Jump to content


  • Content Count

  • Joined

  • Last visited

About Pazzoppe

  • Rank

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Appreciate your great and constructive feedback, Cute. After the pushback on this forum I posted at some other software development sites. Got positive response within minutes and am now moving forward with a demo version. Mock-ups are already done. Cheers
  2. Thanks for your insights! Update: I just dumped Plex because it basically is an audio convertor (which I do not need for my Flac files) and memory hog on my older Synology NAS. It does not offer anything close to the library management capabilities I want. Roon still is a favorite. Their architectural approach is logical and may result in the long term solution that I'm looking for (https://kb.roonlabs.com/Architecture). It's not there yet. Their architecture would even be better if they include the following components: Server software. This software will need to be multi-OS (Win, iOS. Linux, and preferably also Android). My preference is for a Linux stand-alone solution. I started using my NAS but prefer to use this for its data server and backup functions rather than keeping the NAS processor and disk(s) busy with streaming. I do not like a dedicated a PC/laptop to run the server software: too much overhead on the OS side, upgrade issues and hardware issues such as fans. I'll experiment with RaspberryPi / Linux standalone box in coming weeks. Here are the basic server-side software building blocks: Access module. This enables the access to media files using multiple sources and systems; NAS, USB, Cloud, etc. This module will need to updated over time to support changing data input mechanisms. Library creation and management module. Roon architecture offers the basic elements (discovery, indexing, meta, and perhaps some background analysis). They forgot the item that I consider key: Library organization. This is the part where the user can organize the indexed data any way they want, without being forced into the standard album/artist/genre/etc approach. Library organization can be done using pins as done by Pinterest or drag/drop into a series of categories (boards) each representing a part of the audio collection the way the user likes it. Identical to apps management on smartphone/tablet. Creative users can design category screens with subsets of the library to fit their personal preferences (or for DJs to prepare for a concert). Selecting from e.g. 1000 albums is much easier if you have created your own dozen screens with each up to 30 or 40 albums ranked left-right in order of preference, the ability to group albums within one screen icon, and hide tracks on each album (Note: this has nothing to do with AI: intelligence is in the user). Investing a couple of hours to organize your favorite music saves multiple hours of text searching in the future when you want to create new playlists on a daily basis. A Playlist bar on each screen (identical to the taskbars on Win/iOS/Droid systems) allows you to drag/drop items from that or other screens to create the playlist(s). The graphical approach is far more intuitive that the stone-age text approach (I don't think anyone still selects apps on their computers/tablets by typing in the program name on a search screen, so why continuing this old tradition for media files). Of course, it should be possible to (re)organize playlists the same way as organizing the category screens (move up/down / insert / delete etc.) Playback Module. In its simplest form this could be done by pointing to an already existing app on the system. Roon adds Zone playback and DSP capabilities which makes sense as long as it is possible to eliminate these items to minimize processing requirements. For my Flac files there is no need to do any additional processing. My existing hifi system takes care of room acoustics and other local stuff. Output Module: points to the desired DAC location. Can be done using USB or networked/i2s solution. For my trial solution I use Chromecast audio connected via cat6 LAN (thereby avoiding wifi issues). Easily upgraded to higher quality networked DAC later on. Client software: Remote control of the Server software. Needs to run on Android, iOS and Win. Connection into the Server side is network (wifi) based. Client software should only be needed to setup and change the server based modules. It should not have an active role in playing music to avoid that music stops when the remote is not visible for the server. A web browser approach for the client offers the most long-term solution and avoid Android/iOS dependencies. Example: A nice (Lumin-like) stand-alone solution that I installed with my in-laws last year suddenly became useless when their iPad automatically updated to IOS9. Apple eliminated some control capabilities in the new OS in favor of their own hardware solutions. Android is less risky and problems are more easy to fix, but not fully future proof either. What's next? I firmly believe that search/text selection of audio files from a bare bone library index is something we should have left behind when the first iPhone was released ten years ago. After searching for several weeks and asking suggestions on this and other forums I conclude that a graphical library management tool for audio hasn't been built yet. So, I'll build it myself. Initially I can make sure all my audiofiles are indexed in an SQL (lite) database with some PHP and standard graphics tools for library management and playlist creation. Playing can be done via an existing playing app. Most logical place to implement this graphic library management tool is with the Kodi open-source environment. That's phase 2. For hardware I'm considering a HiFiBerry with audiosonics sabre DAC. Cheers
  3. Roon is interesting Freann! What I immediately liked is that they also move a step forward by enabling associations based on usage. The fact that it runs multi-platform / NAS is also a great plus. Not exactly the graphics approach I envisioned but their approach might also save time when creating playlists. Thanks.
