Speculation: Ryzen 3000 series

[jpiniero]

In case that anyone had any doubt that the Adored leak was made up by him

https://www.anandtech.com/show/1385...nt-on-matisse-cpu-tdp-range-same-as-ryzen2000

No APUs, no GPU chiplet.

Pretty disappointing, especially since Intel's already has 6 core mobile "APUs" available and 8 cores coming. I know, it's dubious to put that many cores in a laptop but it does sort of imply that Renoir will be 4 cores tops. But I guess that could change, depending on the release date.

How does that prove that he made it all up ? We don't know who his leak was, and what they may have said.

HIs made up model list implied that the CPU release/announcement was at CES, and that there were GPU chiplets coming too in Q3.

He also pretty obviously made up a model list of Navi.

 

[ozzy702]

HIs made up model list implied that the CPU release/announcement was at CES, and that there were GPU chiplets coming too in Q3.

He also pretty obviously made up a model list of Navi.

^ This. Hearing the AMD fanbois drool over Adored's wishful thinking (not leaks) and then being disappointed when they turn out to be false is hilarious. Steve at Hardware Unboxed was dead on.

 

[Shivansps]

How does that prove that he made it all up ? We don't know who his leak was, and what they may have said.

GPU chiplets was the only thing left that could had been true and it may given more credit to everything else he said, now is clear he petty much made up the whole thing using the reddit leak and that the whole leak was wishfull thinking on his part.

Ive already said this, if he had know that AMD was launching the 240GE at $75 so late at december after his videos he would have never put a $100 6C/12T on his table, that is a clear sign something was up.

 

[Mopetar]

HIs made up model list implied that the CPU release/announcement was at CES, and that there were GPU chiplets coming too in Q3.

He also pretty obviously made up a model list of Navi.

now is clear he petty much made up the whole thing using the reddit leak.

I don't quite get why people are claiming (or even thinking) that he made it up. It's far more likely that someone else made something up and sent it to him.

Attributing malice when other explanations are more suitable is foolish. I think it almost says more about the person making the claims than it does about Adored.

 

[Shivansps]

I don't quite get why people are claiming (or even thinking) that he made it up. It's far more likely that someone else made something up and sent it to him.

Attributing malice when other explanations are more suitable is foolish. I think it almost says more about the person making the claims than it does about Adored.

Maybe, but he is the face here, thats really not excuse.


About the APUs it is clear that AMD will be launching Picasso at desktop as well, and there may be a Zen2 monolithic APU later on according to what AMD said, im not sure if it worth to launch another APU on DDR4. maybe GPU chiplets will be coming for DDR5.

 

[Mopetar]

Maybe, but he is the face here, thats really not excuse.

There is a world of difference between reporting a rumor that turns out to be bogus and manufacturing your own bogus information and trying to pass it off as a new rumor.

Conflating the two in any way is every bit as dishonest and morally bankrupt as what you're accusing him of doing.

 

[Shivansps]

There is a world of difference between reporting a rumor that turns out to be bogus and manufacturing your own bogus information and trying to pass it off as a new rumor.

Conflating the two in any way is every bit as dishonest and morally bankrupt as what you're accusing him of doing.

We have no way to know what happened, what I do know, as with every youtuber, he gains money out of those videos in one way or another. So the mistrust is high, everyone can made things up claiming some "contact" has given the information to him. So i dont undertand why so many people is defending it, they should demanding him to explain what happened.

 

[PeterScott]

We have no way to know what happened, what I do know, as with every youtuber, he gains money out of those videos in one way or another. So the mistrust is high, everyone can made things up claiming some "contact" has given the information to him. So i dont undertand why so many people is defending it, they should demanding him to explain what happened.

This.

I don't care if he made it up, or passed off something someone else made up.

The "leak" was obviously too good to be true, and looked more like fanboy wishful thinking.

IMO it should have been obvious from the start that the information was bogus.

But he makes his money on Views, so he has a vested interest in passing off juicy info, even if (obviously) false.

IMO he deserves to be slagged for hyping what was obviously false info for views, regardless of who made it up.

 

[Tuna-Fish]

I wonder why PC World was told they cannot show the backside of the new Ryzen package. Gordon mentions it about one minute into their video.

That makes me wonder a bit too. Isn't there like nothing whatsover on the bottom of an AM4 package?

Spoiler: am4 from behind

What reason could there possibly be to ban looking at that? Have they stuffed some new power delivery/interesting other crap in the middle hole? Would there even be room on all motherboards? I looked at a couple of them, and what they stuff there seems to vary.

 

[LTC8K6]

You could theoretically mount a chip there?

 

[exquisitechar]

Adored's leaker never said that the APUs would be chiplet based, that was his own speculation, IIRC.

The ES seems to match up pretty well with Adored's "Ryzen 5 3600". So, at least as far as the CPUs are confirmed, "too good to be true" is not proven to be the case for now. A 5.1GHz SC turbo, though...now that remains to be seen.

 

[Spartak]

I'm not sure why people dismiss the adoredTV "leak"?

