Leaderboard

For those who are seeing extra top level folders pointing to live tv, this has been resolved for the 4.8.1 maintenance release that we'll be releasing in the coming days. Thanks.

Is there any way to revert to the previous behavior of missing episodes being listed among all the present episodes in the Season view? Or allowing use to choose the behavior we want. Having to go to the 3 dot menu is really counterintuitive. This really is only useful if you already know there are missing episodes. If someone is watching a series they have no way of knowing if an episode is missing and just continue on with the "Next Episode"

Device Locations Plugin for Emby Purpose This plugin improves observability of users accessing your server remotely. All private ipv4 and ipv6 ip addresses are ignored. help understand if users are sharing their accounts with other people additional stats are added to help understand device behavior when are devices active? This can help tune servers during peak activity which areas have the most activity? This may inform content delivery strategies which users authenticate on devices? Installation Download the dll and move it to the plugin directory Restart your server Requirements Register for a free account at https://www.ip2location.io/ to obtain an API key used for enriching location data If you are using a paid account you can remove the attribution in the plugin settings, otherwise, you need to keep the attribution to remain compliant Update the plugin settings with the API key Enriching data There are 2 scheduled tasks that need to run before full results are available. Sync activity data - Runs hourly and syncs Emby activity events to the plugin Enrich location data - Runs daily and enriches geoIP data to the plugin data. You can update settings to configure the max number of requests to ip2location PER run. This is useful if you're testing and/or have a significant number of authentication events (large user base, etc). The ip2location API allows 30k requests / month. TODO: Add ipv4/ipv6 validation when sync'ing Emby data Add caching when enriching data EmbyDeviceLocations.dll

What I would like to see is user groups in general. Create a Family group of users or a Kids group of users and assign that group, access or restrict that group's access etc Once you have groups, you can then start to think how they could be used in all sorts of places in the server. But there is already an open request for that.

The logic is pretty simple really. Being able to give users the abilities to share playlists with other users where it makes sense, and not otherwise unnecessarily exposing usernames. Not really at all. I've outlined several reasons why this is a desirable feature that don't simply equate to "I don't want my wife to know my mistress is on the server" That kind of wording is extremely dismissive despite numerous valid reasons that have already been provided as to why this would be useful in a variety of situations. Unnecessarily exposing all usernames on the server(especially admin accounts) to all other users is just bad opsec in general, and there are a number of ways this could cause problems and frustration for the server admin or amongst their users. Imagine having friends or relatives that you get along with, but they don't get along with each other. See how enabling this could get abused? Cousin Larry hates cousin Fred, so Larry makes 20 different playlists of Golden Girls and shares them with Fred. Or better yet, say the admin has fail2ban monitoring failed login attempts for emby and Fred thinks it's funny to get Larry's IP banned. Or the admin is using AD authentication (like I am) and little nephew Johnny thinks he's going to try to guess uncle's password. Next thing you know uncle admin is locked out of his computer because there was a lockout policy set in AD after 5 failed attempts. I get that the playlist sharing is an optional feature, but it's a very useful one. It's usefulness is greatly diminished though by the fact that anyone can share anything with anyone when enabled, and can also see all other accounts on the server. You guys are missing the point that giving server admins the ability to better control permissions while still being able give their users the ability to use a highly desirable feature would be a good thing.

Option 29 on page one here. https://emby.media/community/index.php?/topic/89802-et-for-android-and-et-for-windows-ui-match-android-tv-app/

To me, this is a back-door around the options we have in the server to hide users from login screens. Those options become kinda useless if we're just gonna allow anyone to see the entire user base with a right click...

I guess we could add a top option labeled "Original quality" that uses an astronomical bitrate number for the limit but, personally, I think that should come with a warning that it may cause playback problems. Also, it shouldn't be labeled "Direct Play" because other factors may require repackaging or transcoding.

