[Scummvm-devel] Proposed 1.8.0 release schedule

Eugene Sandulenko sev at scummvm.org
Mon Feb 1 13:44:42 CET 2016


On 1 February 2016 at 00:39, John Willis <John.Willis at distant-earth.com>
wrote:
>
> For the last few releases ports have been slow to arrive, often long after
> the main release (I know I am as guilty as anyone). I wonder if there is
> merit on having a real push to get ports in for the release or at least get
> status reports on them before 1.8.0 is branched. A mail directly to
> maintainers maybe?
>
I always was e-mailing the porters directly once release is tagged.

With a little luck and some willing maintainers maybe we can get unofficial
> 1.8.0 test builds out on blogs and whatnot for a wide range of ports on or
> just after the 7th branch use give that period till the 19th to get some
> user feedback from the various communities around the ports/devices.
>
I don't really like the idea. We will not be able to provide support for
these ports unless we have the porter officially and reachable by the team.

It just strikes me that it would be a shame to push into another release if
> the ports fail to materialise. I guess it's the curse of a stable and well
> maintained codebase, a lot of the porters have no need to jump in very
> often but rot and regressions can still creep in and if we slipped a few
> weeks but got a raft of bug fixes for ports from some focused effort that
> is not all told a bad thing. Especially as we will be pushing new platform
> enhanced ports for devices like the RaspberryPi in this release (well I
> hope so anyway).
>
For a long time I was telling the porters that the effort is minimal: build
your port, check it launches with couple of games and package it.

With our architecture once you have an engine running, most of others will
run as well. Those engines which differ either require some specific port
feature like full-color support, or cursor palettes, or are
resource-demanding, like big resolution or heavy computational stuff as
MT-32 emulator or memory requirements.

In my opinion, the primary reason for lower porter engagement is that many
platforms these days are cannibalized either by Linux/Andoid or by iOS.
Also we see two primary architectures, x86 and ARM emerging the market.
Thus, people are not really keen of putting much effort into maintaining
"exotic" ports.

What is a bit sad, is that the Android and iOS ports are underlooked. I put
high hopes into bSr43's involvement with iOS port, so it could be finally
brought up-to-date.


Eugene
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.scummvm.org/pipermail/scummvm-devel/attachments/20160201/cca83871/attachment.html>


More information about the Scummvm-devel mailing list