The 3000 series APU Zen+ launched, and the Zen2 'debuted' at CES. Were people really expecting AMD to launch them right now? That's an assumption that Adored has never claimed.

His line-up with pricing is still highly likely TBH and I can also see AMD launch assembled 8-core APU's for the desktop in 19Q3.

The performance is there, pricing might be higher around entry-level, but at the R7/R9 you can bloody bet AMD will want to cause mayhem to Intel.

 

[Shivansps]

That makes me wonder a bit too. Isn't there like nothing whatsover on the bottom of an AM4 package?

Spoiler: am4 from behind

What reason could there possibly be to ban looking at that? Have they stuffed some new power delivery/interesting other crap in the middle hole? Would there even be room on all motherboards? I looked at a couple of them, and what they stuff there seems to vary.

Maybe, just maybe, the CPU at CES was not AM4.

 

[LTC8K6]

Maybe, just maybe, the CPU at CES was not AM4.

That would make sense but it seems like you could see that anyway?
They would Shirley have told us that it was not AM4.

 

[Shivansps]

That makes me wonder a bit too. Isn't there like nothing whatsover on the bottom of an AM4 package?

Spoiler: am4 from behind

What reason could there possibly be to ban looking at that? Have they stuffed some new power delivery/interesting other crap in the middle hole? Would there even be room on all motherboards? I looked at a couple of them, and what they stuff there seems to vary.

That would make sense but it seems like you could see that anyway?
They would Shirley have told us that it was not AM4.

Not really, if AMD is working on a new socket or a revision of AM4 is not something you want people to know right now.
How far off is DDR5 anyway?

In the same way they did not wanted to show more than 8C on AM4 what whould have killed TR sales.

 

[exquisitechar]

How far off is DDR5 anyway?

2021 for server, 2022 for consumer, I think.

 

[LTC8K6]

Not really, if AMD is working on a new socket or a revision of AM4 is not something you want people to know right now.
How far off is DDR5 anyway?

In the same way they did not wanted to show more than 8C on AM4 what whould have killed TR sales.

What is stopping a reporter from noticing that it's not AM4 and saying so?
They could apparently look at the back, just not show it.

 

[trollspotter]

AM4+ for PCI4.0 or something else? Or it could have just been poorly worded and we are reading too much into what he said.

 

[Mopetar]

How far off is DDR5 anyway?

The most recent rumors suggest the end of 2019 as being the most likely, but I wouldn't be surprised if it slipped into 2020 for mainstream parts.

In the same way they did not wanted to show more than 8C on AM4 what whould have killed TR sales.

They showed a die shot which made it plainly obvious to anyone paying attention there will be 2 chiplet Ryzen parts.

It's also pretty obvious that ThreadRipper is going to be using 4 chiplets at a minimum, and it wouldn't be too surprising for them to release an 8 chiplet version either.

ThreadRipper isn't affected any more by this than AMD's other product lines. Due to Intel's supply shortage there's apparently such a demand right now that everything continues to sell no matter how much companies might Osborne their products.

 

[Mopetar]

We have no way to know what happened

Then why make an accusation such as "he made it up" if when by your own admission you don't know.

If I were to hear that you got into a car accident, it would be pretty presumptuous of me to immediately assume that you must have been drinking and on the phone, which is what lead to the accident.

 

[LTC8K6]

https://www.techquila.co.in/sk-hynix-develops-first-16-gb-ddr5-5200-memory-chip/

Mass production from Hynix in 2020.

 

[LTC8K6]

"Micron still expects to start production of DDR5 memory chips in late 2019. "

Also mass production in 2020.

Anandtech article.

 

[Shivansps]

What is stopping a reporter from noticing that it's not AM4 and saying so?
They could apparently look at the back, just not show it.

It would be hard to figure out right at the moment by just looking at the pins.
This whole thing is making me suspect that the I/O die may support both DDR4 and DDR5, and that may be an "AM5" in the works.

Altrought it would make no sence to have such CPU at CES if you are not going to say anything about it.

 

[LTC8K6]

It would be hard to figure out right at the moment by just looking at the pins.

I guess, but apparently it would be noticed by viewers if they pointed it at a camera?

 

[PotatoWithEarsOnSide]

Personally I'm more inclined towards a mix of 10% IPC and 5% frequency uplift for the CES sample, I guess there were 1-2 other members who thought the sample was running 4.2-4.3Ghz.

I expect another 10% frequency increase at release, so in my book that goes to 4.7Ghz. CB score should be unaffected by IPC-fmax ratio, and should hover above 2250.

PS: it would actually be better if it was all IPC uplift and clocks were ~4GHz, that would almost guarantee a 10% clock increse at launch.

Yup, I was at 4.2-4.3GHz but now I've widened that a little to 4.1-4.3GHz.
The consensus is definitely 4.0-4.4GHz, and anything 4.5GHz and above represents a regression, so simply unlikely.
The talk of the last few months was of 9-13% IPC increase, and that fits the 4.1-4.3GHz range.
In all honesty though, an ES running at above 4.0GHz? When was the last time that we saw something like that?