Chronic Logic

It'd be nice to have joints that could pivot and bend with out breaking links.  Additionally, being able to run the cable through a pully would allow weight on some parts of the bridge to distribute to others.

Quote: from pasqualz on 4:18 pm on Oct. 13, 2001
This very idea is being debated in the New Game Ideas forum! I think that would be the place to continue this thread. By the way, I never thought to actually try making a skyscraper with Pontifex, very smart of you!I tried, but I was trying to make it triangular... didn't work so well.  But the toys you get at the end of the first levels made it all worth while.

/Ryan

I witnessed this same thing tonight. I've posted a response in the ATI Mobility thread.

I don't usually press F8 500 times when using AIM, but whatever floats your boat.

My problem is the blank screen like Gray had when I return. I thought I'd crashed it hard, and reset the machine. I didn't know that I could press ESC multiple times to close P*. This was on an nVidia GeForce 256 DDR.

As I had theorized in an earlier post, I suspected that the tower I had designed was staying together because of the use of cables holding the tower together. I was wrong.

Subsequent testing has shown that the reason my tower doesn't blowup is because of the "short" diagonal joists. The long ones break. This seems to be in direct contradiction with previous observations.

Additionally, I have been able to increase the number of joints on a story without causing it to blow up, but I have to use short diagonals to do it. Longer diagonals cause stress failure.

Perhaps this is why the small joint boxes don't automatically explode. If they did, it wouldn't make for much of a game. http://www.pontifex2.com/iB_html/non-cgi/emoticons/smile.gif" border="0" valign="absmiddle" alt=':)'>

I'm working on a new tower level right now, to see what other factors are at play, but I have sucessfully built a single column in the entire vertical space without causing any broken links. That is 640m high. It is also the most computationally complex level I've seen yet.  My computer is uncharacteristically slow when testing. I'll make it available when I'm done.

Update: At about ũ.6 million, I ran into the same problem as someone else about their bridge disappearing. I hit the links limit, and consequently wasn't able to build beyond the 5th story. I will try and cut back on some building materials and the design. I know you all are counting on me, but it is late, and I'm going to give up for the evening.

Good luck to everyone who submitted a bridge for the contest, the results come out in T-7:48:23.624...

(Edited by baggio at 5:14 am on Oct. 29, 2001)

That's interesting. I hadn't though to build it in the air like that to see what happens. I'm confused now. We've established that short links are weaker. We've also established that you can build a very high, and very strudy tower, built by Calastigro and modified by me.

There are two differences as I see it. Our design doesn't use two long diagnal beams. It features four short beams with a crossbeam joining the front and back -- I added the same pattern to the sides.  Our design also uses a cable to join the top and bottom of the box.

It seems to me, that maybe because the cable allows more stretch, that the cross beams in our design won't break. Check it out:
ftp://chronic:logic@24.20.221.2:100/-%20original%20creations%20-/Baggio/Twist.pxb" target="_blank">ftp://chronic:logic@24.20.221.2:100/-%20ori....ist.pxb
if you haven't already.

Anyway, I'm late for a review, so I can't delve into this further right now, but if I get the chance, I'll do some more testing.

(Edited by baggio at 8:55 pm on Oct. 28, 2001)

See http://pontifex.mendelsohn.de/forum/sthang1.pxb" target="_blank">http://pontifex.mendelsohn.de/forum/sthang1.pxb and sthang2.pxb.
Interesting.  In Sthang2.pxb, I've been trying to figure out how deleting the segment with the delete arrow, prevents the entire structure from collapsing.

I think it is in part due to the elasticity of the cable and the moment of the supported beams.  If you slow down the break, the center beam is the first joint to break.  (Incidently on the joint, and not on the cable as has been mentioned)  The reduced tension on the center, causes an elastic snapback that is felt on the common cable segment.

This translates down the other two cables and gives them an ever so slight tug.  This additional force is enough to break their joints and send the entire structure crashing.  

I find odd that once all this weight is releaved though, the cables do not recoil consistant with the tension that they had been placed under.  

Ask anyone in the Navy, line snapback is a very really, and potentially deadly event.  Those cables should be going every which way.

