Another, another Zen 2 Leak

[zhubaohi]

So recently, a guy on Taobao (Chinese part of Ali-express) claims that he gets his hand on some early samples of zen2, including a few early samples of 3300X and a sample of 3858X.

This guy is selling 3300X for around 200 USD, and 3858X (should be 3850X, it could be a typo on his side) for 1300USD. According to him, the 3300X he is selling has 6 cores/12 threads and 15MB cashe. He also claimed that the base clock speed is 5.0 Ghz, and can be overclocked to 5.8Ghz, but I doubt it. (Could be single-core rather than 6-core).

 

Now this could be a scam, but there are people in Chinese tech community already paid for it, and I'll let you guys know if there's any follow up. This could also indicate that AMD has passed early-sample test stage, and is really for announcement. 

 

My source:

Taobao link will be posted later

 

[alatron978]

5.8 GHZ IS INSANE ON ONE CORE. If it's true all I have to say is RIP intel.

[NunoLava1998]

1 minute ago, alatron978 said:

5.8 GHZ IS INSANE ON ONE CORE. If it's true all I have to say is RIP intel.

I think he meant 4GHz and 4.8GHz. 6GHz is stupid as fuck lol

[NunoLava1998]

Also it's not 3858X it's 3850X

[zhubaohi]

1 minute ago, NunoLava1998 said:

Also it's not 3858X it's 3850X

You're probably right. It's probably a typo on his end, I'm just copying whatever he said.

[Quadriplegic]

He can take that 5.8 GHz typo and stick it in his ass.

I'd say it's a scam, but will keep an eye on this

[NunoLava1998]

1399 chinese yuan = 199$

3999 chinese yuan = 599$

 

That looks pretty high, wouldn't be surprised if it's fake

[cj09beira]

it seems to me the launch might be closer than the rumored July time frame,

and Op thanks for keeping an eye on those forums, 

[Doobeedoo]

5GHz base clock. Wouldn't mind seeing that once again though.

[GoldenLag]

Probably fake. I see no way that it is legitimate 

[TetraSky]

As much as I would love to believe it, there's no way this is true. 

[D13H4RD]

There is no damn way that CPU has a 5GHz base clock 

[cj09beira]

4 minutes ago, D13H4RD said:

There is no damn way that CPU has a 5GHz base clock 

its possible that a extreme edtition cpu can be that fast, just look how much clockspeed vega gained by going to 7nm, some people have it running well over 2100mhz, lets not claim one way or another before we see some kind of proof 

[GoldenLag]

28 minutes ago, cj09beira said:

its possible that a extreme edtition cpu can be that fast, just look how much clockspeed vega gained by going to 7nm, some people have it running well over 2100mhz, lets not claim one way or another before we see some kind of proof 

Its roughly 20% gain over 14 nm. Based on slides from AMD and info from TSMC.

 

You can also do some approximation by making the assumption glofo 14nm is the same as TSMC 16nm. 

[cj09beira]

10 minutes ago, GoldenLag said:

Its roughly 20% gain over 14 nm. Based on slides from AMD and info from TSMC.

 

You can also do some approximation by making the assumption glofo 14nm is the same as TSMC 16nm. 

while true we cant use those figures because they are set at the middle of the frequency curve not the end which is what we care about

[jiyeon]

3 hours ago, zhubaohi said:

According to him, the 3300X he is selling has 6 cores/12 threads and 15MB cashe. He also claimed that the base clock speed is 5.0 Ghz, and can be overclocked to 5.8Ghz

I was fully with the claims until I read that last part. My 14nm 6 cores/6 threads CPU tops at 5.3GHz and he's claiming that a 7nm 6 cores/12 threads CPU can do 5.8GHz.

 

I'll use a Pentium for a week if this becomes true.

[MeatFeastMan]

Well...it doesn't always work out like this but...

 

If we look at the specs of ryzen 1st gen to ryzen 2nd gen, the first gen has a max boost clock of 4ghz, while the max boost on the 2nd gen is 4.3ghz. That's 300mhz for a 2nm node shrink. That means it is around 150mhz per 1nm of shrink.

 

So we should expect 750mhz more from 12nm to 7nm. But, considering we are going to TSMC from a low power solution, it's safe to say we may get 200mhz more maybe. So I'd definitely think it will be at least 5ghz boost. But 5.8ghz is insanity. However it is possible that we could close in on 5.5. AMD will have been working far harder on zen2 than they would have been on zen+. So I wouldn't completely rule out the cpu pulling away from 5ghz to maybe 5.4/5.5.

 

The main improvements will be made on memory most likely, that's the one area where Ryzen could become a real thorn in Intel's side. If they can get it around 4000mhz ram, that could really help them pull away from the 9900k.

 

5.8ghz is crazy and unreasonable unless AMD have thrown the kitchen sink at this release, but I wouldn't rule out going well above 5ghz.

[CarlBar]

I'd buy single core OC might go above 5Ghz, and i wouldn't rule out someone like Derbauer being able to hit 5.8Ghz on a single core non-sub-zero if you can pick which core on the chiplet and have a lot of samples to find the absolute most golden silicon.But not as a regular OC type of thing.

[Taf the Ghost]

5.8 Ghz is only happening on a Chiller, at minimum. That's going to be more LN2 range.

[Master Disaster]

The fact BIOSes exist that support Zen 2 proves they're passed the early sample stage.

[CarlBar]

I wonder how many of the supposed buyers are actually AMD trying to find out if this is a genuine leak and where in the chain they came from if it is?

[Taf the Ghost]

7 minutes ago, Master Disaster said:

The fact BIOSes exist that support Zen 2 proves they're passed the early sample stage.

 

1 minute ago, CarlBar said:

I wonder how many of the supposed buyers are actually AMD trying to find out if this is a genuine leak and where in the chain they came from if it is?

We know the chips are coming in the April to May time range, so it's not impossible the seller has some ES samples. What isn't real is 5.8 Ghz.

[S w a t s o n]

Cmon guys, 16 cores at 5.8GHz in the AM4 package, what could possibly go wrong.

 

*CLC explodes due to steam pressure*

[williamcll]

Looking up online shows up a Baidu forum thread that has just as much speculations. Hard to say where to source is because looking up TaoBao doesn't seem to show anything.

[descendency]

6 hours ago, alatron978 said:

5.8 GHZ IS INSANE ON ONE CORE. If it's true all I have to say is RIP intel.

 

 

The rumors on the street (maybe that are being fed by Intel fanboys or Intel itself) is that Intel's 7nm process is due 2020 or 2021, just about skipping 10nm. I don't really know enough to know how much it matters - but I think 'RIP Intel' is a bit over simplified (also considering they have a >80% market share...)