Jump to content

Transcode in H265


Snaaaake

Recommended Posts

No need for drama, just limit her bandwidth while waiting for this.

 

Thats what I'm doing for the time being, and was already doing.. knocked it down another 25%, but how long should I keep that up? Til I'm feeding her 420p? Cox isint giving her any forgiveness months so its costing money today, I can hold out for another few months but based upon the gaslighting going on about technical hurdles I'm sceptical any traction is going to be seen in the near future.

 

I left another media server software I had been supporting for a decade and moved to emby because of a feature that other software refused to implement for going over 2 years at the time.. Emby supported the feature (Linux NV Transcoding) so I moved, and aint looked back, because some features are just that critical.

 

My mom is an edge case user, she's mostly bedridden after a couple dozen surgeries in the last several years.. She consumes a ton of TV, gets anxiety being alone w/no TV so its always on.. but after years of medical issues she's basically bankrupt and was hoping cutting the cord was a viable option.. I'll gladly spin up some other software just for her alone if it cuts her utility costs in half, no drama about it.

Link to comment
Share on other sites

Nope not at all, Grandma is going to have to understand what codecs are as-is because there's nothing stopping her from dropping files into a library currently that her server needs to support, today.

 

seems to me, that grandma finds an advanced transcoder feature on her server and decides to start dicking around w/stuff she dont understand is a whole ado about nothing when grandma can easily drop readily available files into her libraries that her server dont support right out of the box running all default settings.

 

Grandma says, "I've always just used Roku because it is simple. Never had to use GPU hardware before but if I can serve to my grandkids possibly over my internet connection instead of having them only able to view at home through my mediocre upload speed that would be nice.".

 

Because grandma tried that and h264 is woefully eating up all her bandwidth with just 2 grandchildren watching. Grandma in this case has 6 grandchildren all under 18. So 6 people able to encode all on grandma's meager 6mb upload with 1mb a piece using HEVC and her opti-core processor, cuz grandma is rich.. cmon it's grandma. Spend it or let it get inherited? Spend it... So grandma would need to understand what codecs are but will need to doubly understand how to ration bandwidth per all 6 grandchildren if all 6 were on at once. But it is a smart grandma she picks up new things fast.

 

She just dislikes when Emby breaks and makes her have to maintain her server to allow her grandchildren to watch. She perfers it not break and they can just watch things. She prefers to maintain Emby when she wants and needs to only. So when new features come out she is hesistant to ever try them. So HEVC transcoding might never get used by grandmas and all the trouble it took to get it there. You have to break that gap, bridge that gap, shrink that gap. So fear and wisdom come together.

 

There is no reason to think people would break their server and then couldn't put it back together themselves. Nor is there reason to believe that everyone can put it back together and wouldn't come here to say, "YOU GUYS BROKE MY STUFF!" and huff, puff, and blow the house down. They have a humpty dumpty and we caused it. All the kings horses and all the kings men aren't cheap. They have hourly rates.

 

So it cannot be made of straw or sticks. Has to be made of bricks to fit 6 pigs. Then little red riding hood doesn't have to come to grandma's house to watch Emby she can watch it at home and avoid that big bad wolf altogether. She can skype grandma and use Emby over the internet. Then little red riding hood can go next door to goldilocks and let her and the three bears come over and watch Emby as a group. Little red riding hood becomes super popular and the sleeping beauty mirror starts to notice. Keeping the wicked witch out of little red riding hood is what the text/instructions/etc would do when encoding.

 

No humpty dumpty(broken Emby server) and must stop the wicked witch(user doesn't understand options) then you have a yellow brick road(path to Oz the great, victory, do not look behind the curtain - it just works).  ;)

Edited by speechles
Link to comment
Share on other sites

Yeah lets worry about made up users from a fictional writer and not the real ones you got here.

 

So wait, grandma wants h265 because she dont have good bandwidth.. but dont want h265 because its complicated? Grandma is rich and can afford the hardware, but is hesitant to use new features and wont use it anyhow, but even if she's not going to use it.. its somehow going to prevent it her stuff from working? I'm not sure what your on about, but pass the douchie to the left hand side, pls.

