If all is well, I just sent you the file Johnathon.

By the look of my e-mail app, Its coming through... I'll borrow work's broadband to finnish downloading it, and upload it to the directory I setup on my webhost, when I get in.

---

Ok, the avi is here:

Link to folder:
http://blue-world.co.uk/freehost

Dirrect to AVI:
http://blue-world.co.uk/freehost/screen.avi

It looks good!

---

Just received 'rosetta_graphics_beta_4.80_windows_intelx86.exe'. Any notes on what has changed or what to expect?

- rp

---

This latest beta version does not require the glut32.dll. The text should scale with the window and there is information about the run status and user. If this version runs okay, we will make it the official rosetta version for windows. We are working still working on linux and mac versions.

Still waiting for my first graphics WU - I'm beginning to feel deprived!:( Maybe I'll get one soon now, when the graphics WUs become the official Windows version.:)

---

I know the feeling! At least we've got the avi now!

---

The new beta 4.80 looks even better - really great work :-) . One point though, the visualisation of the molecules in the two boxes in the upper left corner (searching and accepted) sometimes reaches underneath the windows title-bar so that they are not displayed completely.

4.80 graphical runs great here so far and looks nice, too (vista / gfx7800 gtx)

---

--
Florian
www.domplatz1.de

The new beta 4.80 looks even better - really great work :-) . One point though, the visualisation of the molecules in the two boxes in the upper left corner (searching and accepted) sometimes reaches underneath the windows title-bar so that they are not displayed completely.

This is partly intentional. I decided that rather than rescaling (and having the size change frequently) as the protein folds up, the box size would remain constant. But it may be that in some cases this is results in too much being left out of the "view port". If you want to post a screen shot I will take a look.

Two of my machines received some graphics WU's and the graphics look and work nicely! good job

- rp

---

Is there a certain idea behind the distribution of the gfx WUs?

I run R@H on 5 PCs. Interestingly, my slowest ones - PIII600/XP2000+/P42,8HT - all received gfx WUs, but my fastest one - XP3200+/X23880+ - didn't get any. The P4 received 11 (!) gfx WUs in a row, while my X23800+ didn't get any yet.

Are you distributing the WUs according to a certain scheme or is it by a random pattern, like "first come - first serve" ?

---

This is partly intentional. I decided that rather than rescaling (and having the size change frequently) as the protein folds up, the box size would remain constant. But it may be that in some cases this is results in too much being left out of the "view port". If you want to post a screen shot I will take a look.

It's no big issue and I don't mind, I just wanted you to know. It's only the top part of the most extensive "arm" that sometimes is not displayed. It amounts to less than 1 % of the whole image.

I wanted to post here to thank you for the change in layout I saw this morning of my latest graphic work unit.


I really like the addition of work unit completion status and my basic stats on the display.


-Sid


Now for the bad news....

I was watching two graphic displays at once (AMD X2) and when I began browsing to make this post... my computer froze for a moment (maybe 10 seconds) and then continued on.. when I closed the two graphic display windows, all seems to be normal. I repeated this scenerio and got the same result. Now bear in mind I am 'tweaking' a new system and it is possible the pause was due to that (although it doesn't appear so at this time)

I am also seeing unhandled exception errors (access violation) listed on my results when I view them at this site??

<stderr_txt>
# =====================================
# random seed: 1586701
# =====================================

***UNHANDLED EXCEPTION****
Reason: Access Violation (0xc0000005) at address 0x0055B114 read attempt to address 0x78192A84

1: 11/25/05 23:04:34

-Sid

---

Proudly crunching with TeAm Anandtech

4.80 looks nice, really like the new green for the graphs :)
i am not so sure if the new font is better readable though.
and whats the red dot? i have seen more than one at the same time too, not beeing next to each other.

---

4.80 looks nice, really like the new green for the graphs :)
i am not so sure if the new font is better readable though.
and whats the red dot? i have seen more than one at the same time too, not beeing next to each other.

The red (and slightly larger) dots show the Energy and RMSD of the final predictions that your work unit will send back to us. The idea is that you can see if you've produced anything with low RMSD (left hand side of the plot.) They will only appear during the "Full Atom Relax" stage.

I thought the new font was a little harder to ready too. But it has the advantage of not requiring the glut32.dll, and also being scalable. So you can make your window larger, and you should be able to see it more clearly.

ok, thx for the answers.
did i get that right? the red dots show the "best" predictions that my workunit did so far (which would explain why i saw two of them at the same time) and not the "target"?

---

4.80 looks nice, really like the new green for the graphs :)
i am not so sure if the new font is better readable though.
and whats the red dot? i have seen more than one at the same time too, not beeing next to each other.

The red (and slightly larger) dots show the Energy and RMSD of the final predictions that your work unit will send back to us. The idea is that you can see if you've produced anything with low RMSD (left hand side of the plot.) They will only appear during the "Full Atom Relax" stage.

I thought the new font was a little harder to ready too. But it has the advantage of not requiring the glut32.dll, and also being scalable. So you can make your window larger, and you should be able to see it more clearly.

I'm not quite fully qualified to talk about them, since the only one I've seen so far is on one of the "farm" systems in the garage, and I work them via Boincview and VNC. The graphics are sorta visible, but probably not giving the full impact.

Couple of thoughts to put on the back burner for later.

I'm a great one for interactivity (in case you hadn't noticed :) ), so what about allowing some way to set graph (and maybe font) colors. It could be as simple as a .cfg file in the boinc folder:

rosetta.cfg containing

graph_color=#ff0000
graph_hilight=#ffff00
text_color=#00ffff

which would set the graph color to red, hilight spots to yellow, and the text to cyan.

Of course, this waits till AFTER we can spin the molecules and scale the graphs (big grin).

---

[pout and whine]

I wanna see the graphics! I've had only 1 WU so far and it was way back when they were first released in beta. Haven't seen any of the cool changes you all are talking about.

[/pout and whine]

Anyone got any cheese to go with this whine?

Kathryn

---

Kathryn :o)
The BOINC FAQ Service
The Unofficial BOINC Wiki
The Trac System
More BOINC information than you can shake a stick of RAM at.

When I open the graphics, I get a window asking me if I really want to run graphics on a remote machine. All my installations are stand alone. ???

---

When I open the graphics, I get a window asking me if I really want to run graphics on a remote machine. All my installations are stand alone. ???

If I remember correctly (and I probably don't). This is a bug fixed by the latest version of BOINC. 5.2.12. You can get it from the download page on the BOINC site. Sorry, don't have the link handy.

Kathryn

---

Kathryn :o)
The BOINC FAQ Service
The Unofficial BOINC Wiki
The Trac System
More BOINC information than you can shake a stick of RAM at.