While code complexity is definitely a big factor in implementing this fully, at minimum being able to hide admin accounts and perhaps other select accounts from playlist sharing would be an ideal compromise. As far as a full fledged implementation goes, (assuming that it can be done without adding too much complexity or introducing other issues) I think it would be ideal to have playlist sharing options have its own page much like library access controls do. That would provide separation from a UI perspective and make it more intuitive for admins to implement the per user controls for who a user can share playlists with. @TeamB@Luke I also just want to make it clear that I appreciate all the hard work that goes into emby from a development and support aspect. I'm definitely not trying to be that guy that demands that my feature request be fully implemented and fleshed out exactly the way I want, and get buthurt when/if it doesn't. Just trying to make suggestions that can help the versatility/usability and possibly security where applicable. I've been using emby for years and decided I should probably start contributing on the forums to help others and try to help make the product better overall. So again I'm going to say thank you to the whole team, and to all the regulars who help keep the wheels turning here.

@Happy2Play + @YellowPotato: Thanks for the logs. Unfortunately, the only insight they gave is that it's unlikely due to hw detection. Now, both theories are busted and I don't have another one right now. Generally, I don't think it's worth putting much more effort into this, because there's no perspective for Emby Server on Win 7. If it wouldn't be this time, then it would still be very soon to break finally. A net8 update is just around the corner and that will most likely be the final RIP anyway. And even for those with a light-hearted attitude towards security: Running an OS with a multi-year gap in security updates is insane, not brave - so it might be time to update the OS now instead of entering more trouble.

Hello everyone, I confirm the same problem and error in log and event viewer with version 4.8 and Windows 7 x86 updated. A clean installation doesn't solve the problem, but by installing the latest 4.7.x everything works but I don't have time to block the updates and the first time I restart Emby it updates and stops. Regards.

This forum is concerned with the streaming software provided by Emby; it is not the place for phone questions. Paul

Hi, Just upgraded to 4.8.0.80, and I notice a different language used in the activity logs. My activity is in English (probably because all my setup is in English (usgin Kodi+Emby plugin), but my friend have his TV set in French and using emby apps on it, therefore the activity logs on my server is in French for him. Should not be in English since my server is setup in English not French.. should not depend on the device, but on the server I guess! Joel

Think I've found a bug. I use user access to control which folders a user has access to with a library. Although the user profile settings show this to be correct, in practice through the client the user has full access to all folders in a libabry.

Thanks Found it!! That was it. It was transcoding the subs and failing

Brilliant. Thank you. I learnt something. Works well.

Some of the config endpoints now support updates via partial objects, so we will eventually get to the point where that can be done everywhere.

I have yes, emby is in large about the music for me, hence my decade of whinging about playlists on and off mate haha i have seen faves down there, was just an observation that the option isn't in navigation but it is for all the other libraries, anyways thank you again for making playlists work like playlists should onto the android tv bug that's been annoying me forever now, but i'll let ya catch up with all the other 4.8 issues first lol

you have to include the complete payload

Under Network in Admin Panel, uncheck this box.

Hey Team, Just wanted to say thank you very much for this guide! I upgraded to DSM 7 without following any of the early steps, and wiped my Emby settings. The guide was really easy to follow and seems to have fixed everything for me! Thanks you!

Awesome, thanks for letting me know!

Great to hear, thanks Luke!

to be honest, he knew about this issue many months before 4.8 stable released.

Two things to know: The pin is not the same anymore. It is now called a Profile Pin and has a slightly different function, and this is explained on the user password configuration screen To use it on the Shield, install our standard android app, which is also in the store for it: https://emby.media/emby-for-android.html

Right I would strongly suggest not using this feature. It's something we should have never added but it's been hard to get rid of due to the few people who do use it.

Can you please attach the emby server log? There could be something else going on. Thanks.

Hi, yes, Youtube is likely the direction that we'll go in.

Installed via usb version 1.7.4 and the issue is resolved. I’m assuming you’re just waiting on Samsung to approve the new version for the tizen app store? Thanks for the help!

So we're all stuck with the lowest common denominator? Where did I leave my Windows XP CD? Probably with the MS-DOS 5.1 floppies Why do you even try and support users on OS that went EOL 4 years ago? Crazy!

