Improvements to data saver option, servers and recruitment for a dev to work on...

[77BLc9JwJGpLVARkamJx]

Gonna try this data saver and see if those text heavy stuff affected or not in fit height reader setting.

 

[Krsomz]

I turned the data saver on and tried reading a chapter. Each page of the chapter was the same as page 1 throughout. Maybe I’m
doing something wrong.

 

[bane_ecf]

Wow the download rates are much, much better right now. Thanks!

 

[ebullient]

Serious question, why Java? I could do it in any language besides that.

edit: ah, h@h requires it. OK.

 

[Holo]

It doesn't necessarily have to be java but the client code is open source and is written in java. The idea is that you don't reinvent the wheel and modify that code.

But if you want to start from scratch that's also fine, but time is also an important factor.

 

[rspreet]

If it is tied to device, this means I have to turn on the data saver mode on every device, even if I am using same account on all of them. Right ?

 

[pikokola]

@Holo
You need to consider that building the M@H isn't the only step you need to take to make it work.
We need people to run the client too, and looking for dedicated client hoster around the world isn't a small task.
EH give many incentives to run it, the hath perks for example.
I don't want to pester you for more feature to be implemented as an incentive for them, but I believe it also needs to be considered as soon as possible if that's the path you're going to take.

With those being said... Mangaverse when?

 

[AthK]

Ok... Now, after this update 'Data Saver' option no longer works with 'Legacy Reader'... Even after I turn it ON from settings page, the images are loading in PNG format... Is it done deliberately or some bug while upgrading??

 

[adzil]

Even if we decided to take the p2p content delivery route wouldn’t it also require some kind of “p2p reader apps” so users can also read it on their phones?

 

[kuyukotto]

yay! now it's very convenient! I just noticed the reader settings since it's peak hour in asia.

 

[jk-jk]

Will anything be done in the future about all the old uncached content that doesn't download at nearly the same speed (if it even downloads at all) as newer content?

 

[pikokola]

@adzil
No, user access MD central server to fetch the metadata for a chapter/gallery.
If possible the server then automatically chooses a server (p2p server) appropriate to the user to load the image from (i.e closer/having low traffic).
It's the same idea that you could choose from which server you want to load the image from, but instead of running more server, they would let users run their own server to help hosting the image.

 

[ebullient]

Dearest Holo, I love you but you're overthinking things and being stubborn. Swallow the disk space and just go through all your images and convert them to webp. Opt everyone in and let them opt out. Even with data saver on I'm still getting served 1.6MB images. With webp that shrinks to 248 Kb. Most won't notice because the quality is still very high. Right now you're optimizing for the people who care which is a much much smaller group and will work a bit harder to click a button for quality. If it's ios show them something else.

Disk space is cheaper than bandwidth and the p2p headache, just do it, please.

 

[Cookiebytes]

I'm sorry since I'm a little ignorant about this, but wouldn't it be better to have data saver on by default just now, and give people a very bright eye catching banner or something telling them it's on, why it's on and what it does, as well as how to turn it off? I'd imagine many people wouldn't actually notice and if they did and it bothered them they'd find out why really easily. When the situation gets better, you can make the default off again, right? Also, is it really true that webp is that much efficient? Sorry if this is unhelpful

 

[jayapala]

Reporting!!
So I have many devices and try data saver on each of them.

In PC I use Chrome, Edge, and Firefox. All of them seems ok. Loading become faster (~30s to ~5s). Resolution becomes lower but that's expected.

In my android smartphone I use Samsung Internet. Everything seems ok. Similar with PC.

In my iPad, the data saver broke the manga reader in every browser that I used (Safari, chrome, and firefox).
Only the first page shows up and I cannot change page (even when the page indicator shows I already in different page).
When I change to long strip, some pages will be partially cut (usually after around 3/4 of the page).

My guess is that in iPad, all browsers use webkit (safari engine). Therefore, it will broke the manga reader no matter what browser you choose because all of them use the same engine.

 

[ozanty]

@jayapala I use iPad safari to read mangadex. With datasaver one, it takes a while to load the new page and old page does not change till the new page comes. It is kind of annoying but workable. When you change pages, wait more for new page to load. And yeah, it is same with all browsers in iPad becasue Apple forces eveybody to use same engine.
It is broken. I don't know why though.

 

[Holo]

@teasday It's weird that it doesn't work on iOS

 

[thefahre]

With data saver turned on, old chapters now load basically a little bit slower compared to before the pandemic. Not ideal I guess, but I can read now, instead of waiting literally minutes for one image to load

 

[NodiX]

For WEBP supporters as well as doubters, I made a demo to implement cross browser WEBP implementation to show that implementing WEBP shouldn't be that hard. It has features that choose between 4 quality, and fetch images through Mangadex API. If someone asks I can open source it later, so even if Mangadex devs aren't interested in implementing it, someone else just can set it up as additional CDN to offload some of Mangadex bandwidth. For more info head off to my reddit post.

 

[Holo]

@pikokola

That's the game, right?