Motherboard Forums


Reply
Thread Tools Display Modes

White lines in doom 3

 
 





















Tim
Guest
Posts: n/a

 
      08-21-2004, 04:19 AM


I have a sapphire 9800 pro and I'm getting, you guessed it, white lines!
They occur sporadically and appear to extend from objects after a turn, as
if the lines from the object were overdrawn. Any tips or fixes for this?

TIA



 
Reply With Quote
 
Keanu
Guest
Posts: n/a

 
      08-21-2004, 07:16 PM
>Tim wrote in message ...
>I have a sapphire 9800 pro and I'm getting, you guessed it, white lines!
>They occur sporadically and appear to extend from objects after a turn, as
>if the lines from the object were overdrawn. Any tips or fixes for this?
>
>TIA


This is most likely the 'specular glitch'.. Along axix, white pixies.
This is generated from the way 'pixel shaders' are calculating using the
default Doom3 engine.

This should be fixed using the Humus patch.. He's some well known Guru
in the video field, and rearranged the shader's instruction sequence to
be more compatible to the ATI radeon latest families.. I know this is
right for 9800 and X800; Most probably 9600/9700 as well.
It improves things by using the Float capabilities of the card.

Not only it fixes the glitches, but it makes the game run smoother, faster.

I provid a link I found in ZIP format because I don't have Winrar,
and the original link did not work... Thansk to French dudes for hosting it.
http://alexharis37.free.fr/doom/humusd3atitweak.zip
Just unzip in your doom folder and it'll create a glprog folder in the base
folder with interaction.vfp in it ... So this is not gonna break anything.
Nothing else to set..

Doom3 knows how to load these extensions.
Read more about it here:.. (Original versions discussed)
http://www.beyond3d.com/forum/viewto...=342441#342441

Further TIP.. Read about TripleBuffering and get faster framerate and
virtually image no tearing using Vsynch-Off..
Read this group for more doom tips.

Enjoy.
P.S. I don't mind repeating myself... I happy someone else gets the good.
N³o


 
Reply With Quote
 
Tony DiMarzio
Guest
Posts: n/a

 
      08-22-2004, 05:58 PM
You need to stop posting this Humus propaganda garbage as a response to
anyone having problems with Doom3 and display artifacting. The only thing
the Humus mod does is switch a texture lookup to a floating point OP in one
of Doom3's main shaders. In some cases it can increase performance for
Radeon users, however, it turns out that the floating point OP and the
texture lookup are not mathematically equivalent so you end up causing
visual artifacting by using the Humus mod. That's right... CAUSING visual
artifacting with the Humus mod - meanwhile your posting the Humus mod as a
way for people to fix an issue that has nothing to do with the software
configuration.

--
Tony DiMarzio




"Keanu" <> wrote in message
news:34GdnY1woecIDLrcRVn-...
> >Tim wrote in message ...
> >I have a sapphire 9800 pro and I'm getting, you guessed it, white lines!
> >They occur sporadically and appear to extend from objects after a turn,

as
> >if the lines from the object were overdrawn. Any tips or fixes for this?
> >
> >TIA

>
> This is most likely the 'specular glitch'.. Along axix, white pixies.
> This is generated from the way 'pixel shaders' are calculating using the
> default Doom3 engine.
>
> This should be fixed using the Humus patch.. He's some well known Guru
> in the video field, and rearranged the shader's instruction sequence to
> be more compatible to the ATI radeon latest families.. I know this is
> right for 9800 and X800; Most probably 9600/9700 as well.
> It improves things by using the Float capabilities of the card.
>
> Not only it fixes the glitches, but it makes the game run smoother,

faster.
>
> I provid a link I found in ZIP format because I don't have Winrar,
> and the original link did not work... Thansk to French dudes for hosting

it.
> http://alexharis37.free.fr/doom/humusd3atitweak.zip
> Just unzip in your doom folder and it'll create a glprog folder in the

base
> folder with interaction.vfp in it ... So this is not gonna break anything.
> Nothing else to set..
>
> Doom3 knows how to load these extensions.
> Read more about it here:.. (Original versions discussed)
> http://www.beyond3d.com/forum/viewto...=342441#342441
>
> Further TIP.. Read about TripleBuffering and get faster framerate and
> virtually image no tearing using Vsynch-Off..
> Read this group for more doom tips.
>
> Enjoy.
> P.S. I don't mind repeating myself... I happy someone else gets the good.
> N³o
>
>



 
Reply With Quote
 
Keanu
Guest
Posts: n/a

 
      08-22-2004, 08:58 PM
>Tony DiMarzio wrote in message ...
>You need to stop posting this Humus propaganda garbage as a response to
>anyone having problems with Doom3 and display artifacting.
>

Learn to read better... I target specific symptom reports.
white pixels, glitches along axis.