I was away and when back server had fully started and was operational Not sure what was the problem but it looked to have fixed by itself or an upgrade of something due to the update that took time ? Sorry for the noise !

nvm.. sorry Luke.. i have a ASUS PRO WS W680-ACE IPMI... and the IPMI Card disables the iGPU after BIOS Update... so its not a Emby issue... my renderD128 is gone... I think everything is fine i have to fix this

That means you don't have a poster for the movie. Try using the Identify feature on it and see if that helps pull in an image.

I actually went through the web browsers of emby and into the plugins catalogue. The plugin still shows up and you can click to install and click to pay for it. I paid first via paypal, and then installed the plugin as I had used this before. So, a note to emby Team. Atleast someone make a note before others decide to pay for it to not get anything. Im not moaning about the $3. Its just not good service for emby to have a plugin avaiable that you pay for that doesnt work.

Imagine a TV in a living room of a family with two kids a mother and father. The kids accounts don't set the PIN and the kids can open Emby on the TV and freely select their account without any additional auth prompt. Mom and Dad both set a PIN on their account so that, if the kids attempt to switch to their account on the TV, they will have to provide the PIN. Make sense?

2.3 and 2.4 are only compatible with the stable 4.8 release, which is 4.8.0.80 I've made the minimum build more specific on the catalog, so that should stop Emby from trying to install them on your beta 4.8.

Exactly like with phpMyAdmin for those who know. A simple QRcode allowing association with an application that generates random codes like Google Authenticator. It is therefore Emby which generates the QRcode and Google Authenticator which receives it. Easy as pie

......that and the remote access........my data gets from a to b and it all happens in a happy magic cloud...yup....

and that is why i wont use plex

@krislemkeignore me. I found a typo in the GIDLIST on my docker-compose file for the render option. After correcting that (as well as updating to the latest beta), all of the hardware decoder/encoder options showed up and it's working flawlessly now. Transcoding a 4K H265 stream down to a 720p H264 stream used to push the CPU to around 80% of its capacity and now it's < 5% utilization. Quality is still really good even on the high speed settings, so I'm super happy with the performance of this little N100 processor.

@krislemke, what does your transcoding page look like now that it's working? I made the jump to the latest beta version as well and now I can see two software encoders showing up - are you seeing the N100 register as a hardware decoder or encoder? I tried transcoding a few different videos too and never saw it leverage the hardware acceleration during playback.

Hi, we are looking into this. Thanks.

This can be a problem. You often do get a response from Emby folks, but it can be frustrating. So you have balance quality of support with any support. I am grateful to all of the users that lend support to other users.

Great! I've just released a new (beta) version to work with the new 4.8 notifications system (https://github.com/tielvh/emby-ntfy-notifications/releases). Tested with 4.8.0.49

Hi, rookie question from new Emby user - is there a way to change the order of Songs/Albums/Appears on the Artist page? I personally have a strong preference to have the Albums listed first. Thanks!

Hi Emby Team ! I'd like to know if it would be possible to make the Music Artists, Genre, Albums bar editable in the admin. For example I don't need the composers to appear or I'd prefer to see the Genre to be located before the Albums tab. That would make the search of music quicker mainly on my phone as you have to scroll right to make the genres appear. Thanks in advance :)

