You guessed it, beta 6

[Eriol]

oh! sorry, see my open stupid mistake I'm running the screensaver not the text client...

Still getting that uploading error, though

[Brian the Fist]

Just to let y'all know, this beta is pretty well done (i.e. we've seen enough) and I'll be putting up the new one shortly (possibly tomorrow). This one will choose the gen. 0 structure a little differently and will lock in the helices now hopefully allowing some lucky beta tester to reach the true native fold

As for the autoupdate, we're not ready to test that with the beta server yet, sorry.

[Digital Parasite]

Originally posted by Brian the Fist
Just to let y'all know, this beta is pretty well done (i.e. we've seen enough) and I'll be putting up the new one shortly (possibly tomorrow).

If you are going to put it up tomorrow, the earlier the better as someone else pointed out a lot of people will be leaving for the Easter long weekend and I won't be able to get the new beta clients up and running if they are released too late in the afternoon until next week.

Jeff.

[m0ti]

And I just started getting some decent looking folds!

Improved quite a bit in the last few gens, I'm down to 5.86 and at gen 210... just a bit further...

[Paratima]

Fold faster!

[Grumpy]

Time to increase your clock speed m0ti..add an extra multi or two, and do not worry about the smoke coming from your case, it's just excess energy left from the RMSD being lowered

[m0ti]

Gen 228 and at 5.61... you see, it really is dropping. I'll hit that native fold yet!

At least give me time to break the top 10! (hmm, there's no begging icon... guess this is the closest there is: )

[Grumpy]

Meesa catching you m0ti...5.66 Gen 91

Update: The low RMS was Folded on the Dual but Client 1, the one that was real slow of the mark. The 2nd Client has not improved its condition at all...Every 2 or 3 Generations Client 1 gets a little lower.....getting excited

[IronBits]

Originally posted by Brian the Fist
Just to let y'all know, this beta is pretty well done (i.e. we've seen enough) and I'll be putting up the new one shortly (possibly tomorrow). This one will choose the gen. 0 structure a little differently and will lock in the helices now hopefully allowing some lucky beta tester to reach the true native fold

Nice
I'm ready willing and able, ready for duty sir, let er rip!
OH, what happens to the other beta clients work ifn we can't get to them for a day or more?

[Welnic]

Originally posted by IronBits
Nice
I'm ready willing and able, ready for duty sir, let er rip!
OH, what happens to the other beta clients work ifn we can't get to them for a day or more?

That happened to me on the last switch, they get full credit for 24 hours and then half for the next 24 just like the regular client. But you definitely get behind on the low RMS race. Not that you care about that.

[m0ti]

Seem to be stuck at 5.61 and at gen 242. Need to get under 5.58 to make the top 10...argh.

Also I'm impatiently awaiting the new beta... I want to jump on it as soon as it comes out!

[Grumpy]

Yeah, mine is stuck on 5.661, hate that. Looks like this one may have bottomed too. When the new Beta arrives I will let this one go to Gen 250, and get the 2nd Clent to stop and update

[m0ti]

Just wanted to ask something, the work we're doing in later generations is still depndent on a random variable isn't it?

It's not like whatever fold is best at gen 0 uniquely determines the rest of our folding for all the subsequent generations, is it?

[Welnic]

Originally posted by m0ti
Just wanted to ask something, the work we're doing in later generations is still depndent on a random variable isn't it?

It's not like whatever fold is best at gen 0 uniquely determines the rest of our folding for all the subsequent generations, is it?

I took a folder at some point and duplicated it then ran it on two different machines. The results diverged, I think that there is a new random seed every time that the client starts.

I didn't get any credit for turning in work after the other client had already submitted the data. That, combined with not being able to change the filelist so that you can turn in work with another handle shot down my plans to sell generation 200 files to people on the black market.

[m0ti]

Good to hear!

[IronBits]

Originally posted by Welnic
That happened to me on the last switch, they get full credit for 24 hours and then half for the next 24 just like the regular client. But you definitely get behind on the low RMS race. Not that you care about that.

Ya but, it can take 8 days to finish a protein...
Are we to abandon the current protein and go with the new beta and new protein, or will the new beta pick up where this one leaves off?

[Digital Parasite]

Originally posted by IronBits
Ya but, it can take 8 days to finish a protein...
Are we to abandon the current protein and go with the new beta and new protein, or will the new beta pick up where this one leaves off?

All the beta clients have been on the same protein so Howard can evaluate how the changes to the algorithm are affecting the results.

The new beta will have an improved (hopefully) algorithm so you will want to abandon what you have done and start fresh with the new beta.

Jeff.

[AMD_is_logical]

Originally posted by IronBits
Ya but, it can take 8 days to finish a protein...
Are we to abandon the current protein and go with the new beta and new protein, or will the new beta pick up where this one leaves off?

We'll probably have to remove the current work (especially the filelist.txt file) and start fresh.

[Welnic]

Originally posted by AMD_is_logical
We'll probably have to remove the current work (especially the filelist.txt file) and start fresh.

I think the last time it just uploaded the old data as it was still valid for points, and then started on the new stuff. It deleted the old data after uploading it.

It definitely just starts off with the 10000 gen 0 random folds as the first thing that it does with the new client.