>The only thing
>the Humus mod does is switch a texture lookup to a floating point OP in one
>of Doom3's main shaders. In some cases it can increase performance for
>Radeon users, however, it turns out that the floating point OP and the
>texture lookup are not mathematically equivalent so you end up causing
>visual artifacting by using the Humus mod.
>

ROTFLMAO...You are so wrong.

(1) Doom3 does a lot of work through texture lookups, which could
also be done by math instead. On older graphics cards texture lookups are faster.

(2) When Doom3 asks for AF, it only asks for those texture layers
and accesses, which need AF. So the AF cost is not that big.

(3) When you force AF through ATI's control panel, all texture layers get AF,
and also the texture accesses described in (1) get AF
- although it makes no sense there! The result is a *hefty* drop
in performance on ATI cards.

(4) Humus hack replaces the texture lookups in (1) by math. As a result
the problem described in (3) goes away, since math doesn't get AF.
-madshi

Hence perfromance gain on later radeon models; and NEVER a single
report of artifacting. In fact it's quite the opposite as soon as you increase
the resolution. Float is more precise !

http://forums.anandtech.com/messagev...&enterthread=y

>That's right... CAUSING visual
>artifacting with the Humus mod - meanwhile your posting the Humus mod as a
>way for people to fix an issue that has nothing to do with the software
>configuration.
>

LOL... You are so confused it's scary. Maybe you should read what others have to say:
*** READ THIS AND STEP ASIDE Tony clueless DiMarzio ***

As you will notice; Some people noticing glitches see them dissapear; as I did.
Some people don't see any glitches to start with, and others don't see this fixing it.
It's all a matter of using or not certain features of the Doom3 engine.
Chances are that if you notice these kind of glitches; Humus + AFx16 will help.

http://www.videocard-forum.com/ati/R...st_187337.html
"this really made a VERY nice improvement overall.
Very nice smooth performance on high quality settings now with 9800pro
on P4 2.4 (400bus). Definitely improves frame rates, smoothness,
visuals, etc., with no sparkles or atifacting either. "
http://www.elitebastards.com/page.ph...d=1&comments=1
(Although this one did not report any quality improvements, he's been seeing everything else)

http://forums.anandtech.com/messagev...keyword1=humus

*** The original version I had, .pk4 format... DID NOT fix the artefacts; While interaction.vpf did. ! ***
Then you might wonder how many versions and variants exist of that Humux 'tweak'... I have no idea.
I just provided the link to the one I know works for me, and numerous others.
To see the real benefit, it becomes apparent that enabling AFx16 in the Catalyst panel might be necessary.

Although it might not fix everyone's problem...And certainly not your bad temper.
Giving an option to try out is generally welcome.

I don't know what's your problem Tony;.. But you should gather more facts about your assertions.

This French thread summarize the technical details.. I will try to translate and summarize that part;
For the benefit of others.
----------------------------------
http://www.presence-pc.com/news/n4715.html

Once Humus got a hand on Doom3; He started an investigation for ATI performance. All non official.
He eventually linked a performance loss to the pixel shader "interaction.vfp".
He then strated to optimize this for ATI by changing the instructions, and got close to 40% gain.
but that was the early version, and early reports. Which were to evolve later.

It is explained that the pixel shader modelize the light equation (BRFD Bidirectional Reflectance Distribution Function),
uses for most three essential components: Ambiant, which is an omnipresent light source in all directions.
Diffuse; Which represents the light reflected by an uniform ponctual source.
AND SPECULAR: Which represents the object brilliance and mirror properties. Without it every objects would be
mate.

