Jump to content
HWBOT Community Forums

3dmark11...


zeneffect

Recommended Posts

need wrapper for 3dm11 now it seems too...

Ain't nobody got time for that :D PM me with details, and I can come up with something.

Can you also post a comparison score ON/OFF 'tweak' ?

 

Tell futuremark they will fix this

 

This. Now while it's still 'latest', there is a reason why they verify drivers before scores are valid.

Edited by GENiEBEN
Link to comment
Share on other sites

Ain't nobody got time for that :D PM me with details, and I can come up with something.

Can you also post a comparison score ON/OFF 'tweak' ?

 

 

 

This. Now while it's still 'latest', there is a reason why they verify drivers before scores are valid.

 

If they actually have to update the benchmark, it won't help. No one uses the latest version anyway since it gives lower scores than v101.

 

If it's just an update to their validation system, then go for it!

Link to comment
Share on other sites

v1.0.3 gives higher game tests, lower Physics.... depends on your system which version will be most efficient.

 

It would be nice if FM had version number visible somewhere really obvious, then HWB could (if they wanted) enforce the version that could be used, same as wPrime.

Link to comment
Share on other sites

you guys realize that i posted how to do it right?

 

v1.01 btw... though i doubt the version number will change the effect of the exploit. i see alot of missing geometry in 2nd game tests (car + forest) as well as missing geometry in combined score. i will see if this effects nvidia as well tonight.

 

*edit*

 

looks like physics test got a boost too... i ran ultra noob xmp profile @ 2400 with kingston predator 2666 kit.

Edited by zeneffect
Link to comment
Share on other sites

I did notice, thought it's more to it than just that.

So what is the default value with latest amd driver (or any other)?

(on nvidia rig now so I really can't personally check).

 

default is 1. set it to 0 and give it a test. this is with the 12.11 driver. when i get home i will send you a detailed guide as to what i did :)

 

i KNOW its not legal by just watching the screen... though the benchmark isnt "tricked" in anyway... its just the hardware is restricted to the point to where it cant render everything the way it should and skips alot of what is in the game test scenes. ;)

 

tested it last night for a few hours, and its from changin the flip depth. i can get same results with 2-5 but not a high score as 0.

Edited by zeneffect
Link to comment
Share on other sites

Not sure what the discussion really is about. The result is invalid. Yes, you can do all kinds of things that will give you "big numbers" but if the result, as shown on 3dmark.com, doesn't validate, it is fairly meaningless. There is a lot you can run & submit with 3.54 SystemInfo that will not fly with 4.12...

 

We'll get far more strict with the SI version (and benchmark version) as soon as the new installers are finally out.

 

Good point on the version number on the result screen. I'll see if we can do something about it for this next patch... and as for the next 3DMark result screen, oh it will have [redacted] and [redacted] :D

 

Now if you have similar cheat score, submitted with v1.03 and SI 4.12 and 3DMark.com doesn't complain (no anomaly box), I'd be interested to see a link.

Edited by FM_Jarnis
Link to comment
Share on other sites

Does this happen with older drivers and flip queue = 0?

12.11 seems a bit special to me, scores jump way high compared to old drivers, like 15% without any tweaks outside CCC...wonder what they did to get the extra FPS?

 

I tested this driver in 3DMark 11 and Unigine yesterday. my card was Aircooled, the boost is amazing, but card gets much hotter now... And i cant reach my clocks on air, GPU is 20 MHZ slower now, but score 15% Higher, lol... Seems like AMD is buying performance with current :P

Link to comment
Share on other sites

Not sure what the discussion really is about. The result is invalid. Yes, you can do all kinds of things that will give you "big numbers" but if the result, as shown on 3dmark.com, doesn't validate, it is fairly meaningless. There is a lot you can run & submit with 3.54 SystemInfo that will not fly with 4.12...

 

We'll get far more strict with the SI version (and benchmark version) as soon as the new installers are finally out.

 

Good point on the version number on the result screen. I'll see if we can do something about it for this next patch... and as for the next 3DMark result screen, oh it will have [redacted] and [redacted] :D

 

Now if you have similar cheat score, submitted with v1.03 and SI 4.12 and 3DMark.com doesn't complain (no anomaly box), I'd be interested to see a link.

 

http://www.3dmark.com/3dm11/4767105

 

only thing orb seems to care about is tessellation. we dont really seem to care about that here on the bot.

 

i can re-run with the bench with tess on to see if it makes an impact. need to re-update to 12.11... 12.10 SUCKS in comparison lol.

 

*edit*

 

rolled back up to 12.11 beta 3 and now i cant run 3dm11. crashes @ gt1 instantly :/

Edited by zeneffect
Link to comment
Share on other sites

I tested this driver in 3DMark 11 and Unigine yesterday. my card was Aircooled, the boost is amazing, but card gets much hotter now... And i cant reach my clocks on air, GPU is 20 MHZ slower now, but score 15% Higher, lol... Seems like AMD is buying performance with current :P

Just lower the GPU clock then xD

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...