Hi Hrnchamd, I've noticed more folks posting about texture flickering problems and I was wondering if this was something you could take a look at.

Here are some of the threads I've seen:
http://www.gamesas.com/bgsforums/index.php?showtopic=916934
http://www.gamesas.com/bgsforums/index.php?showtopic=964470
http://www.gamesas.com/bgsforums/index.php?showtopic=967104
http://www.gamesas.com/bgsforums/index.php?showtopic=903481

The problem affects my system too. I run Morrowind under Linux with a nVidia 7950GT graphics card. I've found that downgrading the driver to the very old 100.14.19 Linux driver for this card makes the flickering go away. Although I have an nVidia card, many of the people reporting the problem seem to have an Intel mobile graphics chip on their laptop.

To the best of my knowledge, the problem only occurs in exterior cells, I've never seen it happen in interiors.
It seems to be like a known type of graphics problem: http://wiki.activeworlds.com/index.php?title=Z-buffer_fighting.

If this is something you might be able to fix, I'm sure a number of people would be very grateful. Thanks.

Just wondering, but is the EXE Optimizer safe to use along with this? It says it increases the stability of Morrowind, but I'm not sure if the MCP does something similar or if the optimizer patches even actually help in the first place. The game seems to play fine (I ran the optimizer AFTER the MCP) and whatnot, but I can't tell if I made it more stable or less stable. Has anyone done any significant testing of the two together?

The MCP and the Exe Optimizer are compatible, the only thing to keep in mind is to apply ExeOpt last - which you did. If you did it the other way round, the MCP would complain about an unknown exe and refuse to patch it.

Both the MCP and ExeOpt improve stability, but are doing that differently, they address different problems. The MCP fixes problems in Morrowind's data management which, if Morrowind confused its data sufficiently, could lead to crashes. The Exe Optimizer rewrites certain processor commands that seem to be problematic for Intel CPUs.

The effect of the MCP has been tested thoroughly. I don't know how much testing has gone into ExeOpt, but I'm seeing reports of people who regard it as essential (mostly people with Intel CPUs), and others who say it doesn't make a difference. Since ExeOpt doesn't *break* anything and has no bad side-effects, I recommend to use it.

Which was a very bad idea in the first place. You don't play games, not even Morrowind, on a computer that has an Intel graphic chipset; it's just not meant for that.

I know, I had the same issue with my previous laptop. It had an Intel chipset. They're just not worth the trouble. ATI or NVidia yes, Intel no.

We always used to say:-
Intel make excellent Network cards, adequate processors, rubbish motherboards and atrocious graphics cards/chipsets!

That's what they used to say but last I checked Intel had taken the top of the graphics market easily from ATI and Nvidia. That said I'd never play Morrowind or any other game on a laptop again if I can help it.

??? I think it might be time to check again...

If you refer to the GMA chipsets, sure, theoretically. In practice, they've got a lot of problems.

I had the same problem too (I have an nVidia 7950GT)
The problem was caused exactly by z-buffer settings.

This is the way i fixed it, don't know if could help others.
I changed the z-buffer format in MGE (under Global Graphics settings) to D24X8 (the default one D24S8 was the one causing problems to me non only in Morrowind, even in other games)

Thanks. That's good to know, but as yet, MGE doesn't work under Linux, which is what I use.

I don't know if this has been mentioned before, but I found this excerpt in Kirel's Interior Weather mod.

It would be great if these could be fixed.

Can I use this along with the Unnoficial Morrowind Patch 1.64 and/or the Morrowind Code patch?

Yes, it is recommended they both be used together.

Alright, thanks

I know John already answered you, but just to make sure. This /is/ the morrowind code patch.

So, yes. It will work with the Unofficial morrowind patch.

Just an idea here but I spotted Gez's quote over at http://www.tamriel-rebuilt.org/forum/viewtopic.php?t=21341 about the the createmaps command. If you're looking for something to add (:ninja:) hrnchamd, then adding an interior cell check and skipping any interiors might be a nice idea. Hardly essential though.

Hey, I found one constant source of CTD for me even with the Code Patch. It's somewhat easy to reproduce.

Create a monster that has summon Bonelord spells, or bonewalker, anything, and get it to cast the spell near a cell transition door. Now kill it, and before it finishes the death animation, which kills the summons, walk through the door. Bam, instant Crash to Desktop. With an error message.

This is easy to do with BigMod 2 with the Uber Bonelords that summon bonelords...

If you can fix this, I will be very happy.

...any indication for possible date of next release of Cogs Patch?