  4. Good points Cute. The database can be represented in several ways, cross-views, so to speak. The drag/drop relates to creating user-based views that are not tied to the basic mp3/vorbis tags. Users can create these views to structure the music the way they want to see it. From these user views they create the play list(s), or let the software automatically create a play list because the user already combined the music into favorites sections. Some players use a "mood" tag to approach this. JRiver moves a tiny bit in that direction through their custom views. These are, however, created by combining search expressions so it is not possible to add a bunch of handpicked items. In a graphical user environment these views are created by dragging the items onto a custom screen/view. For large quantities of files, a combination of mechanisms will be needed, so first use a search term to reduce the amount of displayed items and drag/drop the preferred items into the custom view. The key reason for creating these custom views is that, once it is done, it makes creating play lists so much easier and faster. Current players require the user to go through the same motions each time they want to create a new playlist. In addition, the user also needs to spend quite some time working with the tagging software to make sure the information fits the tight structure. A highly inefficient approach. A graphical solution eliminates this need because the tags are much less relevant. The retagging effort is only useful for portability purposes. I am not inventing anything new here because this is how some of the e-book software apps work. You're right in your observation that this cannot go down to the track level because of artwork. The graphic aspects stop at the album level, unless there are underlying video clips. For a private collection we probably talk about less than 1000 albums, from which a few hundred favorite albums are selected for display in ten to twenty custom views. One level deeper the user can hide/unhide tracks. SQL (lite) is so easy to implement and comes with so many display related tools and portability advantages that I probably would not bother to find a hardwired alternative, but you are correct in that it is not essential. A similar approach is done by Pinterest, where the user creates custom "boards" by "pinning" multiple images from a collection. That will also work well with audio albums.
  5. Well Kelly, if you did not even get this analogy I suggest you focus on other posts. In my initial post I have given quite a detailed overview of what I expect to see in this market in 2017. The only thing that I could further have included are a couple of mock-up screens but I assumed this was so self-explanatory that I didn't bother to create them. If you want to have a clue: A course in SQL and associated BI /UI presentation layer development might help (assuming you would know about the logic/PHP tier in the middle).
  6. Apologies Moeb, this type of stuff is in my profile but apparently not visible. 1. The "player" for me is the software part between the data storage (NAS, Cloud, USB, etc) and the DAC which turns it into an analogue system for my old but still very pleasing audio set. 2. For music management I've tried many packages. I forgot to mention Picard, Plex, Fritz!Media, Foobar, Monkey. For the time being I use Plex with a very much organized audio folder system on my NAS as well as Kodi with 256G of USB on another system. I use these for music as background. When I want to get a maximum music experience I use a CD. 3. I have an ICT background so am flexible on OS side. The NAS (usually Linux based) has my preference. I have two of them and they already run 24/7 so prefer these over a Windows/iOS solution. Remote control apps need to run on Android and iOS. Especially for this last one: they need to be compatible with multiple releases to avoid that the user can no longer install the app because he/she has an older but still perfectly good functioning iPhone. 4. Budget: Typically these type of players are free, but for a (software) system that has the right functionality and no ads I think it is reasonable to pay a price that is in line with other dedicated audio software products, e.g. Adobe Audition. 5. See 2. I should add the MinimServer that actually offers great search capabilities. JRiver also has enhanced search mechanism using regex but to me that is just an improved way of doing something oldfashioned. Thanks for your suggestions! For sure i will give these a try. Cheers
  7. Appreciate the advice! I may not have done a good job to explain my media player (software) nightmare, so let me rephrase it. You’re away from home and walk into a supermarket to buy these great tasting little sweet green-red apples. You forgot their name and where they came from but they are very easy to find in the fruit department. To your surprise they stop you at the entrance and tell you that they now have a modern shopping experience where you no longer need to go into the store but can order everything on-line with your smartphone. So you type "apples", which results in a list of dozens of different apple types as well as iPhone cables and plugs that is impossible to browse through. If you type “red” you get an even longer list including tomatoes and peppers. They want to you type in the country of origin, so you take a guess and type “Chili” and get a list of 250 products that makes it even worse. You leave the supermarket and try the one next door. Here, they have an advanced shopping experience, allowing you to browse for several characteristics in one search line. How cool! So you type “apple, green, red, swet” and get a “no response” because you forgot an “e” in the last word, but the system is very smart and suggests several anti-transpiration solutions. You try again and the system tells you you need to decide between “red” or “green” and locks up. The supermarket manager looks at your phone and tells you that they do not yet have a 100% fool-proof solution for your recently upgraded phone operating system, but they'll probably have that next year. He suggest you to give it another try. You're getting hungry. After rebooting the user app and finding the proper wifi network again you try “greenish-red” but that does not give a response either. The supermarket manager tells you that your best chance may be "red-greenish", but only in the new beta software version that has not indexed his store yet so you need to wait max 15-20 minutes. You give up and walk around the corner where you immediately see you favorite apple on the fruit stall in front of a small store. Hope this helps to envision why I want to browse my library using a touch-screen drag/drop graphical user interface in line with the technology we have all embraced since iPhone was released in 2007. FYI: a small to medium sized supermarket has around 20,000 product, same as my media collection.
