28-Mar-2024 18:55 GMT.
UNDER CONSTRUCTION
Anonymous, there are 44 items in your selection
[News] PowerPC/MorphOS advances in RC5-72 once more!ANN.lu
Posted on 13-Jun-2004 13:01 GMT by takemehomegrandma44 comments
View flat
View list
And so it happened again! The PowerPC/MorphOS platform lay down one more platform in the RC5-72 space race. This time it was the Alpha/DEC OSF/1 platform that hit the ground. :-)

And it went down with a loud *BLAM*, because at the same time the PowerPC/MorphOS platform sets another personal best when it comes to the number of blocks crunched in a single day:

** 3,836 ** blocks! :-O :-D

http://www.morphos-news.de/guides/rc5-72/dnet-mos-stats.php

The MorphOS now puts itself at the #14 position from the top, and we are actually closing in on the "mainstream" platforms. The curve now gets steeper and the opponents gets tougher, but it's still not at all impossible (at least not the closest two opponents, which we probably will beat before the summer is over, perhaps even with some margins).

However, this will require more effort from all of us (if running the client in the background and letting your computer be switched on when you leave it can be considered an *effort*(?)). While some new team members has been added to the dnet-mos-stats page above, I have also noticed that some of you who helped crunching keys in the beginning now has stopped for some reason. Come back, we need you! :-)

If everyone listed on the dnet-mos-stats page above starts crunching keys again, at least for a little while every day, then it's my belief that we easily can go up to about some 4,500 keys per day without any greater effort (every group of only ten users with G4 will bring in more than 700 blocks every 24h of non-stop crunching (72-96 per user/day)). And if only *a few* additional of you MorphOS users who is reading this, but has not yet joined the effort, pull yourselves together and really joins, then we would reach over 5,000 daily blocks *very easily* (could be a lot more). And then, when MorphOS1.5 is released (at some undefined future date), this figure could very well double over night thanks to the Altivec support in the client (with Altivec, the crunching is almost three times as fast on the same CPU, but not everyone has G4's)!

So this is a "call to arms"! Come and join us! :-)

New users can find some very easy step-by-step instructions here:
http://www.morphos-news.de/guides/rc5-72/index.html

... and when you have got it up and running (and waited some 24 hours or so to get your first stats registered on the server) you could post a link to your stats on this page:
http://www.morphzone.org/modules/newbb/viewtopic.php?topic_id=1491&forum=9&180

So come on people, let's get back to work! Let's save all those poor left-over clock cycles that otherwise would have died without accomplishing anything in their short lives! Let's give them some meaningful existence instead!
:-) :-D

PowerPC/MorphOS advances in RC5-72 once more! : Comment 1 of 44ANN.lu
Posted by Anonymous on 13-Jun-2004 12:07 GMT
Hey!

bPlan is also part of the team, cool :-)
http://www.morphos-news.de/guides/rc5-72/dnet-mos-stats.php?user=437432
PowerPC/MorphOS advances in RC5-72 once more! : Comment 2 of 44ANN.lu
Posted by Hagge on 13-Jun-2004 12:27 GMT
I wouldn't call this news, or well, not news I want to read on ann.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 3 of 44ANN.lu
Posted by Anonymous on 13-Jun-2004 12:30 GMT
In reply to Comment 2 (Hagge):
I agree, it's definitely a post in the forum category, not in the
news category.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 4 of 44ANN.lu
Posted by Thomas Würgler/Pagan on 13-Jun-2004 12:33 GMT
The same goes for all the brand new OS4 pre-release users of course :)

In fact most of the steps on http://www.morphos-news.de/guides/rc5-72/index.html goes for the OS4 release as well. Especially the joining Amiga team part :)

Hopefully an updated kernel with Altivec support will be available soon as an update, but in the meanwhile keep crunching. OGR-P2 is already pretty much as fast as it can be.

