Ryzen 3000 Leaks

[S w a t s o n]

Give me that sweet, sweet 50th anniversary edition AMD.

M O A R
O
A
R

C O R E S
O
R
E
S

[NunoLava1998]

2 minutes ago, CarlBar said:

 

 

Nice puns you two. And yeah the Ryzen 3 was probably the biggest suprise of this set of leaks for me. That if true is basically going to kill far more of intels lineup than even i was expecting.

 

The rest, i kinda got caught out on the combination of upped core count and significantly upped frequency, (the R9's aside which i basically called), but that was because i expected one or the other, not both, but i allways acknowledged both was possibble, i just didn't think they'd do it because of thermals.

The Ryzen 3 is legitimately insane, the highest end one (3300G) would be close to Intel's 8th gen i7's (telling from core count and GHz, it may not) and have Navi integrated, for fucking 129$

[williamcll]

With chips like these and threadripper 2990WX it seems like AMD is still on track on moore's law, unlike intel.

[The Benjamins]

2 minutes ago, VegetableStu said:

realistically the next mountain would be 5nm, but anything smaller we'd be close to fighting with physics

I thought I was somewhere that TSMC has plans for 3nm, I might be wrong tho.

[Deus Voltage]

1 minute ago, VegetableStu said:

realistically the next mountain would be 5nm, but anything smaller we'd be close to fighting with physics

Year 2030: SubAtomic-InfinityFabric-ButterDonnut-MultiChiplet-Zen 10- Base 5.1 GHz ....

 

On a serious note though, I really wonder how the CPU scene will look like in 2030. I just hope I'll be alive by then

[S w a t s o n]

3 minutes ago, The Benjamins said:

I thought I was somewhere that TSMC has plans for 3nm, I might be wrong tho.

 

1 minute ago, VegetableStu said:

wasn't it "4.5"? o_o (wait i think that was AMD's future roadmap)

https://en.wikichip.org/wiki/3_nm_lithography_process

[S w a t s o n]

1 minute ago, VegetableStu said:

huh. page doesn't have much though o_o

Well this stuff is still in concept stage. We're looking at 5 years out before engineering samples I think. Node names are a joke now, I think it's samsung that internally has a "node" at every single nanometer. So like 10nm, 9nm, 8nm, 7nm. Like that. TSMC could decide to call it 4.5nm but it's going to be 3nm class unless it's just 5nm++

[boey]

15 hours ago, Deus Voltage said:

Most of this information is provided by AdoredTV, I highly suggest watching this video for more information.

Edit 1: These leaks were released after the ones that were posted on Reddit (which I cannot find at the moment, if someone can find them, that'd be great!) 

Another edit: AdoredTV stated the following regarding the Reddit leak referenced in his video:

Edit 2: Here is the link to another older leak (Thank you Crunchy Dragon!): 

  Reveal hidden contents

 

 

 

With that being said, say hello to these little beasties!

 

Lower End:

 

Ryzen 3 3300: 6C/12T //  Base: 3.2 GHz //  Boost: 4.0 GHz // 50 W TDP // Price: 99$

Ryzen 3 3300X: 6C/12T //  Base: 3.5 GHz //  Boost: 4.3 GHz // 65 W TDP // Price: 129$

Ryzen 3 3300G: 6C/12T/ 15CU (Navi) //  Base: 3.0 GHz //  Boost: 3.8 GHz // 65 W TDP // Price: 129$

 

Mid Range:

 

Ryzen 5 3600: 8C/16T //  Base: 3.6 GHz //  Boost: 4.4 GHz // 65 W TDP // Price: 179$

Ryzen 5 3600X: 8C/16T //  Base: 4.0 GHz //  Boost: 4.8 GHz // 95 W TDP // Price: 229$

Ryzen 5 3600G: 8C/16T/ 20CU (Navi) //  Base: 3.2 GHz //  Boost: 4.0 GHz // 95 W TDP // Price: 199$

 

Upper Mid Range:

 

Ryzen 7 3700: 12C/24T //  Base: 3.8 GHz //  Boost: 4.6 GHz // 95 W TDP // Price: 299$

Ryzen 7 3700X: 12C/24T //  Base: 4.2 GHz //  Boost: 5.0 GHz // 95 W TDP // Price: 329$

 

High End:

 

Ryzen 9 3800X: 16C/32T //  Base: 3.9 GHz //  Boost: 4.7GHz // 125 W TDP // Price: 449$

Ryzen 9 3850X: 16C/32T //  Base: 4.3 GHz //  Boost: 5.1GHz // 135 W TDP // Price: 499$

 

 

 

Opinion:

I am really looking forward to seeing the benchmarks for these. The CPU scene is getting HOT! AMD is really shaking up the market in a big way! 

