Great result for the project and all it's crunchers !!!

Two days after I read the US government "has no proof aliens excist" I surely believe this private project will proof them wrong in my lifetime...

Well, I think : people don't moan if you can't report your work... ET can wait another day !

For you guys at the lab : keep up the great job !!!

On The Other Hand, the US Government also doesn't have proof that aliens don't exist... ;-)

There is no proof that God exists but but the official motto of the USA is "In God we Trust" so how can the politicians say aliens don't exist?

way to go off topic on the admins thread..

2 billion? I still feel hungry. crunch crunch crunch

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

True, but they are celebrating 1 billion and there are no work units to work on for THREE days. I have my computer set for 10 days of work!!
Something is not right.

That "Something" is time needed to change the code and test it.

Do you remember the Y2K problem?
(dates written as 00 may be interpreted as 1900 or 2000 because only one byte was used for date in some old programs/databases)

This is something similar - for now they can't accept your new results
because you will get/return e.g. result #2,150,000,000 and they do not have "room" with such high a number in their "hotel".

From the First post:
"... This explains why we're a little slow to recover from today's outage:
we need to update code to accept larger numbers."

First off: they're celebrating 2 billion (binary), not [1]... (could be a typo there, you've got the concept in the following 'graphs...)

Second: if a date was stored with only one byte, you'd have to change the assumption every decade, not just every century... (I ran some programs that had only one byte for the year, changing the date to the next decade was easy... [Autocoder]) the "Y2K" problem was because people had assigned 2 bytes, not 4, (not me, I had thought of the problem in 1974!, and assigned 4 digit dates...)

---

.

Hello, from Albany, CA!...

Second: if a date was stored with only one byte, you'd have to change the assumption every decade, not just every century... (I ran some programs that had only one byte for the year, changing the date to the next decade was easy... [Autocoder]) the "Y2K" problem was because people had assigned 2 bytes, not 4, (not me, I had thought of the problem in 1974!, and assigned 4 digit dates...)

[off topic]
Don't confuse a modern ASCII or EBCDIC encoding using a whole byte per character, with the actual storage space required for data. A decimal digit only requires four bits (if you know in advance it's going to be numeric), so two of them can fit neatly side-by-side in a byte - and probably did, given the cost of memory in the early days.
[/off topic]

Tuesday morning we reached 2 Billion BOINC results processed by SETI@home! Or to be more accurate, 2 to the power of 31 results (2,147,483,648). This explains why we haven't yet recovered from the weekly outage: we need to update code to accept larger numbers.

I blame NEZ, if he had not done so much we would not be here yet :¬)

This morning we reached 2 Billion BOINC results processed by SETI@home! Or to be more accurate, 2 to the power of 31 results (2,147,483,648). This explains why we're a little slow to recover from today's outage: we need to update code to accept larger numbers.

Unsigned is not a long term solution!

Congrats to the crunchers and the project.

Been too busy to check forums lately, so figured I would comment from the beginning. This really is great news. It means more people are participating in this project and that makes me happy. Congrats to everyone on every end of the project. this is a huge milestone.

---

"By faith we understand that the universe was formed at God's command, so that what is seen was not made out of what was visible". Hebrews 11.3

Second: if a date was stored with only one byte, you'd have to change the assumption every decade, not just every century... (I ran some programs that had only one byte for the year, changing the date to the next decade was easy... [Autocoder]) the "Y2K" problem was because people had assigned 2 bytes, not 4, (not me, I had thought of the problem in 1974!, and assigned 4 digit dates...)

[off topic]
Don't confuse a modern ASCII or EBCDIC encoding using a whole byte per character, with the actual storage space required for data. A decimal digit only requires four bits (if you know in advance it's going to be numeric), so two of them can fit neatly side-by-side in a byte - and probably did, given the cost of memory in the early days.
[/off topic]

BCD, Richard, binary-coded-decimal. IIRC there's even some support for it in the x86 instruction set. Ah, yes, DAA -- Decimal Adjust After Addition for example.

http://en.wikipedia.org/wiki/Intel_BCD_opcode

wow congrats.

---

Can we say: "That's a wrap" ???

---

Congratulations from Italy! Keep up the good work boys!

---

Emm, guys? 19th over here, no work units. what gives?

Emm, guys? 19th over here, no work units. what gives?

Instead of asking that question in a news thread you should have checked out the Number Crunching forum (where these sort of questions are spose to be asked) and read the More HE Problems ???? thread or just asked your question there. ;)

Cheers.

---

So what did he initially say? 2 billion results?

I see it myself by checking the numbers in the task list.

What is the way of doing it now? Just decimal characters perhaps?

Or maybe 64-bit numbers are being used now?

The next question for project administrator!
Workunit group database last updated:25 Aug 2008 18:07:51 UTC
http://setiathome.berkeley.edu/data_density.html
When will the next update of this table?

Mooncalf: I am sorry I have to respond to this: you cannot prove a negative.

I'm pretty positive there's aliens out there actually. Even if life on earth is a one in a trillion affair, that just means there's GOT to be more lifeforms out there.

Fed up with SETI and lack of available data to crunch -0 always an issue these days !

---

i've got loads of data to crunch , and i'll keep on crunching till we find something.when things go dry i will just wait a bit , then crunch some more
when it is available. i have been maxing out on WUs recently so i can't understand how you have been running short.

I too have been getting a lot of work units lately. Using BOINC Manager check that your downloads aren't stuck in a "Download Pending" state . If so, select some or all work units and click "Retry Now".

---

2 to 32nd power? (For results)

I'm not a scientist, but wouldn't 2 billion results be 2 to the 9th power? 2 + 9 zero's after it.

(Typo I think!)

---

2 + 9 zero's after it is 2 times 10 to the 9th power.

2 to the 31st power minus 1 is 2147483647 (the largest signed number to be stored in 32 bits), a bit more than 2 billion ;-)

Gruß,
Gundolf