For people who are still running older clients: make sure to upgrade to the latest, as the old OGR is not running any more and the latest versions have some performance optimizations.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 5 of 44ANN.lu
Posted by Alkis Tsapanidis on 13-Jun-2004 12:37 GMT
In reply to Comment 2 (Hagge):
New about RC5 were always posted on ANN.lu. There were several reports about the
Amiga team when the RC5-64 contest was still in progress. I don't get what your
problem is. If you don't like some news, just skip them!
PowerPC/MorphOS advances in RC5-72 once more! : Comment 6 of 44ANN.lu
Posted by Anonymous on 13-Jun-2004 12:56 GMT
In reply to Comment 4 (Thomas Würgler/Pagan):
I agree, please join the Amiga RC5 team! This does not make your
blocks crunched on MorphOS go away in any way. Every block crunched
will be credited to ALL of the following:

1. You (the individual), the user/mail address of the one who cruches.
2. The platform (this is handled by the client), like PowerPC/MorphOS.
3. An optional team of your choice, like the Amiga RC5 team!

So you will not loose anything by selecting a team, rather the
opposite, you will all help moving the Amiga team up. And it does not
matter what computer you use to be able to join a team; in the Amiga
RC5 team for instance, a great deal of the blocks are crunched on x86.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 7 of 44ANN.lu
Posted by Anonymous on 13-Jun-2004 12:58 GMT
In reply to Comment 5 (Alkis Tsapanidis):
Exactly, and only 6 posts down below this one there is a post about
the Amiga RC5 team's progress in the OGR-25 contest. If you are not
interested in RC5-72/OGR-25, then don't read posts about it. It's as
simple as that!
PowerPC/MorphOS advances in RC5-72 once more! : Comment 8 of 44ANN.lu
Posted by Bladerunner on 13-Jun-2004 15:44 GMT
In reply to Comment 1 (Anonymous):
Yeah I noticed that too allready some days ago.
But.. what is more impressive, have you seen the numbers per day they are crunching? Well I thought ok, Some Pegys connected, but no, according to a post on Morphzone, it is a single Computer running MorphOS! Peg III G5 Prototype Perhaps?
PowerPC/MorphOS advances in RC5-72 once more! : Comment 9 of 44ANN.lu
Posted by Anonymous on 13-Jun-2004 15:55 GMT
In reply to Comment 8 (Bladerunner):
Yes its said that this is only one computer!
PowerPC/MorphOS advances in RC5-72 once more! : Comment 10 of 44ANN.lu
Posted by Emeric SH on 13-Jun-2004 17:32 GMT
In reply to Comment 6 (Anonymous):
"I agree, please join the Amiga RC5 team!"

I was always an Amiga RC5 Team member when crunching, and I bet most of the MorphOS users are members as well.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 11 of 44ANN.lu
Posted by takemehomegrandma on 13-Jun-2004 17:42 GMT
People, have a look at this:

B-plan has joined the effort: http://stats.distributed.net/participant/psummary.php?project_id=8&id=437432

... and do you know what? Their stats on that page is not from 10 computers, not from 5 computers, not from 2 computers, but from ONE SINGLE COMPUTER! It looks like a new CPU card is in the works! :-D

If you look at the short history of their RC5 crunching career, you will see that it has peaked at 711 blocks in a 24h period(!), which is TEN TIMES as fast as my 1GHz G4 without Altivec, and THREE TIMES as fast as current Pegasos G4's **with Altivec enabled**! :-o :-o :-D

As I said earlier, today the Pegasos/MorphOS platform set a *new personal record* in the amount of blocks crunched, and if you compare these 3,836 blocks that the MorphOS joint effort has produced, you will notice that the 711 blocks FROM THIS SINGLE MACHINE equals to about 18.5% of that total record! :-o :-) :-D

If only 6 machines with this CPU card would have been running non-stop yesterday, the amount of blocks produced would have exceeded that day's TOTAL AMOUNT of ALL OTHER PLATFORMS (34) below the Pegasos/MorphOS current position! :-o