 

6c 12t for $99 and at 50w tdp?? what the fuck? 

[AngryBeaver]

This looks like something you would find on WccfTech. The clock speeds alone make me pretty sure this is fake. They aren't going to move from 2 generations of pretty hard caps to a 3rd gen where that magically disappears. You also aren't going to see boost speeds keep increasing as the core count goes up. If those boost are valid on the 16c then they would have no reason to not use them on a 8 core. They are trying to pull ahead of intel across the board, so that doesn't make sense.

 

I am willing to put money on the fact this is completely fake.

[williamcll]

27 minutes ago, VegetableStu said:

realistically the next mountain would be 5nm, but anything smaller we'd be close to fighting with physics

you could always increase surface area

[TigerHawk]

16 minutes ago, AngryBeaver said:

This looks like something you would find on WccfTech. The clock speeds alone make me pretty sure this is fake. They aren't going to move from 2 generations of pretty hard caps to a 3rd gen where that magically disappears. You also aren't going to see boost speeds keep increasing as the core count goes up. If those boost are valid on the 16c then they would have no reason to not use them on a 8 core. They are trying to pull ahead of intel across the board, so that doesn't make sense.

 

I am willing to put money on the fact this is completely fake.

Same. I feel like some of the information may be accurate but on the wrong SKU.

 

Take the 3600X for example

 

Ryzen 5 3600X: 8C/16T //  Base: 4.0 GHz //  Boost: 4.8 GHz // 95 W TDP // Price: 229$

 

now change it into this

 

Ryzen 7 3700X: 8C/16T //  Base: 4.0 GHz //  Boost: 4.8 GHz // 95 W TDP // Price: 329$

 

Suddenly its a lot more believable as a SKU. It is a reasonable upgrade over the previous gen and I'd say still a good buy for people looking to upgrade.

 

With this lineup, there are none that don't have SMT enabled. There is no way they would just trash all the CPUs with the SMT module that didn't form correctly. That is the entire reason you get CPUs like Ryzen 3 1200s with 4 cores 4 threads and such. It's a way to still make money off of dies that had about 60-70% of it malformed in the manufacturing process.

[asus killer]

the cheap bastard in me... 6C/12T for 99$ yeahhhh

[Big Head Tech]

Steve here from PC Budget Solutions and I want to chime in.  For those that don't know I am a HUGE AMD advocate ever since Bulldozer was such a failure.  During that time most PCs that came from my office were Intel.  Until Ryzen came around where I went 75% with Ryzen and 25% with Intel.  I have hope that AMD can push speeds to near 5GHz with a 5% IPC boost which should push them near or slightly past Intel.  If these leaks would be true, Intel would honestly, become the new Bulldozer.  AMD would have a 10% IPC advantage, matching or exceeding clock speeds, wreaking havoc on price and core count.  14nm yields couldn't match this and with the struggle of yielding good results from 10nm, Intel would be in serious trouble for 2 years, up to possibly 5 years.  I would be exceed but don't go getting a 2nd mortgage though.  That being said, if AMD does release anything close to these, me and my team will do our real world reviews of the entire lineup!

[AngryBeaver]

29 minutes ago, Turtle Rig said:

Looks like May 2019 were going to have a 32 thread desktop for 500 dollars and all running at 5.1Ghz ....... drools

 

Ryzen 3 3300	3.2GHz	4.0GHz	6/12	50W	$99	CES
Ryzen 3 3300X	3.5GHz	4.3GHz	6/12	65W	$129	CES
Ryzen 3 3300G	3.0GHz	3.8GHz	6/12	65W	$129	Q3 2019
Ryzen 5 3600	3.6GHz	4.4GHz	8/16	65W	$178	CES
Ryzen 5 3600X	4.0GHz	4.8GHz	8/16	95W	$229	CES
Ryzen 5 3600G	3.2GHz	4.0GHz	8/16	95W	$199	Q3 2019
Ryzen 7 3700	3.8GHz	4.6GHz	12/24	95W	$299	CES
Ryzen 7 3700X	4.2GHz	5.0GHz	12/24	105W	$329	CES
Ryzen 9 3800X	3.9GHz	4.7GHz	16/32	125W	$449	CES
Ryzen 9 3850X	4.3GHz	5.1GHz	16/32	135W	$499	May 2019

This is from another thread on this topic, but I think it is relevant here. My apologies Turtle for cross quoting.

