Suggestion

JD_Mortal

New Member
There is a few things I would like to suggest, some which I believe should not be difficult to add...

1: Better sound "Noise gate" options...
- Needs more explanation or better terms to describe the settings and sliders. They are great if you are in the "industry", but not self explanatory or suited terms to what they actually do.
- Needs a buffer-on feature and a soft-on feature. Because the hard-on is... (Can't say this with a strait face)... not filling... the recordings with the sounds coming out of our mouths. "ould ike o ay ank ou or our ard ork" (Would like to say thank you for your hard work... but if you can't hear me say it, then it is useless.)
- Auto-level dampening. (Peak-reduction) {Lowers volume when sound is saturated and clipping}
- Auto-level "mute" or [close threshold]
- Auto-level "unmute" or [open threshold]

Because there is a horrible lag in reaction, the hard-on, even with the fastest "timer" ms, is not sufficient. 25ms may fire up to 500ms off target, due to all computers 500ms timer-interrupts. Same with 200 and 500, but 500 can skip and hit 1000ms if it happens on the next cycle. Using a buffer will give you nano-precision, if you want it. (P.S. MP3 has a volume option, so you don't actually have to alter the actual volume. Thus, this whole option for sound can be done post-process, and the final "result" can then be optionally saved as "modified volume with original sound", or "modified sound and volume", which would actually remove the silent parts and physically alter the sound file output data. Each data-packet in a stream has a volume-level value. When not present, it just uses the default "full-volume" of the "first set volume". If present in each packet, it adjusts to become the new volume.)

The buffer-on feature and soft-on, would act similar to the hard-on setting used now. However, the buffer-on is just a buffer of sound, so when the detection is triggered, it is actually opening the time prior to the time of the trigger. Being "soft", would simply be a progressive volume increase, from mute, instead of a blunt "mute/unmute" or hard opening, or hard-on. (You don't use an actual buffer. You simply post-edit the volume data, prior to the activation, once recording stops. Then, before you actually write the sound data, you can remove those muted spots, which would have the unclipped beginnings of words spoken. Unlike the hard-on setup, where you just can't get those missing sounds back, once clipped-away.)

2: On-screen indicators... (Overlays that do not get recorded, if recording the screen, in full-screen.)
- Recording/Stopped [Stream/File/Replay]
- Mute/Unmute [Mic/Desktop/Source] {Also note, Mute and Unmute should have the ability to be two separate buttons. Without anything to indicate if sound is being recorded, you don't know if it is toggled-on or toggled-off.}

3: Experimental "2nd-mic cancellation".
- Use of a second microphone to "negate" some "ambient noise" and "echo" from a source-mic.

Little known fact, a "professional microphone" is not just one made of good quality parts. Professional microphones have a three-tier setup. One mic-face records what is fed into it. Another component, behind that face, negates what is NOT fed into it (echo/reverb/ambiance). Then another is sealed away from external noise, used to negate (tapping on the mic, electronic noise/hum, and handling the mic/cord/stand. The third element here is not as important as the second one, to our purposes here. We want to reduce the "shhhhhhhhhhhhhh" ambient noise, the "BANG" loud noises not spoken into the feed-mic, and echo/reverb from sounds coming off walls or out through our speakers. (Face a second microphone away from you, behind the feed-mic, and overlay that track, after you reverse the wave, as a "noise reduction map". Easy to do in post-editing and even on-the-fly, with MP3 format. AAC. We just have to adjust the level, but the dual recording or active mixing, ensures that it is "in sync".

4: Multiple sources, as a project, saved in a folder...
- The ability to, say... Record the desktop, while also recording a source-screen, while also recording a camera-source, while retaining isolation of the "game sound", the "microphone sound", and actual "windows mixer" sounds. (Using similar methods described above, to reduce the "game-sounds", and "windows sounds" from the "recorded mic track", would be a big bonus.)

