NotebookForums.com › Forums › Notebook Manufacturers › Dell Forums › Dell Home (Inspiron, XPS, Studio) › 6800 Go Successful Bios Flash
New Posts  All Forums:Forum Nav:

6800 Go Successful Bios Flash - Page 19

post #361 of 427
Wow this is some interesting BIOS hacking. Has anyone actually verified that the voltages of 6800 Go are controlled by the BIOS? That's about the only difference between the NV42 non Ultras and the Ultras.

If anyone ignites their laptop I want photos!
post #362 of 427
Thread Starter 
Quote:
Originally Posted by swaaye
Wow this is some interesting BIOS hacking. Has anyone actually verified that the voltages of 6800 Go are controlled by the BIOS? That's about the only difference between the NV42 non Ultras and the Ultras.

If anyone ignites their laptop I want photos!
A very intersting question. It would seem logical, but it might be driver controlled. I have always assumed that the driver's powermizer used flags that were defined to represent voltages in the bios, but it could be direct control.

Driver voltage control would be a good thing to rule out for certain.
post #363 of 427
Ran at 487/1200 for over four hours in EQ2 last night. I'm going to try 500/1250 next I think.

3dmark05:
450/1060 = 5140(ish)
475/1200 = 5550
487/1200 = 5650 (88C in RivaTuner monitor log using ATITool)
500/1200 = 5752
500/1250 = 5850

I think the formula here is for every 50 MHz of memory of 12.5 MHz of CPU, you're going to get another 100 3DMarks in 3Dmark05.
post #364 of 427
question, can someone provide a link as to where I can buy the geforce 6800 go ultra gfx cooler? i'm interested in flashing my non-ultra too (after some of you brave people try it first of course)!
post #365 of 427
Quote:
Originally Posted by johnnobts
question, can someone provide a link as to where I can buy the geforce 6800 go ultra gfx cooler? i'm interested in flashing my non-ultra too (after some of you brave people try it first of course)!
You cant buy it without the ultra card. The heatsink is a non serviceable part.
post #366 of 427
You know if the Sager and Dell bios are completely different in hex view its because they are different versions. Have any of you ever looked at a driver file between versions? Hardly any of the hex addresses match up so I'd assume the same is true for the two bios's. If someoen has the balls (I don't since this is my only pc atm) I'd be curious to see what happens when you force flash the sager bios.
post #367 of 427
Effect of overclocking and driver selection on real-world game performance (FarCry).

Resolution: 1600×1200
Ultra quality option, Direct3D renderer
Level: Research, demo: Research.tmd
Pixel shader: default model
Antialising: None
Anisotrophic filtering: 2×
HDR: level 6
Geometry Instancing: disabled
Normal-maps compression: disabled

Driver 76.50 stock settings
Score = 30.72 FPS (Run 1)
Score = 30.75 FPS (Run 2)
Average score = 30.73 FPS

Driver 77.30 at 475 core 1250 memory settings
Score = 26.93 FPS (Run 1)
Score = 27.22 FPS (Run 2)
Average score = 27.07 FPS

Driver 76.50 at 475 core 1250 memory settings
Score = 31.60 FPS (Run 1)
Score = 31.84 FPS (Run 2)
Average score = 31.71 FPS

Driver 76.50 at 500 core 1250 memory settings
Score = 32.79 FPS (Run 1)
Score = 33.04 FPS (Run 2)
Average score = 32.91 FPS

Verdict:
(1) 77.30 appalling performance in FarCry.
(2) 76.50 nice performance in FarCry.
(3) 475/1250 is only 3% faster than 460/1060.
(4) 500/1250 is only 7% faster than 460/1060.