 

Look, I get it.. I'm a software engineer too, when I fuckup it impacts millions of devices and nearly as many paying customers.. there's at least 4 layers of QA from the code I write to production, and that still dont catch everything.. do we let that fear paralyze us and prevent us from doing our jobs? nope.. Are we forced to compromise on the perfect design for whats actually obtainable? Yes all the time because if we dont introduce capabilities the competition offers the whole company and everyone's livelihood will suffer as a result.. when we can introduce new capabilities none of the competition has, everyone wins as a result.. this is basic dev shop stuff and if Emby had any decent Project Managers they coulda ran the numbers years ago and determined the cost to benefit ratio of being one of the first to support HEVC output woulda been worth the investment and grown the user base.. but now as it stands, its on the verge of causing a decline in user base (and thus MONEY) because the competition is further ahead on this killer feature.

 

I'll continue to support Emby, I was attracted to it because it supported power users and implemented highly desirable features before repeated turd polishing.. however, the message received in this thread is counter to what I've seen as the culture here overall, here the the power users are being sacrificed at the altar of the plebs. One of the longest running Feature Requests here, that never gets far from the top of the subforum, with some of the best justification of any desirable features.. just gets nonsensical excuses, no commitments, and generally far more pushback from Devs/Mods/Admins than would ever be expected.

 

If Emby's Modus Operandi is to cater to the lowest common denominator, then I've been greatly mislead by great plethora of options, flexibility, and capabilities of Emby.. I didnt realize stuff like LDAP Plugins, IPTV, multi-arch/platform transcoding, reverse proxy hosting, etc only exist because your fictional Grandma can figure em out w/out asking for any additional support.

Link to comment
Share on other sites

Yeah lets worry about made up users from a fictional writer and not the real ones you got here.

 

If Emby's Modus Operandi is to cater to the lowest common denominator, then I've been greatly mislead by great plethora of options, flexibility, and capabilities of Emby.. I didnt realize stuff like LDAP Plugins, IPTV, multi-arch/platform transcoding, reverse proxy hosting, etc only exist because your fictional Grandma can figure em out w/out asking for any additional support.

 

Okay.. how else do you quantify the users who don't come to this forum? How do you add in the majority of users of Emby as they are the majority the ones not here right now seeing this post. Who speaks for them?

 

The real ones we have here are also important. But we aren't even discuss importance of users. We are discuss the time it takes to tackle an issue that is so large you need to outsource to get there. I am guessing you understand? But that also isn't the case..

 

It is more than cost-to-benefit and more benefit-to-cost as far as that goes. The users here understand that. The users who do not come here will not. Who speaks for them? The silenty majority.

 

I was trying to add levity but clearly you need to show your educational background? It isn't "easy" when you need to keep that silenty majority who never gives a clue to their wants, needs, desires when you add feature requests like this. 

 

So they must design the UI and basically how to present the ability to the user beyond just a checkbox. The rest I am not on. I just do Roku. Do not take my own opinion as something to run with. It is just MY opnion. Mine to own. I really do want the same as you. But I do not want the added stress of people throwing Roku issues at me caused by directly adding HEVC support in a sloppy, rushed, or hasty manner. Nor do you. Everything has been very open and convenient and nothing was taken away. 

 

Like all feature requests which is all this is. These things take time. Some more time than others. It simply is what it is. You cannot make up your own terms or get upset. It is irrational to get upset. Just say it sucks we don't have HEVC yet and I agree. That is all. It isn't some conspiracy to not do it just because only power users want this. My analogy showed how a grandma would benefit from it too...

 

side-note: serious question, please do no take this the wrong way. I am just curious. How large a team do you think Emby is? The real answer will likely surprise you and I myself do not even know the answer. So that is the suprise. That is also part of this and must be factored into this equation.

Edited by speechles
Link to comment
Share on other sites

Oh I'm not upset, just merely disappointed.. this seems like just the thing Emby woulda had for a while now.. and I fully appreciate the focus the devs spend on fixing bugs over adding features.. quick way to drive users away is not addressing issues.. another thing that landed me on emby was the fast dev candace and vast majority of change log being nothing but fixes.. but at same time cant let perfect be the enemy of good either.

 

As you are a Client Dev, I'm not convinced that this would be any significant work load for you.. Certain Roku's can already direct stream HEVC files correct? Emby Clients are capable of determining this capability already, and have been able to for quite a while now? I'm failing to understand why this feature is anything more than entirely server side? People have proposed options for the clients to force HEVC, but such a feature would come after this one, and such a feature could still be implemented on server side to force feed a user/device HEVC when supported without any additional capabilities from existing clients as far as I can tell.