Humus realized that in D3, various components of the equation were read in texture buffers; and that these textures
were the main source of ATI memory trashing... The SPECULAR component in particular since their values vary
a lot over surface objects. These variations requires access to non-linear textures and could impair the texture cache.

The Diffuse light components are computed from the tool that generates the levels, using costly, time consuming
radiosity mathematics. On teh other hand; Specular light is simple compared to that, and if ATI cards appear to
be texture trashing more than Nvidia, they are in turn more capable to REALLY compute their component values
instead of reading a pre-computed values in a texture buffer... It's how Humus perceived the potential fix.

Humus changed the pixel shader instructions:
"TEX R1, specular, texture[6], 2D;" from "interaction.vfp" with "POW R1, specular.x, specExp.x;".

This makes the overall approach to the fix; Which later became applicable to other part of the pixel shader.

Between the first and last version of the fix; Humus got the opportunity to discuss with John Carmack.
This resulted in a patch that can really do an excellent job; but closer to a 20% performance gain ratio.
Without any loss in quality.
--------------- (Forgive me for I am not a pro translator.)

Although most of these threads do not report any specific FIX to artefacts.
I am not alone to have noticed this interaction.vpf fixing also this strange occurance of white
pixels dancing along some 'virtual' axis.. It's been reffered to 'specular glitch earlier; but may
be unrelated... This part of the effect might depend on a lot of 'options' you chose to activate
or not ingame or through Catalyst drivers.

I rest my case.

>--
>Tony DiMarzio
>
>
>

(...)
>> Not only it fixes the glitches, but it makes the game run smoother,

>faster.
>>
>> I provid a link I found in ZIP format because I don't have Winrar,
>> and the original link did not work... Thansk to French dudes for hosting

>it.
>> http://alexharis37.free.fr/doom/humusd3atitweak.zip
>>> Just unzip in your doom folder and it'll create a glprog folder in the

>base
>> folder with interaction.vfp in it ... So this is not gonna break anything.
>> Nothing else to set..
>>
>> Doom3 knows how to load these extensions.
>> Read more about it here:.. (Original versions discussed)
>> http://www.beyond3d.com/forum/viewto...=342441#342441
>>



 
Reply With Quote
 
Tony DiMarzio
Guest
Posts: n/a

 
      08-23-2004, 01:47 AM
Are you infatuated with Humus or something? You speak of him so highly and
with such regard.

Now.... where were we... Oh yeah... I'm supposed to respond to that
"cluster**** of pasted beyond3d.com forum quotes" you call a post.

1. Don't ramble on about subject material of which you have no clue. It
doesn't make you sound smart or knowledgable.

2. I'm curious... Why when you discovered the "Holy Grail" of Doom3 fixes
(Humus' mod) and scoured the forums for information about what it does and
how it affects performance, did you decide NOT to read the posts from Humus
and Carmack himself stating that the modifications were not mathematically
equivalent and caused artifacts in the shader such that certain object edges
would appear bright white. There were also screen shots depicting the
glitches Humus' mod would cause. Humus himself admitted that this code
should not be used in its current state. ------ and you tell me "Learn to
read better". Here is just one of many examples of posts from the same
thread where you got your information from stating how this patch affects
IQ.

"""""""""""""""""""""""""""""""""""No such thing as a free lunch

Just FYI this patch sacrifices IQ for performance, if you don't want
to lose IQ I wouldn't apply the patch. It also causes artifacting in many
places.

from the B3D thread:

Quote:
Originally Posted by Chalnoth
Chalnoth wrote:
After some looking, I don't think it's mathematically
equivalent, Humus (I found a place similar to the one kyleb posted, and
compared how different specular exponents affected the wall and the
chaingun...the wall with the texture lookup always looked dull, with the
chaingun shiny...but if I made the wall dull with a low specular exponent,
then the chaingun looked way too bright).

There may be a different per-material or per-object specular
lookup table.




Quote:
Originally Posted by kyleb
nasty bright white areas with humus' code:




Quote:
Originally Posted by danzg7
here's another example of this tweak not working correctly




Quote:
Originally Posted by micron
got some pretty serious tearing while using this tweak. Vsync
helped a little, but it was still pretty bad.




