It's been brought up several times, and perhaps you're just quoting,
but "get a decent processor" is an ridiculous and inflammatory thing
to say when the server would work absolutely fine on most of the
processors specifically mentioned in the past if they were running
windows.
This is a valve issue, and fingers shouldn't be pointed at the linux
admins for trying to run a server in linux that would run on windows.
Quoting Ondřej Hošek <[EMAIL PROTECTED]>:
On 02.11.07 14:48 Uhr, Miano, Steven M. wrote:
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:hlds_linux-
[EMAIL PROTECTED] On Behalf Of John Martindale
Sent: Thursday, November 01, 2007 2:12 PM
To: [email protected]
Subject: [hlds_linux] Error
Bad data found in model "dispenser_toolbox.dmx" (bad bone weights)
Bad data found in model "dispenser_gib1.smd" (bad bone weights)
Bad data found in model "dispenser_gib2.smd" (bad bone weights)
Bad data found in model "dispenser_gib3.smd" (bad bone weights)
Bad data found in model "dispenser_gib4.smd" (bad bone weights)
Has anyone addressed this issue, or found a work around for it yet?
Seems pretty quite for a usually quick to respond list on this issue....
That's because we've been repeatedly hearing of this issue for a long,
long, long time. We know it's pretty boring to sift through all the
archives, but as long-time list members, we can't tell who's new and who
already should know. (I guess a discussion forum would alleviate these
problems, since you get access to the archives and a search function
right away.) Sorry.
The probable cause is lack of SSE2. "Remedy": get a decent processor.
Yeah, sorry again.
Alfred Reynolds of Valve said that lack of the FCMOV instruction causes
it. I've discounted this by writing a program in assembler which
actually executes such an instruction, and it didn't end with an
"Illegal Instruction" error (as it would if the processor didn't have
FCMOV) even on the bad-bone-weight-fraught machines. AFAIR, we have
received no response from Valve since then.
It's either a mistake in the hand-written math code by Valve or an error
in GCC. We're not sure which, as Valve has neither come clean nor
pointed a finger at the GCC people. AFAIK, Windows machines are not hit
by this error, so it might be GCC-specific assembly or a GCC bug.
I hope very much that it's being fixed right now. A Valve engineer did
come around and tell us he has a machine to replicate this bug on, but
we haven't heard of his progress.
~~ Ondra
_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