Let's pray together and invoke the name of Hrnchamd... :bowdown:

--------

Here are a few interesting bugs (described below) noticed by a French modder, abg, who spent much time experimenting the AI of NPCs & companions.

To reproduce these bugs, the best way is to make a NPC following you, using a script like this one (for example, on Fargoth):

Begin CompanionExpress_scriptShort CompanionIf ( GetCurrentAIPackage != 3 )   If ( GetDistance, player <= 400 )	  Set Companion To 1	  AIFollow, player, 0, 0, 0, 0   EndIfEndIf

Another very efficient way to check these bugs is to use a companion mod with the share inventory feature, and then to check from time to time the charge amount of the magic items possessed in his inventory.

bug 1
NPCs quickly loose the benefit of magical items with constant effects. If they keep a magic item on them (a ring, etc.), the equipped item has no more magic effect on them. It is very easy to check. For example, add an magic item to your NPC:

AddItem, "Montain spirit", 1

(this artefact is an armor with a constant magical shield effect, so easy to notice either it is active or not)
Then, go for a walk with your NPC companion, sooner or later, his armor will loose the shield effect. It is not only a visual flaw, and the same bug occurs with different magical items. It is hard to determine how and when exactly the bug occurs, but it is not connected with the 72 hours bug.

bug 2
NPCs do not use magical items which are not fully charged (at 100%).
To check this, add an item:

AddItem, "cruel sparkbolt ring", 1

Then, start a fight. Your companion will use the ring only once, although it has many others charges. So he won't use it any more, excepted if the item is fully recharged.

bug 3
Magical items carried by NPCs do never regenerate. Even after 24h, a magical ring used a few times will keep its status bar unfilled.

bug 4
NPCs have suicidal tendencies: they will use on them any magical item with negative effects! To provoke this bug:

AddItem, "amulet of 6th house", 1

And then start a fight. The Amulet of 6th House drains the fatigue of the wearer.

bug 5
Many magical effects (like spells) are totally ignored by the AI. A list of those unused magical effects could be established, it's not easy, but some have been identified: spells like restore health, conjuration, etc. The same bug occurs with Bloodmoon conjuration spells, they are never used by NPCs (while NPCs use Morrowind conjuration spells).

To be more specific, abg states that these bugs are related to the combat AI, and not to scripted AI using "cast" or "explodespell".

Hoping that Hrnchamd will be back soon and able to fix these AI flaws.

-----------

@ Psyringe: Since you opened a new thread for MCP each time, would you agree to make a list of all the bugs reported in the first post (as you've done for MGE), thus if Hrnchamd comes back he will be able to see at a glance the entire list of (reproducible) bugs reported since the first release? Of course, I would be glad to help establishing this list.

The CE bug occurs when you change cells. Emma and company did research on this a few years ago, you can read about it on her http://forums.elricm.com/emma/viewtopic.php?t=515.

Hi Mordicus, hi all,

As stated Mordicus, I spent some time exploring the use of magic by AI. Because of my lame english, I shamefully left Mordicus do the dirty work ! :bolt:
Some comments :


It's the same thing... I mean : this little script gives inventory sharing...


As I understand the sentence, it's the opposite (they can use it only if charged to 100%)



The NPCs use these effects in the form of spells but not in the form of enchanted items... (any NPC, during a fight, could cast "hearth heal" but he will never use an amulet (or any item) with this effect)
I would try to compile a list of effects used by the IA and I'm interested in the decision making process (why do they use this effect rather than another and why some are never used ?)

Thanks a lot Mordicus (l'anglais est ma croix, ?a ne sera pas la premi?re fois que je me ridiculiserai en public ? cause de lui )

Oops, sorry abg, I forgot the "NOT" (edited my post):
NPCs do not use magical items which are NOT fully charged (at 100%).

As I am posting again, thank you Jac for the link to Emma's forum.

The NPCs ignore Restore Health on enchanted objects because they only use damaging enchantments that work on the enemy. You can have them use a Restore Health on Self enchantment if it is paired with a Fire Damage on Touch, for example (I wrote about this topic http://www.gamesas.com/bgsforums/index.php?act=findpost&hl=&pid=13818159).

That's something I've got on my ToDo list for quite some time now, but I've never found the time to do it. Any help in preparing a list would be much appreciated.

How about fixing the spell bug? It seems that you can never get the highest magnitude listed for a spell with a variable magnitude. I have experienced this consistently (never getting healed 5 points by the Engraved Ring of Healing [heal 1-5] for example), and its listed on the Enchant page at the UESP under the "Constant Effect Tip" section.