Question Zen 6 Speculation Thread

[Win2012R2]

I was hoping for AMD to make a fork soon though.

It's in AMDs interest to have it gone.

 

[Tuna-Fish]

Yes they did and it's not surprising lol who is going to fund it when they don't have to fund it I was hoping for AMD to make a fork soon though.

Clear Linux did it's purpose, the optimizations in it mostly made into other distros. It doesn't have to keep going.

 

[Bigos]

It's in AMDs interest to have it gone.

Right, the fastest distro for AMD CPUs - have it gone.

I would rather say both Intel and AMD should work on upstreaming all of the optimizations that made their CPUs faster on Clear Linux. Hopefully most stuff has already been done.

 

[Win2012R2]

Right, the fastest distro for AMD CPUs - have it gone.

It might be the faster for AMD, but if it was giving edge to Intel CPUs - which was its stated objective, then it's better for AMD for it to be gone, so fork is unlikely in my view.

Not saying it's right for the end users, but if even Intel can't sustain Clear Linux, then AMD won't either - most distros are catching up with higher defaults anyway.

 

[511]

It might be the faster for AMD, but if it was giving edge to Intel CPUs - which was its stated objective, then it's better for AMD for it to be gone, so fork is unlikely in my view.

Not saying it's right for the end users, but if even Intel can't sustain Clear Linux, then AMD won't either - most distros are catching up with higher defaults anyway.

those were possible due to Clear Linux having those defaults now what is going to happen

 

[Win2012R2]

those were possible due to Clear Linux having those defaults now what is going to happen

But was it defaults (that any distro can make) or was it actual changes to code to run better on Intel?

 

[511]

But was it defaults (that any distro can make) or was it actual changes to code to run better on Intel?

It was just of optimization for Intel platforms that can be ported over to other distros and applied to AMD as well

 

[LightningZ71]

It wasn't providing a CLEAR advantage to them in any way, and with AMD being ahead at the moment, the optimizations were only amplifying that fact. Effectively, they were just helping AMD marketing. In addition, many of the things they were pushing wound up in the kernel, and there's not a whole lot there to do for now.

 

[Io Magnesso]

After all, it's ridiculous that Zen6 is aiming for 7GHz.

 

[Io Magnesso]

I don't want to believe what Leeker says

 

[Geddagod]

After all, it's ridiculous that Zen6 is aiming for 7GHz.

I don't even think mid 6GHz is all that unbelievable. But 7GHz+ does seem outlandish, yes.

 

[Tuna-Fish]

I don't even think mid 6GHz is all that unbelievable. But 7GHz+ does seem outlandish, yes.

I'd just remind everyone again that 7GHz is supposedly the frequency target, and that AMD has set very aggressive frequency targets (which they then miss) for the entirety of Lisa Su's tenure. If the result is 6.5GHz, 7GHz was probably the target.

 

[RnR_au]

If the result is 6.5GHz, 7GHz was probably the target.

How do we know that 6.5GHz wasn't the target?

 

[Io Magnesso]

To be honest, I think 6.5GHz would be difficult.
If it comes out even at 6GHz, I think it's a good hand

 

[Tuna-Fish]

How do we know that 6.5GHz wasn't the target?

... If it hits 6.5, we know that 6.5 wasn't the target because for the past decade AMD has set unrealistically high frequency targets that they always miss?

If it only gets 6GHz, then maybe 6.5 is the target.

The part I'm annoyed with is that half the internet is going "omg 7GHz AMD wins forever", like, it's not going to hit 7GHz, and the other half is going "MLID is a clown and lying, 7GHz can't possibly be true" ... well he is a clown, you got me there, but he also very clearly has had a bunch of internal AMD documents before anyone else. He probably saw a 7GHz target somewhere and either didn't understand what it means or didn't bother to explain to his viewers that no, for AMD 7GHz target probably doesn't mean 7GHz chips. But 7GHz target is credible to me, based on the better litho and AMD being "optimistic" about frequencies as always.

 

[Geddagod]

... If it hits 6.5, we know that 6.5 wasn't the target because for the past decade AMD has set unrealistically high frequency targets that they always miss?

If it only gets 6GHz, then maybe 6.5 is the target.

The part I'm annoyed with is that half the internet is going "omg 7GHz AMD wins forever", like, it's not going to hit 7GHz, and the other half is going "MLID is a clown and lying, 7GHz can't possibly be true" ... well he is a clown, you got me there, but he also very clearly has had a bunch of internal AMD documents before anyone else. He probably saw a 7GHz target somewhere and either didn't understand what it means or didn't bother to explain to his viewers that no, for AMD 7GHz target probably doesn't mean 7GHz chips. But 7GHz target is credible to me, based on the better litho and AMD being "optimistic" about frequencies as always.

His claim is that it's targeting above 7GHz. Him saying it's targeting 7GHz is him hedging. He also claims 7GHz is "kinda in the middle" of where they are trying to achieve. I would rewatch his segment on it from June loose ends, it's only 5-10 mins.
Hitting mid 6GHz while targeting above 7GHz would be a worse miss than AMD targeting 6GHz and hitting 5.7GHz for Zen 4.

 

[yuri69]

I'd just remind everyone again that 7GHz is supposedly the frequency target, and that AMD has set very aggressive frequency targets (which they then miss) for the entirety of Lisa Su's tenure. If the result is 6.5GHz, 7GHz was probably the target.

Even if the result is 6.150GHz the story could run as "AMD targeted 7GHz but RDNA3-ed"... The thing is, outside world would never know.

 

[marees]

Even if the result is 6.150GHz the story could run as "AMD targeted 7GHz but RDNA3-ed"... The thing is, outside world would never know.

I still remember MLID crying "AMD lied" because RDNA 3 didn't match his leaks

 

[511]

I still remember MLID crying "AMD lied" because RDNA 3 didn't match his leaks

Wait till you watch his latest Intel video where he says he doesn't leak Intel stuff cause Intel fails to achieve what he was told lol(which were his fabricated leaks).

 

[Josh128]

RDNA 3 was a peculiar case because we actually have official AMD slides saying that it was engineered for >3GHz, so we know 100% for sure that was the target, yet it fell short of that by ~20% or so.

RDNA 3 GHz

 

[mostwanted002]

Any confidence in AMD doing dual IMCs on Zen6?

https://twitter.com/x/status/1947641226204786878

In the light of some mobos (1 so far, not much) doing revisions around memory traces.

 

[Kaffeekenan]

Any confidence in AMD doing dual IMCs on Zen6?

https://twitter.com/x/status/1947641226204786878

In the light of some mobos (1 so far, not much) doing revisions around memory traces.

If so, what would be the advantage of such a dualcontroller?

 

[mostwanted002]

Intel has been doing dual IMC for a while. One IMC per channel, allowing better stability at higher speeds on DDR5 platforms.

 

[MS_AT]

AMD doing dual IMCs on Zen6

Don't tell anyone but:

View attachment 127004
Just so we get it clear, Raphael and Granite Ridge ALREADY HAVE 2 memory controllers. One per channel, handling two sub-channels, just like Intel. Why this is news is crazy to me. Even RTG is covering it like it's a novel feature.

 

[511]

Don't tell anyone but:

You should mark leak as a spoiler especially such confidential info