I wanted to continue on some of the ideas we had way back on the Audio fingerprinting segment detection which could be used for things like Intro detection, commercial detection or even areas of interest detection in audio and thus in video streams that have audio. Chromaprint project : https://github.com/acoustid/chromaprint The idea is to use a tool like ffmpeg to extract what is called a chromaprint of the audio, this is a stream of data that represents the fingerprint of the audio over time. You can then compare this simpler representation against other chromaprints to find similar sounding audio segments. We started some of this work over a year ago on this thread: https://emby.media/community/index.php?/topic/48304-show-intro-skip-option/&do=findComment&comment=952613 And found some interesting ways to use the data. The first being the ability to detect segments of the video using multiple episodes of a tv show the intro theme music that could then be marked to allow auto skipping. I found this work interesting as it is kind of what I do, look for patterns in very very large data sets on a daily basis. So I went back and had a look at my original python that compared some of the bitstreams and found it was very slow, fine for a POC in small scale but very slow. The main cost of doing this is in the bitstream comparisons needed to find aligning (similar) audio segments in two potentially large chromaprint bitstreams. The approach I were using to find the bit difference counts was brute force and inefficient. I went on the hunt for a more efficient hamming difference and fond a bunch of bit diff work that pointed to a bit shift approach that is very much faster. So why all the fuss about this one aspect, well it is super important if you want a performant solution as a LOT of the heavy lifting in finding audio overlaps is comparing a lot of data and if you can do that as fast as possible then it will allows you more options, more exhausting searching etc I have implemented a simple hamming diff test using my original brute force apporach and the new bit shift approach and it does make a big difference, about 86% or 6.8 time faster. This bit shifting approach comes from stackoverflow, from a post about fast bit Hamming calcs in Matlab (a data science tool) https://stackoverflow.com/questions/1024904/calculating-hamming-weight-efficiently-in-matlab I just adapted it to python for my test but it could be any language that can do bit shifting. here is the python code: import time # simple brute force def count_bit_diff(left, right): count = 0 for i in range(0, 32): if ((left >> i) & 1) != ((right >> i) & 1): count = count + 1 return count # https://stackoverflow.com/questions/1024904/calculating-hamming-weight-efficiently-in-matlab # http://graphics.stanford.edu/~seander/bithacks.html#CountBitsSetNaive def hamming_diff(left, right): # w = bitand(bitshift(w, -1), uint32(1431655765)) + bitand(w, uint32(1431655765)); # w = bitand(bitshift(w, -2), uint32(858993459)) + bitand(w, uint32(858993459)); # w = bitand(bitshift(w, -4), uint32(252645135)) + bitand(w, uint32(252645135)); # w = bitand(bitshift(w, -8), uint32(16711935)) + bitand(w, uint32(16711935)); # w = bitand(bitshift(w, -16), uint32(65535)) + bitand(w, uint32(65535)); distance = left ^ right distance = ((distance >> 1) & 1431655765) + (distance & 1431655765) distance = ((distance >> 2) & 858993459) + (distance & 858993459) distance = ((distance >> 4) & 252645135) + (distance & 252645135) distance = ((distance >> 8) & 16711935) + (distance & 16711935) distance = ((distance >> 16) & 65535) + (distance & 65535) return distance # confirm approaches are the same through the 32 bit address range points = 20 skip = int(0xFFFFFFFF / points) loops = 0 while loops <= points: right_val = int(loops * skip) diff01 = count_bit_diff(0, right_val) diff02 = hamming_diff(0, right_val) loops += 1 print("%s %s" % (diff01, diff02)) # do some timing comparisons # brute force started = time.perf_counter() for x in range(0, 2000000): diff01 = count_bit_diff(0, x) finished = time.perf_counter() print(f"time01 {finished - started:0.4f} seconds") # faster way started = time.perf_counter() for x in range(0, 2000000): diff01 = hamming_diff(0, x) finished = time.perf_counter() print(f"time02 {finished - started:0.4f} seconds") Output time01 19.2995 seconds time02 2.8164 seconds That is some nice improvements. Now that 4.7 is getting close and this version has the new ffmpeg with the chromaprint module compiled in I might get back on this to have a play with what else this sort of thing can be used for. I know the core Emby will soon have intro detection and skipping built in so that might not be as pressing but I am sure there are other areas of investigation. I am not going to write a plugin, this post is just for reference to continue on some of my thoughts on this. I will probably just add to this thread as I play with this. Repo with a few of the projects and tools i have built. https://github.com/faush01/ThemeService

I would find it helpful to know this information at the album level (and album reports interface) - not so much at the track level. Could greater info in this FR be provided without significant bloat to the DB or impact to UI browse performance? (some may have an order of magnitude more tracks than episodes and movies) But albums are made up of multiple tracks;how might album info be derived? From the 1st track? Would be highly unusual (negligible?) for tracks on same album to be ripped at different settings? or use "mixed" label to denote when track fields are not 100% consistent within the album. Interesting info for me is bit depth, sample rate, channels, codec, size

I also would like to hide tabs from the music library, for example, I want to hide the "Favorites" and "Album Artists" tabs.