 

I suppose my expectations for emby may be a bit too high, I adopted this software purely because if its superior hardware transcoding capabilities that blow everything else out of the water.. too see this feature request languish for years, with very little traction is disheartening.. HEVC Encoding is just the kind of killer feature that lets people do more with less, exactly the same capabilities that HW Transcoding currently affords.. and when used appropriately, actually requires less resources, not more as so many in this thread have argued.. just seems to me this coulda been a higher priority than it has been.. I know the day will come when I see this hit the beta channel, and that will be a glorious day.. but with the tone you guys are giving us here, it doesn't seem like its going to be anytime soon.

 

might as well lock this thread down as its merely going to get more contentious as more people have to recon with data caps and poor broadband capabilities that translate IRL into a hidden tax to operating a growing Emby service, from a purely financial standpoint just one single remote user paying +$50/mo for datacap removal would take just 6mo to exceed the cost I paid for my Nvidia Quadro P2000 card, which in turn is more than capable of saturating my Comcast GigE uploads (~40Mbit) outputting HEVC.. and that will translate into more paid subscriptions, after all its a premium feature that people will gladly give you a portion of the money saved to obtain.. how many FR's on this site have such potential? I've yet to see another.

Edited by nayr
Link to comment
Share on other sites

Happy2Play

It is a feature request that will eventually come.  But ultimately a feature that would potentially be useless to me unless I were to add additional hardware.  And that is what everyone will have to understand not just the ones requesting this feature.

Link to comment
Share on other sites

Understand what? That @@Happy2Play finds it useless? 

 

I dont use Cover Art Plugin, Cloud/Floder Sync, Alexa/Home Integration, LDAP, or Podcasts, I find them useless too.. I paid for a premium subscription primarily for hardware transcoding, which amazingly enough required additional hardware.. its those premium features that support this project right? Is that not why HW transcoding became a premium feature in the first place was to increase subscriptions? is money supporting the project useless? If I give money because of a feature, and that money is used to improve a feature you do use, is it still useless to you?

 

its low quality posts just like yours @@Happy2Play that contribute only to the contention and aggravation in this thread.. yes, we got it.. its a feature that will eventually come, thats been clear for years now.. thank you for your service.

Edited by nayr
Link to comment
Share on other sites

Happy2Play

That hardware will be required for this operation as I have one server that will serve this media but will not transcode it at all.

Edited by Happy2Play
Link to comment
Share on other sites

my previous Phenom X6 could no longer serve much anything on x264 either, was unable to keep up with number of transcodes using software alone.. every core fully pegged and the server was sucking down 150W for just cpu.. added HW Transcoder card, system was then basically idle, dont use half the watts it did before and transcodes start up nearly as fast as direct encodes..

 

My choice at the time was switch to emby, spend $300 on a gfx card or build a whole new server for far more money than that.. I bought the gfx card, limped the server along for another year and then build the server I really wanted when I could afford it, and moved the video card over to it.. 

 

Might be time to consider retiring that decade old Xeon, just the power savings alone over the next year or two would more than pay for a new i5 system that supports HEVC transcoding out of the box.. It would crush that xeon in virtually every performance metric.. I've worked alot with video recording servers professionally.. (think security cameras), tossed countless old xeon servers in the garbage and replaced em w/refurbished dell desktops w/Quicksync support.. for like $300 I was replacing a 300W+ monster with a 35W lightweight, that could handle more than twice the cameras with less load and bottlenecks.. Even in parts of the country where electricity was cheap, they would pay themselves off within a year in just utility costs.. there's a reason why any PC recycler will give you stacks of old xeon servers for pennies per pound.. they cost more than they are worth to operate.

 

There's other gains to be had too, modern CPU's have lots of new instructions to help aid with video and network loads making everything more responsive, your battery backup will now last 2-3x longer, idle power consumption is fantastic now days, more cores are standard which is great for server loads, NVM SSDs which are cheaper/faster than SATA variants, etc.