Quote:
Originally Posted by Smurfie
This could the start of a bad precedence. ATI employee comes in
with IQ lossy tweak. Nvidia might see ATI doing IQ lossy tweak and introduce
their own IQ lossy tweak.

Then IQ lossy tweak wars start, and we get flat shaded polygons
in the end.

I just hope ATI remembers their mantra on IQ.




Quote:
Originally Posted by CrunchTime
well upon further inspection, the shaders with this patch look
terrible.

Almost all of the doors have this ugly aliased white look to it,
basically that is why we get a FPS increase, in my case, 8 fps...

The IQ is dropped dramatically which is why JC knows what hes
doing



Better off reducing filtering quality to bilinear and/or reducing AF
level, that will give you a performance boost and less IQ loss/artifacting
than ati's shader replacement code for doom3.

--------------------------------------------------------------------------
Last edited by tranCendenZ : Aug 8, 2004 at 11:42 AM.


"""""""""""""""""""""""""""""""""""""""""

3. The artifacting that people have referred to as "snow" "white speckles"
"red dots" among other things, is ultimately the result of the Radeon
hardware, core and or memory, running too hot. Plain and simple - it is not
a software problem. The software is merely exposing weaknesses in people's
systems.

4. In my experience, people who "Rest their case" before they've proven
anything are impossible to reason with and aren't worth the response. I just
couldn't resist replying to your garbage though I guess.

--
Tony DiMarzio




"Keanu" <> wrote in message
news:24idnd4Q0cSHZrXcRVn-...
> >Tony DiMarzio wrote in message ...
> >You need to stop posting this Humus propaganda garbage as a response to
> >anyone having problems with Doom3 and display artifacting.
> >

> Learn to read better... I target specific symptom reports.
> white pixels, glitches along axis.
>
> >The only thing
> >the Humus mod does is switch a texture lookup to a floating point OP in

one
> >of Doom3's main shaders. In some cases it can increase performance for
> >Radeon users, however, it turns out that the floating point OP and the
> >texture lookup are not mathematically equivalent so you end up causing
> >visual artifacting by using the Humus mod.
> >

> ROTFLMAO...You are so wrong.
>
> (1) Doom3 does a lot of work through texture lookups, which could
> also be done by math instead. On older graphics cards texture lookups are

faster.
>
> (2) When Doom3 asks for AF, it only asks for those texture layers
> and accesses, which need AF. So the AF cost is not that big.
>
> (3) When you force AF through ATI's control panel, all texture layers get

AF,
> and also the texture accesses described in (1) get AF
> - although it makes no sense there! The result is a *hefty* drop
> in performance on ATI cards.
>
> (4) Humus hack replaces the texture lookups in (1) by math. As a result
> the problem described in (3) goes away, since math doesn't get AF.
> -madshi
>
> Hence perfromance gain on later radeon models; and NEVER a single
> report of artifacting. In fact it's quite the opposite as soon as you

increase
> the resolution. Float is more precise !
>
>

http://forums.anandtech.com/messagev...&enterthread=y
>
> >That's right... CAUSING visual
> >artifacting with the Humus mod - meanwhile your posting the Humus mod as

a
> >way for people to fix an issue that has nothing to do with the software
> >configuration.
> >

> LOL... You are so confused it's scary. Maybe you should read what others

have to say:
> *** READ THIS AND STEP ASIDE Tony clueless DiMarzio ***
>
> As you will notice; Some people noticing glitches see them dissapear; as I

did.
> Some people don't see any glitches to start with, and others don't see

this fixing it.
> It's all a matter of using or not certain features of the Doom3 engine.
> Chances are that if you notice these kind of glitches; Humus + AFx16 will

help.
>
>

http://www.videocard-forum.com/ati/R...st_187337.html
> "this really made a VERY nice improvement overall.
> Very nice smooth performance on high quality settings now with 9800pro
> on P4 2.4 (400bus). Definitely improves frame rates, smoothness,
> visuals, etc., with no sparkles or atifacting either. "
> http://www.elitebastards.com/page.ph...d=1&comments=1
> (Although this one did not report any quality improvements, he's been

seeing everything else)
>
>

http://forums.anandtech.com/messagev...keyword1=humus
>
> *** The original version I had, .pk4 format... DID NOT fix the artefacts;