If all 93 registered Pegasos crunchers on the MorphOS dnet stats page ([url=http://www.morphos-news.de/guides/rc5-72/dnet-mos-stats.php]http://www.morphos-news.de/guides/rc5-72/dnet-mos-stats.php[/url]) would have had one of these babies each, and had them running non stop for 24h, it would sum up in 66,123 blocks (:-o), which would put the Pegasos at #5 of ALL PLATFORMS when it comes to daily blocks submitted, overtaking (by far) platforms like Sparc/Solaris, PowerPC/Linux, X86/OpenBSD, X86/NetBSD, etc! :-o :-D

Do you get the picture? :-D

BTW, for reference: http://www.morphzone.org/modules/newbb/viewtopic.php?topic_id=2354&forum=9
PowerPC/MorphOS advances in RC5-72 once more! : Comment 12 of 44ANN.lu
Posted by JKD on 13-Jun-2004 17:44 GMT
In reply to Comment 11 (takemehomegrandma):
Maybe they participate with a PowerMac G5 and not a Pegasos ;-)

Steve
PowerPC/MorphOS advances in RC5-72 once more! : Comment 13 of 44ANN.lu
Posted by Hagge on 13-Jun-2004 18:49 GMT
In reply to Comment 5 (Alkis Tsapanidis):
"New about RC5 were always posted on ANN.lu. There were several reports about the
Amiga team when the RC5-64 contest was still in progress. I don't get what your
problem is. If you don't like some news, just skip them!"

Yeah, because it was cool to see the Amiga name, it was cool to be number 3 on RC5.

However, now there are lots of crap "competitions" like this one, and the result is probably not in the top and who the fsck cares when others run different "benchmarks".
My problem is I want to check ann for new news, REAL news, not new crap like this stuff.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 14 of 44ANN.lu
Posted by Hagge on 13-Jun-2004 18:51 GMT
In reply to Comment 7 (Anonymous):
"Exactly, and only 6 posts down below this one there is a post about
the Amiga RC5 team's progress in the OGR-25 contest. If you are not
interested in RC5-72/OGR-25, then don't read posts about it. It's as
simple as that!"

That one is probably just as boring.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 15 of 44ANN.lu
Posted by Alkis Tsapanidis on 13-Jun-2004 20:18 GMT
In reply to Comment 13 (Hagge):
Ok, I'll say it once again. Skip the damned items you find boring! Don't read
them! Is that *SO* difficult!? I found both items interesting.
If you have nothing useful (or humorous or in any way interesting) to say, just
don't comment. Posting that you don't like a news item just because you don't
care about it's subject is NOT interesting.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 16 of 44ANN.lu
Posted by Johan Rönnblom on 13-Jun-2004 20:31 GMT
In reply to Comment 11 (takemehomegrandma):
I think it's more likely that they have a couple of machines, and that
there's been some confusion. Piru suggests three machines in that
thread you link to, which seems reasonable speed-wise.

Btw.. I wouldn't expect them to use any new in-development hardware to
run OGR.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 17 of 44ANN.lu
Posted by Johan Rönnblom on 13-Jun-2004 20:32 GMT
In reply to Comment 16 (Johan Rönnblom):
RC572 I mean of course, not OGR. Sorry.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 18 of 44ANN.lu
Posted by priest on 14-Jun-2004 05:37 GMT
In reply to Comment 2 (Hagge):
I partly agree...

MorphOS users (at least most of them) are part of the Amiga community.
So I'm more interested of the Amiga RC5 team status...
PowerPC/MorphOS advances in RC5-72 once more! : Comment 19 of 44ANN.lu
Posted by priest on 14-Jun-2004 05:40 GMT
In reply to Comment 8 (Bladerunner):
"Peg III G5 Prototype Perhaps?"

More likely a MOS "runtime" for PowerMacG5 ....
PowerPC/MorphOS advances in RC5-72 once more! : Comment 20 of 44ANN.lu
Posted by Gunnar on 14-Jun-2004 06:01 GMT
G4 and Altivec question.

I wonder if we could use the Altivec enabled version now?
Wouldn't it wourk even without MOS 1.5?