Edited by nayr
  • Like 2
Link to comment
Share on other sites

As you are a Client Dev, I'm not convinced that this would be any significant work load for you.. Certain Roku's can already direct stream HEVC files correct? Emby Clients are capable of determining this capability already, and have been able to for quite a while now? I'm failing to understand why this feature is anything more than entirely server side? People have proposed options for the clients to force HEVC, but such a feature would come after this one, and such a feature could still be implemented on server side to force feed a user/device HEVC when supported without any additional capabilities from existing clients as far as I can tell.

 

When there is a problem with playback of HEVC the simple answer to users on Roku is transcoding. Now if we transcode to HEVC on these problematic Roku this can/will/is problematic. Now the issue is confounded by the fact for the problem Roku only HEVC transcoding is being offered or something to that extent. Then in this case it does present a problem.

 

The Roku app will need to be re-written for its fallback through different encodings it can use when transcoding as well as different playback methods when we get to this point. All the apps will need to have their fallback mechanisms rewritten to support h265 and h264 separately as differing fallback methods. Right now the singular fallback method for transcoding is h264.

 

Also the Roku is problematic with _some_ HEVC files encoded. Especially concerning SDR10 and handles that poorly and can't handle it at all really. But my point is when encoding HEVC the clients all have unique HEVC caveats and dealing with each of those also has to be hand-held delicately. So yes, every single app will have to test its robustness against any transcoding changes it is just the nature of the beast.

 

Change is good though. I am 100% with you that we need to get there eventually. It is just hard to pintpoint when because I have no idea the roadmap. I am just here trying to help giving the best advice I can and that is trust, peace, and harmony. We can all get along and get there together. We just have to remember that it isn't something small this feature request is asking for. It is a big change. With big change comes apprehension, then defensiveness, then denial, then acceptance, then realization. So we are almost there. It just takes time.

 

Everything happens for a reason. Reasons are big motivators. The reasons for this feature request are well documented.

Edited by speechles
  • Like 1
Link to comment
Share on other sites

but with the tone you guys are giving us here, it doesn't seem like its going to be anytime soon.

 

Do not let the past be any indicator of the future. Anytime soon is not a time frame it is ambiguous. Same as the time frames given to you to implement this feature request.

 

If you want to argue about time frames pick a date and deadline. Say by this date you will have your cake. When that cake isn't delivered by the date the cake is a lie. We are not liars. We do have cake.

 

The best managers of projects can only do so much with time. Time is your biggest enemy and largest antagonist. It will always be against you, used to judge you, and used as a gauge of value. But it isn't actually so cut and dry.

 

What you see isn't half, or a quarter, or even 10% of what is in development. You are privy to just a small opportunity to see/hear some of what there is being worked on because we are honest. But there is competition in this space. We cannot share everything. It is very competitive. If the competition has this in a fairly robust manner than anytime soon will be sooner than you think. You just need patience. Less anger and just be rational and let this just play out and see where it goes. It has a happy ending like all fairy tales. :)

Edited by speechles
Link to comment
Share on other sites

wiwolavida

I wonder what comes first. This feature being implemented or me getting ftth and making this feature mainly obsolete...

Link to comment
Share on other sites

I wonder what comes first. This feature being implemented or me getting ftth and making this feature mainly obsolete...

 

Such negativity and you wonder why you chase developers away from commenting. It makes them feel they are stepping into dog mess and then leaving footprints of that mess everywhere. Instead try to get them interested in getting this feature implemented. As @@nayr has done. Not sarcastically taking cheap shots just for the sake of cheap shots. Part of the solution or part of the problem?

 

Btw @@nayr you are a fine gentleman and worthy of this debate. 

Edited by speechles
Link to comment
Share on other sites

daedalus

Such negativity and you wonder why you chase developers away from commenting. It makes them feel they are stepping into dog mess and then leaving footprints of that mess everywhere. Instead try to get them interested in getting this feature implemented. As @@nayr has done. Not sarcastically taking cheap shots just for the sake of cheap shots. Part of the solution or part of the problem?

 

Btw @@nayr you are a fine gentleman and worthy of this debate. 

 

there are people that simply finally have had enough and the devs had put them off

every "cheap shot" just shows how frustrated they are

  • Like 1
Link to comment
Share on other sites

Such negativity and you wonder why you chase developers away from commenting. It makes them feel they are stepping into dog mess and then leaving footprints of that mess everywhere. Instead try to get them interested in getting this feature implemented. As @@nayr has done. Not sarcastically taking cheap shots just for the sake of cheap shots. Part of the solution or part of the problem?