While interaction.vpf did. ! ***
> Then you might wonder how many versions and variants exist of that Humux

'tweak'... I have no idea.
> I just provided the link to the one I know works for me, and numerous

others.
> To see the real benefit, it becomes apparent that enabling AFx16 in the

Catalyst panel might be necessary.
>
> Although it might not fix everyone's problem...And certainly not your bad

temper.
> Giving an option to try out is generally welcome.
>
> I don't know what's your problem Tony;.. But you should gather more facts

about your assertions.
>
> This French thread summarize the technical details.. I will try to

translate and summarize that part;
> For the benefit of others.
> ----------------------------------
> http://www.presence-pc.com/news/n4715.html
>
> Once Humus got a hand on Doom3; He started an investigation for ATI

performance. All non official.
> He eventually linked a performance loss to the pixel shader

"interaction.vfp".
> He then strated to optimize this for ATI by changing the instructions, and

got close to 40% gain.
> but that was the early version, and early reports. Which were to evolve

later.
>
> It is explained that the pixel shader modelize the light equation (BRFD

Bidirectional Reflectance Distribution Function),
> uses for most three essential components: Ambiant, which is an omnipresent

light source in all directions.
> Diffuse; Which represents the light reflected by an uniform ponctual

source.
> AND SPECULAR: Which represents the object brilliance and mirror

properties. Without it every objects would be
> mate.
>
> Humus realized that in D3, various components of the equation were read in

texture buffers; and that these textures
> were the main source of ATI memory trashing... The SPECULAR component in

particular since their values vary
> a lot over surface objects. These variations requires access to non-linear

textures and could impair the texture cache.
>
> The Diffuse light components are computed from the tool that generates the

levels, using costly, time consuming
> radiosity mathematics. On teh other hand; Specular light is simple

compared to that, and if ATI cards appear to
> be texture trashing more than Nvidia, they are in turn more capable to

REALLY compute their component values
> instead of reading a pre-computed values in a texture buffer... It's how

Humus perceived the potential fix.
>
> Humus changed the pixel shader instructions:
> "TEX R1, specular, texture[6], 2D;" from "interaction.vfp" with "POW

R1, specular.x, specExp.x;".
>
> This makes the overall approach to the fix; Which later became applicable

to other part of the pixel shader.
>
> Between the first and last version of the fix; Humus got the opportunity

to discuss with John Carmack.
> This resulted in a patch that can really do an excellent job; but closer

to a 20% performance gain ratio.
> Without any loss in quality.
> --------------- (Forgive me for I am not a pro translator.)
>
> Although most of these threads do not report any specific FIX to

artefacts.
> I am not alone to have noticed this interaction.vpf fixing also this

strange occurance of white
> pixels dancing along some 'virtual' axis.. It's been reffered to 'specular

glitch earlier; but may
> be unrelated... This part of the effect might depend on a lot of 'options'

you chose to activate
> or not ingame or through Catalyst drivers.
>
> I rest my case.
>
> >--
> >Tony DiMarzio
> >
> >
> >

> (...)
> >> Not only it fixes the glitches, but it makes the game run smoother,

> >faster.
> >>
> >> I provid a link I found in ZIP format because I don't have Winrar,
> >> and the original link did not work... Thansk to French dudes for

hosting
> >it.
> >> http://alexharis37.free.fr/doom/humusd3atitweak.zip
> >>> Just unzip in your doom folder and it'll create a glprog folder in the

> >base
> >> folder with interaction.vfp in it ... So this is not gonna break

anything.
> >> Nothing else to set..
> >>
> >> Doom3 knows how to load these extensions.
> >> Read more about it here:.. (Original versions discussed)
> >> http://www.beyond3d.com/forum/viewto...=342441#342441
> >>

>
>







 
Reply With Quote
 
Keanu
Guest
Posts: n/a

 
      08-23-2004, 08:39 AM
>Tony DiMarzio wrote in message ...
>Are you infatuated with Humus or something? You speak of him so highly and
>with such regard.
>

He's a guru in his field and respect him for all I've read and
thankful for his help.

