The OpenBSD team has released OpenBSD 4.5. There have been lots of changes and bug fixes, but it’s a rather daunting list that doesn’t really lend itself towards a summary (hint), but I guess if you use OpenBSD you are perfectly capable of figuring this out yourself. You can get the new release from the download page.
Just compiled it last week from CVS. Seems to be working well, except python wants me to compile X11 before it will compile from ports (and there is no flavor for no_x11). Seems that I can just download the X11 .gz install files now! Horray.
Other than that, it’s running beautifully on my VIA C7 MiniITX system
Go OpenBSD!
Looks like some much needed features and drivers have been added, especially for netbooks. I wonder how well ACPI works now? I’m also very happy to see some WPA support in the Atheros driver… I just wish they hadn’t been so pigheaded about WPA support in general for so long.
We don’t need any swine jokes during these dark times, sir 😉
No one was being pigheaded… it requires people to
actually do the work. Code doesn’t write itself.
Yet.
OpenBSD would have been something I would have loved to run on my Acer Aspire One but the lack of support for the AR5700EG chipset really puts a spanner in the works. I hope that they’ll bring support for the said chipset.
Unfortunately, it doesn’t look like it’d be much use on my eee 1000he either. The Atheros AR928x chips aren’t supported by OpenBSD yet.
I got one of those too and I want to slap OpenBSD on there. I don’t think powersaving is really a big deal anyways (other than dimming the display), since those Atoms only clock down to save a watt or two (max), while the GMA950 is the true beast and (always) runs @ 10-11Watts. =( I may need to change my wireless card over anyways – to support one that OSx and OpenBSD will support.
It’ll be nice having an OpenBSD laptop/netbook that will last (realistically) 5-7 hours per charge.
I’m not so much concerned with using ACPI to save power, although it does help a little to throttle back the CPU in the 1000he. I’m much more interested in whether suspend and/or hibernate would work properly now, as that’s been both NetBSD’s and OpenBSD’s sticking point for many years when it comes to laptop usage, though FreeBSD is quite a bit better in this regard now.
ACPI is not necessary to do CPU frequency scaling for most systems (depends mostly on the particular CPU). Some bits were commited just after 4.5 to work towards suspend/resume support and it is hopeful that by 4.6 it’ll be working on a few of the laptops that are common within the developer ranks (lots of ThinkPads, various Dell’s, EeePCs and some others). Also I find it funny that almost all OpenBSD developers use laptops [with OpenBSD] and yet going to conferences and such that a lot of FreeBSD/NetBSD developers use Mac laptops. Anyway, I use a laptop and have done so for many years and wouldn’t run any other OS just because of the current lack of suspend/resume.
Different priorities. For me, suspend and resume are must haves on a laptop, not optional features. Hibernate, on the other hand, is much less important to me.
So many new features you didn’t make it to the front page *sigh*
There are many features towards desktop(more like laptop) use in this release and more are coming for the next one. Some things are still missing, but now you can even play FPSs(OpenArena) pretty well.
Of course it is not for your grandma, but if you are admining OpenBSD servers and routers, it only makes sense to have an OpenBSD desktop to test configurations and build custom packages for your machines if you need them.
I think 4.6 will attract a lot of new users.
Just out of curiosity – what is happening in OpenBSD 4.6 that could attract new users? I’m looking through the OpenBSD website right now and I don’t see any plans or goals for the 4.6 release so I am unfortunately out of the loop on what will be appearing.
They are completely rewriting the install script. Don’t expect graphical installers or such, but it’s going to be a whole lot easier for new users. Advanced sound support has improved a lot since 4.4 and by 4.6 (some fixes weren’t ready for the feature freeze) it should be stable together with 3D support for Open chipsets like intel and some ATI cards.
Other WIP features are not so sure to ship yet, but if some of them make it to the next release. It won’t be the year of the OpenBSD desktop
, but the devs expect(or fear) more users will try it out than now.
I wish they could co-operate either with Debian or Gentoo to have a fully working and easily manageable ports system. I hope.
If you want “easily manageable” which means a GUI then go use Debian / Gentoo. This OS isn’t intended for people unwilling to use their brain or read a little documentation.
Exactly. I started learning on Mandrake Linux back in grade 9. I then proceeded to try Red Hat and then sticking with Debian for a while. My friend introduced me to OpenBSD 3.0 at first (just after pf was introduced and back when I thought the installer was really difficult. Ha!).
The amount I learnt from getting my hands in OpenBSD was astronomically more than any of those OS’ and it clearly isn’t an OS designed for someone who wants to take the “Easy Train”. There are some other BSD derivatives out there (based on OpenBSD as well I believe) that do a better job for your average Joe Winston.
Edited 2009-05-01 20:43 UTC
Why? What’s wrong with the *BSD port system? The last thing I’d want to see is that monster dpkg shoving its way into *BSD systems as well, especially when *BSD already has a much faster and lighter package management system.
God no, I hope they never do. Ports kick apt in the nads and spits in portage’s face.
I can’t be knowledgeable on every subject we cover. I don’t know anything about OpeBSD, NetBSD, or any of the other BSDs, nor does my interest really lie there. As such, a changelog simply means jack all to me. That’s why NetBSD provides a nice summary of the most important changes that I can use. You know, like ever other software project has been doing since, I don’t know, Jesus walked the earth.
OpenBSD just gives me a damn changelog that might as well be written in Chinese. If you want this on the frontpage, then help us out by writing a nice item for it. Like I said, you can’t expect me to know everything or have an interest in everything.
No there aren’t too many features and the changelog isn’t too big, instead there is too much hype in FOSS world.
Coding wasn’t the main holdback of WPA in OpenBSD. Many of the top developers didn’t want WPA implemented for a while, giving the reason that WPA wasn’t anywhere near as secure as IPSEC or an OpenVPN-based setup. In their minds, that was what any respectable user should be using… and for a large infrastructure, imho, they’re absolutely right, but a lot of users didn’t need or want to deal with that or needed to connect to WPA-enabled networks. . They only relented a few releases back (4.3 is when the initial code for WPA began to be integrated), and that’s what I’m referring to when I say they were being pigheaded about the issue. There would’ve been no point coding WPA for OpenBSD if Theo and the rest of the top people were going to reject it.
This was NOT an official stance from the project.
WPA support didn’t exist as there was no acceptable code to be integrated. It was written, now there is support.