*********************
EDIT: results without HDR
*********************

Resolution: 1600×1200
Ultra quality option, Direct3D renderer
Level: Research, demo: Research.tmd
Pixel shader: default model
Antialising: 4×
Anisotrophic filtering: 16×
HDR: disabled
Geometry Instancing: disabled
Normal-maps compression: disabled

Driver 76.50 stock settings
Score = 49.45 FPS (Run 1)
Score = 50.52 FPS (Run 2)
Average score = 49.98 FPS

Driver 76.50 at 500 core 1250 memory settings
Score = 57.09 FPS (Run 1)
Score = 57.82 FPS (Run 2)
Average score = 57.45 FPS

Verdict without HDR:
500/1250 is 14.9% faster than 460/1060
post #368 of 427
HDR severely limits these cards and isn't used in too many games. Disable it and run comparisons with AA/AF cranked up and see if it makes a difference - I'm betting the extra fillrate and bandwidth will definitely come in to play.
post #369 of 427
I'm still running @ 525/1250 with no issues for about 24 hours now.
post #370 of 427
Quote:
Originally Posted by cornholio
I'm still running @ 525/1250 with no issues for about 24 hours now.
Try running 3dmark03 and tell me what happens during the nature test.
post #371 of 427
I am updating real-world gaming results for FarCry with no HDR, will be edited in my previous post above.
post #372 of 427
Quote:
Originally Posted by RamIt
Try running 3dmark03 and tell me what happens during the nature test.
Will Try
post #373 of 427
Quote:
Originally Posted by yagiska
Yes. Only because no one has done it, and there is no way to know for certain what would happen.

That being done, I'd love to someone do it. Since you have the ultra cooler, you've got the greatest chance for not frying/melting something.

However, there's no telling if the card will take it. There is some argument for it, being that both cards have the same core, however, the ultra requires more watts, and I suspect that your new bios won't like the 9300's stock power supply.

Just remember this, nothing is every discovered without experimentation. The only way we'll ever know if it will work and be advantegous is if someone tries it. Just take precautions if you decide to.

You've got the cooler, you plan to start with lower clocks. Just make sure you know how to flash it back if your video dies and you have no display. If you melt something though, don't blame me, because I'll tell you right now it's risky.
I will get my hands on a 130w AC adapter.
I AM willing to try, but lower clocked first, to see if it handles GDDR1!

Just need to get the cooler first...
post #374 of 427
Can you run 3dmark2005 at the different speeds... but with 1920x1200 for the resolution?

I'm curious if it makes a bigger difference at higher resolutions.
Stock XPS2 I just got 2927 at 1920x1200 with the 76.50 drivers
post #375 of 427
Quote:
Originally Posted by RamIt
Try running 3dmark03 and tell me what happens during the nature test.

It crashes on nature for me too @ 525/1250.
post #376 of 427
I dont get it. It runs farcry, doom3 and hl2 fine at 530-1250 but crashes out of 3dm03 in the nature test at 500-1250
post #377 of 427
Quote:
Originally Posted by RamIt
I dont get it. It runs farcry, doom3 and hl2 fine at 530-1250 but crashes out of 3dm03 in the nature test at 500-1250
My new 450/1100 bios don't crash it... I don't know what would cause that....
post #378 of 427
UPDATE: real-world game performance in FarCry.
Amazing results with 80.40 driver without HDR

Resolution: 1600×1200
Ultra quality option, Direct3D renderer
Level: Research, demo: Research.tmd
Pixel shader: default model
Antialising: 4×
Anisotrophic filtering: 16×
HDR: disabled
Geometry Instancing: disabled
Normal-maps compression: disabled

Driver 76.50 stock settings
Score = 49.45 FPS (Run 1)
Score = 50.52 FPS (Run 2)
Average score = 49.98 FPS

Driver 76.50 at 500 core 1250 memory settings
Score = 57.09 FPS (Run 1)
Score = 57.82 FPS (Run 2)
Average score = 57.45 FPS

Driver 80.40 beta at 500 core 1250 memory settings
Score = 64.33 FPS (Run 1)
Score = 66.35 FPS (Run 2)
Average score = 65.34 FPS

Verdict without HDR:
(1) 500/1250 is 14.9% faster than 460/1060 for 76.50 driver
(2) Amazing performance for 80.40 driver
post #379 of 427
Well, I just set mine to 525c1250 and it crashes at the end of 3dmark05(not 03).

I am hoping to get this score as high as I can then when my new system gets here I wont have to worry about it.

Btw, I talked to Dell and they are sending me a new gen 2 due to some other issue I am having.
post #380 of 427
Post your 3DMark05 differences too...
New Posts  All Forums:Forum Nav:
  Return Home
NotebookForums.com › Forums › Notebook Manufacturers › Dell Forums › Dell Home (Inspiron, XPS, Studio) › 6800 Go Successful Bios Flash