Which BIOS will Unlock X800GTO Pipes X800GTO

  • Thread starter Thread starter dang49
  • Start date Start date
D

dang49

Hi, I have ( for about 3 weeks) a Sapphire X800 GTO for AGP ( with Rialto
chip). I have been reading quite a lot about the possibility of unlocking
the 4 cripled pipes of the R430 chip so I ran ATI Tool and the card reports
the required "0xFFFFFFFF" and "0xFFFFFF9F" for the die and substrate fuses.
The RAM chips are covered with heatsink but I will assume I have at least 2
ns chips.

My question is that since the GPU is a design for the PCI Express bus do I
flash a BIOS designed for the PCI Express? (i.e. X800 XL BIOS) or does the
Rialto chip play a role in the BIOS code?

Anyone try this with the same card?

Regards,
Dan
 
As I understand it, none of the AGP versions of the X800GTO card can be
unlocked, only the PCIe versions (and possibly only the GTO^2 at that).

Kendt
 
Got it. All 16 pipes unlocked by editing the bios as per the good
insrtuctions here
http://forums.techpowerup.com/showthread.php?t=8060

I don't have any benchmark apps but the average FPS increased from 190 to
223 (no overclock) with the ATI Tool 3D display running.
Click to expand...

Also this forum is on the rv490 core i dont know if your gona flashup
the thing but to me it's suspicious. does your card posses the dual
bios technologies?
 
Before you load and save from RaBiT you need to edit a copy of the original
bios with a hex editor. I download an editor named "010 Editor" for a try
and seems to work OK but there are many available. In my BIOS the byte at
address 0x7A was 0x61. The "1" here disables one quad ( 4 pipes). I edited
0x7A to read "60" and then saved the file. I then loaded and resaved with
RaBiT which I think only needs to produce a correct checksum.

Of course before any of this happens you need to verify that the fuses are
not hardware locked per
http://www.techpowerup.com/articles/overclocking/vidcard/127

I'm not sure if this is relevant but there seems to be several versions of
the Sapphire X800GTO AGP board with possibilities for R430 and R480 cores.
My version is R430 on a red ATI X800 XL type board. Some of the reviews I
have seen show the card with a blue board (like the PCI Express version)
that looks shorter than mine and has a completly different heatsink. Some
boards are hardware locked and cannot be modded with just a BIOS change.

Regards,
Dan
 
Before you load and save from RaBiT you need to edit a copy of the original
bios with a hex editor. I download an editor named "010 Editor" for a try
and seems to work OK but there are many available. In my BIOS the byte at
address 0x7A was 0x61. The "1" here disables one quad ( 4 pipes). I edited
0x7A to read "60" and then saved the file. I then loaded and resaved with
RaBiT which I think only needs to produce a correct checksum.

Of course before any of this happens you need to verify that the fuses are
not hardware locked per
http://www.techpowerup.com/articles/overclocking/vidcard/127

I'm not sure if this is relevant but there seems to be several versions of
the Sapphire X800GTO AGP board with possibilities for R430 and R480 cores.
My version is R430 on a red ATI X800 XL type board. Some of the reviews I
have seen show the card with a blue board (like the PCI Express version)
that looks shorter than mine and has a completly different heatsink. Some
boards are hardware locked and cannot be modded with just a BIOS change.

Regards,
Dan
Click to expand...

Yep, If my card is laserlock it can't be modded but i would like to
take the chance one time, not now but in a near future.

The thing that would be find to me is that the hex editor show me the
same number and letter than the bios rv480 481 in the rv430 witch can
be modded if not laserlock. Also i've read that 0xFFFFFFFF means that
it can be modded whit the rv480. Is this the same with the rv430?

Some say that msi card rx800 128meg are laser lock but not the rx800
256meg. who knows why? maybe it because the 128mb version has some
lower timming? even if its gddr1.

Well i will google a little more.
 
I don't know about dual bios technologies or RV 490 cores. My card is a
Sapphire X800GTO AGP8x and according to ATI Tool, has a R430 core.
ATIFlash -i does report two cards; zero = Rialto, one = R430. My flash
command "atiflash -p -f 0 16pipes.bin".
 
Yes, sort of. As I read it, the important bits for the fuses are in all
X800 type boards.
CONFIG_DIE_FUSES = ####F### and
CONFIG_SUBSTRATE_FUSES = #######F
So a register does not have to specifically read "FFFFFFFF" but needs the F
in the indicated positions.
Click to expand...
Well i've download the atitool 2,5 and find out that i do not have the
F in both setting its a 7 and a E.

I guess that i only have to check out the gpu processor to find out if
there is a laser cut... maybe later... :-(
 
DDC said:
Yep, If my card is laserlock it can't be modded but i would like to
take the chance one time, not now but in a near future.

The thing that would be find to me is that the hex editor show me the
same number and letter than the bios rv480 481 in the rv430 witch can
be modded if not laserlock. Also i've read that 0xFFFFFFFF means that
it can be modded whit the rv480. Is this the same with the rv430?
Click to expand...

Yes, sort of. As I read it, the important bits for the fuses are in all
X800 type boards.
CONFIG_DIE_FUSES = ####F### and
CONFIG_SUBSTRATE_FUSES = #######F
So a register does not have to specifically read "FFFFFFFF" but needs the F
in the indicated positions.
 
ATI tool reports all gto2 cores as 430, probably the same for all if there
are gto 480 cores.
It's been proven that 480 cores reported as 430 cores are 480. Does that
make sense? I'm tired.

Mike
 
Perhaps a good way to find out is to OC (max clock in ATI Tool). From what
I've been reading the R480 should easily clock past 500 with stock cooling
but the 430 will start to suffer at about 440 MHz and never get past 470
MHz. Or you could lift the HSF and see if it says its R480.

Regards,
Dan
 
Back
Top