This would allow, for instance, us to record our screen 1:1 at whatever the resolution is... while recording the game being played, at whatever resolution it is set to play/record at... while recording us or some external thing, at whatever resolution it is at... and all be in sync, for editing. (Like recording yourself playing a game, with a green-screen. Before getting into the game, which you use to fill the whole screen later, you are showing your desktop and you launching the game, which may be launching on another monitor. The sounds being all one multi-track file, or individual tracks, which all align to one-another and the video. Instead of all pre-mixed into one file that would actually have crappy mp3 conversion as it has full-range sound from all devices at various levels which would take hours to setup in advance.)

5: Slightly better intuitive interface and some kind of "step guidance", to setup or use the program.

I figured it out, without much effort, but some things seem redundant, uninformative, not informative enough, over-informative, out of place, and there are a lot of "must do first" things that you don't find unless you dig around through the pages/tabs/tree, in the settings. (Plus some missing options when it comes to the recording part, with unhelpful warnings about sources and keys not being setup for things that you are not even using, or trying to use, or irrelevant to the thing you are doing. For instance, the keys not being setup for the buffer-video-capture, produces a warning when trying to do a normal unbuffered recording. Wrong place to mention that warning, and there was no option to just record, only record and buffer.)

Also note... Buffer recording is not "right". A ten second buffer setting records 20 seconds of video. One would assume that it is going to record 10 seconds +whatever. (Whatever being the minimum-video time, or a gracious 1-2 seconds post-stop recording time, or pre-buffer recording extension.) Again, without something to indicate that the "buffer is ready for recording" {buffer active/on}, then we are left guessing if it is actually ready to capture. Same with it not having any indication that it is "now saving" or "captured".

Also, is "record from buffer" the same thing as "record" when recording without the buffer? (Seems redundant, unless it is trying to say/state... "Start full recording, including the buffer time captured". Would word that as this... "record with buffer", not "from", or "record, including buffer", or "record from buffer start". If you are recording from the buffer, you are saying that you are going to buffer the buffer or record the recording.)

6: Include "$seq" (sequence or sequential values) for saving files. Possibly also $SEQ$4 to pad with 4 leading zeros. (Insertable where we format them to be, as opposed to the odd appending that programs or windows generates.)

7: List the compatible file formats. (Containers)
- "MP4, FLV", above, or near, or in a pop-up where we have to pick locations and names for the files. Threw me off when my video recorded as MP4, but the buffered vid started saving as FLV. Quick search found those are the only formats used.

==============================

Done with my odd-ball suggestions.

Love this program. Can't wait to see what it holds for the future. It has already blown most others out of the water, leaving them as chum, from a technical perspective, but it is not quite a yacht for normal drunk consumers to party on. Many are sure to fall off the boat in the most shallow water, or even on land, until they get sober and try boarding again.
 

JD_Mortal

New Member
Also wanted to add... another suggestion, in the form of a question. (Because I am unsure if you are using anything below.)

Is there any code taking advantage of Radeon's direct code access or Nvidia's video-processing code, or the new video-compression code built-in to all the new 3xxx and 4xxx CPU/GPU components?

Taking advantage of the built-in functions of already processed streams, or parallel streams, within the cards themselves, is easy and reduces CPU demand to near zero. (Thus, you can potentially get even better encoding, where the GPU would not also have additional code to assist.)

One last thing...

Some kind of "time-lapse" option, for slow and ultra-slow recordings. (1-2FPS for video, and 1sec-10min for some image format. I don't think video likes lower than 2FPS, as it will just record lower to 2FPS anyways, or get buggy on certain devices.) I do a lot of time-laps stuff, for lets-play video. I record about 2 FPS and run the video 8x to 16x, which translates into around 16FPS-32FPS after conversion. However, it allows me to record for 8-32 hours, and remain under 4GB on most full-data capture programs. If that could be extended to 64-128 hours, with this "on the fly" compression format, that would help a lot. (Or have the ability to read sequential images into a video, as individual frames.)
 
Top