AMD Zen 2 (Ryzen 3000) - *** NO COMPETITOR HINTING ***

[Journey]

Whats your guys' opinions on this article?

Some of it could be right, but it is equally guessing rather than having had evidence to show that the rumour is impossible, more than improbable. Take for example the comment about reducing the price of the 6c/12t parts, for the last several months AMD have been seen to have 6c/12t CPU's on the market at the $130 range, and 8c/16t parts at $159-199 and that is without the smaller 7nm dies. I don't particularly find a $100 6c/12t CPU improbable at all, in fact I find it very likely, low end CPU's sell in way more volume than high end, and when you start looking at the TCO to a business, the ability to introduce a desktop with more performance at less cost is something very valuable and gains, and margins are something that inevitably will come with volume once your brand is recognised as the best choice.

AMD may well be aiming for a total shift change in the market place, aiming for maximum impact in both value, performance and technology whilst they have the chance to make inroads, and show the potential they have to offer. Using the current trend of year after year of stagnation with very similar products at the same pricing, it doesn't work if you are trying to gain market share from the current incumbent, or it does but way more slowly than they have time for. Set the goal posts high, so that the leader has to follow and better you or know they will keep losing out.

I am sure CES will provide some interesting information, I don't think everything we have seen to be 100% accurate, but only time will tell.

 

[Hagar]

Well, Seasonic is one of those marketing based PSU manufacturers after all

And it is now 8 years old, no plans to change either.

 

[beany_bot]

Ryzen 3000 is genuinely the most excited I have been about PC stuff since I think the AMD 290x released. or perhaps the first quad cores landed.

Intel and Nvidia have just sucked all the joy out of it for me over the last ~5 years.

 

[DragonQ]

Whats your guys' opinions on this article?
https://www.extremetech.com/computi...mds-ryzen-3000-series-are-too-good-to-be-true

Apparently its just not possible for these rumours to be true based on a couple of different things, not sure how to take it! it could just be an intel fanboy but it does sound like he has some valid points, one of them being that the 3000 series Ryzen 3 is basically the 2000 series Ryzen 7, and its being sold for half the price? And everything beneath 6 cores is now no more? Im not sure, but its probably best to take all this information with just a grain of salt anyway

The hype is fun but I think it's more likely that the current core counts will remain the same in the new series and that a Ryzen 9 12c/24t chip will be released later on to counter Intel's rumoured 10c/20t variant coming late next year. I think the 16c/32t rumours are just really unlikely based on the fact that'd be cutting into Threadripper sales too much, plus it gives them no room to expand with Ryzen 4.

 

[Matkinson]

Tbh, for me, it's the most excited i've been for pc components ever. It's mostly been tiny increases each year, which were nice but...

 

[beany_bot]

The hype is fun but I think it's more likely that the current core counts will remain the same in the new series and that a Ryzen 9 12c/24t chip will be released later on to counter Intel's rumoured 10c/20t variant coming late next year. I think the 16c/32t rumours are just really unlikely based on the fact that'd be cutting into Threadripper sales too much, plus it gives them no room to expand with Ryzen 4.

It doesn't cut into TR at all when they release the new TR with 32 and 64 cores....
We might not be used to it but this is the pace the CPU market used to move at!

 

[beany_bot]

Tbh, for me, it's the most excited i've been for pc components ever. It's mostly been tiny increases each year, which were nice but...

Athlon 64, VooDoo 3DFX GPU's.
Absolute game changers.

 

[muziqaz]

And it is now 8 years old, no plans to change either.

Indeed, I have Seasonic 1250 watt PSU, I think I bought it for 2x 290x, and it was released in 2012. So yeah, still ticking in my secondary system. If it wasn't for Superflower I would be Seasonic only

 

[muziqaz]

It doesn't cut into TR at all when they release the new TR with 32 and 64 cores....
We might not be used to it but this is the pace the CPU market used to move at!