MOS 1.5 will save the Altivec registers when tasks are swaped.
This is of course needed to allow many programs to use altivec.
But as long as we only run one program that uses altivec couldn't we
run it simpy run it without saving the rgisters aka on MOS 1.4 ?


Cheers
Gunnar
PowerPC/MorphOS advances in RC5-72 once more! : Comment 21 of 44ANN.lu
Posted by Sammy Nordström on 14-Jun-2004 09:50 GMT
In reply to Comment 19 (priest):
...or, maybe they have setup a Linux cluster on old traded in Pegasos I boards? :-P
PowerPC/MorphOS advances in RC5-72 once more! : Comment 22 of 44ANN.lu
Posted by Some Farker on 14-Jun-2004 12:55 GMT
Yawn.

Call me when someone edits a major motion picture on a Pegasus running morph-os, or when a landmark game is released for the platform.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 23 of 44ANN.lu
Posted by Alkis Tsapanidis on 14-Jun-2004 13:45 GMT
In reply to Comment 22 (Some Farker):
This is valid for all minor platforms.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 24 of 44ANN.lu
Posted by ujb on 14-Jun-2004 13:52 GMT
In reply to Comment 21 (Sammy Nordström):
Not bad Sammy, BUT no Linux cluster, but a MorphOS Cluster. And of course not old PegIs but a cluster of three neat PegII G4.
Yes that's MOS clustering and this is one of the firt viewable benfits of underlying Quark which can track the resources.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 25 of 44ANN.lu
Posted by JKD on 14-Jun-2004 13:56 GMT
In reply to Comment 24 (ujb):
eh? dnetc stats are rolled up by id (usually an email address)...no clustering is required at all.

I have 3 machines regularly participating - Powermac Cube 1GHZ G4, Dual CPU PC P3 - 2x1GHz, Peg2 G4 1GHz....the stats server runs up the totals submitted under my email address.

oh....you were joking? Never mind ;-)

Steve
PowerPC/MorphOS advances in RC5-72 once more! : Comment 26 of 44ANN.lu
Posted by ujb on 14-Jun-2004 14:12 GMT
In reply to Comment 25 (JKD):
No, even I am a person of humorous sense I am not joking here. It *is* a cluster!
PowerPC/MorphOS advances in RC5-72 once more! : Comment 27 of 44ANN.lu
Posted by Anonymous on 14-Jun-2004 16:07 GMT
all people with AmigaOS4 on their AmigaONEs should also be running the
latest distributed net client - thats available on Official clients page

and they should also be part of the Distributed Amiga Team (was known as Amiga RC5 Team.

- renamed because we always had a vision for the future of this...we're now, with G3/G4/970(aka G5), in a position to seriously considering other distributed efforts...eg SETI, Folding@Home etc

PS the numbers for the MorphOS 'team' arent too hot, i recall doing 531 blocks a few weeks back in a single day (my average is about 120 when systems are quiet)
PowerPC/MorphOS advances in RC5-72 once more! : Comment 28 of 44ANN.lu
Posted by takemehomegrandma on 14-Jun-2004 17:56 GMT
In reply to Comment 27 (Anonymous):
> PS the numbers for the MorphOS 'team' arent too hot, i recall doing 531
> blocks a few weeks back in a single day

You should update yourself at this page:

http://www.morphos-news.de/guides/rc5-72/dnet-mos-stats.php

At the bottom of that page you can "show all" ...


> (my average is about 120 when systems are quiet)

I too am (now) doing 120 on average. I am using a combo of one Peg G4 @ 1GHz *without* Altivec (with altivec it's almost three times as fast), and one Peg G3 @ 600MHz. That combination has peaked at 144 blocks in a 24h period, but it seems like 120 is the average (when the G4 is used for "normal" usage and the client is running in the background).

http://stats.distributed.net/participant/psummary.php?project_id=8&id=433721

BTW, of course I am dedicating my blocks to the Distributed Amiga Team as well! :-)
PowerPC/MorphOS advances in RC5-72 once more! : Comment 29 of 44ANN.lu
Posted by JKD on 14-Jun-2004 20:21 GMT
In reply to Comment 26 (ujb):
Well...I guess we'd all like to see pictures and this 'distributed' client/server version of the MOS dnetc client then :-D

One of the Genesi guys has said it's 3 Peg II G4 systems...