 

Btw @@nayr you are a fine gentleman and worthy of this debate. 

 

Here's some positivity to counteract that negative stuff:  You guys have a great product - so much better than your competition.  Can't wait for this new feature to be rolled out, making Emby even more awesome!

  • Like 1
Link to comment
Share on other sites

SikSlayer

Crap, I had a whole long rant about how much I love Emby and all this and I lost it due to a simple mis-click.  :angry: (Stupid auto-save didn't kick in in time)

 

I wanted to dog pile on the positivity train WGB123 started. I LOVE Emby, I've been using it since before it was called Media Browser in the Vista days. I want to see it grow and improve.

I STILL feel shitty that it took me so long to be a lifetime subscriber. Pretty sure I made a thread on the subject.

 

I positively HATE that Plex gets the press and love for something you guys pioneered first.

 

But put simply, your most involved and users and paying customers feel like they are being talked down to. We see feature requests either answered with canned responses, or what comes off as apathy. There (is/was?) a whole thread about it.

 

But it's speechles's posts that has me posting myself. Please don't think for a second that I and others don't understand where the developers are coming from. It's understandable that you don't want to show your hand before you play the cards (or however the phrase goes), but this HEVC thing is a TEXTBOOK example of what I and others have an issue with. As you say, it's well documented why this feature is needed, but seemingly useless debate comes first for what should have been obvious even when it was first requested.

 

I trust that HEVC will be implemented as requested sometime soon, but at this point I hope even more so that this issue isn't repeated for AV1 specifically, and that this is a positive learning experience for both users and developers going forward.

 

PS: I just happened to read more into on the AV1 subject, and I feel I have a more realistic expectation on when it's feasible (more like 2021/2022 than late 2020). I'm also willing to bet Nvidia won't have FULL HW decoding (let alone encoding) on their video cards until MAYBE AFTER their next line of cards on whatever die shrink they go to next. It might be mixed first. By then, Intel might finally get out of their own rut, and have decoding support the year after that.

Edited by SikSlayer
Link to comment
Share on other sites

KarlDag

Understand what? That @@Happy2Play finds it useless?

 

I dont use Cover Art Plugin, Cloud/Floder Sync, Alexa/Home Integration, LDAP, or Podcasts, I find them useless too.. I paid for a premium subscription primarily for hardware transcoding, which amazingly enough required additional hardware.. its those premium features that support this project right? Is that not why HW transcoding became a premium feature in the first place was to increase subscriptions? is money supporting the project useless? If I give money because of a feature, and that money is used to improve a feature you do use, is it still useless to you?

 

its low quality posts just like yours @@Happy2Play that contribute only to the contention and aggravation in this thread.. yes, we got it.. its a feature that will eventually come, thats been clear for years now.. thank you for your service.

You have a terrible attitude. Your use case is not everyone's, and other people want other things just as much.

 

Devs said it's coming, if you can't wait then move on.

Edited by KarlDag
Link to comment
Share on other sites

daedalus

You have a terrible attitude. Your use case is not everyone's, and other people want other things just as much.

and just because he has a different use case he's not allowed to be upset for not seeing progress on what he paid for?

 

he's just a year here and already knows what others know for years

 

 

Devs said it's coming, if you can't wait then move on.

we have waited and still wait for partly dead simple things for literally years, i totally can understand him

Edited by daedalus
  • Like 1
Link to comment
Share on other sites

wiwolavida

Is there an update on when this feature will be released so I can convert my videos to h.265?

 

The update was: "we might be working on it."

 

If you only want to convert your videos, you should use handbrake. It supports HW acceleration. You just create a profile and drop your videos in a folder watched by handbrake. It will convert them automatically. It's available for all OS and Docker.

 

Also this feature is for transcoding your files to offer h265 with a certain bitrate for playback and syncing. It's not indented for reencoding your whole library to h265.

Edited by wiwolavida
Link to comment
Share on other sites

If you could use it for syncing then probably you could also use it for converting as well and therefore it could be used for reencoding. ;-)

  • Like 1
Link to comment
Share on other sites

  • 3 weeks later...

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...