Bugged CPUID Verification?

[(sin)morpheus]

Someone has achieved 4.018GHz with a venice 3200!

Here is the validated screen.

http://valid.x86-secret.com/show_oc.php?id=19402

[G H Z]

[Waus-mod]

I saw it this afternoon. posted it at the wr database. his mem speed is verryyyy high!

[kakaroto]

something tells me this looks bugged.

[conrad.maranan]

401.9 MHz with 3-3-3-8 timings is looking kind of suspect to me. I don't know about this one.

[(sin)morpheus]

Sorry, didn't see a thread for it and figured it deserved one.

I don't think it's a fake, how would it get validated if it was a fake?

[Ref]

That's clearly a BUG.
No way can anyone reach DDR800 on DDR1 with such timings.

[AlterBridge86]

CPU-z has bugs which cause things to be faked..i saw a validated Athlon64 at like 7ghz..just a glitch in CPU-z. but since its actually reporting it, it can be "validated". When Franck finds it i'm sure he'll look into it.

[(sin)morpheus]

I see, I haven't seen any fake cpu-z validated screenies yet, because I'm fairly new to the cpu-z validation scene.

It would be scary if that isn't a bug.

[charlie]

no way...

changed title to reflect situation...

[dreamworkz]

man...I wish the guy would show some pics getting to that wr mark. Like pedro and ricky they show pics of their monitor screens which proves that they did get to wr mark. I wish that guy would come on this forum or any oc forum and post some pics cause thats quite some achievement. Cause if not, people start to question if he did or not. I was suprised to see it was done with a venice and not a fx-55. and the ram timings and ram speeds is questionable too. I hope its a bug....cause ricky and pedro rock!!!

[(sin)morpheus]

Thanks for changing the title for me.

[Ref]

It is NOT real.
If someone wanted to get a record, he would NOT use such timings and memory speed, because CPU usually OCs easier with low memory freq.

And IF it was real, someone would give some normal name, not Rob Williams and more info about the record ... and he would post it on some forum... NONE of this happend. It is a bug.

[Sucka]

Without knowing who that is, i'm skeptical. I am not going to call cheat, but without hearing from this guy i just don't know...

[metro.cl]

It is NOT real.
If someone wanted to get a record, he would NOT use such timings and memory speed, because CPU usually OCs easier with low memory freq.

And IF it was real, someone would give some normal name, not Rob Williams and more info about the record ... and he would post it on some forum... NONE of this happend. It is a bug.

locked multi reminds you of something?? it a venice 3200+ so max multi is 10 that's why the ddr800 mems

still waiting to see what it is

[shadowing]

I highly doubt this is real. It seems like a cheat...

There is really no way to reach DDR800 speed with that voltage and timings. What kind of chip is even capable of doing that?

[bottone]

Do the AMD64 Memory controllers go higher in Single Channel mode?

I see he's only using a single stick..., I never ever tried single channel in my 939 system (whats the point??)

Although I still suspect it may be a bug.

[don_vercetti]

cynical bunch aren't we. Just because he isn't on THESE forums doesn't mean he doesnt exist, these aren't the only OC'ing forums. Isn't the WR on HTT something mad like 500mhz? Ultimately, i'm sure a DFI COULD reach that speed, theoretically with one stick of TCCD, highish volts.

But i guess that guilty until proven innocent, huh.

[cpuz]

Before we remove it, I need to know what is wrong (if something is wrong actually), otherwise obviously it will come back.

Edit : OK, I contacted Rob, and he explained me this was a BIOS issue. It is fixed now, and the entry was removed from the DB.

[Deathspawner]

Yeah, sorry about that guys.

I only uploaded it to show people on another forum to get help to fix the problem. I had no idea that it would get logged or in some high-score list. I thought the only way people would see it is to have the link handed to them.

[Deathspawner]

It is NOT real.
If someone wanted to get a record, he would NOT use such timings and memory speed, because CPU usually OCs easier with low memory freq.

And IF it was real, someone would give some normal name, not Rob Williams and more info about the record ... and he would post it on some forum... NONE of this happend. It is a bug.

How the heck is Rob Williams not a normal name? IT'S THE NAME I'VE USED FOR 21 YEARS!

don_vercetti, thanks for the back up, hehe. I have visited these forums for a couple years, but have just never signed up until now.

[don_vercetti]

No problem, some people tend to be a bit small minded.