![]() |
OAS Playout.......... - a potted history |
![]() |
More or less why I wrote it - read on if you enjoy a laugh.... or skip to Pt 2 where it gets technical.
![]() |
My interest in all things radio dates back a good few years, as some of these rather appalling photos will testify to - some early attempts at some sort of "studio", those with good eyesight may spot the complete lack of any sort of mixing equipment and the "mike stand" closely resembling a bit of string.
We're in the dot-matrix printer era here..... this from an old Star LC10 still connected to my Pentium P266. Anyone know where u can get hold of printer ribbons for it these days?? |
Incidently, ImageFM (such as it was?) still continues under the banner of Image Communications, custodian of which is my long standing friend Sir. Jim of Hart who is currently building a "real" studio in his converted garage - you can read all about how it's going along with a load of other stuff by diving off there.
| Much of this site is/was dedicated to the old Dragon computer and I'll bet we were one of the first people to be sampling and playing our jingles off an 8-bit micro. My brief foray into hardware yielded (amongst other things) the 8-bit sampling hardware I bolted onto my Dragon64, along with a custom ram disk block for needed to hold samples of any significant length. The precursor to OAS Playout was then a Dragon GUI (with mouse!) sampler system complete with echo and delay capabilities. It made use of just about every bit of memory on the Dragon with about 1K to spare. And we still used string for our mike stands..... |
![]() |
Much of our radio was influenced by early (and not so early) commercial radio and to a certain extent still is today. My words of wisdom on some of these (along with the odd sample) can be found on my local radio page.....
![]() |
We also built a couple of mixers along the way - this last one (built out of the remnents of the one in the previous photo) I still use today for trials. The advent of MP3 as a common (CD quality) playback medium (forced me to upgrade my original 486DX2 anyway) and the whole sampling thing not being the CPU intensive process it was on my old Dragon opened up the possibiltiies of producing something I wish I could've done 15 years ago - forget all those tape decks we mucked around with trying to get decent quality jingles done however writing software for a living makes you less inclined to want to spend your spare time doing so. However after much threatening to do it I finally made a start last year. |
please note the lack of string.
Before I started down the development route I had a trawl round the 'net & asked friends for suggestions on packages already out there - I tried a few and came away disappointed. I had a couple of things in mind, based mostly on my previous experience and what I'd like to have used in my younger days:
|
I started looking around for MP3 playback resources, this took some time because I was finding the Microsoft API stuff either very limited (the legacy WaveOut stuff) or overly complex (some of the DirectX, MediaPlayer stuff). Eventually I found and downloaded the xaudio MP3 libraries which despite the documentation lacking somewhat, managed to get a basic player up and running very quickly. The libraries are very powerful and reasonable simple to use, unfortunatly they aren't free and the licensing issue is now causing headaches in producing a commercial product (more on this on the main development page). I wrote a brief 'spec' and rolled out a first 'trial' release for initial comments to a couple of friends who had agreed to test it for me and feed back comments of their own. |
trial1, March 2001 Only the twin main players at this point. Originally the aim was to keep the display to 800x600 - it's now a minimum 1024x768. |
Here's my original spec:
Overview -------- I'm not proposing to develop an automated playout system - there seems to be enough of those around although once the basics are in, it could probably lend itself to being enhanced - this is more a manual system. It's also probably going to look a bit sparten to start with - I'm not targetting the brightly coloured cd-look here - functional is the end result. From a visual point of view, there should be 2 distinct playout panels, each one independently coupled to a sound card (or possibly each o/p from a sblive - it does get 2 device enumerations after all?). In the middle is some kind of 'jingle jukebox' where you can very quickly pick a jingle (or several) and play them out immediately. All music/jingles to be held in a shareable database. Playout Consoles ---------------- Each one allows 'n' tracks or jingles to be cued up ready for play. With no user intervention, hitting 'play' fires the lot out in sequence. But there will be buttons to override this - specifically kick another track in the sequence off !now! and possibly some fade in/mix between the two. Jingle Central -------------- A sizeable bit of the display will list in some kind of straightforward manner all the available jingles/snippets/soundbites etc. At any time, you can highlight one or many, then fire them out in sequence or on a keypress to any o/p channel. A controllable preset will automatically drag the volume down of any track being played on the console for the jingle to take precedence. Database -------- Might be a seperate program used to update this. Offers a quick an easy way to bulk drop in a load of audio files (drag n drop typically) and assign them a load of defaults. On the mp3 side, will pull out the name 'tag' info automatically. The key additional other data to be held with each file will be: - preset volume level (a must) - start time (avoiding any silence present at start) space for such things as: end time, fade in/out but these are lending themselves more towards automation - I figure in our use we will be doing more of a 'hands-on' approach to linking stuff together. jrb, 25/03/01

trial4, April 2001 - the jingle consoles are in, as is a lot of the key functionality by this point. Clearly I'm unsure about what's going to go in the middle at this point......
I had a couple of folk evaluating these early trials - me old mate jim who has done some studenty radio type stuff and who at the time was doing IT support for the GWR radio group - so he know's a thing or two I reckoned.... He was forever trying to play as many tracks simulateously out of one sound card, telling me to add colour and "make the buttons bigger, that way the jock's are less likely to arse it up". Some of the comments and suggestions he made came round again recently as I was setting up for the Hastings Rock trial - so cheers mate.
The other guy taking a look has done hospital radio in the past and more recently on the Hastings Rock RSL - he was the catalyst to get OAS Playout on the station so again, thanks simon!
|
trial7, May 2001 - visually and functionally now very close to the finished product, Beta1 rolled out in June as the more subtle features were added throughout my summer 'vacation'. |
Up until the official 'beta' releases Playout worked purely from file directories - the Load buttons in fact activated a standard Windows Load File dialog. In the screen shot opposite you can say I finally took jim's advice and made the buttons bigger.... I took some time out now to start work on the database side - a friend suggested using a general purpose database to hold the music and after a brief foray into the world of ODBC drivers and database connectivity established that I could in fact get my C code to talk to a database fairly easily. The fact that MS-Access could also hook into my C DLLs meant I could get an interface up and running fairly quickly to bulk import all my MP3 files into the system. The Beta's started shipping with two DLLs - one for the 'database engine' and one for the legacy 'files only' - the later now forms the freeware version of Playout. |
Early June 2001 I managed to dislocate my knee, leaving me stuck on my butt for most of the summer. A positive spin from this was the amount of time I could devote to Playout - the Beta 2 release added a lot of subtle features that many people aren't aware of including the right mouse click menus, double click 'fast loading' and property dialogs. It also got left running a lot so many of the reliability issues were sorted out. I also worked the database front end further, adding support for ID3v2 tags c/o of the OpenSource ID3 tagging library. A pre-compiled copy of the tagging (as used with OAS Playout) is available on my main Software page.
|
By the time the opportunity came to trial it on Hastings Rock (Feb 2002) I was surprised and a tad pleased at how few changes they wanted - much of the extras they wanted were things Jim and I had discussed earlier, a favourite being keyboard support, something I'd always pretty much dismissed. Nevertheless a few fairly major functions were added as a result of this notably around the area of cue points - the package now automatically determines the 'best' fade in/out point eliminating the need to manually set these points for most tracks. Check the main development page for current activities and feel free to drop me a mail with any thoughts or suggestions for the future. |
OAS Playout v1.0 - the date/time was a long standing request from jim which finally went in during a late Beta. |
(c)June 2002, OnAStickSoftware, mailto: playout@onasticksoftware.co.uk