did a lot of research on the cameras (after I impulsively bought the Apeman, duh), focusing on video (that's what I want). Every time I saw video I liked, it was inevitably a Browning, so that's what I got (Recon Force Advantage). I realized that this is not a top of the line camera as such, but didn't expect that. What it does is take some really nice video without me having to be there.
Yea, non of the cameras have the pixels advertised. The native pix of the Browning I believe is 6MB.
After also getting some great videos overnight, as I expected, I'm really happy with my camera. I still have to experiment with the settings, it will take me a night for each setting to see what I get.
What I do have a problem with right now is the camera noise on the video. I have a lot of birds on these videos, and they sing a lot. The audio is rather low, and when i turn it up, you get a constant noise from the camera, sounds like a mix of heavy wind and traffic.
Now I'm trying to reduce that noise out in VSDC (upgraded to pro), and I'm pretty much getting nowhere. I'm trying with the 'gate filter', but I guess I have no clue, so I either have loud noise, or nothing at all.

[johnmeyer]

Audio from a trail camera is going to be pretty bad. The microphone is the cheapest possible, since audio is not very important to hunters. Also, the camera is outdoors and has no "dead cat" or foam filter to reduce the noise from wind.

Getting rid of background noise will be difficult. I own iZotope RX which is a fairly high-end audio restoration tool, and even with its various noise filters, you can only do a partial reduction of background noise like wind. To do this restoration, you take a sample of the background noise, during a moment when that is all there is, and then you apply the noise reduction filter, using that sample, to the rest of the audio. The problem is that if you try to cut out too much, you'll get all sorts of artifacts. A noise gate won't introduce artifacts, per se, but it will give you audio where the background noise comes up to full volume during the sections when you have actual audio, and then when the audio want to keep isn't present, you get dead silence.

The best practice for this type of noise reduction is to use the noise sample, but only reduce about 30% of what you eventually want to get rid of. Then, take a new noise sample on that partially noise-reduced audio, and do another partial reduction. You get fewer artifacts this way.

You might try Audacity. I've never used it, but many people recommend that as a free alternative to the rather pricey iZotope.

[JytteC]

I do have Audacity, but having to split it up, then work it in another software... that's way more time than I'll have on a regular basis.
Yea, I know this is very basic equipment, and the important part is the video, still, since it has those filters I thought it was worth a shot, if only I could figure out what to do with it.
I opened another thread on this audio problem though.
In the end of the day I ,and whoever else is watching, may just have to live with it.

PS, even if I DID have a 'dead cat', chances are it would be gone the next morning LOL critters are nosy.

[johnmeyer]

PS, even if I DID have a 'dead cat', chances are it would be gone the next morning LOL critters are nosy.

Just to be clear, since "dead cat" may not have a German equivalent, I am referring to microphone covers, like these:

Google Image "Dead Cat Microphone" search

[JytteC]

LOL, I knew what you meant. And I am not German, I just live here at present.
I've seen critters tear off and carry off these covers in a hurry, they do look tempting I guess.