[Brian the Fist]

For this coming version (almost there..) please try NOT deleting filelist.txt. It SHOULD do the following:

- Discard the most recent generation from the current beta (the one still going on now), which you are presumably somewhere in the middle of
- upload all buffered data from this current beta (if any and if connected to net), giving full credit if within 24 hrs etc.
- start fresh from gen. 0 with new algorithm.

If you are not networked, it should keep buffered all but the last generation as mentioned above, and still start folding from gen. 0 with the new algorithm. When you connect, it should upload the buffered stuff from the old version, giving you credit if under 48 hrs, and then continue with uploading the new stuff.

As you can tell this is a little tricky to accomplish all with one filelist.txt but I seem to have got it working, and that's why you guys and gals are all testing it for me
So let me know if there's any issues with the above once Ive posted the next version

P.S. Please remember, this IS a test only, so scores don't matter, nor does updating the moment I release the new version. Just grab it and install it whenever you get the chance and you will be doing us a great service.

[Welnic]

Originally posted by Brian the Fist
-----snip------
P.S. Please remember, this IS a test only, so scores don't matter
-------snip-------

[Paratima]

No worries, Welnic. He was only kidding!

[Angus]

I dang near had a heart attack!!!!

[Grumpy]

So for the update when it arrives, we just overwrite the files in the Client Directory and do not delete anything

[Paratima]

Originally posted by Grumpy
So for the update when it arrives, we just overwrite the files in the Client Directory and do not delete anything

That's the way I read it.

Just remember, this is a beta release, so at any time it could go all wahooni-shaped! In which case, you won't be the only one with questions.

[Grumpy]

What happens with the error logs, are they uploaded automatically or do we have to post them. One of my Dual clients has died as it cannot contact the server. The other can contact and is fine. Tried restarting..dies..tried recovers..dies..same message, cannot contact server . What do I delete to get it running again :bs:


Of course, The Doctor could be updating the Beta and thus the Server would be down

[Paratima]

There doesn't seem to be any updating going on and I don't run duallies...

And no, the client doesn't phone home with error logs. Post the pertinent part of the log here and either Howard will happen along or one of the bunch that runs dual and you'll get some help.

Also, I think this may have been reported before. Try checking back through the bug-reporting threads. (We REALLY need an organized FAQ thingie here.)

[Grumpy]

So, is anyone able to upload, my other Client and my XP1800 are not uploading..I think the Server is down

Oh, heres the log ..

========================[ Apr 15, 2003 10:16 PM ]========================

========================[ Apr 15, 2003 10:30 PM ]========================

========================[ Apr 17, 2003 6:51 AM ]========================
ERROR: [010.003] {taskapi.c, line 1217} [ReadServerResponse] Timeout waiting for response, got 7260 chars.
ERROR: [777.000] {ncbi_http_connector.c, line 217} [HTTP] Error writing body at offset 12288
ERROR: [777.000] {ncbi_http_connector.c, line 117} [HTTP] Retry attempt(s) exhausted, giving up
ERROR: [010.003] {taskapi.c, line 1217} [ReadServerResponse] Timeout waiting for response, got 0 chars.
ERROR: [000.000] {foldtrajlite2.c, line 3886} Error during upload: NO RESPONSE FROM SERVER - WILL TRY AGAIN LATER
ERROR: [010.003] {taskapi.c, line 1217} [ReadServerResponse] Timeout waiting for response, got 10164 chars.
ERROR: [777.000] {ncbi_http_connector.c, line 217} [HTTP] Error writing body at offset 8192
ERROR: [777.000] {ncbi_http_connector.c, line 117} [HTTP] Retry attempt(s) exhausted, giving up
ERROR: [010.003] {taskapi.c, line 1217} [ReadServerResponse] Timeout waiting for response, got 0 chars.
ERROR: [000.000] {foldtrajlite2.c, line 3886} Error during upload: NO RESPONSE FROM SERVER - WILL TRY AGAIN LATER

========================[ Apr 19, 2003 8:25 PM ]========================

========================[ Apr 19, 2003 9:56 PM ]========================

========================[ Apr 19, 2003 10:01 PM ]========================

========================[ Apr 19, 2003 10:05 PM ]========================

[Grumpy]

So, is anyone able to upload, my other Client and my XP1800 are not uploading..I think the Server is down

Update..everything seems to be uploading after all, just the 2 other Clients were having a think

The one that crashed seems to be working again..I deleted my handle and re-entered the data, then restsrted..now it looks OK Strange things are afoot...

[Grumpy]

Client 1 had more probs, had to delete filelist.txt to get it running Here is the log.

========================[ Apr 19, 2003 11:19 PM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 1:08 AM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 2:39 AM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 4:09 AM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 5:40 AM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 7:11 AM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 8:41 AM ]========================
FATAL ERROR: [000.000] {foldtrajlite2.c, line 2821} Unable to find file ws5azxgu_0_ws5azxgu_protein_156_0000022.val; cannot continue - replace file and start again, or manually delete filelist.txt

========================[ Apr 20, 2003 9:07 AM ]========================

========================[ Apr 20, 2003 9:08 AM ]

========================

Update..looked at Task Manager, 9 copies of Foldtrajlite running

:shocked:

[DViD]

Originally posted by Grumpy
Client 1 had more probs, had to delete filelist.txt to get it running Here is the log.

.....

Update..looked at Task Manager, 9 copies of Foldtrajlite running

:shocked:

Weird, I don't have any error message in the log... GUI 2.2beta3, client beta7...
Oops, this is beta6 topic