Now don't worry Tony; I'm done with what you call propaganda.
I've got other things to do; and people can find all the help with
Google on this subject..

>Now.... where were we... Oh yeah... I'm supposed to respond to that
>"cluster**** of pasted beyond3d.com forum quotes" you call a post.
>

Not really.

>1. Don't ramble on about subject material of which you have no clue. It
>doesn't make you sound smart or knowledgable.
>

What I found obvious in your previous reply is that you take too much
for granted. You presume that it will not help for some reasons, while
I could post the before and after snapshots and show you personally.
But it's not necessary, since people will probably try this patch anyway,
and see if it helps their particular case.

I might have sounded less smart in that reply, than I usually am.
You deserve some credit not to believe in what seem improbable.

>>

>2. I'm curious... Why when you discovered the "Holy Grail" of Doom3 fixes
>(Humus' mod) and scoured the forums for information about what it does and
>how it affects performance, did you decide NOT to read the posts from Humus
>and Carmack himself stating that the modifications were not mathematically
>equivalent and caused artifacts in the shader such that certain object edges
>would appear bright white. There were also screen shots depicting the
>glitches Humus' mod would cause. Humus himself admitted that this code
>should not be used in its current state. ------ and you tell me "Learn to
>read better". Here is just one of many examples of posts from the same
>thread where you got your information from stating how this patch affects
>IQ.
>

Then you've shown me something... I can see below "no such thing as free lunch";
I see where it's all going. Then all I can say is that I still like MUCH more
the after Humus patch than the before.. Funny thing is that although I have noticed
this very slight artefact; It's far less potent than what I had before. The 'new'
artefacts might be true; But I've tried again without Humus and thare are numerous
occasions where we can see 'dotted' with lines around some objects in a similar
manner; barely different. With or without.

Then it's a compromise. An imperfect one, but it has the advantage of giving me a
better framerate overall; Humus and all.

>"""""""""""""""""""""""""""""""""""No such thing as a free lunch

(SNIP)

I'll have to snip all the attached data from my reply.
It appears clearly that you are right about the imperfection of the tweak.
According to 'these' guyz anyway.
I fell for the early reports and did NOT checked every other posts.

I see "micron" reported tearing with Vsynch On. I'm sorry to say
micron must have a strange setup. There's no tearing at all here.
Where do we draw the line between what we read and experience for ourselves.

Then Kyleb talk about 'nasty bright areas'.... Again, I can't see any problem.
Maybe what he calls nasty is not so obvious.
Could it be that they were using an early version of the patch ?...

In fact most of the terrible comments I read "shaders look terrible",
"almost all doors with ugly aliased".. Are *not* noticed here.

>3. The artifacting that people have referred to as "snow" "white speckles"
>"red dots" among other things, is ultimately the result of the Radeon
>hardware, core and or memory, running too hot. Plain and simple - it is not
>a software problem. The software is merely exposing weaknesses in people's
>systems.
>

Never denied that fact.

>4. In my experience, people who "Rest their case" before they've proven
>anything are impossible to reason with and aren't worth the response. I just
>couldn't resist replying to your garbage though I guess.
>

You did good.
Nothing is absolute in the area of troubleshooting. I learned something,
and will keep an eye on the IQ matter.

Now another example where nothing is obvious.. Many report that Doom3
engine is essentially 32bit color depth. While I have experienced the game
at what looked like 16bit depth, and starting D3 from a 32bit desktop would
perform slower than from a 16bit one. It was also visually noticeable.
I then found and tried: (no quotes around the number)
seta r_colorbits 16
Put in my autoexec.bat; Configured desktop to 32bit and did a doom3 timedemo.
I got back the performance I was getting from a 16bit desktop; While the game
visuals would be at 32bit depth quality.

It's a fact for me, while many argue r_colorbits is not supported anymore just
by trying it flat from the console. Maybe they haven't tried setting it from autoexec.cfg
first.
Maybe this strange game behavior is due to very specific conditions, installation...
It's not always obvious; and anyone else than myself might consider that r_colorbit
story impossible, pure fiction.

I do have to apologise, for you did prove your point; And I lost my temper when you
called it propaganda.
Is the Humus patch as bad as these people say ?... Does not look bad to me at all.
In the end you win on both accounts. I'm done with this topic; I'll let time be judge
and people google for themselves.