It does cut into million other things, like dual channel, AM4 socket compatibility, becoming Intel with so many different CPUs in the market that you can't figure out which one is faster. AMD needs to keep it simple. 8 cores tops for AM4, which is plenty for gaming/enthusiast market until let's say Zen 3/4 arrives and software/games catch up a bit more, leave 12cores+ for Threadripper to rip a new one to Intel for pro enthusiast market.

 

[jigger]

It does cut into million other things, like dual channel, AM4 socket compatibility, becoming Intel with so many different CPUs in the market that you can't figure out which one is faster. AMD needs to keep it simple. 8 cores tops for AM4, which is plenty for gaming/enthusiast market until let's say Zen 3/4 arrives and software/games catch up a bit more, leave 12cores+ for Threadripper to rip a new one to Intel for pro enthusiast market.

So AMD should adopt Intel's strategy...

 

[muziqaz]

So AMD should adopt Intel's strategy...

which strategy? If they go 16 cores on AM4, they will be using Intels strategy to segment the market with incompatible chips, that one motherboard can run certain CPU while other motherboard with similar chipset cannot run new CPU, or adding more SKUs and being like Intel where you need an encyclopedia of IT to decide which cpu/motherboard to buy

 

[jigger]

which strategy? If they go 16 cores on AM4, they will be using Intels strategy to segment the market with incompatible chips, that one motherboard can run certain CPU while other motherboard with similar chipset cannot run new CPU, or adding more SKUs and being like Intel where you need an encyclopedia of IT to decide which cpu/motherboard to buy

Offer little and charge a lot. The tech industry is really simple. You offer the best you can for the least amount of money. Old tech must always die.

 

[Journey]

and being like Intel where you need an encyclopedia of IT to decide which cpu/motherboard to buy

The build you own PC market is tiny, and is far from complicated just because you have to check a motherboard vendor list to see if a specific CPU would work. I'm guessing you were not around building computers when we had 386/486, with Cyrix/AMD/Intel and having to look at if your new sound card might not work in the ISA slot you have for it as you didn't have enough IRQ's left over once you took into account the others devices. If reading a QVL is list most difficult thing about having a socket that can be upgraded, then bring on the complication.

 

[muziqaz]

Offer little and charge a lot. The tech industry is really simple. You offer the best you can for the least amount of money. Old tech must always die.

Yes intel offer little in quality, compared to AMD, but I mean quantity wise they flooded the market with so many different/same SKUs, it takes awhile to get your head around. AMD needs to keep it simple, and introducing another 4 SKUs (at the top end) into their AM4 line up which might or might not be compatible with older AM4 motherboards is not making things more simple

 

[Deleted member 209350]

Some of it could be right, but it is equally guessing rather than having had evidence to show that the rumour is impossible, more than improbable. Take for example the comment about reducing the price of the 6c/12t parts, for the last several months AMD have been seen to have 6c/12t CPU's on the market at the $130 range, and 8c/16t parts at $159-199 and that is without the smaller 7nm dies. I don't particularly find a $100 6c/12t CPU improbable at all, in fact I find it very likely, low end CPU's sell in way more volume than high end, and when you start looking at the TCO to a business, the ability to introduce a desktop with more performance at less cost is something very valuable and gains, and margins are something that inevitably will come with volume once your brand is recognised as the best choice.

AMD may well be aiming for a total shift change in the market place, aiming for maximum impact in both value, performance and technology whilst they have the chance to make inroads, and show the potential they have to offer. Using the current trend of year after year of stagnation with very similar products at the same pricing, it doesn't work if you are trying to gain market share from the current incumbent, or it does but way more slowly than they have time for. Set the goal posts high, so that the leader has to follow and better you or know they will keep losing out.

I am sure CES will provide some interesting information, I don't think everything we have seen to be 100% accurate, but only time will tell.

