• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

[ ChipHell ] QuadFire R9 390x Firestrike Extreme: 38875

KaRLiToS

Golden Member
SOURCE
Source #2


f6352df2e9314fc4a6eebbf5cb09209c.png


20812ac0f1474b70baee2f8811b6a037.jpg


Poor Translation
Today there are users " cooldogman "In Baidu Post Bar gives a figure known as running sub R9 390X Quad Crossfire, with the Core i7-5960X, 3DMark Fire Strike Extreme project reached 38,875 points.
This is what concept? In the current 3DMark official rankings, the best result is four parallel 29,207 from Core i7-5960X, GTX 980, and are significantly overclocked, this four R9 390X cards can lead a full 33%!
Judging from the screenshot, 3DMark for identifying new card already in place, the core frequency to 1455MHz Ultra , memory frequency of 5GHz, but not the frequency of the processor, but also to see the rumors of HBM high-bandwidth memory.
Puliao who said that this achievement and the last leaked ran 3DMark 11 points, are from Fujian friends Xel'Naga hair test, but used the same graphics card and driver samples from a Japanese friend in Taiwan to work in Shanghai .
 
So, who's ready for another game of "find the real benchmark?"

The fact that the translated quote mentions the previous fake 3DMark 11 benchmark raises some suspicion... Not really sure what it is trying to say, but I think they are playing it off like that one was legit?
 
Well, if that's the 390x in the photo, I guess someone could play Little Detective and try to match it to an existing card to figure out if it's legit.
 
You want to know what confirms that it's fake? The fact that 3DMark knows what card it is. Also the fact that it decimates the current world record, which is less than 30k.
 
Maybe the picture in the ChipHell OP is not related to the benchmark score but the PCB look like the HD 7970 reference PCBS

2mo1rvs.jpg
 
Just look at the outputs on the bracket. Fake picture, its a 7970 for sure.
 
Just look at the outputs on the bracket. Fake picture, its a 7970 for sure.

Not to mention that even if by some extreme coincidence the PCB looked that similar to the 7970 PCB, that still doesn't explain the fact that the "leaked" picture clearly shows the connections for GDDR5 modules (HBM memory would be on package).
 
Not to mention that even if by some extreme coincidence the PCB looked that similar to the 7970 PCB, that still doesn't explain the fact that the "leaked" picture clearly shows the connections for GDDR5 modules (HBM memory would be on package).

And based on this, the die would be way too small.
 
I thought AMD was moving away from physical bridges for Crossfire (at least for the top-end parts)? If that's the case then how convenient that that part of the card was left out I'm the supposed image.

Edit: missed the earlier post on this. Can't delete my post... oh well
 
Last edited:
It's way too early for any legitimate leaks. It's time to face the facts......

Every leak so far is bogus.
 
Back
Top