"It is better to fail in originality than to succeed in imitation." -Herman Melville
N³o



 
Reply With Quote
 
Tony DiMarzio
Guest
Posts: n/a

 
      08-23-2004, 04:08 PM
No hard feelings. I am sorry for losing my temper as well. I just don't want
people to unnecessarily get their hopes up for what Humus' modification can
do for them and any issues they might be having with Doom3.

Humus' modification is still a good thing to try. It's one of those things
that must be decided on in a case-by-case basis by the end user. The extent
of the Image Quality problem is purely subjective. Some will notice... some
will not. For the people that don't notice or that do notice and just don't
mind, this is a no-lose fix which could potentially increase FPS.

IMHO, attempting to fix graphical artifacts in Doom3 should always first
start with the hardware and systems level stuff - cooling, over-clocking,
drivers, PSU, BIOS setup, etc etc etc... Humus' fix should only be attempted
after those options have been attempted either successfully or
unsuccessfully.


"Keanu" <> wrote in message
news:S6qdnbo0TqzSArTcRVn-...
> >Tony DiMarzio wrote in message ...
> >Are you infatuated with Humus or something? You speak of him so highly

and
> >with such regard.
> >

> He's a guru in his field and respect him for all I've read and
> thankful for his help.
>
> Now don't worry Tony; I'm done with what you call propaganda.
> I've got other things to do; and people can find all the help with
> Google on this subject..
>
> >Now.... where were we... Oh yeah... I'm supposed to respond to that
> >"cluster**** of pasted beyond3d.com forum quotes" you call a post.
> >

> Not really.
>
> >1. Don't ramble on about subject material of which you have no clue. It
> >doesn't make you sound smart or knowledgable.
> >

> What I found obvious in your previous reply is that you take too much
> for granted. You presume that it will not help for some reasons, while
> I could post the before and after snapshots and show you personally.
> But it's not necessary, since people will probably try this patch anyway,
> and see if it helps their particular case.
>
> I might have sounded less smart in that reply, than I usually am.
> You deserve some credit not to believe in what seem improbable.
>
> >>

> >2. I'm curious... Why when you discovered the "Holy Grail" of Doom3 fixes
> >(Humus' mod) and scoured the forums for information about what it does

and
> >how it affects performance, did you decide NOT to read the posts from

Humus
> >and Carmack himself stating that the modifications were not

mathematically
> >equivalent and caused artifacts in the shader such that certain object

edges
> >would appear bright white. There were also screen shots depicting the
> >glitches Humus' mod would cause. Humus himself admitted that this code
> >should not be used in its current state. ------ and you tell me "Learn to
> >read better". Here is just one of many examples of posts from the same
> >thread where you got your information from stating how this patch affects
> >IQ.
> >

> Then you've shown me something... I can see below "no such thing as free

lunch";
> I see where it's all going. Then all I can say is that I still like MUCH

more
> the after Humus patch than the before.. Funny thing is that although I

have noticed
> this very slight artefact; It's far less potent than what I had before.

The 'new'
> artefacts might be true; But I've tried again without Humus and thare are

numerous
> occasions where we can see 'dotted' with lines around some objects in a

similar
> manner; barely different. With or without.
>
> Then it's a compromise. An imperfect one, but it has the advantage of

giving me a
> better framerate overall; Humus and all.
>
> >"""""""""""""""""""""""""""""""""""No such thing as a free lunch

> (SNIP)
>
> I'll have to snip all the attached data from my reply.
> It appears clearly that you are right about the imperfection of the tweak.
> According to 'these' guyz anyway.
> I fell for the early reports and did NOT checked every other posts.
>
> I see "micron" reported tearing with Vsynch On. I'm sorry to say
> micron must have a strange setup. There's no tearing at all here.
> Where do we draw the line between what we read and experience for

ourselves.
>
> Then Kyleb talk about 'nasty bright areas'.... Again, I can't see any

problem.
> Maybe what he calls nasty is not so obvious.
> Could it be that they were using an early version of the patch ?...
>
> In fact most of the terrible comments I read "shaders look terrible",
> "almost all doors with ugly aliased".. Are *not* noticed here.
>
> >3. The artifacting that people have referred to as "snow" "white