 

See the items I highlighted in red. Look at the base clocks. I can understand them wanting to raise the boost clocks on higher end models, but the base clock doesn't make sense.

 

As core count increases normally you see a lower base clock, because of the amount of heat and power generated on these chips when all cores are running full out. The base clock is the lowest clocks would drop in that scenario. Lets compare it to their threadripper line which would be the best comparison to this.

 

 

As core count is reduced we see the base block come up as well as the boost clock... the only exception is the 2920x and I think this was more of a pricing move to push people to the 2950x. I marked out the older threadrippers. But if you look you will see the base clock works the same way there.

 

This is why I believe all of this "Leaked" information is all make up crap. There might be a slight grain of truth somewhere, but it isn't on actual cpu specs.

[The Benjamins]

6 minutes ago, TigerHawk said:

Same. I feel like some of the information may be accurate but on the wrong SKU.

 

Take the 3600X for example

 

Ryzen 5 3600X: 8C/16T //  Base: 4.0 GHz //  Boost: 4.8 GHz // 95 W TDP // Price: 229$

 

now change it into this

 

Ryzen 7 3700X: 8C/16T //  Base: 4.0 GHz //  Boost: 4.8 GHz // 95 W TDP // Price: 329$

 

Suddenly its a lot more believable as a SKU. It is a reasonable upgrade over the previous gen and I'd say still a good buy for people looking to upgrade.

 

With this lineup, there are none that don't have SMT enabled. There is no way they would just trash all the CPUs with the SMT module that didn't form correctly. That is the entire reason you get CPUs like Ryzen 3 1200s with 4 cores 4 threads and such. It's a way to still make money off of dies that had about 60-70% of it malformed in the manufacturing process.

If AMD can't get higher core count at the $300~ price point that would be a big let down for me. they can double cores on servers at a expected same price bracket but we can't see any improvement on desktop, that seems like a intel thing.

 

If it is what you said, it is a failure.

[TigerHawk]

2 minutes ago, AngryBeaver said:

This is from another thread on this topic, but I think it is relevant here. My apologies Turtle for cross quoting.

 

See the items I highlighted in red. Look at the base clocks. I can understand them wanting to raise the boost clocks on higher end models, but the base clock doesn't make sense.

 

As core count increases normally you see a lower base clock, because of the amount of heat and power generated on these chips when all cores are running full out. The base clock is the lowest clocks would drop in that scenario. Lets compare it to their threadripper line which would be the best comparison to this.

 

 

As core count is reduced we see the base block come up as well as the boost clock... the only exception is the 2920x and I think this was more of a pricing move to push people to the 2950x. I marked out the older threadrippers. But if you look you will see the base clock works the same way there.

 

This is why I believe all of this "Leaked" information is all make up crap. There might be a slight grain of truth somewhere, but it isn't on actual cpu specs.

Not to mention the R9 3850X releasing not too long after TR2 would completely cannibalize the 2950X, which I believe is the best selling TR SKU right now? and it would have higher base and boost clocks? With a MUCH lower cost of entry on the AM4 socket instead of TR?

[TigerHawk]

2 minutes ago, The Benjamins said:

If AMD can't get higher core count at the $300~ price point that would be a big let down for me. they can double cores on servers at a expected same price bracket but we can't see any improvement on desktop, that seems like a intel thing.

 

If it is what you said, it is a failure.

I don't think so. I think they would just not want to cannibalize the threadripper platform by putting too many cores on regular desktop.

 

Unless, that is, they are planning to do away with HEDT altogether and kill threadripper off. Just go Ryzen and EPYC. Maybe reuse the threadripper name for R9s or something.

[AngryBeaver]

1 minute ago, TigerHawk said:

Not to mention the R9 3850X releasing not too long after TR2 would completely cannibalize the 2950X, which I believe is the best selling TR SKU right now? and it would have higher base and boost clocks? With a MUCH lower cost of entry on the AM4 socket instead of TR?

Well the TR line still has advantages like PCI-E Lanes and quad channel memory.

[The Benjamins]

1 minute ago, TigerHawk said:

I don't think so. I think they would just not want to cannibalize the threadripper platform by putting too many cores on regular desktop.