Very well put, and I most certainly agree. I for one feel like the rumours are more likely to be true than not, simply to get their Ryzen brand out a bit more and gain a big market share that currently intel dominate. I feel like once they have gained the market share, their pricing will change (probably be closer to intel's than they are now) so if you're going to take advantage of these cpu's, its better to do it now while they are still quite lowkey

 

[muziqaz]

The build you own PC market is tiny, and is far from complicated just because you have to check a motherboard vendor list to see if a specific CPU would work. I'm guessing you were not around building computers when we had 386/486, with Cyrix/AMD/Intel and having to look at if your new sound card might not work in the ISA slot you have for it as you didn't have enough IRQ's left over once you took into account the others devices. If reading a QVL is list most difficult thing about having a socket that can be upgraded, then bring on the complication.

I didn't know what a computer was back in those days, but I was already around
So you are saying todays grey sheep of the mass that is earths population with attention span of 10 seconds is more wise than hardcore geeks back in those days?

 

[jigger]

Yes intel offer little in quality, compared to AMD, but I mean quantity wise they flooded the market with so many different/same SKUs, it takes awhile to get your head around. AMD needs to keep it simple, and introducing another 4 SKUs (at the top end) into their AM4 line up which might or might not be compatible with older AM4 motherboards is not making things more simple

I suppose we have to define simple.

 

[Journey]

I didn't know what a computer was back in those days, but I was already around
So you are saying todays grey sheep of the mass that is earths population with attention span of 10 seconds is more wise than hardcore geeks back in those days?

I'm saying if the barrier of reading a document stops you building a computer then you should just quit trying to build it and move on. AMD could have 500 SKU's, ture enthusiasts and the build-your-own PC crowd/community would work through them and short list what is best in no time at all, and forums would be filled with what/how/why scenarios, just like they are now.

The internet didn't exist back in the days when it was semi-hard to build and configure a PC, and we have to rely on computer clubs, magazines, and our own wits of problem solving to get things working as intended in order to get your fix of Prince of Persia, or Doom, or in my case Ultima.

Easy, simple, stupid proof, dumbed down etc. these are things that it seems people all want these days, and yes I get it, but at the same time it's sad to see people worrying about having to read something, like having the jigsaw already finished when they take it out of the box.

 

[LePhuronn]

Whats your guys' opinions on this article?

I think the guy's trying so hard to curb the hype train he's dismissing the bigger picture of how Zen 2 is designed and the potential it brings. So much so he cites AdoredTV's leaked numbers as the basis for his article, but then shows Overclock3D's numbers instead; AdoredTV filtered out some of the really crazy stuff and his prices were actually a bit higher than the Reddit post that was largely debunked. So right there I kinda question the mindset going in.

A lot of what AdoredTV has been talking about recently comes from looking at how scalable the Zen 2 design is, and by extension how much redundancy there is in the silicon should anything fail. He's speculated that the chiplets are easily salvageable even with a half the cores defunct, and the full EPYC I/O die could be split into quarters to produce a much smaller dual-channel part. ExtremeTech seems to say "OMG have you seen the size of that I/O die? That would never fit in AM4! And AMD need to redesign the memory connections in Threadripper 2 so of course none of this will work!".

ExtremeTech just seem to be thinking "well, this is how AMD did it, so that's how they will continue to do it" whilst seemingly ignoring how we've seen Zen 2 conceived and implemented already. He's swung the pendulum too far the other way and it's annoying.

 

[subbytna]

I'm saying if the barrier of reading a document stops you building a computer then you should just quit trying to build it and move on. AMD could have 500 SKU's, ture enthusiasts and the build-your-own PC crowd/community would work through them and short list what is best in no time at all, and forums would be filled with what/how/why scenarios, just like they are now.

The internet didn't exist back in the days when it was semi-hard to build and configure a PC, and we have to rely on computer clubs, magazines, and our own wits of problem solving to get things working as intended in order to get your fix of Prince of Persia, or Doom, or in my case Ultima.

Easy, simple, stupid proof, dumbed down etc. these are things that it seems people all want these days, and yes I get it, but at the same time it's sad to see people worrying about having to read something, like having the jigsaw already finished when they take it out of the box.

Exactly