Major stompage on the horizon

[jasong]

Tomorrow, at our present rate, we'll stomp US-D. In about a month-and-a-half we should be the top team.

If you look at our output, we beat all the other teams by an EXTREME margin, about 5 to 1. :D :|party|: :cheers:

[birdman2584]

hmmm, gotta love that, i think ill join in on some of the :stomp: fun.

[rsbriggs]

US-D has indeed been stomped. :thumbs: :elephant:

However, in response to a Free-DC stompage being only 350k points away, UT Austin has upped their output considerably and will start pulling away from us unless a few others join in..... :swear: :help:

(Ahem... And to a certain someone.... Merging accounts to run away only delays things. You can run, but you can't hide.... )
:spank:

[black_civic55]

:sack: :nicow: :sofa:

[em99010pepe]

Be afraid, be very afraid.....:neener:

Carlos

[rsbriggs]

Well, you're safe for a while. I'm NOT liking the production I'm getting relative to Lexx. He's running Linux, I'm running Windows. I believe that we have a similar number of machines running, but he outproduces me by 4x. :geezer: I really need to verify that the O/S is what is making the difference.

So, tonight I start the following experiment. I'm going to shut down all copies of EON, except on one AMD 2600+ dual boot machine that I have. For a couple of days, it will be the only machine I have running EON, and it will run under Windows XP.

Once I seem to have stable production of some-number-of-units-per-day running under Windows, I'm going to switch the box over to the current Fedora core 4, and run for a couple of days. If I still don't like what I'm seeing, I may experiment with some different flavors of Linux.

I may repeat the experiment after that with an Intel box.

However, once I find the combination that works the best, watch out.
:swear: This would be a whole lot easier if they would just include some sort of benchmark in Eon itself.... :swear:

[jasong]

My title was apparently more predictive than I thought!!! (http://www.free-dc.org/forum/showthread.php?s=&postid=87815#post87815)

[em99010pepe]

Originally posted by jasong
My title was apparently more predictive than I thought!!! (http://www.free-dc.org/forum/showthread.php?s=&postid=87815#post87815)

Jasong,

The link is wrong. Should be:

My title was apparently more predictive than I thought!!! (http://www.free-dc.org/forum/showthread.php?s=&threadid=9491)

rsbriggs,

I think Lexx has more than 20 machines.
There's no excuse to leave eOn. I'm also running it on Win XP machines. Are you afraid of the dark side of the force?

Carlos

[rsbriggs]

I'm not leaving Eon at all - go back and read my post again....

[em99010pepe]

Originally posted by rsbriggs
I'm not leaving Eon at all - go back and read my post again....

for the moment.....

Carlos

[Mustard]

No, Lexx does not have more than 20 machines running eon. :) The norm is 17.

You also need to take a look at chipsets on the motherboard. Motherboards with VIA chipsets run eon pretty good. You can quickly tell now systems are running in general by keeping an eye on the "time" element of the stats pages. The work units change quite a bit, and you can't tell really between systems or operating systems due to the lack of an "internal benchmark" work unit as some other projects have included in the past.

Take a look at vgriffin and look at the time on that. He is running some powerful systems. When he kicks in, there isn't any way FDC can keep up with UT without major increases in horsepower. While we are a mil up on UT at present, look back on how fast vgriffin moved up and you will see that 1 mil is not much of a lead at all really.

The current FDC #1 cruncher outperforms me by around 20% when he kicks all his stuff on. I do not know what he was running for motherboar/chip sets, or O/S's, but he puts out some serious stuff. When PCZ was crunching, he also put out some serious numbers, but I have no idea of what he was running. Bok has also run some very serious numbers on eon in the past.

However the reality is that FAD is currently the project of coice for the major percentage of heavy hitters. A lot of that is due to the stampede. But then I wouldn't be surprised to see FAD stay in the gunsights of the medical project cuncher/participants as many have fallen out with D2OL/TSU.

Another factor hitting eon is that the project is still in the process of getting the client/server bugs worked out, The outtages have been a contributing factor for a lot of participants/teams to not become heavily involved in crunching.

We are close to topping 5 mil, which would be nice to see. :)

[Bok]

I think we'll see 5M today based on current output.....

Bok :cheers: