pipeline testing

Discussion in 'RivaTuner Advanced Discussion forum' started by analogking, Jan 31, 2003.

  1. analogking

    analogking Guest

    Does anyone have any schematics for the 9500. I found someone who can test the card for defects to try to put an end to the software vs. hardware issue.

    Thanks
     
  2. Unwinder

    Unwinder Ancient Guru Staff Member

    Messages:
    17,198
    Likes Received:
    6,866
    Are you naturally stupid? The problems appeard with _hardware_ modded cards when sowtware mod was _not_ available to public. How can it be related to software?! What's a point of spreading fames if you don't understand the things you are talking about?
    This is hardware issue. Period. You may hope that I'll 'fix' the 'bug' in the script and the checkers will disappear. But it's dead hope and it will never materialize. Instead of flaming you coud just peek in the script and try to realize how does it work. Maybe than you'll understand that it cannot cause the problem.

    And I really doubt that the guy you've found will be able to test the HSR unit/pipelines for defects.
     
  3. analogking

    analogking Guest

    mix in a clue dr. russia

    If you would like to be rude or not accept possibilities to your ideas, that is fine. Erase this message. The flaw could be related to many things besides the pipelines. If you would like the readings I get from the card that would be fine, I would be more than happy to send them out.

    I am not looking to change my card to a 9700. I am much more interested in research involving resale of flawed chips.

    Why do you think I asked the question about underclocking?

    Does underclocking have anything to do with the pipelines?

    If you lack the technology to test the card please dont make assumptions.

    My question was not addressed to you.

    If I have a question for you I shall ask it.

    thanks
     
  4. analogking

    analogking Guest

    Re: Re: pipeline testing


    If anyone would like information from the person who can test the pipes I would be happy to ask them to send you the info.

    Arizona is a home to Intel and Motorola. Apparently they know quite a bit about how boards are made.
     

  5. analogking

    analogking Guest

    Re: Re: pipeline testing

    why am I asking for hardware information? To prove the point. My purpose is to show a hardware defect on the card, not in the software.

    so please once again, I am looking for a schematic on the 9500.
     
  6. Robbeh

    Robbeh Guest

    I know you are keen to defend the excellent software that you probably spent days creating, but please stop complaining to people who try to find solutions. Some people have discovered that if you soft mod a card that arleady has 8 pipes it exhibits the same checkerboarding. I can guarantee its not a bug in your scipt, but it might be ati hard coding some protection into their drivers. Anologueking, ive seen so much speculation into the solution its nice to see someone bringing some expertise to the problem. Ill have a dig around and see if i can find the schematics.
     
  7. Robbeh

    Robbeh Guest

    Sorry looks like you'll need an insider at ATI. They state on the site "ATI components and schematics are considered proprietary and are not available to end users".
     
  8. Unwinder

    Unwinder Ancient Guru Staff Member

    Messages:
    17,198
    Likes Received:
    6,866
    I'm not gonna defend software which can have troubles. But I will react to absolutely stubid ideas in the same way as I did it. I really don't like when the ppl try to discuss the things that are beyond of their knowledge.

    Do you know what is placebo effect? Dou you know what do the script do? Do you know that it doesn't change anything on the boards with 0x4e44 DID and only absolute noobs can report that it cnanged anything? I doubt that you do. Basically the script do the following:

    9700 conversion mode:
    PCI DeviceID = PCI DeviceID & 0xf0f0 | 0x0e04.
    9500 conversion mode:
    PCI DeviceID = PCI DeviceID & 0xf0f0 | 0x0104.

    Nothing more, nothing less. Try to apply this formula to R9700PRO ID. ;)

    It's really fun to ask ATI schematics, registers specifications and so on in public forums. This info is not (and will not ever be) public.
     
  9. analogking

    analogking Guest

    It's all about the hardware

    Ok, to make this clear as day.

    The problem has been stated that artifacts occur due to hardware problems.

    The questions I have are based on an error occuring due to a hardware defect.

    1. Can the specific point of error be identified?
    2. Is the defect repairable?
    3. Is the new PCB design using the same components as the old design (not including the design change)?

    Could it be possible that the chip has been altered the same way a celeron has?

    And finally, what are the odds I get a response starting with "Are you naturally stupid, or Peepul should not talk about what they dont understand"
     

Share This Page