Whilst pointless for dnetc (since the keyserver handles the distributed part really well so far) it would be an interesting demonstration of some new (hitherto unseen) features of MOS/Quark....

The story would have been far more exciting if it were a dual CPU Peg II...or foreshadowed the imminent release of MOS1.5 w/full Altivec support of course ;-)
PowerPC/MorphOS advances in RC5-72 once more! : Comment 30 of 44ANN.lu
Posted by Anonymous Orc on 15-Jun-2004 06:41 GMT
reminds me...must get my DEC Alpha up and running again. Expect trouble ahead! ;-)
PowerPC/MorphOS advances in RC5-72 once more! : Comment 31 of 44ANN.lu
Posted by Sammy Nordström on 15-Jun-2004 16:04 GMT
In reply to Comment 24 (ujb):
I see, the Q/Box has finally matured to the point of beeing able to run it's own tasks without depending on the A/Box...
PowerPC/MorphOS advances in RC5-72 once more! : Comment 32 of 44ANN.lu
Posted by JKD on 15-Jun-2004 16:36 GMT
In reply to Comment 31 (Sammy Nordström):
I think technically Quark must have always been able to do that, since it can run the A/Box as a task for one ;-)

It's interesting, one of the challenges to maintain compatabiity is which space things run in. I believe one of the reasons (and it's a major one) that the system appears dead once something crashes in 'A/BOX' space is because that's where the drivers run and without proper MP or resource tracking you cannot be sure that any code is not compromised nor can it be easily 're-started'.

Whether there's currently anything else 'under the hood' that Quark does is anybody's guess...we all know it's a microkernel - hence the extension to Q/Box but I don't think anything is publically shown or known about the definitive approach.

As I said....I wanna see the code and some nice pictures :D

Steve
PowerPC/MorphOS advances in RC5-72 once more! : Comment 33 of 44ANN.lu
Posted by Sammy Nordström on 15-Jun-2004 17:11 GMT
In reply to Comment 32 (JKD):
>I think technically Quark must have always been able to do that, since it can >run the A/Box as a task for one ;-)

No, the A/Box doesn't run without the A/Box... :-P

>As I said....I wanna see the code and some nice pictures :D

Me too! Although, pictures would be enough for me. Besides, I don't expect MorphOS to go open source anytime soon... ;-)
PowerPC/MorphOS advances in RC5-72 once more! : Comment 34 of 44ANN.lu
Posted by JKD on 15-Jun-2004 20:42 GMT
In reply to Comment 33 (Sammy Nordström):
> No, the A/Box doesn't run without the A/Box... :-P

erm...ok. Something external needs to init the system and bootstrap the A/Box...could be done in firmware but we are told that the Quark microkernel runs underneat the A/Box...but I know you were joking anyway ;-)

>Me too! Although, pictures would be enough for me. Besides, I don't expect MorphOS to go open
> source anytime soon... ;-)

I was thinking more of this as special version of dnetc that takes advantage of kernel features allowing distributed processing. Of course, if Quark supports some form of Multi Processing without a special dnetc executable...then that's a whole different concept. Dnetc already detects multiple CPUs in one machine..but it would be intriguing (well, at least to limited knowledge people like me) if it automatically detected CPUs that weren't physically located in the same machine and was able to distribute tasks over a network :D

It's a poor application, since distributed server/client thing works well but technically intriguing that MOS might support some kind of clustering.

I'm sort of accusing ujb of BS'ing, at least I don't believe him..but ya know, innocent until proven and all that!