I don't think there is a problem... you just need to install an ICD. If you have a normal ATI card, get the drivers from ATI. I think every card they make now has a basic ICD. Some of them are not the Official drivers though, they are beta.

If you have a Rage Mobility, you are at the whem of the laptop manufacturer. The Rage Mobility chipset will support OpenGL as stated on ATI's website, but it is up to the laptop manufacturer to implement that in the drivers. Harrass them, not ATI or CL.

You may want to try http://www.glsetup.com" target="_blank">GLSetup.com to see if it helps. The site appears to be down at the moment, but I suspect it will return shortly. If not, there are older versions mirrored at various 3D graphics sites. Do a search for GLSetup.

http://www.rageunderground.com/drivers.shtml" target="_blank">Rage Underground also provides unofficial drivers with and ICD. This is unfortunately a problem with ATI, and there isn't much that CL can do about it.

(Edited by baggio at 3:46 am on Oct. 31, 2001)

(Edited by baggio at 3:48 am on Oct. 31, 2001)

I was trying it on an ATI Rage Mobility P last night.  It looked exactly the same as my nVidia GeForce 256 DDR with the default drivers from Windows XP.  Neither installation had OpenGL support.  New Det. drivers from nVidia provide a full OpenGL ICD.  The drivers for my ATI do not.

IF you get a screen that has no red, and is very slugish -- You DON'T have an OpenGL ICD for your graphics card, and are relying on the software version in Windows.  That version is very minimal, and not intended to replace one from your video card manufacturer.

Why is the freaking reset button so close to the submit?!?!?

I'm not going to rewrite my doctorial thesis again, but this is the jist of it. If you can't get an OpenGL ICD for your video card, but you have DirectX, try http://www.scitechsoft.com/gldirect.html" target="_blank">GLDirect and see if it solves your problem.

I tried this tonight on an ATI Mobility laptop, and I did not have a menu. Perhaps this will work for other cards and/or manufacturer's.

There was another site I had found that supported the ATI OpenGL... http://www.soundcardcentral.com/rage3d/rageltman.htm" target='_blank'>http://www.soundcardcentral.com/rage3d/rageltman.htm If this doesn't have what you need, I don't know what to tell you. It seems like these problems are driver problems.

When I had an ATI, it seemed like they were taking forever to get OpenGL compatible drivers to the market. I no longer use my ATI in part because of that. Perhaps these are not complete driver sets, and/or perhaps CL is using some sledom used function. Hopefully these drivers will take care of your issue. If they don't you've given it you best, and I'd submit a formal complaint to ATI first, then perhaps CL.

I don't think the problem is CL's though, for the simple fact that P* runs on other videocard makes.

(Edited by baggio at 10:26 pm on Oct. 31, 2001)

Perhaps in the olden days of yore... In Windows, those limitations don't really exist though. I'm pretty certain that this is somehow and artificial limit, and not because of an archaic segmented memory problem. That was one of the reasons for dos extenders back before windows.

Anyone remember any programs that used Dos4gw?

I had the same thing happen while building my colossal tower. I'm planning 16 stories, and I ran into the too many links problem on the 6th! Building has ground to a halt until the problem is fixed.

I have observered in the mean time, that while the links may be gone, the verticies are still there. What I mean is, in the edit screen, move the sun until it is night. Then when you test the "bridge" you will see the verticies suspended where links used to be.

The second thing I noticed that removing extranious cable, that wasn't really providing any support was not providing me with additional beams of steel. For that reason, I suspect that the limit may be on verticies, and not beams.

I hope this is a limitation that Alex lifts soon. If I don't have the building completed soon, I'll begin to owe the city money, and the workers are being paid, even if there aren't building materials. I had to pay them time and a half yesterday, it being halloween and all.

I've noticed a similar thing while running and using some of the toys for completeing all the levels on hard.  For me it was because I had the simulation set to slow.  For some reason, it was not registering any clicks.  I changed to Fast or Normal speed and I could click again.

Just to add to the me too fray.  XP, or any other version of windows for that matter, does NOT come with OpenGL drivers.  You MUST install drivers from the video card manufacturer in order for OpenGL to work properly.  There is a software version that Microsoft provides, but it is only good for the screensavers at best.  With video drivers that add OpenGL support, Pontifex runs perfectly.

/Ryan