speckles"
> >"red dots" among other things, is ultimately the result of the Radeon
> >hardware, core and or memory, running too hot. Plain and simple - it is

not
> >a software problem. The software is merely exposing weaknesses in

people's
> >systems.
> >

> Never denied that fact.
>
> >4. In my experience, people who "Rest their case" before they've proven
> >anything are impossible to reason with and aren't worth the response. I

just
> >couldn't resist replying to your garbage though I guess.
> >

> You did good.
> Nothing is absolute in the area of troubleshooting. I learned something,
> and will keep an eye on the IQ matter.
>
> Now another example where nothing is obvious.. Many report that Doom3
> engine is essentially 32bit color depth. While I have experienced the game
> at what looked like 16bit depth, and starting D3 from a 32bit desktop

would
> perform slower than from a 16bit one. It was also visually noticeable.
> I then found and tried: (no quotes around the number)
> seta r_colorbits 16
> Put in my autoexec.bat; Configured desktop to 32bit and did a doom3

timedemo.
> I got back the performance I was getting from a 16bit desktop; While the

game
> visuals would be at 32bit depth quality.
>
> It's a fact for me, while many argue r_colorbits is not supported anymore

just
> by trying it flat from the console. Maybe they haven't tried setting it

from autoexec.cfg
> first.
> Maybe this strange game behavior is due to very specific conditions,

installation...
> It's not always obvious; and anyone else than myself might consider that

r_colorbit
> story impossible, pure fiction.
>
> I do have to apologise, for you did prove your point; And I lost my temper

when you
> called it propaganda.
> Is the Humus patch as bad as these people say ?... Does not look bad to me

at all.
> In the end you win on both accounts. I'm done with this topic; I'll let

time be judge
> and people google for themselves.
>
> "It is better to fail in originality than to succeed in

imitation." -Herman Melville
> N³o
>
>
>



 
Reply With Quote
 
Keanu
Guest
Posts: n/a

 
      08-23-2004, 07:31 PM
Tony DiMarzio wrote in message ...
>No hard feelings. I am sorry for losing my temper as well.


So many variants with Doom3 settings Catalyst versions, Shader tweak;
Even at least three ways to use this tweak; Manually in pak000.pk4,
z_humus.pk4 (which I know is not very good), and the latest I found;
Interaction.vfp. Then it's not guaranteed you will notice anything depending
on Anisotropic Filtering 16x or not; From catalyst, or not, and so forth and so on.

I could have been mistaking on a number of things.
One of them for sure is jumping to the guns, and resting my case much too early.
.... Could it be that dark thing we drink called Expresso..
Plus I'm trying to quit smoking..... Again.... Yeah, I know.. Lame.

... Or is Doom3 really starting to get to me...

I keep the sledgehammer on my desk and....
I've received a big crate from Martian Buddy; Or was it Humus ? Can't recall.
Oh.. I think I heard voices in the Kitchen...

Damn lights went off again..
Bye for now.. I'll see who's making that noise upstairs.

>I just don't want people to unnecessarily get their hopes up for what
>Humus' modification can do for them and any issues they might be
>having with Doom3.
>

(SNIP)
>IMHO, attempting to fix graphical artifacts in Doom3 should always first
>start with the hardware and systems level stuff - cooling, over-clocking,
>drivers, PSU, BIOS setup, etc etc etc... Humus' fix should only be attempted
>after those options have been attempted either successfully or
>unsuccessfully.

(SNIP)
<AUTOSENDER TIMEOUT: message send; user autologoff>


 
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
playing divx -> s-video -> tv and white border Peter Dell 0 02-03-2008 08:06 PM
HP PSC 1510 Won't Print Black & White Photos patetc HP 0 01-12-2008 02:15 PM
display has vertical lines and then scrambled characters ll Dell 9 08-28-2007 09:16 PM
vertical lines dell inspiron 9300 François Geoffriau Dell 1 07-11-2007 01:19 AM
P4P800S / Radeon 7000 black and white boot up (vertical scrolling) problem Terry Asus 2 01-06-2007 07:02 PM


All times are GMT. The time now is 03:41 PM.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43