  8. You're right Kelly. The Lumin D1 is an excellent product. If only they would offer the graphical library solution that has become the standard for iPhone and iPad (or android equivalent) products. I've spent weeks retagging my audio and organizing my files in a custom folder structure to cope with the rudimentary capabilities of present players. But I prefer to have a state-of-the-art operation solution. I discussed it with Kodi developers. They are considering to create a graphic add-on to improve library management. It basically boils down to adding a flexible user presentation/input layer tied into the existing SQL database. Not rocket science as you can see from the multiple solutions existing for e-book management. Reason for my post is to see if it perhaps already is out there.
  9. Thanks Elvia, I checked out JRiver. They offer several interesting display views but no flexible library management. the last software that more or less cam close to that was MusicMatch Jukebox in 2008, before Yahoo started eliminating this and other great features.
  10. Hi Kelly, I know Lumin and many other of these 'boxed' solutions have great reviews in the press. Soundwise they are excellent. But when I read the complaints on the forums from the users about having to reboot the system and/or the remotes because they keep on crashing or lose wifi connection, or users who beg to get another upgrade for last year's system that is now considered obsolete by the manufacturer I lose interest. For the price I would actually have expected Lumin to pack a high-end graphical user interface with their system. To me they appear in the hardware category. I want expertise in the software area. Cheers
  11. Thanks Cute, It looks like you've put a lot of work in managing audio quality in your software. Unfortunately this is not what I am looking for. My interest is in the graphical management is the music library. I see current media players as similar to 'smart' phones before 2007. I want an approach that is similar to what iPhone did when it introduced their fist phone ten years ago. Cheers
  12. After flaccing my CD collection I am trying to move to a media player solution. Given the limitations of boxed solutions I opted for the NAS–Player/server–DAC approach to connect into my older stereo system. On the storage and DAC side I am fairly OK. I start with Synology NAS and use Chromecast Audio at the other end. Once that operates smooth I will start replacing elements until quality gets sufficientlly close to what I get from my CDs. Key open issue is the player software. Being used to full graphics software solutions I am surprised by the lack of evolution in this area. I’ve looked at a dozen players named on this forum. Many of them appear to be great in expanding on all kinds of side functions like DSP but none of them seems to master the core capability of music selection. They get stuck on the basic mp3/vorbis tagging functions, with at best the ability to include more complex search capabilities. Done by typing text. Huh? No drag/drop on touch screens? I thought we went past that when touch screens were introduced almost fifteen years ago. In trying to fit the medieval tagging model I've already wasted more time retagging my collection than I spent flaccing it. I want a graphical solution. I don’t want to search 10,000 files typing in text in a smartphone or tablet. I want to search audio files the same way as I browse CDs in my cabinet, or books in my library. I have thirty pianists performing Debussy preludes. In my cabinet I group my favorites into one section of the Debussy collection, and rather than typing in some unspeakable pianist name I know from the color on the back side of the album that it is this performance that I want to hear. I can do exactly this with my e-book collection. Why not the same for audio? Player requirements: Present the indexed information initially in one or more overview screens/pages (albums, artists, etc.) on my remote (smartphone/tablet) identical to the app overview on android. Alphabetic is good enough. Next I want to create a series (10+) of personal screens/pages each focused on my preferred music display. One for Debussy, one for African Jazz, one for Metal, another one for recently bought music, etc. If I wanted to recreate my wooden CD cabinet that should be feasible as well. From the general overview screens I can now drag/drop music icons into the personal screens to build my music collection. Some icons I may want to use in more than one personal screen. Nothing new: We all do this with apps on our tablets. Of course these are only graphical representations of a database. The actual files stay where they are. Just like on any tablet screen I can further create grouped icons. Getting rid of a music icon (which could be an album or song) from a screen should be as easy as dragging it to a “remove” icon, like on an android (or iOS) screen. If an icon is an album I can click on it to open it and hide those songs I always hated but had to suffer through because they were part of the album. Again, the original files are not touched. If I want to delete the audio file I can drag it to a garbage icon in the general alphabetic music screen. Each screen/page has one or more “playlist/queue” icons allowing me to drag/drop an icon for playing or create a list for future use. I can quickly browse my entire collection by swiping back and forth between my screens to select items for the queue. Speech control is an obvious adder functionality. Import/export: XML-based. No different than other database structures. As you can see, these items are all very standard and have been implemented in many forms. Take a look at the software solutions that exist for e-books and photo albums. It should be straightforward to create something like this for audio. Have I overlooked a player with this capability or are we still in the stone age of players? Appreciate your suggestions! Cheers
  • Create New...