 

Unless, that is, they are planning to do away with HEDT altogether and kill threadripper off. Just go Ryzen and EPYC. Maybe reuse the threadripper name for R9s or something.

This is what I always thought for gen 3

AM4: 4(?)-12c

TR4: 12-48c

SP3: 8-64c

 

I don't see how that is overlapping, and TR4 will have more I/O and Ram channels.

[AngryBeaver]

Also looking at the 16core R9 3850x vs the 2950x Threadripper. Look at the TDP requirements. I know the smaller process will reduce the power requirements, but they are also increasing the clocks which will counter-act that.. so we have the R9 using only 135w TDP vs the 180 of the 2950x?  That also doesn't seem realistic.

[GoldenLag]

7 minutes ago, TigerHawk said:

Not to mention the R9 3850X releasing not too long after TR2 would completely cannibalize the 2950X, which I believe is the best selling TR SKU right now? and it would have higher base and boost clocks? With a MUCH lower cost of entry on the AM4 socket instead of TR?

Threadripper really isnt selling well to begin with. (they are HEDT, AKA high margin low volume parts) also you have the IO advantage of the Threadrippers which you cant ignore.

 

we dont know the release date of 7nm and its afaik quite a bit after threadripper 2 launched. i dont see how this is too much off an issue. 

[Mihle]

38 minutes ago, AngryBeaver said:

This looks like something you would find on WccfTech. The clock speeds alone make me pretty sure this is fake. They aren't going to move from 2 generations of pretty hard caps to a 3rd gen where that magically disappears. You also aren't going to see boost speeds keep increasing as the core count goes up. If those boost are valid on the 16c then they would have no reason to not use them on a 8 core. They are trying to pull ahead of intel across the board, so that doesn't make sense.

 

I am willing to put money on the fact this is completely fake.

I bet the core counts is correct, but the price and clock speed might not be. (the reddit leak had same core counts, but less clock speed)

 

BTW
AdoredTV (and I, it seems likely) guess that the 8 core 3600(x) is 2 chiplets that have 4 cores each turned on, but the 3600G is a full 8 core chiplet with the other switched with Navi GPU chip, and therefore lower clocks.
Also, the 3850X is probably 50 year anniversary edition, and most likely very low volume, maybe even limited edition.

[AngryBeaver]

2 minutes ago, Mihle said:

I bet the core counts is correct, but the price and clock speed might not be. (the reddit leak had same core counts, but less clock speed)

 

BTW
AdoredTV (and I, it seems likely) guess that the 8 core 3600(x) is 2 chiplets that have 4 cores each turned on, but the 3600G is a full 8 core chiplet with the other switched with Navi GPU chip, and therefore lower clocks.
Also, the 3850X is probably 50 year anniversary edition, and most likely very low volume, maybe even limited edition.

If this information was true I don't think that is the case on your assessment.

 

The 3600 and 3600x would be the normal chips, where as the 3600g would be an 8 core chip with a built in GPU. These would be the replacements for the 2200g and 2400g would be my guess.

 

The 3850x Would just be a higher binned chip with increased clocks. Again this is assuming all of this crap information was true (which it isn't)

[Mihle]

23 minutes ago, TigerHawk said:

Not to mention the R9 3850X releasing not too long after TR2 would completely cannibalize the 2950X, which I believe is the best selling TR SKU right now? and it would have higher base and boost clocks? With a MUCH lower cost of entry on the AM4 socket instead of TR?

TR have more memory channels and more PCIE lanes. If 3800X specs is sort of true I bet its memory starved in memory heavy tasks that TR is less so/isnt.

[Mihle]

7 minutes ago, AngryBeaver said:

If this information was true I don't think that is the case on your assessment.

 

The 3600 and 3600x would be the normal chips, where as the 3600g would be an 8 core chip with a built in GPU. These would be the replacements for the 2200g and 2400g would be my guess.

 

The 3850x Would just be a higher binned chip with increased clocks. Again this is assuming all of this crap information was true (which it isn't)

I bet they have chiplets and IO die as Epyc Rome does. One 14nm IO die with 2 7nm chiplets, 8 core max each. It would make the leak make much more sense. IO is less bonuses scaling down and its harder to scale them down.

Not saying the leak is true, just saying it isnt necessary false. (I personally bet core count is correct but clock speed isnt)
We just have to wait and see,