Steve
PowerPC/MorphOS advances in RC5-72 once more! : Comment 35 of 44ANN.lu
Posted by Anonymous on 16-Jun-2004 05:48 GMT
In reply to Comment 34 (JKD):
QNX does just that. Q/Box may do something similar.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 36 of 44ANN.lu
Posted by Anonymous on 16-Jun-2004 08:42 GMT
I am not too sure if and which cluster features are built into MOS.
http://www.morphzone.org/modules/MorphOS/kernelinfo.php speaks about "Distributed computing", but...

Bplan should do a 24h burn-in test for all Pegasos with the dnetc
under MorphOS. Then we would see some serious numbers and be sure
nothing crashes due to errors in HW.

I am not too impressed by a setup made of 3 G4-1GHz machines.
They promised Pegasos to be a Dual-capable board with the prospect
to deliver a CPU card with 2x 867MHz G4 in 2002. Nothing happened,
we didn't even see a prototype although there could be a Dual-1.5GHz
now, that smokes everything else in RC5 (except the new 2.5GHz PowerMacs).
:-]
PowerPC/MorphOS advances in RC5-72 once more! : Comment 37 of 44ANN.lu
Posted by Gunnar on 17-Jun-2004 08:10 GMT
In reply to Comment 36 (Anonymous):
Hi

IHMO: Amiga OS or MOS can make proper use of two CPUs.
A dual G4 would only be good for Linux at the moment.

MOS/Amiga OS can us 64 Bit.

So the best machine for MOS/AmigaOS is a single 1.5 Ghz G4. IMHO.

The current G4 Pegasos board are nice and very good.
Maybe they will bring out a faster G4 in the future too. :)

Cheers
Gunnar
PowerPC/MorphOS advances in RC5-72 once more! : Comment 38 of 44ANN.lu
Posted by Peter H. on 17-Jun-2004 11:32 GMT
Seems some people left the Distributed Amiga Team.

http://stats.distributed.net/team/tmsummary.php?project_id=8&team=30562
PowerPC/MorphOS advances in RC5-72 once more! : Comment 39 of 44ANN.lu
Posted by Anonymous on 17-Jun-2004 16:12 GMT
In reply to Comment 37 (Gunnar):
>IHMO: Amiga OS or MOS can make proper use of two CPUs.
>A dual G4 would only be good for Linux at the moment.

You mean can not

>MOS/Amiga OS can us 64 Bit.

For MorphOS, not insinde the A-Box. In the Q-Box probably.

Bye
PowerPC/MorphOS advances in RC5-72 once more! : Comment 40 of 44ANN.lu
Posted by katos1 on 17-Jun-2004 16:57 GMT
In reply to Comment 38 (Peter H.):
distributed Pegasos team?

56 days and only 1200+ blocks?

Maybe this PERSON never joined the distributed amiga team at all.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 41 of 44ANN.lu
Posted by Peter on 17-Jun-2004 17:04 GMT
In reply to Comment 40 (katos1):
Not every member cracking since 56 days for Pegasos Team. As I
understand it was going public yesterday.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 42 of 44ANN.lu
Posted by Peter on 17-Jun-2004 17:05 GMT
In reply to Comment 40 (katos1):
Sure, some where in the amiga team before.
PowerPC/MorphOS advances in RC5-72 once more! : Comment 43 of 44ANN.lu
Posted by katos1 on 17-Jun-2004 17:07 GMT
In reply to Comment 40 (katos1):
my bad... 5 people!! this will really hurt the Distributed Amiga team a LOT!!! ;-)
PowerPC/MorphOS advances in RC5-72 once more! : Comment 44 of 44ANN.lu
Posted by katos1 on 17-Jun-2004 17:13 GMT
In reply to Comment 42 (Peter):
@peter,

you are probably correct. but what is wrong with starting a Pegasos Team?
I own a PegII and I crunch for Team Amiga. Next week one of my friends may beg me to join his team... and I just might for kicks. Who cares what team you are on. every block my computer completes adds to the total of my chosen os, thats all I care about. Given a little time and a newly available client for OS4 and Amiga one owners can watch their os #'s grow.

Enjoy,

katos1
Anonymous, there are 44 items in your selection
Back to Top